~ Penn CareyLaW"
UNIVERSITY

of PENNSYLVANIA

Public Law and Legal Theory Research Paper Series
Research Paper No. 24-04

“Mass Incarceration”
Myths and Facts: Aiming Reform
at the Real Problems
Paul H. Robinson

UNIVERSITY OF PENNSYLVANIA CAREY LAW SCHOOL

Jeffrey Seaman

UNIVERSITY OF PENNSYLVANIA

This paper can be downloaded without charge from the Social Science Research Network
Electronic Paper collection: https://ssrn.com/abstract=4698071.

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

“Mass Incarceration” Myths and Facts:
Aiming Reform at the Real Problems
Paul H Robinson1 and Jeffrey Seaman2
Abstract
Few claims have won such widespread acceptance in legal academia as the
“mass incarceration” narrative: the idea that the rise in America’s prison
population over the last half century was fueled largely by the needless and unjust
imprisonment of millions of criminal offenders due to punitive changes in
sentencing. To many academics and activists, the question is not how accurate the
mass incarceration narrative is, but how mass incarceration can be ended. This
Article argues the “mass incarceration” narrative is based on a series of myths and,
as a result, many proposed reforms are based on a misunderstanding of America’s
past and present carceral practices. A more accurate understanding is needed to
produce effective reform.
The central myth of the mass incarceration narrative is that exceptional and
unjustified punitiveness largely explains America’s significant increase in prison
population since the 1960s. This explanation overlooks the numerous nonsentencing factors that increased incarceration: a near doubling in U.S. population,
higher crime rates, increased justice system effectiveness, deinstitutionalization of
the mentally ill, new and tightened criminalizations, worsening criminal offender
histories, and more. While this Article makes no attempt at statistical precision,
these non-sentencing factors can easily explain most of America’s elevated
incarceration compared to the 1960s—a fact in direct conflict with the mass
incarceration narrative. Additionally, while some punishments have increased in
severity since the 1960s, most of these increases are likely to be seen as moving
sentences closer to what the community – and many incarceration reformers –
would believe is appropriate and just, as in cases of sexual assault, domestic
violence, stalking, human trafficking, firearm offenses, and child pornography,
among others.
Comparing America’s prison population to foreign countries, as the mass
incarceration narrative often does, similarly overlooks the contributions of many of
these non-sentencing factors and incorrectly assumes that a higher American per
capita incarceration rate always reflects a problem with American, instead of
foreign, practice. While America can certainly learn from foreign countries, the
reality is that many foreign sentencing practices have sparked chronic and
widespread dissatisfaction abroad. It may be that the dispute over incarceration
practices is more a dispute between the elites and the community than a dispute
between the U.S. and other democracies’ populations.
While all decarceration reformers should welcome a clearer picture of
America’s incarceration practices, it is hard not to conclude that many mass
1

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

incarceration myths were created deliberately by those who oppose not only
incarceration but punishment generally. For these activists, the mass incarceration
narrative is primarily a means toward eliminating punishment, a goal that is
difficult to pursue directly because it is so contrary to the views of the general
population and even a majority of academia.
This Article is not pro-incarceration. It subjects the mass incarceration
narrative to much needed scrutiny precisely because reforming incarceration
practices is necessary. The criminal justice system should strive to deliver just
punishment in the most societally beneficial way, which we believe means
increasing the use of non-incarcerative sanctions. The myths of the mass
incarceration narrative frequently lead activists to overlook non-incarcerative
reforms that deliver just punishment—a tragic failure because such reforms would
have much stronger popular support than the anti-punishment or unsophisticated
anti-prison reforms now pushed by the mass incarceration narrative.
Part I of the Article describes the mass incarceration myths that have
become so broadly accepted. Part II reviews the facts of American incarceration
practice, which contradict many, if not most, aspects of the narrative. Part III offers
our reform proposals, which we believe more accurately address the problems in
current incarceration practice. Central to those proposals are the use of creative
non-incarcerative sanctions that still deliver punishment proportional to a nuanced
assessment of each offender’s moral blameworthiness.
________________
Table of Contents
I. Mass Incarceration Myths
A. The Myth of Prison Composition
B. The Myth of Continual Increase
C. The Myth of Foreign Comparisons
D. The Central Myth: Exceptional and Unjustified U.S. Punitiveness
II. Mass Incarceration Facts
A. American Incarceration: Trends and Composition
1. The Problem of Data
2. America’s Rising and Falling Prison Population
3. Prison Composition
B. Non-sentencing Factors in the Rise of the U.S. Prison Population
1. Population Increase
2. Increased Crime
3. Deinstitutionalization of the Mentally Ill
4. Increased Criminal Justice System Effectiveness
5. Changing Criminal Histories
6. New Criminalizations and Increased Enforcement
Drug Offenses
Child Pornography and Sexual Abuse
2

Electronic copy available at: https://ssrn.com/abstract=4698071

5
5
6
6
7
9
9
9
11
12
14
15
15
16
17
19
20
21
22

Robinson & Seaman

1 15 24

Human Trafficking
New White-Collar Crimes
Sexual Assault
Domestic Violence
Stalking
Weapons Offenses

23
23
24
24
25
25

7. Other Non-Sentencing Factors
25
8. The Total Effect of Non-Sentencing Factors on Incarceration
26
C. Sentencing Factors in the Rise of the U.S. Prison Population
27
1. Changes in the Chance of Receiving a Prison Sentence
27
2. Changes in Time Served in Prison for an Offense
28
4. How Unjust Were These Sentencing Changes?
31
5. Who Bears Responsibility for Punitive Excesses in Punishment Post-1960? 33
D. American vs. Foreign Countries’ Prison Populations: A Problem on Which Side of the
Ocean?
34
1. The Claims of American vs. Foreign Countries’ Incarceration Practices
34
2. Important Factors: Crime, Criminal Justice Effectiveness, Institutionalization,
and Criminal Histories
36
Crime Rate
Criminal Justice System Effectiveness
Civil Commitment of the Mentally Ill; Criminal Histories

3. How Exceptional Are American Sentencing Policies?
Chance of Convicted Offenders Receiving a Prison Sentence
Time Served in America vs. Other Democracies

36
37
38

38
39
39

4. Are the Lenient Sentencing Practices of Foreign Countries Desirable?
40
III. Aiming Reform at Real Incarceration Problems Rather Than Myths
41
A. Decriminalizing or Reducing Penalties as Needed to Track Shifting Community Views
43
B. Abolishing Unnuanced Punishment Policies in Favor of Rules That Capture People’s
Nuanced Judgments of Justice
43
C. Employing More Non-Incarcerative Punishments Consonant with Desert
44
D. Reforming Prison to Make It Less Destructive to Prisoners
46
IV. Conclusion
47
________________
Few claims have won such widespread acceptance in legal academia as the “mass
incarceration” narrative: the idea that the rise in America’s prison population over the last half
century was fueled largely by the needless and unjust imprisonment of millions of criminal
offenders due to punitive changes in sentencing.3 To many researchers and activists, the
question is not how accurate the mass incarceration narrative is, but how mass incarceration
can be ended. This Article argues the “mass incarceration” narrative is based on a series of
myths and, as a result, many proposed reforms are based on a misunderstanding of America’s
past and present carceral practices. A more accurate understanding is needed to produce
effective reform.
3

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

The central myth of the mass incarceration narrative is that exceptional and unjustified
punitiveness largely explains America’s significant increase in prison population since the 1960s.
This explanation overlooks the numerous non-sentencing factors that increased incarceration: a
near doubling in the U.S. population, higher crime, increased justice system effectiveness,
deinstitutionalization of the mentally ill, new and tightened criminalizations, worsening criminal
offender histories, and more. While this Article makes no attempt at statistical precision, these
non-sentencing factors can easily explain most of America’s elevated incarceration as compared
to the 1960s—a fact ignored by the mass incarceration narrative. Additionally, while some
punishments have increased in severity since the 1960s, most of these increases are likely to be
seen as moving sentences closer to what the community—and many incarceration reformers—
would believe to be appropriate and just, as in cases of sexual assault, domestic violence,
stalking, human trafficking, firearm offenses, and child pornography, among others.
Comparing America’s prison population to foreign countries, as the mass incarceration
narrative often does, also overlooks the contributions of many non-sentencing factors and
incorrectly assumes that a higher American per capita incarceration rate always reflects a
problem with American, instead of foreign, practice. While America can certainly learn from
foreign countries, the reality is that many foreign sentencing practices have sparked chronic and
widespread dissatisfaction abroad. It may be that the dispute over incarceration practices is
more a dispute between the elites and the community than a dispute between the U.S. and
other democracies’ populations.
While all decarceration reformers should welcome a clearer picture of America’s
incarceration practices, it is hard not to conclude that many mass incarceration myths were
created deliberately by those who oppose not only incarceration but punishment generally. For
these activists, the mass incarceration narrative is primarily a means toward eliminating
punishment, a goal that is difficult to pursue directly because it is so contrary to the views of the
vast bulk of the population,4 and even a majority of modern academia.5 Those opposed to
punishment ignore the reality that delivering a just punishment as society sees it is essential not
only because it can deter and protect, but also because the community believes doing justice is
a moral imperative. The elitist impulse to impose a criminal justice system that ignores this
societal imperative is not only deeply anti-democratic, but also bound to fail as such a “justiceless” system will lose moral credibility with the community and thereby lose its ability to shape
norms and gain compliance.6
This Article is not pro-incarceration. It subjects the mass incarceration narrative to much
needed scrutiny precisely because reforming incarceration practices is necessary. The criminal
justice system should strive to deliver just punishment in the most societally beneficial way,
which we believe means increasing use of non-incarcerative punishment. The myths of the mass
incarceration narrative frequently lead activists to overlook the development of a wider range of
non-incarcerative reforms that deliver just punishment – a tragic failure because such reforms
would have much stronger popular support than the anti-punishment or unsophisticated antiprison reforms now pushed by the mass incarceration narrative.
Part I of the Article describes the mass incarceration myths that have become so broadly
accepted. Part II reviews the facts of American incarceration practice, which contradict many, if
not most, aspects of the standard narrative. Part III offers our reform proposals, which we
believe more accurately address the problems in current American incarceration practice.
4

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

Central to our proposals is the use of creative non-incarcerative sanctions that still deliver
punishment proportional to a nuanced assessment of each offender’s moral blameworthiness.

I. Mass Incarceration Myths
The popular narrative of “mass incarceration” claims America’s prison population
exploded over the last half century because of the needless and unjust imprisonment of millions
of criminal offenders due to punitive changes in sentencing. Even the term sounds sinister, with
“mass incarceration” suggesting a carceral state herding millions of undifferentiated, often poor
and minority, offenders behind barbed wire and prison walls. To many academics and activists,
remedying the sin of mass incarceration is one of the most pressing social justice issues of our
time. Indeed, some have gone so far as to compare American “mass incarceration” to slavery or
Stalin’s Gulag Archipelago.7
Given the weightiness of the claims, it is worth investigating how accurate the mass
incarceration narrative is. We have no desire to promote the status quo—unjust imprisonment
is certainly a serious problem whenever it occurs. But our analysis suggests the popular mass
incarceration narrative is not a good foundation on which to seek prison reform because the
narrative is riddled with more myth than fact.
The myths of prison composition, continual increase, and foreign comparisons serve as
support for the central myth of exceptional and unjustified punitiveness, which unites the
various claims of the mass incarceration narrative into a coherent, yet incorrect, causal story.

