Docsity
Log inSign up
Docsity
Prepare for your exams
Prepare for your exams

Study with the several resources on Docsity

Earn points to download
Earn points to download

Earn points by helping other students or get them with a premium plan

Guidelines and tips
Guidelines and tips
Sell on Docsity
Sell on Docsity
Log inSign up

Prepare for your exams

Study with the several resources on Docsity

Find documents

Prepare for your exams with the study notes shared by other students like you on Docsity

Search Store documents

The best documents sold by students who completed their studies

Search through all study resources
Docsity AINEW

Summarize your documents, ask them questions, convert them into quizzes and concept maps

Explore questions

Clear up your doubts by reading the answers to questions asked by your fellow students

Earn points to download

Earn points by helping other students or get them with a premium plan

Share documents
download point icon20 Points

For each uploaded document

Answer questions
download point icon5 Points

For each given answer (max 1 per day)

All the ways to get free points
Get points immediately

Choose a premium plan with all the points you need

Study Opportunities

Choose your next study program

Get in touch with the best universities in the world. Search through thousands of universities and official partners

Community

Ask the community

Ask the community for help and clear up your study doubts

University Rankings

Discover the best universities in your country according to Docsity users

Free resources

Our save-the-student-ebooks!

Download our free guides on studying techniques, anxiety management strategies, and thesis advice from Docsity tutors

From our blog

Exams and Study
Go to the blog

Critique of Incapacitation as a Primary Motivation for Mass Incarceration in the US, Study notes of Criminal Justice

Heriot-Watt UniversityCriminal Justice

The limitations and ethical concerns of using incapacitation as a primary justification for mass incarceration in the US. It highlights the inefficiencies of scale, the importance of selectively incapacitating high-risk individuals, and the need to consider the relative incapacitative effects of different policies. The document also explores the role of deterrence, rehabilitation, and retribution in sentencing, and the challenges of measuring the impact of incarceration on crime rates.

What you will learn

  • What is selective incapacitation and how does it differ from traditional incapacitation policies?
  • What are the limitations of using incapacitation as a primary motivation for mass incarceration in the US?
  • How can policymakers identify high-risk individuals for selective incapacitation?

Typology: Study notes

2021/2022

Uploaded on 09/27/2022

gilian
gilian 🇬🇧

4.6

(11)

228 documents

1 / 18

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
Incapacitation
Shawn D. Bushway*
People who are incarcerated are incapacitated: they do not
commit as many crimes as they would have in the absence of
incarceration. The best modern estimates for the size of the effect
are modest, in the neighborhood of two to five serious crimes
per year of prison time. These effects are larger if incarceration
is used in a more targeted way for higher-rate offenders, but
will inevitably decline as incarceration is used more heavily.
This chapter reviews the research and presents the following
basic recommendations for policy: (1) incapacitation should
not be relied on as a primary motivation for a broad-based
incarceration regime; (2) incapacitation cannot be used to
justify the current levels of incarceration in the United States;
(3) “release valve” policies to reduce the prison population in
the short term should focus on releasing individuals who are
at lowest risk for offending; and (4) policymakers should be
aware of the relative incapacitative effects of different policies,
even if their main motives do not include incapacitation.
INTRODUCTION
There are many different purposes of sentencing in criminal law, including
the utilitarian goals of deterrence, rehabilitation and incapacitation, and
the retributive goal of just deserts.1 Incapacitation reduces crime by literally
preventing someone from committing crime in society through direct control
during the incarceration experience—or, more bluntly, “[a] thug in prison
can’t mug your sister.”2 This directness is the main attraction of incapacitation.
1. Michael Tonry, Purposes and Functions of Sentencing, 34 CRime & Just. 1 (2006). See Jeffrie
G. Murphy, “Retribution,” in the present Volume; Daniel S. Nagin, “Deterrence,” in the present
Volume; Francis T. Cullen, “Correctional Rehabilitation,” in the present Volume.
2. Ben Wattenberg, Circling Crime Hawk, wash. times (June 10, 1999).
* Professor of Public Administration & Policy and Professor of Criminal Justice, University
at Albany, State University of New York. This chapter is based in large part on Shawn D. Bushway,
Incapacitation, in enCyCloPedia of CRiminology and CRiminal JustiCe 2443 (2014).
37
pf3
pf4
pf5
pf8
pf9
pfa
pfd
pfe
pff
pf12

Partial preview of the text

Download Critique of Incapacitation as a Primary Motivation for Mass Incarceration in the US and more Study notes Criminal Justice in PDF only on Docsity!

Incapacitation

Shawn D. Bushway*

People who are incarcerated are incapacitated: they do not

commit as many crimes as they would have in the absence of

incarceration. The best modern estimates for the size of the effect

are modest, in the neighborhood of two to five serious crimes

per year of prison time. These effects are larger if incarceration

is used in a more targeted way for higher-rate offenders, but

will inevitably decline as incarceration is used more heavily.

This chapter reviews the research and presents the following

basic recommendations for policy: (1) incapacitation should

not be relied on as a primary motivation for a broad-based

incarceration regime; (2) incapacitation cannot be used to

justify the current levels of incarceration in the United States;

(3) “release valve” policies to reduce the prison population in

the short term should focus on releasing individuals who are

at lowest risk for offending; and (4) policymakers should be

aware of the relative incapacitative effects of different policies,

even if their main motives do not include incapacitation.

INTRODUCTION

There are many different purposes of sentencing in criminal law, including

the utilitarian goals of deterrence, rehabilitation and incapacitation, and

the retributive goal of just deserts.^1 Incapacitation reduces crime by literally

preventing someone from committing crime in society through direct control

during the incarceration experience—or, more bluntly, “[a] thug in prison

can’t mug your sister.”^2 This directness is the main attraction of incapacitation.

  1. Michael Tonry, Purposes and Functions of Sentencing, 34 CRime & J ust. 1 (2006). See Jeffrie G. Murphy, “Retribution,” in the present Volume; Daniel S. Nagin, “Deterrence,” in the present Volume; Francis T. Cullen, “Correctional Rehabilitation,” in the present Volume.
  2. Ben Wattenberg, Circling Crime Hawk , wash. times (June 10, 1999).
  • Professor of Public Administration & Policy and Professor of Criminal Justice, University at Albany, State University of New York. This chapter is based in large part on Shawn D. Bushway, Incapacitation , in enCyCloPedia of CRiminology and CRiminal JustiCe 2443 (2014).

While it is not impossible to commit a crime in prison, the possibility is limited

by the direct control exerted by the correctional system.^3

The size of these benefits depends directly on the offending behavior of

those individuals who are incarcerated. Incapacitation benefits will be larger for

policies that manage to incarcerate higher-rate criminals. Most criminologists

believe that criminal offending is highly skewed among the offending

population, with a relatively small minority of all offenders responsible for

the majority of all crimes.^4 Selective incapacitation focuses on the idea that

policymakers can prospectively identify the most active offenders prior to their

period of peak activity, and prevent a great deal of crime through “selectively”

incapacitating these high-risk individuals. 5 Legal scholars sometimes object

to the idea of selective incapacitation on the legal or ethical grounds that the

policy is at least implicitly “punishing” the offender for future crimes not yet

committed, rather than the crime for which the person has been convicted.^6

Whatever the ethics or legal support for this idea, selective incapacitation

also implies that there will be declining marginal returns for incarceration,

a least with respect to incapacitation. If society starts by incapacitating the

highest-rate and most frequent offenders, additional incarceration will generate

reduced benefits from incapacitation as society incarcerates lower-rate, less

frequent offenders.^7 A concise way of saying this is that there are inefficiencies

of scale—the impact gets smaller the more incarceration a society uses. The

incapacitative impact of incarceration is also inherently time-limited. A prison

cell can only incapacitate a criminal for the time that he is in prison.