A. The Myth of Prison Composition
Ask the average American what the largest contributor to mass incarceration is, and they
will likely respond “the War on Drugs.” One of the many myths of the mass incarceration
narrative is that America’s prison population has swelled largely through the imprisonment of
drug possessors, as well as other less serious first-time and non-violent offenders. As one
scholar puts it, “Though explanations [of mass incarceration] differ, almost all analysts agree
that a major cause has been the "War on Drugs."8 That perspective is often promoted by
researchers. As a piece published by Brookings opines: “Drug crimes have been the
predominant reason for new admissions into state and federal prisons in recent decades.”9
Similarly, the claim that much of mass incarceration could be solved by simply changing
the way non-violent offenders are punished is popular. For example, a report by the Brennan
Center suggested the prison population could be safely reduced by 25% simply by releasing
those prisoners who are “almost all non-violent, lower-level offenders.”10 That same report
suggested another 14% could be released as their incarceration had “little public safety
rationale.”11
In discussions of mass incarceration, it is common to find reference to the supposedly
draconian punishments meted out to low-level offenders. For example, one writer diagnosing
the causes of mass incarceration in the U.S. writes: “The U.S. even utilizes life sentences for firsttime, nonviolent offenders, and it is the only nation in the world that sentences children to
LWOP [Life Without Parole].”12 Perhaps because it is politically more palatable to advocate for
reducing the punishment of first-time offenders, non-violent offenders, and drug possessors, it
5

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

is these categories of prisoners that get the most rhetorical space in the mass incarceration
literature. But as Part II shows, this categorization of the prison population is deeply misleading.
Even informed proponents of reducing prison populations admit that blaming America’s high
prison population on the treatment of non-violent offenders is incorrect. For example, the
Prison Policy Initiative labels as a “myth” the idea that “Releasing ‘nonviolent drug offenders’
would end mass incarceration.”13

B. The Myth of Continual Increase
Another myth the mass incarceration narrative promotes is a distorted view of current
incarceration trends. Many in the public would be surprised to learn that American prison
populations have been falling for 15 years. Mass incarceration activists sometimes overstate the
size of prison increases or obscure the direction of trendlines to increase the urgency of their
cause. Consider the ACLU’s description of mass incarceration from 2022: "Mass incarceration in
the United States has ballooned over the last 30 years. Although the United States has only 5%
of the world’s population, it has 25% of the world’s prison population. There are five times as
many people incarcerated today than there were in 1970."14 One would be forgiven for
assuming this “ballooning” was a constant or increasing problem over the last 30 years; in
reality, the trend for 13 of those 30 years was a decline in prison population.
Ignoring the current trend of decreasing incarceration is not purely a matter of
drumming up political urgency. Acknowledging the steady decline in prison population might
raise questions about the cause of past incarceration increases—if prison populations are falling
without drastic changes in sentencing policy, perhaps the previous rise in imprisonment had less
to do with punitive sentencing than the mass incarceration narrative suggests. Whatever the
causes of “mass incarceration,” it has clearly peaked and is receding—a fact the mass
incarceration narrative would prefer to ignore.

C. The Myth of Foreign Comparisons
The mass incarceration narrative draws strength from contrasting America’s high prison
population with that of foreign countries, especially other democracies. The (incorrect) claim
that America has the largest prison population in the world is repeated loudly, but there is little
effort to examine non-sentencing factors when making international comparisons. Instead, the
narrative assumes that any difference in per capita incarceration rates is proof of exceptional
and unjustified American punitiveness. These comparisons often mislead by overlooking other
relevant factors such as crime rates, criminal justice system effectiveness, the criminal histories
of convicted offenders, and the treatment of mentally ill offenders, among many other possible
factors that contribute to incarceration rate differences. Often detailed crime and sentencing
data simply does not exist to make a fair comparison between countries,15 but this rarely stops
the mass incarceration narrative from leaping to conclusions.
Perhaps even more importantly, such comparisons almost never ask the inconvenient
question of whether higher American per capita incarceration rates reflect a problem in other
countries, which may have unjustly low prison populations compared to what their
circumstances and citizens would call for. Instead, the standard narrative uses foreign
6

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

comparisons to paint a portrait of an exceptionally punitive U.S. surrounded by more
enlightened countries. For example, consider the Brennan Center’s perspective:
The U.S. “punisher’s brain” is absent from European justice models, which emphasize
rehabilitation and resocialization. Germany’s Prison Act, for example, specifically states
that the very purpose of incarceration is to help prisoners lead lives of “social
responsibility free of crime” upon release.16
Such comparisons ignore whether the American public would find alternative
punishment models just. Indeed, there is no consideration of whether foreign countries’
populations find such models just. The mass incarceration narrative’s use of foreign
comparisons reveals a clear strain of anti-democratic elitism: a belief that certain enlightened
experts know what is needed and should dictate punishment policy while the common rabble
(those who must live with the consequences) are kept safely away from having significant
influence over the justice system.
While foreign countries can offer America’s justice system ideas to consider, such
countries’ punishment practices should not be assumed to be ideal. In many cases, it is a myth
to believe copying foreign countries’ punishment practices and per capita incarceration rates
would be possible or desirable.

D. The Central Myth: Exceptional and Unjustified U.S. Punitiveness
The central myth underpinning the mass incarceration narrative is that the predominant
factor driving the increase in American incarceration from the 1960s through the 2000s was
exceptional and unjustified American punitiveness, manifested through changes in sentencing
policies. As the Brennan Center explains:
America can’t shrink its reliance on mass incarceration until we confront our approach
to punishment. [The problem is] our deep-rooted impulse to punish people in ways that
are far beyond what could be considered proportionate. [We must] rein in the punitive
excess of the criminal legal system.17
A report by the National Academies on the causes and consequences of mass
incarceration declares: “America has earned the dubious distinction of being the world’s leader
in incarceration,” and the report decries “America’s unusual preference for harsh punitive
measures that eventually resulted in . . . staggering numbers [of prisoners]. . . . This form of
‘American exceptionalism’ is exceptional in troubling ways, with uniquely harsh and damaging
consequences . . . .”18
This central claim of the mass incarceration narrative is not just that a few misguided
policymakers made a mistake in setting some punishments too high, but rather that the
problem stems from an exceptionally punitive American public whose intuitions about what
makes a just punishment simply cannot be trusted. As critics of American imprisonment
describe in their article Why Is America so Punitive?:

7

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

The harsh laws we have today weren’t created by crunching numbers to produce a cost
benefit analysis… They were created because people believed that harsh punishment
was the right approach, or because the rights of offenders didn’t matter, or didn’t matter
as much as the rights of victims.19
The narrative places primary blame for the rise in America’s prison population squarely
on the shoulders of the American public and their misguided intuitions of justice. Once again, an
anti-democratic elitism emerges. As the same source continues:
Criminal justice policy in the United States is a populist issue … whereas in Europe it is
largely left up to a class of professional experts. Europeans would rarely, if ever, let
voters decide how to punish offenders as California did in 1994 when the voters passed
the infamous three strikes law.20
The claim of exceptional and unjustified American punitiveness leads to the conclusion
that it was largely punitive sentencing changes that created mass incarceration and so the
solution lies largely in reversing those changes. As the National Academies’ report on the causes
of mass incarceration confidently declares:
The increase in U.S. incarceration rates over the past 40 years is preponderantly the
result of increases both in the likelihood of imprisonment and in lengths of prison
sentences—with the latter having been the primary cause since 1990. These increases,
in turn, are a product of the proliferation in nearly every state and in the federal system
of laws and guidelines providing for lengthy prison sentences for drug and violent crimes
and repeat offenses, and the enactment in more than half the states and in the federal
system of three strikes and truth-in-sentencing laws.21
Even when the narrative spreads blame to non-legislative actors, it still points back to punitive
laws. For example, consider one diagnosis of the role of prosecutors in mass incarceration:
Prosecutors do not exist in a vacuum. Their importance in driving mass incarceration is a
function of state legislatures passing 'harsh laws' intended to send more people to
prison and keep them there longer. Tough sentencing laws, including mandatory
minimums, coupled with more restrictive back-end release policies provide prosecutors
with the leverage to secure favorable plea bargains and long prison sentences. Remove
the tough sentencing laws, and you will see changes in charging practices.22
But did sentencing changes driven by an exceptionally punitive public really cause the
rise in America’s prison population? And do current criminal punishments showcase unjust
punitiveness? As section II demonstrates, they did not and do not. The myth of exceptional and
unjustified American punitiveness overlooks the role of non-sentencing factors in the rise of
incarceration, overstates punishment increases, and falsely assumes that any punishment
increases that did occur were inappropriate. In addition to being simplistic and false, this myth’s
implications are disturbing and damaging to the prospects of useful reform. If America’s high
8

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

prison population is the result of exceptional and unjustified punitiveness on the part of the
public, it follows that the necessary reforms to reduce incarceration will likely offend the
public’s punitive sense of justice, thus generating backlash. In other words, if the reforms are
producing a backlash, that is proof positive that reformers are doing the right thing. This
perspective is not only deeply anti-democratic but guarantees failure in the long run. As Part III
makes clear, the aim of incarceration reformers should be to bring criminalization and
punishment practices into line with public understandings of justice, while substituting nonincarcerative punishments where possible. Reforms driven by this goal are far more likely to
gain widespread adoption and successfully reduce per capita incarceration rates than reforms
based on the myth that the public’s sense of justice is a problem to be solved instead of a
mandate to be served.

II. Mass Incarceration Facts
This Part questions the accuracy of the mass incarceration narrative by examining trends
in American incarceration and the factors that contributed to America’s increased prison
population since the 1960s. Each of the mass incarceration myths identified in the previous Part
are countered here with facts. Our aim is to provide a clearer picture of American incarceration
and to highlight how much the mass incarceration narrative fails to consider in its confident
pronouncements. At the same time, we have no wish to be contrarian for its own sake—there
are elements of the narrative that are based on fact, and we will acknowledge those
throughout.