This prison cell might be accomplishing other goals. There are existing

theories of sentencing that present unified goals of rehabilitation or

retribution.^8 However, most current sentencing regimes represent a relative

  1. Serious crimes in prison are included in most measures of reported crime and therefore most modern measures of incapacitation account for serious crimes in prison. However, minor crimes in prison are often handled through administrative mechanisms, and maybe undercounted in official measures of crime. Nevertheless, most researchers assume that the net suppression is positive. For an alternative viewpoint, see Guyora Binder & Ben Notterman, Penal Incapacitation: A Situationist Critique , 54 am. l. Rev. 1 (2017).
  2. Alex R. Piquero, David P. Farrington & Alfred Blumstein, The Criminal Career Paradigm , 30 CRime & Just. 359 (2003).
  3. PeteR w. gReenwood & allan abRahamse, seleCtive inCaPaCitation (1982).
  4. Kathleen Auerhahn, Selective Incapacitation and the Problem of Prediction , 37 CRiminology 703 (1999); Binder & Notterman, supra note 3.
  5. Anne Morrison Piehl et al., The Crime-Control Effect of Incarceration: Does Scale Matter? , 5 CRiminology & Pub. Pol’y 245 (2006).
  6. See, e.g. , Paul H. Robinson & John M. Darley, The Utility of Desert , 91 nw. u. l. Rev. 453 (1997); noRval moRRis, the futuRe of imPRisonment (1974); see also Cullen, supra note 1.

38 Reforming Criminal Justice

to describe their offending prior to the current term of incarceration.^15 These

responses are then used to generate estimates of the annual amount of crime

committed while the person is free, which is then used to generate an estimate of

the benefits of a year of incarceration in terms of the number of crimes prevented.

Initially, research using this approach reported extremely high benefits

from incapacitation, on the order of almost 200 felony crimes per prisoner

year. 16 Almost immediately, scholars identified some serious flaws with this

approach. 17 Data on self-reported crimes is highly skewed, with a few offenders

reporting a great many crimes. As a result, the average number will grossly

overstate the marginal benefit from incarcerating the next person. In addition,

these self-reported crimes were occurring right before the person went to

prison, arguably the peak (and inflated) period of activity during a person’s

“criminal career.” More recent research has reached a consensus around 15 to

20 felony or Uniform Crime Report Part I crimes (murder and non-negligent

homicide, rape, robbery, aggravated assault, burglary, motor vehicle theft,

larceny-theft, and arson) per year in prison, on average. 18 However, there is no

way of knowing whether a given policy change will incapacitate the average

offender, or someone who commits less crime on average.

There is also no way of knowing whether the person placed into prison was

simply replaced by someone else who would have otherwise not committed

crime. The possibility of replacement is most plausible in the case of drug

crimes, where dealers could be replaced by others. The problem of replacement

is similar to the problem of displacement in the case of place-based crime

prevention. Crime might be reduced in a particular area by increasing police

presence, for example, thus making a “crime generating” place less capable of

generating crime.^19 But if criminals simply go to another place, their crimes

may be displaced to the new area and overall crime rates may not be reduced

by this policy. By the same token, crime is not reduced by incapacitation if

a person is incarcerated and is promptly replaced by another individual who

now commits the same crimes the other person would have committed absent

incarceration. It is difficult to identify the extent of replacement empirically.^20

  1. zedlewski, supra note 11.
  2. Id.
  3. Franklin E. Zimring & Gordon Hawkins, The New Mathematics of Imprisonment , 35 CRime & delinQ. 169 (1988).
  4. Alex R. Piquero & Alfred Blumstein, Does Incapacitation Reduce Crime? , 23 J. Quantitative CRiminology 267 (2007).
  5. David Weisburd et al., Does Crime Just Move Around the Corner? A Controlled Study of Spatial Displacement and Diffusion of Crime Control Benefits , 44 CRiminology 549 (2006).
  6. Binder & Notterman, supra note 3.

40 Reforming Criminal Justice

The “top down” approach gets around the problem of replacement by

focusing on the total amount of crime committed in a place rather than on

the crime committed (or not committed) by a particular person. A change in

the number of crimes committed in a certain place that can be directly tied

to changes in the number of people in prison will generate an estimate of the

incapacitation effect that is net of replacement. While this approach has the

twin advantages of controlling for replacement while linking policy directly

to the outcome of interest, it also faces numerous empirical challenges. Places

with higher crime rates also tend to put more people in prison. Failure to

account for this problem will lead to estimates of the incapacitation effect that

are too low. Attempts to causally identify the impact of prison on crime must

therefore break this link by identifying variation in the incarceration rate that is

independent of the crime rate. This independent variation is called exogenous

variation by social scientists. Experiments can cause this exogenous variation,

although it is hard to create experimental variation in incarceration.