A. American Incarceration: Trends and Composition
Before turning to survey general trends in American incarceration and current statistics
on prison composition, we think it important to say a few words about the availability of data.
1. The Problem of Data
The problem of unavailable data bedevils attempts to carefully analyze the causes
behind America’s per capita incarceration rate rise. Given how much is written on mass
incarceration, and the criminal justice system more broadly, one might assume that detailed
historical data on America’s use of prison must be widely available to support various sweeping
assertions. In fact, it often is not. The U.S. government has persistently suppressed, omitted, or
neglected to collect and release vital criminal justice system statistics. These problems of data
collection and publication go far back. It was only in 1979 that congress even established the
Bureau of Justice Statistics (BJS) whose stated mission is: “to collect, analyze, publish, and
disseminate information on crime, criminal offenders, victims of crime, and the operation of
justice systems at all levels of government.”23 Since 1979, the picture has remained spotty, with
the BJS routinely failing to regularly gather or disclose essential statistics.
State governments also commonly fail to gather or disclose criminal justice data relating
to prison use, such as time served data for specific offenses. Interestingly, it appears that large
deep blue states (with all branches of government under Democratic control) are more
9

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

restrictive in making specific time served data publicly available than large deep red states (with
all branches of government under Republican control).24 Given the mass incarceration narrative
is more popular among Democrats, it may be that some state governments are reluctant to
provide data that might disconfirm elements of the mass incarceration narrative.
The lack of data means researchers must attempt to build general pictures from
information resembling Swiss cheese. Sometimes the missing statistics are extraordinarily basic.
For example, something as simple as the percentage of convicted offenders receiving a prison
sentence is difficult to determine pre-1980s.25 The same is true post-2006 when the BJS
inexplicably chose to stop reporting national conviction numbers.26 Important factors relating to
imprisonment (such as criminal offenders’ prior criminal histories) are often entirely absent
from historical data.
The collection and release of data should be a non-partisan issue, and many groups have
noted missing criminal justice system statistics as a problem. As several organizations writing
publicly note about the Bureau of Justice Statistics, “In recent years, numerous regularly
released publications—as well as some special projects—have fallen substantially behind
schedule or not been published at all.”27 Examples of data collection programs the BJS has
abandoned or delayed indefinitely include: Survey of Inmates in Local Jails (last data available in
2002), the National Judicial Reporting Program (last data available in 2006), and the Justice
Assistance Data Survey (last data available in 2010) among many others. Other data collection
programs have been one-off despite their seemingly vital nature to policymakers.28 Consider the
Recidivism Survey of Felons on Probation which was conducted once in 1989 and never again.29
Insufficient funding may hamper some data collection, but the BJS also deliberately
chooses not to release data that it has collected.30 Chronic reporting omissions suggest that
political considerations may be at play in publication decisions. As a result, there have been calls
to reinstate the Director of the BJS as a senate-confirmed post, a requirement removed during
the Obama administration in 2012.31
The spotty data on America’s criminal justice system means there is sometimes too little
published data to prove a claim beyond reasonable doubt or construct a precise model. Indeed,
the lack of data is one reason this article does not attempt to lay forth a precise and
comprehensive explanation of America’s prison population growth. Comprehensive statistical
analysis is sorely needed, but almost impossible to perform with current publicly available data.
The fact that proponents of the mass incarceration narrative often make little effort to collect
more data is perhaps telling of what that data might show. But while data is limited, it is still
enough to expose the myths and false certainties of that narrative. What is certain is that
America’s increased per capita incarceration rate compared to the 1960s is a multicausal
phenomenon, and there are numerous factors utterly ignored by the mass incarceration
narrative that deserve greater attention and study. While we do offer rough suggestions of how
much a factor may have contributed to prison population increases, and we can certainly
identify factors that were more important than others, we attempt no scientific precision in our
estimations. If there is one thing every scholar of the justice system should agree on, it is this:
we need more data.

10

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

2. America’s Rising and Falling Prison Population
This article examines the population of sentenced offenders in state and federal prison,
which is also the primary focus of the mass incarceration narrative. This excludes America’s local
jail population, but that exclusion does not greatly affect our critique of the mass incarceration
narrative. Comparing prison vs. jail per capita incarceration rates reveal they largely move
together, as jail is a feeder for and reflector of prison populations. One reason we choose to
focus on state and federal prisons is that historical data is easier to come by for such offenders
compared to jail populations. Additionally, examining jail populations would raise the question
of pretrial confinement—which is not driven by punitiveness in sentencing but rather by public
safety concerns. To the extent mass incarceration activists critique unjust pretrial release
conditions, we present no criticism of those critiques here. In fact, we have written elsewhere in
support of more non-incarcerative measures to expand pretrial release without greatly
increasing risk of flight or public safety threats.32
Table 1 shows in mostly 5-year increments from 1925 to 2021 the U.S. total prison
population (state and federal), prison population per capita, and, when available, total new
imprisonments and total new imprisonments per capita (though new imprisonment data is
dependent on definitions and may not be comparable between certain years). As is clear from
the table, the U.S. prison population began to rise sharply in 1973 (when there were 204,000
prisoners or 95 per 100,000 residents) and reached its highest absolute point in 2009 (when
there were over 1.6 million prisoners or 504 per 100,000 residents).33 This amounts to a 780%
increase in absolute prison population and a 530% increase in per capita terms over the 19732009 period. This 36-year-long upward trend then reversed, and prison populations have
declined steadily since. In 2021, the prison population was down to 1.2 million prisoners, or 350
per 100,000 U.S. residents—a 25% decline in absolute terms and 30% per capita decline from
2009.
The mass incarceration narrative usually takes 1973 as its starting point to measure a
“normal” prison population rate pre-mass incarceration, and it often takes a year around 2009
as its ending point to maximize the stated increase. We believe it makes more sense to begin an
analysis of the prison population in the early 1960s. The late 1960s and early 1970s were
marked by social unrest, the beginning of a crime surge, and the Vietnam War, which may have
siphoned off young men who might otherwise have ended up in prison. Starting an analysis in
1960, before all these societal transformations, seems to give a better benchmark for what
“normal” incarceration levels looked like.34 Additionally, it is important to choose a recent year
like 2021 as an end point for the analysis as it captures the recent decline in prison population.
Therefore, the relevant question is what factors are responsible for the 2021 U.S. prison
population being 300% larger, on a per capita basis, than in 1960? According to the mass
incarceration narrative, the explanation lies largely in increased and unjustified American
punitiveness. Specifically, a higher percentage of offenders are being sent to prison and
prisoners are receiving unjustly longer sentences. If sentencing policy was returned to the
1960s, then the prison population would shrink back to comparable levels—or so the story
goes. Indeed, some activists point to 1960s sentencing practices as the solution to mass
incarceration.35

11

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

Table 1. U.S. Prison Population Over Time
Year

Total
Prison
Population

192537
1930
1935
1940
1945
1950
1955
1960
1965
1970
1973
1975
1980
1985
199041
1995
2000
200842
200943
201144
2015
2021

91,669
129,453
144,180
173,706
133,649
166,123
185,780
212,953
210,895
196,429
204,211
240,593
315,974
481,616
773,905
1,125,874
1,381,892
1,608,282
1,615,487
1,598,968
1,526,603
1,204,322

Prison Pop.
per 100k
U.S.
residents
79
104
113
131
100
109
112
117
108
96
95
111
138
201
292
411
477
506
504
492
459
350

Total New
Imprisonments

New
Imprisonments
per 100k U.S.
residents36

103,59838

58

190,01439
182,61740
271,366
474,128
562,724
666,077
750,392
736,796
676,793
612,952
412,00645

89
81
113
191
214
237
246
240
217
191
124

3. Prison Composition
In addition to clarifying what per capita prison population increase has occurred, a
reasoned examination also needs to clarify the kind of offender who is and is not actually in
prison. Consider this recent summary:46

12

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

Table 2. Composition of U.S. prison population

Those who take the mass incarceration narrative at face value may be surprised to learn
that prison is not filled with non-violent, first-time drug possessors. As the chart labeled Table 2
shows, most of those in prison are much more serious offenders. Drug offenders comprise just
17% of the total prison population. Moreover, most of these drug offenders have committed
crimes worse than simple possession. There are 34,000 state prisoners serving sentences for
drug possession, compared to 98,000 state prisoners serving sentences for more severe drug
crimes (i.e., drug trafficking). Even more lopsidedly, of the 69,000 federal drug offenders, 99% or
more are serving sentences for drug trafficking. In other words, just 17% of drug offenders in
prison are serving sentences for drug possession (many of which may have been plea bargained
down from more serious offenses).47 Only 3% of the total prison population is made up of those
sentenced for drug possession. For those who believe mass incarceration could mostly be
solved by simply releasing drug possessors (or even all drug offenders including traffickers),
these numbers will come as a nasty surprise.
The mass incarceration narrative’s myth of prison composition also runs into the
inconvenient fact that the prison population is largely made up of those sentenced for violent
crimes or recidivism. For example, a 1996 BJS study examining a representative sample of
711,000 imprisoned felons found “that fully 94% of state prisoners had either committed one or
more violent crimes (62%) or been convicted more than once in the past for nonviolent crimes
(32%).”48 More recent data confirms this picture. In 2021, 62% of state prisoners were convicted
of a violent crime, and if anything, the criminal histories of convicted offenders have worsened
over time.49 Almost no one goes to prison for a first-time, non-violent offense. As John Dilulio
noted: “It’s the hardcore doing the hard time.”50
13

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

This description has only become truer over time. As Table 3 shows, the percentage of
the state prison population convicted of violent offenses increased from 41% to 62% from 1960
to 2021—appearing to contradict the notion that the prison population is larger today because
of an increased share of non-violent offenders receiving prison sentences. Ironically, one would
be more likely to meet a non-violent offender in prison before the era of “mass incarceration”
than after it.
Table 3. Percentage of Violent Offenders in State Prison Population Over Time
Year

1960
1974
1980
1992
2001
2010
2021

Violent Offenders
as % of state
prison population51
41%
52%
58%
47%
49%
53%
62%

This does not mean prison is necessarily an ideal punishment for all violent offenders or
serial property and public order offenders (a category which includes weapons offenses), but it
does expose the myth that America’s prisons are crammed with non-violent, first-time offenders
caught in the clutches of a carceral state. The nature of America’s prison population must be
acknowledged by reformers who wish to seriously change American incarceration practices.
Releasing all drug offenders would mean releasing primarily drug traffickers who have profited
from destroying the lives of others. Releasing all non-violent offenders would mean turning
loose mostly hardened recidivists who have had a second or third chance already and used it to
commit more crimes against their communities. Again, this does not mean there is nothing that
can be done to reduce prison populations, but these facts must be faced squarely without aid
from the comforting myth that America’s prisons are a place where pot smokers mingle with
shoplifters and first-time burglars.

B. Non-sentencing Factors in the Rise of the U.S. Prison Population
This section examines factors behind the rise of the U.S. prison population that are
largely unrelated to sentencing – and the punitiveness it supposedly reflects. To the extent that
non-sentencing factors explain most of the rise in U.S. prison population, it contradicts the mass
incarceration narrative’s assumption that the increase is caused by unjustified and exceptional
American punitiveness.52

14

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

1. Population Increase
An obvious but sometimes ignored factor in the rise of the U.S. prison population has
been the increase in the U.S. population. Between 1960 and 2020, the U.S. population
increased by 85%, from 179,323,175 in the 1960 census to 331,449,281 in the 2020 census.53
Assuming a constant crime rate and incarceration policy, one would expect the prison
population to increase proportionally to the population. In other words, one would expect an
85% increase in the U.S. prison population over this period purely due to population growth. Of
course, many serious proponents of the mass incarceration narrative deal in per capita terms,
which account for population changes, but it is not hard to find absolute prison population
change still being cited by some to overemphasize the scale of the increase.54 It should not need
to be said that an incarceration increase caused by overall population growth is perfectly
justified.
2. Increased Crime
A fact strangely missing from standard accounts of mass incarceration in the latter half
of the 20th century is mention of the enormous crime surge that also took place during the
same period. Between 1960 and 1990, the per capita violent crime rate increased by over
350%.55 Perhaps mass incarceration activists forget to mention this crime surge because it
suggests much of the increase in incarceration had nothing to do with American punitiveness
but rather with American criminality. Holding sentencing and criminal justice system
effectiveness constant (the percentage of reported crimes converted into
convictions/imprisonment), a higher crime rate naturally leads to a higher incarceration rate.
While crime rates peaked in the 1990s, the crime rate today remains significantly higher than
that which prevailed in the 1960s before the start of the crime or incarceration surge. In 2019
(before the violent crime surge beginning in 2020),56 the violent crime rate was 379.4 per
100,000 people compared with 160.9 in 1960—an increase of over 135%.57 Property crime was
also higher in 2019 than in 1960 with 1,726 property crimes reported per 100,000 in 1960
compared to 2,110 in 2019—an increase of 22%.58
These higher crime rates—especially the higher violent crime rates—guarantee a higher
incarceration rate than in the 1960s, and such an increase is desirable so long as crime remains
at elevated levels. To get a rough sense of how much increased crime could affect incarceration
rates, let us assume that about 55% of the 1960 prison population was sentenced for property
offenses and 40% were sentenced for violent offenses.59 If the per capita violent offense rate
increased by 135% and property offense rate by 22%, we would expect an overall per capita
incarceration rate increase of 66%, assuming no changes in sentence severity or criminal justice
system effectiveness.60 Of course, if criminal justice system effectiveness increased or
sentencing for crimes became stricter, the effect of increased crime on the incarceration rate
would be multiplied.
The previous calculation does not even consider the greatest crime increase of all from
1960 to the present: drug crimes. As noted previously, some mistakenly assume the ‘War on
Drugs’ initiated in the 1970s is solely responsible for the increase in drug offenders in prison
compared to their much smaller (around 5%) share in 1960. Under this misperception, the use
of illicit drugs stayed steady, but the government suddenly intervened with harsher
15

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

punishments in the ‘70s. In fact, this is only partially correct—the manufacture, distribution, and
possession of illicit drugs were banned well before the 1960s, but a massive spike in drug usage
led to greater enforcement and higher penalties. However, even without the stricter legislation
associated with the War on Drugs, the population of drug offender in prison would still have
swelled from 1960s levels simply because of the greater prevalence of drug crimes. This is
important when considering the specific effect of the War on Drugs in increasing incarceration
(something considered later in this article).
Unlike for property or violent crimes, where police receive victimization reports, drug
offenses are generally not a reported crime, making it necessary to turn to other sources in
estimating the increasing incidence of drug crimes in the population. In 1960, less than 5% of
the U.S. population had ever tried any illicit drug.61 By 2020, that number was 50%, a ten-fold
increase. Additionally, in 2020, over 20% of Americans used an illicit drug or abused prescription
drugs within the last year, and 13.5% were illicit drug users within the last month.62 In other
words, it seems likely the rate of illicit drug use (and therefore drug crimes) in the population
increased by somewhere between 5-10 times between 1960 and 2020. If we assume drug
offenders made up 5% of the prison population in 1960, and the rate of drug crimes
conservatively only quintupled, then we should expect another 20% increase in the per capita
prison population just from increased drug crime without any changes to drug enforcement or
sentencing policy; if the drug crime rate increased by tenfold, we should expect a 45%
incarceration rate increase.63 Again, these increases would be expected even if the criminal
justice system punished drug offenders in the same way and at the same rate as it did in 1960
before “mass incarceration.” Holding all other factors constant, it would appear increases in the
underlying rate of drug, property, and violent crime would be expected to cause at least an 85%
per capita incarceration rate increase since 1960—a factor largely ignored by the mass
incarceration narrative.64
3. Deinstitutionalization of the Mentally Ill
A relevant and overlooked factor in America’s crime and incarceration surge in the latter
half of the 20th century was the widespread deinstitutionalization of the mentally ill. During the
first part of the 20th century, private and state mental hospitals and asylums absorbed hundreds
of thousands of mentally ill Americans. However, changes in medicine, societal perception, and
law in the 1960s and ‘70s led to the shuttering of most of these institutions, and society was
forced to deal with the mentally ill in other ways. Instead of being locked up in mental hospitals,
some mentally ill ended up incarcerated in prisons instead, after being freed from civil
commitment only to run afoul of the criminal law. Deinstitutionalization meant in many cases
that penal institutions simply took over for mental institutions. While America’s treatment of
the mentally ill has often been tragic and frustrating, the increase in incarceration caused by
deinstitutionalization obviously had nothing to do with increased punitiveness or a change in
criminal justice policy, but rather with the unintended consequences of closing mental
institutions.65
The scale of deinstitutionalization should not be underestimated. In 1955—at its peak—
the population of state mental hospitals was almost 560,000. In 2015, despite the U.S.
population almost doubling since 1955, the number of people in state mental hospitals was
16

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

around 35,000.66 In per capita terms, that equates to an over 95% reduction in the rate of
institutionalization for the mentally ill in state mental hospitals. While most of those
deinstitutionalized posed no criminal threat to the community, there was a portion that did.
While precise quantification is difficult, a non-trivial proportion of current prisoners would have
been institutionalized in mental hospitals in the past. For example, one study suggested that
anywhere from 4.5 to 14% of state prisoners in 1996 would have been institutionalized in
mental hospitals before deinstitutionalization—48,000 to 148,000 prisoners in absolute terms.67
Similarly, another study suggested that 40,000 to 72,000 prisoners in 2000 would have been
institutionalized in the past.68 Prisons are full of mentally ill offenders. By one count, 14% of
state prisoners meet the qualifications for “serious psychological distress,” and 43% have a
history of mental health problems.69 It appears that anywhere from 5 to 15% of currently
incarcerated offenders in prison might have been confined in mental institutions in 1960.70
It should be no surprise, then, that there is an inverse relationship between
institutionalization and crime/incarceration. Graph 1 shows the relation between
deinstitutionalization in mental hospitals and incarceration in prison.71
Graph 1. Institutionalization in Mental Hospitals and in Prisons

--.

Figure 1. ~tes of institutionailiution in tht United St&tH (ptr 100,000 .1dulU), 1934-2001

4. Increased Criminal Justice System Effectiveness
Not all incarceration increases should be seen as problematic. If prison populations rise
because of improved justice system effectiveness, with no change in sentencing policy, the
increase represents progress. One factor broadly overlooked in the mass incarceration literature
is criminal justice system effectiveness: the rate at which criminal offenders are caught and
convicted.
If police and prosecutors became twice as effective at catching and punishing rapists
tomorrow, the prison population of rapists would soon double without any change in rape
sentencing practices, but few would consider this increase problematic. If the goal of a
successful criminal justice system is to catch and justly punish every serious offender, then
increasing effectiveness is cause for celebration, and the resulting increase in imprisonment
17

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

reflects a problem with the ineffectiveness of past practices. Reducing incarceration through
degrading the effectiveness of the justice system at catching criminals (as opposed to reducing
the punishments for convicted criminals) would be downright perverse as it purchases lower
prison populations through denying justice to innocent victims and increasing future
victimizations.
Perhaps because the rate of justice failures is so high, governments are wary of
publishing statistics on criminal justice system effectiveness. However, usable statistics exist for
some things, such as total crime, arrests, and new imprisonments – but the middle steps
between arrest and imprisonment, particularly prosecution effectiveness, are harder to gauge.
There are no nationwide historical statistics on the rate at which prosecutors accept and
prosecute cases “cleared” by police or the percentage of filed cases that result in conviction.
Even today, statistics on prosecution are often murky. However, general trends in criminal
justice effectiveness are still discernable.
In 1960, FBI statistics showed a total of 3,384,200 crimes reported to police. That same
year, there were approximately 95,368 new prison admissions resulting from crime, thus
representing 2.8% of reported crime.72 By 2018, statistics showed 8,402,881 total reported
crimes and approximately 495,698 new imprisonments resulting from crime, or 5.9% of
reported crime.73 This equals a 110% increase in the rate of new imprisonments compared to
reported crimes.
Clearly, the justice system is getting better at turning reported crime into
imprisonment—but why? Contrary to the mass incarceration narrative, most of the answer does
not appear to come from sentencing (i.e., judges sentencing a higher percentage of convicted
offenders to prison). While data on the percentage of convicted offenders sentenced to prison is
hard to find pre-1980s, the picture becomes clearer in the following two decades and suggests
the higher percentage of imprisonments is mainly due to changes in arrest or prosecution as
opposed to harsher judicial sentencing. For example, between 1986 and 2006 the share of
convicted state felony offenders receiving prison sentences slightly decreased from 46% to 41%,
even as the prison population increased substantially and new imprisonments as a percentage
of total crime rose.74 In 1986, there were approximately 13,211,869 crimes reported to police
and approximately 202,995 new prison admissions resulting from crime, or 1.5% of reported
crime.75 In 2006, the numbers were 11,401,511 reported crimes and approximately 615,200
new imprisonments resulting from crime, or 5.4% of reported crime, a substantial increase from
1986 despite convicted offenders receiving prison sentences at a lower rate.76
When looking at the data from 1960 to the present, a general trend becomes
apparent.77 The devastating crime wave starting in the late 1960s caught the criminal justice
system unprepared, underfunded, and understaffed. The result was a steep drop in justice
system effectiveness as criminals got away with crimes at ever higher rates, leading to more
crime in a vicious cycle. Finally, significantly increased funding and personnel, as well as more
effective policing and prosecution strategies, allowed the justice system to catch up with the
crime wave and recover its previous effectiveness. As the crime wave abated beginning in the
1990s, criminal justice system effectiveness continued to increase as crime rates fell, creating a
virtuous cycle.
More detailed statistical analyses in existing literature bear out these surface level
observations. One study found that between 1990 and 2006, the “true conviction” rate for all
18