“Natural” experiments can also create exogenous variation. For example,

Steve Levitt observed that some states in the U.S. were forced by the courts to

reduce their incarceration levels due to charges of overcrowding.^21 Initially, at

least, states under court sanction could meet this mandate only by releasing

prisoners. Likewise, the Italian government routinely releases up to 35% of its

prison population through periodic “collective pardons.”^22 This pardon process

creates variation in the incarceration rate over time and in different places,

because the pardons release varying amounts of prisoners to each Italian

region. These two studies generated similar estimates of 15 to 20 felony crimes

prevented per year in prison. It is interesting that this number aligns well with

the individual estimates from the “bottom up” approach.

In a similar way, Lofstrom and Raphael used county-level variation caused by

California’s Public Safety Realignment Act (Realignment), a 2011 law designed

to reduce prison overcrowding in response to a court order. 23 I have argued that

Realignment is roughly comparable to the policy change observed by Levitt. 24

  1. Steven D. Levitt, The Effect of Prison Population Size on Crime Rates: Evidence from Prison Overcrowding Litigation , 111 Q.J. eCon. 319 (1996).
  2. Alessandro Barbarino & Giovanni Mastrobuonie, The Incapacitation Effect of Incarceration: Evidence from Several Italian Collective Pardons , 6 am. eCon. J. 1 (2014).
  3. Magnus Lofstrom & Steven Raphael, Incarceration and Crime: Evidence from California’s Public Safety Realignment Reform , 664 annals am. aCad. P ol. & soC. sCi. 196 (2016).
  4. Shawn D. Bushway, Evaluating Realignment: What Do We Learn About the Impact of Incarceration on Crime? , 15 CRiminology & Pub. Pol’y 309 (2016); see also Levitt, supra note 21.

Incapacitation 41

it uses data from individuals, and allows researchers to estimate a distribution

of benefits from incapacitation. The main advantage of the second, “top down”

approach is that it provides a useful counterfactual, comparing the crime-

control effects of two different policies that can be described substantively.

However, it can only hint at variation in incapacitation effects because the

estimates are for places, not people. More broadly, but for the same reasons,

this approach cannot prove that the effects estimated from this approach are

due only to incapacitation.

In each of the aggregate papers, the deterrent threat of incarceration has

also changed as has the potential for specific deterrence and rehabilitation. For

example, an individual contemplating crime in a state with a court-imposed

cap on incarceration might plausibly assume that his chance of incarceration is

less if he was to commit a crime. Individuals who are part of collective pardons

might not be able to complete rehabilitation programs, or might decide that

incarceration is not as bad as they had previously thought, and therefore

commit more crimes than they would have without the collective pardon.

While Vollaard argued that the observed impact in the Netherlands was due

to the incapacitation of chronic drug-addicted offenders who were no longer

“deterrable” or amenable to treatment, he also acknowledged that it is still at

least possible that there was also deterrent value to the statute. Evaluations of

three-strike laws in the U.S. using individual-level data have plausibly identified

deterrence for individuals exposed to the risk of the third strike.^30

Kessler and Levitt attempted to differentiate between incapacitative

effects and deterrent effects by looking at the timing of effects on crime. 31

For example, three-strikes laws impose long prison sentences for multiple or

repeat offenders. These individuals would have been incarcerated even without

the three-strike provisions: the difference is that they may now have a 10- or

15-year sentence instead of a 5-year sentence. As a result, there will be no

additional incapacitative benefit to these particular laws in the short run. An

immediate change in the crime rates after the implementation of the law can

then be plausibly considered to be a deterrent rather than an incapacitative

effect. Long-term effects can be plausibly attributed to both deterrence and

incapacitation. They find both short-term and long-term effects, suggesting

evidence for both incapacitation and deterrence.

  1. Eric Helland & Alexander Tabarrok, Does Three Strikes Deter? A Nonparametric Estimation , 42 J. hum. ResouRCes 309 (2007).
  2. Daniel Kessler & Steven D. Levitt, Using Sentence Enhancements to Distinguish between Deterrence and Incapacitation , 42 J.l. & eCon. 343 (1999).

Incapacitation 43

Miles and Ludwig argued that reliable and valid estimates of incapacitation

are too difficult to obtain, and that time would be better spent generating

estimates of the aggregate effects of prison using natural experiments. 32

Although these aggregate measures would not enable the researcher to isolate

the mechanism by which crime declined, it would give policymakers clear

guidance with which to conduct cost-benefit analyses of prison. In a cost-

benefit analyses, the costs of incarceration are compared with the monetized

benefits of preventing a certain numbers of crimes. The monetary costs of

crime are generated using direct accounting methods, compensatory damages

from civil lawsuits, and methods that attempt to put an estimate on individuals’

willingness to pay to avoid victimization. From the perspective of Miles and

Ludwig, generating a clean estimate of the crime-reducing benefits of a policy

would then provide room to concentrate on generating good estimates of the

costs and benefits of such a policy. Whether the effect of incarceration was due

to incapacitation or deterrence or some other process would be both unknown

and irrelevant.

The logic of this argument is compelling—ultimately, what policymakers

need to know is the treatment effect of incarceration. The fact that incarceration

can reduce crime through multiple mechanisms is interesting, but not

particularly important if the policy choice involves more or less incarceration.

However, once policymakers start thinking about the nature of the incarceration

experience, the mechanisms become important. Mueller-Smith has found

compelling evidence that the spell of incarceration increases the offending

rate of an individual after he is released from jail in Harris County, Texas.^33

Bhuller et al. found evidence of exactly the opposite effect for individuals who

have served time in Norway. 34 It almost goes without saying that Norway has

radically different incarceration practices than the jail system in Harris County,

Texas. Both studies found evidence of incapacitation. Without breaking down

the treatment impact of incarceration into its component parts, policymakers

will have little insight into how they can improve upon standard practice.

Fortunately, researchers have begun to develop a new, third approach

to estimate the separate incapacitative benefit of incarceration. This new

approach relies on individual-level data, like the bottom-up approach, but, as

  1. Thomas J. Miles & Jens Ludwig, The Silence of the Lambdas: Deterring Incapacitation Research , 23 J. Quantitative Res. 287 (2007).
  2. Michael Mueller-Smith, The Criminal and Labor Market Impacts of Incarceration (Aug. 18, 2015) (unpublished manuscript).
  3. manudeeP bhulleR et al., inCaRCeRation , ReCidivism, and emPloyment (Nat’l Bureau of Econ. Research Working Paper No. 22648, 2016).

44 Reforming Criminal Justice

Vollaard.^38 They found estimates of around four crimes a year, which is double

that of Wermink et al., but much lower than the estimates from Vollaard. One

possible explanation is that Vollaard studied an earlier, more restrictive version

of the program that was particularly targeted. Another possibility, supported

by additional analysis by Tollenaar et al., is that the programs under study also

affected behavior after release—so, as suggested earlier, Vollaard’s estimates

might have included more than just incapacitation effects.

One problem with this matching approach is that it controls only for observable

differences. Emily Owens fashioned an identification strategy that should control

for both unobserved and observed differences between those who are incarcerated

and those who are not.^39 It also fits nicely between the two approaches, because

she uses individual-level data and a natural experiment to estimate the causal

effect of incapacitation for a subsample of people affected by the policy. She takes

advantage of a technical change in Maryland sentencing guidelines that had a

substantial effect on the sentence for a subset of sentenced offenders.

The change involved the use of juvenile records in sentencing decisions.

Until 2001, these records were included in the criminal history of all individuals

up to the age of 25; after 2001, the age for which juvenile histories counted was

lowered to 22. Thus some of those aged 23-25 received shorter sentences than

they would have received in the earlier years. Owens estimates that this reduced

the average sentence under the Maryland guidelines system by about 25%

(about 9 to 18 months). She also found that, during this time period when they

were at liberty because of the change in rules, youths sentenced after 2001 were

arrested on average 2.5 times per year. Taking account of the specific offenses for

which they were arrested, and the ratio of recorded arrests to recorded offenses

of the same type, she estimates that they were responsible for 1.5 index crimes

per year. This provides a relatively precise estimate of their recorded criminal

activity during a period when they would have been incarcerated under the

previous rules. Like the Dutch estimates, the estimate that Owens develops of

crimes averted is smaller by an order of magnitude than the Levitt estimate of

15 to 20 crimes previously cited in the literature.

Two other recent studies have followed Owens in using exogenous variation

to compare those in prison to otherwise similar people on probation or parole.