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

crimes (measured by crime victimization data and FBI conviction rates) rose from 1.24% to
1.95% of total crime—a 57.2% increase over this period alone.7879
Increased criminal justice system effectiveness compared to 1960 is clearly one reason
America has a higher prison population today. The main driver of this increased effectiveness
appears to be changes in the prosecution success of arrests as opposed to increased arrest
clearance rates.80 In other words, a criminal is not much more likely to be arrested for a crime
today than in 1960, but once arrested, he is more likely to be successfully prosecuted. There are
numerous possible explanations for this increased prosecution success, including better police
investigations and increased prosecution resources, but whatever the cause, the result is clear.
One study found that “between 1994 and 2008 filings [of criminal charges] per arrest rise from
0.374 to 0.573." Since the conviction rate did not change much, that amounts to an almost 55%
increase in prosecution effectiveness over that 14-year period alone.
Considering the 110% increase in the rate of new imprisonments compared to reported
crimes over the 1960-2018 period, justice system effectiveness may have doubled, but we will
conservatively assume only an increase in effectiveness of 67%.81 This means we should expect
a 67% per capita incarceration increase from increasing system effectiveness compared to 1960
even if we hold all other factors constant. Of course, this increased effectiveness interacts with
other factors such as the increased crime rate to explain even larger increases in incarceration
with no change in sentencing policy. For example, if we assume increased crime rates in 2021
would result in an 85% per capita incarceration increase compared to the 1960 rate, then also
increasing criminal justice system effectiveness by 67% would lead to an expected per capita
incarceration rate increase of 209%—meaning increased crime and justice system effectiveness
already explain more than two-thirds of the 300% per capita incarceration rate increase from
1960 to 2021.82
Despite its significant effects on incarceration, justice system effectiveness is almost
never seriously discussed in the mass incarceration narrative because it would cast some
incarceration increases as a success to be celebrated instead of a problem to be solved. To the
extent system effectiveness is mentioned, it is portrayed in a negative light, implying that police
are making wrongful arrests and prosecutors are generating numerous wrongful convictions.83
However, research shows the percentage of wrongful convictions is tiny,84 and there is no good
reason to think the rate of wrongful convictions or wrongful arrests has significantly increased
due to greater system effectiveness at punishing crime generally.
Some proponents of the mass incarceration narrative deplore increasing justice system
effectiveness because they prioritize low prison populations over punishing crime (or because
their real agenda is anti-punishment). While the public would welcome a doubling of justice
system effectiveness, many mass incarceration activists would oppose such a change. This
ideological prioritization of low prison populations over doing justice in cases of crime is
especially bizarre since the vast majority of serious crime already goes unpunished.85
5. Changing Criminal Histories
One almost completely ignored factor in incarceration increases is the fact that the
average criminal offender has changed over time. Repeat offenders are punished differently
than first-time offenders in America—a difference in treatment most people would find
19

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

completely appropriate even if they strongly disagree with the severity of current repeat
offender sentences. Repeat offenders’ blameworthiness is greater as they are engaging in
“nose-thumbing” against the law86 and showcasing a hardened bent toward criminality through
rejecting previous chances at reform. As a result, repeat offenders are more likely to receive a
prison sentence upon conviction, and the length of that prison sentence is likely to be longer. 87
It is important to note that repeat offenders were sentenced differently than first-time offenders
long before the era of “mass incarceration.”88 Therefore, incarceration increases caused by an
increasing share of convicted criminals being repeat offenders cannot be chalked up solely to
increased punitiveness in sentencing. While post-1960s laws certainly increased the punishment
for some repeat offenders, an underlying increase in repeat offenders would have generated
incarceration increases under pre-1970s sentencing policies.
Unfortunately, detailed historical data on convicted offenders’ criminal histories is
extremely hard to find. We are unaware of any useful comparisons from before the 1990s.
However, data does exist from the 1990s through 2000s, and the trend there is enlightening for
the overall period since 1960. According to a study examining felony defendants in large urban
counties,89 the percentage of defendants with 10 or more prior arrests jumped from around
20% in 1990 to 36% in 2009. The percentage with 10 or more felony arrests jumped from 9% to
17%. The data on prior convictions is even more shocking. In 1990, 64% of felony defendants
had no prior felony conviction, while 36% did. Only 4% had 5-9 prior felony convictions, and
only 1% had 10 or more. In 2009, 40% of felony defendants had no prior felony conviction, a
drop of 24 percentage points. Even worse, 15% had 5-9 previous felony convictions, and a
stunning 14% had 10 or more felony convictions—a 1300% increase in the presence of the most
serious recidivists.
There was clearly a severe increase in the criminal records of offenders from 1990 to
2009.90 That this contributed to the rise in incarceration over that period (even as the crime rate
fell) is beyond question. Extrapolating from the limited period of data available, it seems likely
there was a significant increase in the criminal records of offenders between 1960 and today—a
factor that would substantially contribute to a higher incarceration rate.91
6. New Criminalizations and Increased Enforcement
The mass incarceration narrative is eager to point to the War on Drugs’ expanded
criminalization and enforcement against illicit drugs as a contributor to incarceration increases,
but it forgets to mention other significant changes to criminalization that have expanded the
scope of incarcerable behavior. Since 1960, governments at the state and federal level have
created entirely new crimes and tightened their enforcement of many existing laws. Except for
drug laws, most of these changes in criminalization and enforcement are entirely
uncontroversial, and any increases in incarceration caused by them are likely to be seen as
appropriate by most of society. These changes to criminalization are not examples of
exceptional American punitiveness. Indeed, all the crimes discussed in this section are also
crimes in most comparable countries.

20

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

Drug Offenses

As anyone aware of the mass incarceration narrative knows, America’s criminalization of
both drug trafficking and possession has increased the prison population. While the ‘War on
Drugs’ began in the 1970s, America first sought to control drugs at the federal level in 1914 with
the Harrison Anti-Narcotics Act. Even before the 1960s, the government recognized illegal drug
use as a threat and repeatedly tightened punishments. As the Drug Enforcement Agency notes:
In the first half of the 1950s the average length of narcotics sentences in 86 U.S. district
courts had doubled, from two years to four years, largely because of the penalty
provisions of the Boggs Act. After the Congress unanimously passed the Boggs-Daniel
Narcotic Control Act of 1956, the average length of sentences rose in the next two years
from four to six years. The Act provided a mandatory minimum sentence of five years in
prison for a first offense of illegally selling narcotic drugs or marihuana, and from 10 to
40 years.92
These new laws did not result in massive incarceration increases because illegal drug trafficking
and usage were still a relatively minor problem. In 1960, drug offenders made up around 5% of
the total state prison population.93 By 1965, however, there were already almost 4,000 drug
offenders in federal prison, making up 17.9% of the federal prison population.94
This history is important because it shows that regardless of any punitive changes
associated with the ‘War on Drugs,’ existing criminalization of drugs would have guaranteed a
substantial drug offender population in prison simply because of increasing drug crime, a point
noted previously. If we assume the criminal justice system made no changes to drug sentencing
but simply kept the same rate of enforcement as it had in 1960 before the ‘War on Drugs,’ then
a conservatively estimated fivefold increase in the rate of illicit drug use would be reflected in a
400% rise of drug offenders in prison. Roughly adjusting for the increases in other types of
crime, one would expect drug offenders to make up about 13% of the prison population today.95
Their actual share is 15%--about what one would expect if the increase in underlying drug
crimes was a little over fivefold, as seems likely.
Of course, this picture is not perfectly accurate as the surge in drug usage was not met
with the same level of enforcement. It seems likely a smaller percentage of illegal drug users
were caught, but those that were caught suffered somewhat more severe penalties due to
sentencing changes post-1970. The epidemic of drug usage in the 1960s and 70s led to stricter
punishments in the ‘70s, ‘80s, and ‘90s as policymakers seized at sentencing as a utilitarian tool
to stop the drug surge. The infamous ‘War on Drugs’ was declared in 1971 by Richard Nixon and
led to a series of state and federal laws ratcheting up punishments for drug crimes. For example,
Nelson Rockefeller, New York’s Governor, oversaw the passage of a mandatory minimum
sentencing law that mandated a minimum 15 years in prison for those convicted of selling 2
ounces or possessing 4 ounces of certain controlled substances. That said, much of the ‘War on
Drugs’ did not consist in increasing penalties or criminalization but simply spending greater
resources on enforcement to keep up with the surge in drug crimes.
Today, there is substantial debate whether criminalizing drugs instead of pursuing
legalization and regulation was the correct policy choice. Without minimizing the importance of
such debates, it is important to remember that the initial decision to criminalize drugs was not a
21

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

result of exceptional punitiveness on the part of the American public and policymakers. Drug
prohibitions are a feature of governments worldwide, and the Single Convention of Narcotics
was first adopted in 1961 and later amended in 1972. The treaty has been ratified by 186
nations worldwide and commits signatory states to combatting the drug trade and possession
through criminal penalties.96 The enormous costs of drug abuse made governments’ decision to
tighten drug laws and ramp up criminal penalties worldwide understandable, even though many
today have come to see criminal justice systems as ill-equipped to deal with substance abuse.
The size of the problem and its societal cost are certainly hard to ignore. One study found that,
“Substance misuse and substance use disorders costs the U.S. and local communities
approximately $442 billion each year."97 The decision to combat dangerous and addictive drugs
through criminal means is understandable, even if some of the punishments for drug crimes
have come to be seen as unjustified, or the entire policy of criminalization a mistake.
Contrary to a strain of conspiratorial thinking prevalent in the mass incarceration
narrative that views the criminalization of drugs as resulting from a racist plot to imprison
people of color, the truth is more mundane.98 The surging cost to society of drug usage led to a
public demand for government intervention. During the decades the ‘War on Drugs’ was at its
height, it commanded clear public support. As one researcher summarizes the polling data:
“Since 1969, the first year Gallup asked about illegal drug use, Americans have grown
increasingly more concerned about the effects of drugs on young people. For instance, in 1969,
48% of Americans told Gallup that drug use was a serious problem in their community. In 1986,
a majority of Americans, 56%, said that the government spent "too little" money fighting drugs.
By 1995, 31% said drug use was a "crisis" and an additional 63% said it was "a serious problem"
for the nation as a whole.”99
The effect of the ‘War on Drugs’ on incarceration should not be ignored, but neither
should it be overstated. Drug criminalization by itself can explain only 15% of the prison
population today.100 However, changes to drug criminalization and punishment associated with
the ‘War on Drugs’ and American punitiveness post-1970 can explain a much smaller
percentage, since pre-1970 drug criminalization and enforcement policies would still result in a
substantial drug offender prison population today.
It is also important to remember the ‘War on Drugs’ is in steep decline. Between 2009
and 2019, the numbers of people admitted to and held in state prisons for drug offenses both
fell by about a third.101 All told, there are a little over 200,000 sentenced drug offenders in U.S.
prisons. As previously discussed, the vast majority of these (over 165,000) are in prison for more
serious drug offenses than possession, such as trafficking. Moreover, many of these offenders
have committed other crimes.102 Simply decriminalizing the possession of drugs would barely
reduce the prison population, and even assuming a complete legalization of drugs, it is likely
that many of the current 200,000-or-so imprisoned drug offenders would land, or remain,
behind bars on different charges. Becoming a drug trafficker does not tend to be a one-off crime
but is commonly part of a general criminal orientation, which seeks to live and make money
outside the law.
Child Pornography and Sexual Abuse

While drug criminalization has become controversial, there have been many other new
criminalization and tightened enforcement decisions since 1960 that command broad support
22