Mueller-Smith and Bhuller et al. use random assignment to judges to identify

  1. N. Tollenaar et al., Effectiveness of a Prolonged Incarceration and Rehabilitation Measure for High-Frequency Offenders , 10 J. exPeRimental CRiminology 29 (2014).
  2. Emily G. Owens, More Time, Less Crime? Estimating the Incapacitative Effect of Sentence Enhancements , 52 J.l. & eCon. 551 (2009).

46 Reforming Criminal Justice

the impact of prison vs. probation.^40 Some judges sentence offenders more

harshly than others, so random assignment to judges creates a situation where

some offenders get exogenously determined prison rather than probation. As

in the Owens study, the offending of the individuals with probation can be

used to estimate the incapacitative benefit from prison during the time that the

individual would have otherwise been in prison.

The incapacitative impacts of incarceration have at times been ignored

by criminologists who focus solely on the differences in recidivism for those

assigned to prison versus those who are assigned probation, essentially starting

the clock at release rather than from sentencing. Clearly, this ignores one of

the potential benefits from incarceration. Mueller-Smith, who studied Harris

County Texas, and Bhuller et al., who studied Norway, find modest benefits

from incapacitation that are more similar to Owens than Levitt.^41

Of course, incapacitation is not the only consequence of incarceration.

Mueller-Smith found that incarceration actually increases offending after

release,^42 while Bhuller et al. found that incarceration decreases offending after

release relative to probation.^43 Obviously, Harris County and Norway are very

different places with very different correctional philosophies. But, the difference

in estimates after incarceration point to the importance of remembering that

incapacitation is only one of the potential consequences of incarceration.

The possibility exists that these effects could be much bigger for targeted

policies focusing on high-rate offenders. Of course, the reality is that only a

few people commit crime at a high rate. In the Netherlands, a country of 16

million, only 4,000 people are even eligible for the habitual-offender label in a

given year, and even fewer actually receive the penalty. Nevertheless, Vollaard

showed that Dutch cities that used the habitual-offender law more liberally

also had smaller crime reduction per prison year,^44 and Tollenaar et al. found

much smaller estimates for the Dutch law that widened the scope of the original

habitual-offender law. 45 The reality of a strong positive skew is impossible to

avoid—the only way to incarcerate more people is to incarcerate offenders who

commit fewer crimes. This leads inevitably to diminishing marginal returns

from increased incarceration. In the U.S., which has seen a four-fold increase

in the incarceration rate over the last 40 years, researcher after researcher has

  1. Mueller-Smith, supra note 34; Bhuller et al., supra note 35.
  2. Mueller-Smith, supra note 34; Bhuller et al., supra note 35.
  3. Mueller-Smith, supra note 34.
  4. Bhuller et al., supra note 35.
  5. Vollaard, supra note 30.
  6. Tollenaar et al., supra note 39.

Incapacitation 47

enough to create benefits that outweigh not only the costs of incarcerating

the accurately identified high-risk offenders, but also the costs generated by

incarcerating the false positives.^52

There may also be structural or institutional limits to the ability of researchers to

prospectively identify high-risk offenders in the current environment.^53 Empirical

models require data on individuals followed over many years to validate risk-

prediction models.^54 This is not a big problem in places like the Netherlands, since

even the prolific offenders in the Netherlands who were subject to the habitual-

offender law had spent very little time in prison prior to receiving the sentence

enhancement. However, in the U.S., these people would have been incarcerated

for substantial periods of time, drastically reducing the amount of time in which

their behavior could have been observed. As a result, it would have taken them

longer to accumulate the same number of offenses, and prolific offenders will be

less obvious in the U.S. than they will be in the Netherlands.

Analysis of selective incapacitation policies is also complicated by the

fact that incarceration and criminal justice actions may affect the offending

of the incarcerated individuals through specific deterrence, stigmatization,

or incapacitation. And these treatments are being assigned in a non-random

way to the convicted population. In this context, in which a regime is already

trying to implement a treatment, Jeffrey Smith and I have made it clear that

it is hard to evaluate the impact of any variable on subsequent offending

without an explicit model of what the criminal justice actors are already trying

to accomplish. 55 Almost all risk-prediction models focus on recidivism after

release—but the person doing the sentencing presumably cares about the

behavior of the individual from the time of sentencing, which would include

incapacitation. But, if incapacitation is ignored or not modeled, we will not get

a true measure of risk.

To the extent we know what the actors are trying to do, we can more easily

interpret the causal impacts of the various actions. Such information is often

not available, and we need to make strong assumptions to make much progress

  1. Richard Berk, Asymmetric Loss Functions for Forecasting in Criminal Justice Settings , 27 J. Q uantitative CRiminology 107 (2011) [hereinafter Berk, Asymmetric Loss Functions ]; Richard Berk, Balancing the Costs of Forecasting Errors in Parole Decisions , 74 alb. l. R ev. 1071 (2011) [hereinafter Berk, Balancing the Costs ].
  2. Stephen D. Gottfredson & Laura J. Moriarty, Statistical Risk Assessment: Old Problems and New Applications , 52 CRime & delinQ. 178 (2006).
  3. Elaine P. Eggleston et al., Methodological Sensitivities to Latent Class Analysis of Long-term Criminal Trajectories , 20 J. Quantitative CRiminology 1 (2004).
  4. Shawn Bushway & Jeffrey Smith, Sentencing Using Statistical Treatment Rules: What We Don’t Know Can Hurt Us , 23 J. Quantitative CRiminology 377 (2007).

Incapacitation 49

on the question of risk assessment and selective incapacitation, particularly for

those offenders who are heavily involved in the criminal justice system—the

highest-risk offenders. While there is substantial literature on risk prediction, 56

very little of this research takes this problem—that the decisions are endogenous

with respect to the risk—into account. In light of that fact, researchers and

policymakers should be aware that the endogeneity of treatments (with respect

to risk) understates the power of risk-prediction models by suppressing the true

unobserved risk of the person through treatment, including incapacitation.^57

Despite the presence of these challenges and ethical concerns, 58 the use of

risk prediction in the U.S. criminal justice system—for sentencing, correctional

placement, probation supervision and parole release—has exploded and

shows no sign of abating.^59 Although not all risk prediction is used for selective

incapacitation, many of the explicit goals of the Risk, Needs, and Responsivity

(RNR) model that now dominates the risk-prediction field are at least

consistent with selective incapacitation.^60 For example, a central tenet of the

RNR approach is the identification of low-risk offenders who will not offend

even without treatment or supervision. This is selective incapacitation at the

other end of the distribution—why incarcerate or otherwise restrict people

who are at low risk for offending. 61 The logic of selective incapacitation is the

same whether the focus is identifying high-rate or low-rate offenders. It is also

far more attractive, and potentially easier, to identify the larger group of low-

risk people than it is to identify the small group of high-risk people. In an

era when policymakers are seeking to reduce incarceration, using risk tools to

identify the lowest-risk individuals to release so as to minimize potential crime

increases makes good sense. 62

III. INCAPACITATION OUTSIDE THE CONTEXT OF INCARCERATION

The logic of incapacitation need not be limited to the policy of incarceration.

For example, research in economics has considered the incapacitative impact

of school, which keeps youths out of the community and potentially reduces

  1. See Monahan, supra note 52.
  2. Matthew Kleiman et al., Using Risk Assessment to Inform Sentencing Decisions for Nonviolent Offenders in Virginia , 53 CRime & delinQ. 106 (2007).
  3. Kelly Hannah-Moffat, Actuarial Sentencing: An “Unsettled” Proposition , 30 Just. Q. 270 (2013).
  4. John Monahan, A Jurisprudence of Risk Assessment: Forecasting Harm Among Prisoners, Predators, and Patients , 92 va. l. Rev. 391 (2006).
  5. Andrews et al., supra note 50.
  6. Auerhahn, supra note 6.
  7. Kleiman et al., supra note 58.