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

today. One example are laws targeting child pornography and sexual abuse. The criminal justice
system was practically uninvolved with such conduct in the 1960s, and laws mandating
reporting of potential child abuse were only enacted in 1974,103 though truly serious efforts by
the criminal justice system to combat the problem would take longer. While child sexual abuse
has always existed, its importance to policymakers has risen, perhaps fueled by the internet
aiding in grooming children and disseminating child pornography.104 As a result, a wave of
expanded criminalization and enforcement has led to far more people incarcerated for child sex
offenses than in the past. The resulting increase in imprisonment is substantial. There were an
estimated 127,282 individuals incarcerated in 2021 at the state level for “sex offenses involving
children” and another 12,850 such offenders incarcerated at the federal level.105 This represents
roughly 11% of America’s 1.2 million prison inmates. Few would argue the expanded
criminalization and enforcement of child sex offenses is unjustified or reflects exceptional
punitiveness. Rather, most would see it as progress in combatting a scourge that has always
existed but until recently was not treated by the justice system with the gravity it deserves.
Human Trafficking

Another area of expanded criminalization since the 1960s is human trafficking. The
Trafficking Victims Protection Act of 2000 was the first comprehensive federal legislation
targeting human trafficking.106 Previously, trafficking could only be prosecuted under a series of
patchwork older laws regarding slavery and involuntary servitude that were often too narrow to
clearly apply. The new legislation added “provisions prohibiting forced labor, trafficking with
respect to peonage, slavery, involuntary servitude, or forced labor, and sex trafficking of children
or by force, fraud, or coercion” as well as criminalizing any attempt to engage in such activities.
Historically, around 500 or more human traffickers are convicted each year in the federal
system.107 Almost all such traffickers are sentenced to prison, with 86% of federal trafficking
defendants in one recent year sentenced to 5 or more years.108
New White-Collar Crimes

Since the 1960s, federal and state governments have moved to create more white-collar
crimes with statutes targeting environmental and financial offenses, among others. For
example, in 1982, the EPA and DOJ created specialized units to prosecute environmental crimes,
and the Clean Water Act was amended in 1987 to create felony crimes for some environmental
offenses such as knowingly polluting waters without a permit. In 1990, the Clean Air Act was
amended to add “numerous felony provisions.”109 Environmental advocates have hailed the
increasing use of criminalization and incarceration as appropriate responses and deterrents to
those who willfully cause grave environmental damage.
Identity theft is another white-collar crime that ballooned in recent years with the aid of
the internet. In 1998, Congress passed the first federal identity theft specific statute, the
Identity Theft and Assumption Deterrence Act.110 Since, then prosecutions have increased with
many resulting imprisonments. “Identity theft offenders accounted for slightly more than two
percent (2.2%, n=3,694) of the federal prison population as of September 30, 2016.”111
Financial crimes have also been increasingly defined, penalized, and enforced. For
example, in 1986, Congress passed the Money Laundering Control Act which made money
laundering a federal crime for the first time.112 According to U.S. Sentencing Commission Data,
23

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

over 70% of those convicted of fraud and related crimes at the federal level were sentenced to
some form of incarceration.113 While these new white-collar crimes and increased enforcement
have only had a small impact on the prison population, their effect should still be noted to
showcase just how much the criminal justice system has seen its scope—and therefore the
scope for incarceration—expand in recent decades.
Sexual Assault

While there have always been laws criminalizing rape and many kinds of sexual assault,
the post-1960 period saw an increase in public attention to sexual violence and an expanded
focus on enforcing the law. Behaviors that were once normalized or accepted are now rightly
viewed as criminal by society. The #MeToo and #TimesUp movements were a recent
manifestation of a broader trend to take sexual crimes more seriously. It makes sense that as
women gained more political, economic, and social power over the last half-century, rape and
sexual assault would be addressed more rigorously by the justice system. The resulting increase
in imprisonment reflects progress, and seeking to reduce the rate of punishment for sexual
crimes is a step in the wrong direction.
New legislation has substantially expanded criminalization around sexual assault in the
last 50 years. As strange as it may seem to our moral consciences today, marital rape was not
considered a rape crime 50 years ago and could only be prosecuted under assault laws. “In
1976, however, Nebraska became the first state to make marital rape a crime. By 1993, marital
rape was a crime in all 50 states.”114 Additionally, it was only in 1975 that congress passed “rape
shield” laws to change the federal rules of evidence to prevent a victim’s sexual history from
being used in court to undermine or humiliate the victim.115 Prior to such changes, the ability to
drag a victim’s sexual history through court served as a deterrent to victims seeking justice. New
technologies have also opened new avenues for sexual abuse, resulting in new criminalization.
For example, “Thirty-eight states … have enacted revenge porn laws, criminalizing the
distribution of sexually explicit images or videos without the individual’s consent.”116
These changes reflect a broader societal change—often implemented in updated state
laws—that see “lesser” forms of sexual assault as increasingly equivalent to rape and includes a
recognition that rape is a crime that can affect people of any gender. Considering that some
researchers estimate that less than 1% of rapes lead to the rapist spending time in prison,
activists should be looking for ways to continue expanding the prison population of rapists, not
decreasing it.117 As one Stanford researcher has noted about comparing the current U.S.
incarceration rate with lower pre-1970s rates, "That low rate was in part a function of not taking
rape, spousal abuse, and other male violence against women seriously — and we should not
seek to go back there.”118
Domestic Violence

The criminal justice system has also moved to take domestic violence more seriously
over time. In the past, domestic violence was considered an interpersonal issue beyond the
purview of the criminal law. “In fact, many police departments had ‘hands off’ policies prior to
the 1970s, and police training manuals actually specified that arrest was to be avoided
whenever possible in responding to domestic disputes.”119 Since then, states have tightened
domestic violence laws and police have stepped up enforcement. While most domestic violence
24

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

arrests do not lead to a felony assault prosecution, and so such offenders do not receive prison
sentences, a substantial number of those sentenced to prison for aggravated assault committed
that assault in the context of domestic violence. One study found that “of the approximately
1,500 defendants charged with felony assault during May 2000 in the State courts of 11 large
counties, about a third were charged with family violence."120 Of those convicted, 83% were
sentenced to prison or jail.121 Between 2000 and 2002, domestic violence offenders convicted
under federal law for “an interstate domestic violence offense” received even stricter
punishments, as “91% received a prison term with a median length of 60 months.”122 Given that
146,000 state prisoners are serving sentences for assault, the effect of increased domestic
violence criminalization and enforcement on increasing incarceration should not be ignored.
Stalking

Stalking is an old behavior only recently criminalized by specific laws targeting the crime.
Stalking is widespread and ranges in seriousness, but it affects a disturbingly large segment of
the population. By one count, “Approximately 3.4 million people are stalked each year in the
United States, and 1 in 5 women and 1 in 10 men will be stalked in their lifetime.”123 California
was the first state to pass a specific anti-stalking statute in 1990, but all states have followed
with their own anti-stalking laws. The first federal anti-stalking statute was passed in 1996.124
According to a 2009 study by the Department of Justice, stalking results in over 50,000
offenders being jailed or imprisoned each year.125
Weapons Offenses

While states and the federal government regulated weapon possession long before the
1960s, the increase in violent crime beginning in the late 1960s led to calls for taking weapon
offenses more seriously. The result was new gun control laws at the state and federal level and
increased enforcement against weapon offenders. At the federal level, these new laws included
the Gun Control Act of 1968, the Firearm Owners Protection Act of 1986, the Brady Handgun
Violence Prevention Act of 1993, and the Public Safety and Recreational Firearms Use
Protection Act of 1994.126 New state laws tightened gun control and added additional penalties
for crimes committed with deadly weapons. By 1992, 26,000 offenders were convicted under
state laws regulating weapons, with 66% receiving an incarcerative sentence. The average
prison sentence for weapon offenders sentenced to prison was 4 years.127 According to a 2016
study, 21% of state and federal prisoners had “possessed or carried a firearm when they
committed the offense for which they were serving time in prison.”128 Over 153,000 arrests
were made for weapon offenses in 2019 by state and local police,129 and there are around
39,000 prisoners serving time in state prison for a weapons offense.130 At a time when gun
violence is rightly considered a serious issue, there is likely to be significant support for
maintaining the system’s current criminalization and punishment of weapon offenses.
7. Other Non-Sentencing Factors
The above factors are only a selection of the most important ones to consider; they are
by no means an exhaustive list of possible non-sentencing factors that affect prison populations.
Consider two other examples: changes in general life expectancy and parolee behavior.
25

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

Life expectancy in society has increased significantly since the middle of the 20 th century,
meaning that those serving life sentences are likely to survive longer, possibly contributing to
rising prison populations (especially since those with LWOP sentences received during the crime
wave of the 1960s-1990s will remain in the prison population longer). In 1950, U.S. life
expectancy was about 68 years.131 In 2021, it was around 79 years, an increase of 16%. More
specific to the prison population of mostly male and increasingly elderly lifers, the life
expectancy of a 65-year-old man in 1960 was 12.8 more years. In 2021, it was 17, an increase of
over 30%.132 We make no claims about how much gains in life expectancy may have translated
to the prison population, and thus increased the share of prisoners serving life or extremely
long sentences, but it seems likely to have had a non-trivial impact.
Another important non-sentencing factor is the behavior of parolees or other supervised
releasees who may be returned to prison for violating their release terms. The share of new
prison admissions caused by parolees violating the terms of their release (either through
committing a new crime or a technical violation such as failing to meet monitoring conditions)
has increased substantially since 1960. In 1960, the number of new admissions caused by
release revocations was only 15% of what new convictions were.133 By 2018, the number of new
admissions caused by release revocations was 41% of what new convictions were.134 While this
increase in revocations clearly contributed to rising incarceration, its effect is hard to untangle
from other factors that both increase incarceration and the total number of parolees. The
increasing number of parolees returned to prison for committing new crimes is largely a
downstream effect of larger prison populations generating larger parole populations that then
reoffend in line with general crime and recidivism increases.135
However, one possible distinct parole-related factor is whether parolee behavior has
changed over time to be less compliant. For example, if current releasees are more likely to
ignore monitoring conditions than they were in 1960, this would lead to incarceration increases.
Unfortunately, the answer is difficult to determine not only due to a lack of data but also
because the discretionary judgements of parole officers are important in what kind of violations
qualify for readmission to prison. We make no claims about how this factor may have affected
incarceration, but the question deserves research.
8. The Total Effect of Non-Sentencing Factors on Incarceration
Proponents of the mass incarceration narrative usually make no attempt to consider
how a wide variety of non-sentencing factors contributed to the increase in prison population
because such factors have nothing to do with supposed American punitiveness. Instead,
proponents of the narrative commonly mislead their audience into believing that increased
incarceration is itself proof of increased punitiveness.
Contrary to that narrative, the above non-sentencing factors likely explain a large
majority of the increased per capita incarceration rate when comparing 1960 to 2021. In fact,
one can plausibly explain most of the incarceration increase without reference to increasing
punitiveness in sentencing.136 Increased crime and criminal justice system effectiveness alone
likely account for more than 200% of the 300% per capita increase in incarceration. New
criminalization and enforcement likely accounts for at least a 50% increase, though this overlaps
with the increase already counted from crime (since new criminalization increases total crime
26

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

even with no underlying changes in societal behavior). The effect of deinstitutionalization and
criminal offender histories likely accounts for at least another double digit increase in
incarceration. Our best guess is these non-sentencing factors account for at least 200-250% of
the 300% per capita incarceration rate rise between 1960 and 2021. This does leave some
(small) room for sentencing factors, but as discussed in the next section, it is difficult to
distinguish the effect of increased punitiveness in sentencing from a non-sentencing factor like
changing criminal histories in determining why a harsher sentence was handed down. As a
result, this estimate is conservative. What is clear is that one can explain most of America’s
elevated per capita incarceration rate compared to the past with factors ignored by the mass
incarceration narrative.

C. Sentencing Factors in the Rise of the U.S. Prison Population
The mass incarceration narrative is quick to blame punitive sentencing whether in the
form of mandatory minimums, truth-in-sentencing laws (that abolish early release on parole),
or higher guideline punishments for America’s incarceration increase. All these changes fall
under two broad sentencing factors that affect incarceration. First, the system may change the
chance of a convicted offender receiving a prison sentence vs. non-incarcerative punishment,
and second, the system may change the time served in prison for offenders sentenced to prison
(either by changing its sentence length policies or by changing its early release on parole
policies). The mass incarceration narrative does have a kernel of truth in that punishment for
some offenses has increased since 1960, and this increased punishment likely contributes to the
higher U.S. prison population today. Where the narrative misleads is in claiming that these
sentencing factors explain most of the incarceration increase and in claiming that these
punishment increases reflect unjustified punitiveness. In fact, sentencing factors likely explain
less than a fifth of the 300% per capita incarceration rate increase between 1960 and 2021, and
much of this sentencing-related increase would be seen as appropriate by most people,
regardless of political persuasion.
1. Changes in the Chance of Receiving a Prison Sentence
Perhaps the most important way sentencing changes could impact incarceration is if a
significantly higher percentage of similar defendants convicted of the same crime received
prison sentences than had in the past. Frustratingly, we are unaware of published national data
on this seemingly simple question before 1986. A 1991 study by Patrick Langan of the BJS was
also unable to obtain such data, noting that while the imprisonment rate per 100 arrests had
increased, this could be either a consequence of increased prosecution effectiveness (more
arrests being turned into convictions) or judges more frequently handing out prison
sentences.137 While the percentage of convicted defendants receiving prison sentences may
have increased somewhat over the 1960-1986 period, it would appear neither to be a massive
increase nor entirely tied to harsher sentencing laws as changes in the characteristics of
convicted defendants themselves might account for changes in their likelihood of receiving a
prison sentence (e.g., the pool of convicted defendants may have been increasingly tilted
toward more serious crimes or perhaps seriousness of criminal histories was increasing).138
27

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

Fortunately, good data exists from 1986 to 2006 for measuring the percentage of state
felony defendants who received prison sentences. (The BJS has not released national data since
2006, but there is little reason to think there has been any subsequent increase in the
percentage of convicted offenders sentenced to prison given the steady decline in incarceration
since 2009.) Given that the 1986-2006 period reflects the height of “tough on crime” sentencing
laws, one would expect a substantial rise across this period in the chance of a defendant
receiving a prison sentence. In fact, there was barely any change at all. In 1986, 46% of
convicted felony defendants in state courts received a prison sentence, 21% received a jail
sentence, and 33% received non-incarcerative punishments (usually probation).139 In 2006, 41%
of convicted felony defendants in state courts received prison sentences, 28% received jail
sentences, and 31% received non-incarcerative sentences.140 The chance of receiving a prison
sentence actually declined even as the prison population surged over this 20-year period. Even
considering all incarcerative sentences more broadly, there was no significant change in the use
of incarcerative vs. non-incarcerative sentences for convicted defendants. This suggests if
punitive sentencing policy significantly contributed to incarceration increases, it did not do so
through greatly increasing the chance of a convicted offender receiving a prison sentence.
2. Changes in Time Served in Prison for an Offense
Even if sentencing laws do not appear to have had a large effect on the chance of
receiving a prison sentence, perhaps they increased the length of prison sentences enough to
still greatly increase incarceration levels as the mass incarceration narrative assumes. Before
analyzing this claim, it is important to draw a distinction between sentence length and time
served. Any analysis that simply cites sentence length data is practically worthless as what
matters for both punishment and prison populations is the actual time served in prison for
sentenced offenders, not the sentences publicly imposed in court that are dramatically reduced
by early release on parole or “earned credit” schemes that give prisoners more than one day
credit for each day served. Overall, state prisoners on average serve only 46% of their maximum
sentence length.141
Time served data comes in two varieties: mean data and median data. Median data is
most representative of the punishment served by the “average” offender, while mean data is
sensitive to changes in punishment for offenders at the extreme end of the sentencing range.
Given that mandatory minimums and repeat offender laws only apply to part of the prison
population, it is reasonable to assume mean data would be most likely to confirm the mass
incarceration narrative.
Table 4 shows mean time served in state prison for a selection of years from 1923 to
142
2018. The results are striking. The average time served for all offenders in 2018 was 32
months—the same length as in 1953 long before mass incarceration was a concern. Moreover,
it is only 4 months longer than in 1960. Even as the prison population surged between 1960 and
1990, the mean time served among state prisoners did not change at all (staying at 28 months).
The supposedly draconian sentencing laws passed from the 1970s onward appear to have
barely made a dent in time served. (One explanation is that when faced with a mandatory
minimum sentence for one offense, defendants strategically plead guilty to an offense without a
mandatory minimum.143)
28

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

Table 4. Mean Time Served of State Prisoners Upon First Release in Months144
Year
1923145
1935(m)147
1945(m)
1953
1960
1986148
1990149
1992150
1996151
2000152
2002
2007
2018153

All
Offenses
27
22
39
32
28
33
28
N/A
30
30
29
29
32

All
Violent
N/A
N/A
N/A
N/A
N/A
N/A
46
43
45
56
48
50
58

Murder

Rape

Robbery

77146
54
117
116
121
N/A
92
71
95
158
93
143
214

42
40
52
51
45
66
62
65
66
88
80
92
115

37
43
70
49
42
57
48
44
46
55
54
56
58

Sex
Assault
26
28
39
38
N/A
36
35
45
58
45
49
60

Assault

Drug

Burglary

Fraud

26
20
34
28
25
41
30
29
33
38
31
32
30

11
14
23
23
31

26
20
45
30
25
31
29
N/A
31
29
31
27
27

21
14
29
23
17

Auto
Theft
N/A
18
31
24
21

20
N/A
22
17
17
17
17

20
N/A
24
17
21
17
15

20
N/A
24
23
22
21
21

There are, of course, exceptions for specific crimes. The mean time-served for certain
crimes of violence (murder, rape, sexual assault, and robbery) appears to have increased. Most
dramatically, the mean time served for rape jumped from 3.75 years to 9.58 years. (Many
people will see this as a desirable change). Mean time served for murder in 1960 was 10 years;
in 2018, it was 17.8 years. (Similarly, 10 years for intentionally taking another person’s life may
be seen as wholly inappropriate by many.) Robbery increased from 3.5 to 4.8 years. At least for
these crimes, the mass incarceration narrative has a kernel of truth as punishments really did
increase. But this does not show “mass incarceration” by a “carceral society” but rather reflects
societal changes regarding the seriousness of a short list of serious offenses. The question of
whether these changes were desirable or undesirable, just or unjust, is addressed in the next
subsection.
If mean time served data fails to confirm the mass incarceration narrative’s central claim
regarding the effect of punitive sentencing changes, median time-served data does much worse.
Table 5 presents median time-served data for state prisoners. The results are extraordinary
compared to the expectations one might reasonably have if one believed states moved to a far
more punitive sentencing policy. Median time served for all state prisoners in 1960 was 21
months compared to 15.6 months in 2018, a 25% decrease. Moreover, the data is remarkably
consistent across the entire period—there was no surge in median time-served during the
period of surging incarceration from the 1970s through the early 2000s.

29

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

Table 5. Median Time Served for State Prisoners Upon First Release in Months
Year

All

1926154
1930
1935
1940
1945
1953156
1960
1967
1970
1974
1978
1982
1986158
1993159
1997
2000
2005
2009
2018161

19
19
17
21
30
22
21
19
18
18
18
16
17
12
17
17
16
16
15.6

Murder

90
102155
89
121.3157

66
81

94160
210

Robbery

Rape

31
35
38
47
55
37
34
31
30
27
25
25
29
25
31
36
39
35
38.4

24
27
30
31
38
36
30
34
35
32
34
36
38
44
48
63
72
74
86.4

Agg
Assault
17
17
17
19
25
27
20
15
18
16
20
15
16
15
19
20
19
18
16.8162

Burglary

Larceny

20
21
19
21
30
24
20
17
16
16
15
14
15
13
19
19
16
15
17

17
18
14
15
21
18
17
13
13
14
13
10
11
9
13
12
12
11
11

Once again, there are exceptions for individual crimes. The median time served for rape
increased from 38 months in 1960 to 86.4 in 2018. Murder increased from 121.3 months in
1960 to 210 in 2018. However, these large changes in median do not appear for other violent
crimes as they did in the mean data. The change for robbery was a movement from 34 months
in 1960 to 38.4 months in 2018, and median time-served for assault decreased from 20 months
in 1960 to 16.8 in 2018.
Considered together, mean and median time-served data suggests those prisoners
receiving the harshest sentences have indeed experienced an increase in time served even as
most prisoners have not. But why? One contributing factor was likely increased statutory or
guideline sentences for serious offenders, an effect trumpeted by the mass incarceration
narrative, but there is an important confound in criminal histories as already noted above in the
section on non-sentencing factors. Even if the exact same guidelines and sentencing practices
from the 1960s were in effect in the 1990s or 2000s, criminals would receive longer sentences
because the criminals themselves had changed—i.e., they had been convicted of more previous
crimes meaning they would be sentenced as hardened recidivists. Recall that in 1990, 64% of
felony defendants in large urban counties had no prior felony conviction, but in 2009, only 40%
had no such conviction.163 Moreover, the proportion with 10 or more such convictions jumped
from 1% to 14% during that period. Such changes in criminal histories are bound to increase
time served and prison populations, but they are not the result of increased punitiveness.
Even if one ignores worsening criminal histories as part of the explanation for longer
sentences for serious crimes, the increased mean and median sentences for some serious
offenses hardly supports the core claim of the mass incarceration narrative that increased
punitiveness caused America’s incarceration rise. While these changes at the highest end of the
sentencing distribution likely contributed to America’s elevated prison population compared to
1960, their contribution appears minor compared to the significant contributions of nonsentencing factors, as discussed above.
30