50 Reforming Criminal Justice

1. Incapacitation should not be relied on as a primary motivation for a

broad-based incarceration regime. Although incapacitation is real,

and there will be some modest decrease in crime associated with most

incarceration, incapacitation as an idea is not sufficiently robust to

motivate and sustain a systematic sentencing regime. Serious legitimate

questions exist about the ethics of selective incapacitation as a primary

motive for sentencing.^65

2. Incapacitation cannot be used to justify the current levels of

incarceration in the United States. The offender population has a distinct

distribution with respect to offending rates. This distribution is skewed,

with a few high-rate offenders accounting for the majority of the offenses. 66

Most incarceration policies, even one that assigns every crime the same

probability of a prison sentence, will selectively incarcerate the higher-rate

offenders. 67 However, the nature of the offender distribution means that

increased incarceration will have diminished returns to scale in terms of

incapacitative benefit. The evidence is clear-cut that current high levels

of incarceration have captured a wide swath of the offender population,

including those that offend at a low rate. In real terms, this means that the

average benefit to a prison cell in terms of crimes prevented has dropped

at least in half since the 1970s, and probably more. Simply put, the benefits

from incapacitation cannot support the current levels of incarceration in

the U.S., even if a person was to believe that incapacitation was the proper

(and only) goal of sentencing.

3. “Release valve” policies to reduce the prison population in the short

term should focus on releasing individuals who are at lowest risk for

offending. Not all prison-reduction policies will have the same costs in

terms of increased crime due to reduced incapacitation. Higher-risk people

have some observable characteristics that can be used to reliably identify

higher rates of offending. Most notably, age and number of prior offenses

are good predictors of future crime.^68 Type of crime, despite heightened

concerns about violent offenders, is not a good predictor of future crime.^69

Although incapacitation should not drive incarceration policy more

  1. Binder & Notterman, supra note 3.
  2. maRvin e. wolfgang et al., delinQuenCy in a biRth CohoRt (1972); Piquero, Farrington & Blumstein, supra note 4.
  3. Piquero & Blumstein, supra note 18.
  4. Paul Gendreau et al., A Meta-Analysis of the Predictors of Adult Offender Recidivism: What Works! , 34 CRiminology 575 (1996).
  5. Id. ; PatRiCk a. langan & david J. levin, buReau of JustiCe statistiCs, u.s. deP’t of JustiCe, R eCidivism of PRisoneRs Released in 1994 (2002).

52 Reforming Criminal Justice

broadly, evidence about the size of the incapacitative benefit should play

a role in one-time “release valve” decisions to release prisoners. Such

considerations would result in more releases of older offenders, even those

who are serving long sentences for serious crimes. More broadly, crime

control (or incapacitation) should not be used as an explanation or defense

for long, determinate prison sentences, such as life without parole.^70

4. Policymakers should be aware of the relative incapacitative effects

of different policies, even if their main motives do not include

incapacitation. Retribution requires longer periods of incarceration

for violent offenses. This policy might have demonstrably lower crime-

reduction benefits than a policy that focuses shorter prison sentences

on young, high-rate property offenders. Retribution scholars might

legitimately not care about this potential differential impact. Nonetheless,

these effects are real, and should inform the policy decisions about the

use of incarceration in real-life situations. Risk-assessment tools can play

a role in helping identify the relative rates of offending for individuals

involved in the criminal justice system. Care should be taken to accurately

assess the relative costs of different kinds of errors^71 and the impact of

current practice on observed risk.^72 Care should also be taken to make use

of tools that can mitigate the potential for these tools to have a racially

disparate impact.^73

5. In certain specific cases, crime can be reduced in places through the use

of limited short-term spells of incarceration to incapacitate very high-

rate property offenders. The Dutch experience with the limited use of

short spells of incarceration aimed at very high-rate property offenders has

demonstrated that targeted incarceration policies can be used selectively

to reduce crime. 74 These policies almost inevitably rely on deterrence and

even rehabilitation of those same offenders to achieve the full benefit of

incarceration. Moreover, the Dutch experience has also highlighted the

very real (and costly) potential trap of these policies—initial success

almost inevitably leads to increased use—and rapidly declining benefit.

This kind of targeted use of relatively short periods of incarceration for

high-rate property offenders should not be confused with the types of

  1. See generally Erik Luna, “Mandatory Minimums,” in the present Volume.
  2. Berk, Asymmetric Loss Functions , supra note 53; Berk, Balancing the Costs , supra note 53.
  3. Bushway & Smith, supra note 56; Eggleston et al., supra note 55.
  4. Devin G. Pope & Justin R. Sydnor, Implementing Anti-Discrimination Policies in Statistical Profiling Models , 3 am. eCon. J. 206 (2011).
  5. Vollaard, supra note 30.

Incapacitation 53