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

4. How Unjust Were These Sentencing Changes?
There is no question that changes to sentencing law were made from the 1970s through
the 1990s and that these changes made some contribution to lengthening time served for at
least some offenders. But how should we view this change—as the sinister fruit of exceptional
and unjustified punitiveness? Or as a move toward more just sentencing for the most serious
offenses?
Consider the time served data. Were the changes in time served from the 1960s through
today largely unjust or undesirable? The answer appears to be no. The two crimes for which
punishment increased the most are rape and murder. As suggested above, it seems unlikely the
increase in time-served for rape suggests something problematic about current punishment
practices. In 1960, the median imprisoned rapist received less punishment (2.5 years) than the
median imprisoned robber (2.8 years). We are confident in saying few reasonable people in
America today would feel comfortable with such a punishment scheme in which seizing the
cash register from a store clerk is punished more harshly than forcibly raping her. The fact that
the median imprisoned rapist in 2018 spends 7.2 years behind bars while the robber spends
only 3.2 is testament to progress in America, and likely due in no small part to women claiming
more equal power in society. Would any person who truly cares about justice seek to reduce
incarceration in the U.S. at the price of returning the punishment for rape to what most women
(not to mention men) would rightly see as a slap on the wrist for such a serious crime? We think
not.
Similarly, it seems unlikely the community would see the 2018 mean punishment for
murder—17 years in prison—as an injustice. People would likely see a 10-year punishment for
intentionally killing another person (the mean in 1960) as inappropriately lenient, undervaluing
both human life and the suffering brought upon the victim’s family and community. Considering
the country still debates the question of the death penalty vs. life imprisonment, it seems
almost certain most Americans would prefer the 17-year punishment for willfully ending
another’s life.
The increase in time served for other crimes is much smaller, or even negative. It is
difficult to say what exact time served in prison people would find just for a given crime. There
is certainly variation, and mitigating or aggravating circumstances mean there is no single just
punishment for a general offense type. High-quality opinion research has not recently been
done on the topic (perhaps because progressive academic researchers fear the results may
show the public does not support their desire for laxer punishments). However, the U.S.
Sentencing Commission conducted a study in 1994 that attempted to determine representative
public opinion on how federal criminal cases should be sentenced. The study involved face-toface interviews and 42 vignettes describing various criminal cases.164 The results were striking,
as mean public recommended sentences were higher than guidelines sentences for almost all
crimes, and median public recommended sentences were also higher than guideline sentences
for most crimes. The mean public recommended sentence for street robbery was 9.2 years, for
example.165 The public would likely have been disappointed to learn the mean time served for
robbers in 1996 was 3.8 years and has only risen to 4.8 years in 2018. Unfortunately, the study

31

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

did not ask for recommended murder or rape sentences, but it is virtually certain both would
have been much longer than actual time served.
Of course, the American public may have become less punitive over time as total crime
has fallen from its peak in the 1990s. However, it is unlikely their perceptions of a just sentence
have shifted so dramatically that current time served for most crimes would appear to be an
injustice—if anything, it is far more likely, if they were shown the time served data, most
Americans would feel many criminals are getting unjustly lenient sentences.
One of the only recent data points is a 2021 poll which found that 28% of the public
believed convicted criminals spent too much time in prison while 32% believed they did not
spend enough, and 37% believed they spent “about the right amount of time.”166 Unfortunately,
such a question is so unnuanced as to be practically meaningless. For one, the public’s view on
the question is shaped by perceptions of how much time prisoners spend behind bars and not
on realities. Without showing respondents actual time served data, it would be impossible to
get a reliable assessment of their view of current practice.
Indeed, given the drumbeat of “mass incarceration,” one might expect that at least some
significant portion of the population (those exposed to the false narrative) have a grossly
exaggerated view of the system’s current incarceration practices. A person hearing the generally
accepted narrative would believe the system routinely throws first-time, non-violent offenders
behind bars for many years—which would be more than enough reason to say prisoners spend
too much time behind bars. One could also imagine that a person on the political right might
answer “not enough time” simply from watching politicized coverage about “soft on crime”
policies in left-leaning jurisdictions. Additionally, people’s judgements are likely to differ
depending on the crime. For example, it is almost certain that what the public perceives as a
just sentence for drug dealing or possession has shifted in the direction of leniency (or even
outright legalization) since the ‘90s. By contrast, this is likely not the case for the public’s
perception of what is a just punishment for murder or rape.
Our point is not to argue the American public would demand longer incarceration terms
for many prisoners if they were aware of time served data (though that may be the case), but
rather to show that most of the post-1960 increases in time-served punishments are unlikely to
be seen as an injustice by society. A rise in the per capita incarceration rate caused by an
increase in time-served punishments that bring prison terms closer to what the public views as
just deserts is not a sign of a punitive overreaction—it is simply a sign of the justice system
catching up to where it should have been all along.167
It seems likely that many promoters of the mass incarceration narrative fully understand
that most American punishment practices are not unjust in the eyes of society. That,
presumably, is why they blame “American public punitiveness” for the incarceration increases.
Ironically, as shown previously, public desires for longer sentences—no matter how strong they
may have been—actually had little effect on incarceration, as most of the incarceration increase
was caused by non-sentencing factors. But what public opinion does indicate is that reforms
designed to reduce the level of punishment across most crimes are unlikely to gain widespread
support. This is why, in our view (see Part III), working to change the form of punishment, as
opposed to the level of punishment, is a more useful reform program.
None of this is to say that changes to sentencing policy post-1960 did not result in many
unjustly harsh sentences in absolute terms or that there is no room for reform in the direction
32

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

of leniency. Even if we acknowledge, as seems likely, that many mandatory minimum sentencing
laws brought time served punishments closer in line with public views of justice, this does not
mean such laws are ideal or do not cause serious injustices. For example, three-strikes laws and
other mandatory minimum sentencing for repeat offenders may result in many offenders
getting their “just deserts,” which they would not have previously received, but it also means
that there will be some William Rummels who receive life sentences for stealing $230.168 The
use of mandatory minimums short-circuits the important assessment of each defendant’s moral
blameworthiness.
Additionally, it is undeniable that some sentencing changes in the post-1960 period were
driven by utilitarian crime-control thinking that sought to reduce crime through sheer
deterrence instead of providing a just punishment (with deterrence as a fruitful byproduct). This
too routinely generated sentences unproportional to a careful analysis of the offender’s
blameworthiness. (More on this in the next subsection.)
In other words, we share many common criticisms of the system’s punishment practices
over the last several decades, but none of those reservations take away from the falsity of the
standard mass incarceration narrative. The prevailing narrative of punitive changes to
sentencing policies resulting in massive increases in time served and surging prison populations
is wildly misleading at best.
5. Who Bears Responsibility for Punitive Excesses in Punishment Post-1960?
To the extent that some changes in sentencing practices in the post-1960 period
produced excessive sentences—which we do believe is the case—who should bear the blame?
Does primary responsibility lie with an “exceptionally punitive” American public, as the mass
incarceration narrative suggests? This is certainly the prevailing academic narrative, but the
truth is more complicated and much more damning to academics. Policymakers adopted and
the public accepted a number of explicitly utilitarian crime-control sentencing laws in the post1960 period. These laws produced some sentences that clearly and regularly conflicted with
desert (such as mandatory minimums for drug crimes, which could lead to certain drug dealers
or possessors spending more time in prison than some rapists or murderers). But why did
policymakers pursue such utilitarian solutions in the face of rising crime? The fundamental
reason is because academics in the first half of the 20th century successfully unmoored the
criminal justice system from the concept of “just deserts” in favor of utilitarian crime-control
models that shifted the criminal justice system to distributive principles for punishment based
upon general deterrence, incapacitation of the dangerous, or rehabilitation.
Academics pushed a “scientific” utilitarian-based system of punishment that sought to
free society from the “barbaric” notion of retributivist just deserts.169 This meant the legal
system could give grossly inadequate punishment under a theory of utilitarian rehabilitation, as
with rapist-turned-serial-killer William Bonin, who was let loose to kill after spending only five
years in prison for more than a half-dozen brutal rapes of young men and boys.170 The utilitarian
perspective also justified life imprisonment for William Rummel for a minor fraud offense under
a “three-strikes” law because, as a repeat offender, he needed to be incarcerated for as long as
he was dangerous, despite the fact that a life sentence was grossly disproportionate to his moral
blameworthiness.171 Most academics were unwilling to face up to the truth that their support of
33

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

such distributive principles of “coercive crime-control” left the criminal “justice” system’s doors
wide open to terrible injustices and failures of justice. The “barbaric” notion of just deserts that
they derided was the only principle capable of defending equally from excesses of leniency and
harshness.
Of course, there is plenty of blame to go around for the unjustly harsh utilitarian
punishments of the ‘War on Drugs’ and the ‘War on Crime’ more broadly, but it is only fair that
the lion’s share goes to the academics and experts who first suggested “justice” had no place in
the justice system. Far from being a lesson in exceptional and unjustified American punitiveness,
the passage of unjustly harsh utilitarian punishments should more rightly be seen as a lesson in
the folly of letting misguided academic ideas corrupt the justice system by disconnecting it from
the community’s shared notions of justice. Today’s anti-punishment activists in academia should
take note to avoid repeating the mistakes of the past.

D. American vs. Foreign Countries’ Prison Populations: A Problem on Which
Side of the Ocean?
The mass incarceration narrative’s central myth of exceptional and unjustified American
punitiveness does not simply claim American criminal punishments are unjustly punitive but
also that they are exceptionally deplorable compared to other more enlightened countries.
America’s per capita incarceration rate is routinely criticized by comparing it to other countries,
but what do such comparisons actually reveal? There is certainly some truth to the mass
incarceration narrative’s comparisons: America does appear to use prison as a punishment
more often than other Western democracies, and America should take note of the wider use of
non-incarcerative punishments in some foreign countries. However, the mass incarceration
narrative rarely pauses to consider the confounding factors that make prison population
comparisons difficult, and it never stops to question whether America’s higher per capita
incarceration rate compared to many other countries reveals a problem in foreign countries
instead of America’s justice system. Such motivated comparisons often fail to see how flawed
foreign justice systems are and how much they may have to learn from the U.S.
1. The Claims of American vs. Foreign Countries’ Incarceration Practices
The fact that America has such a high prison population is cited as a compelling
justification for reform by mass incarceration proponents: “By virtually every measure, the
United States incarcerates more of its people than any other nation in the world.”172 Similarly:
Not only does the U.S. have the highest incarceration rate in the world; every
single U.S. state incarcerates more people per capita than virtually any independent
democracy on earth. To be sure, states like New York and Massachusetts appear
progressive in their incarceration rates compared to states like Louisiana, but compared
to the rest of the world, every U.S. state relies too heavily on prisons and jails to respond
to crime.173

34

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

Some mass incarceration researchers see America’s exceptionally high prison populations as a
legacy of slavery and a general barbarity in America as compared to Europe. Consider James
Cullen writing for the Brennan Center:
Of course, de Tocqueville also saw much to criticize in the young United States, including
its commitment to slavery. That legacy continues to haunt the country today, even as
most of the world has adopted punishment systems more in line with what de
Tocqueville hoped to find. Today, the U.S. incarceration rate is nine times higher than
Germany, eight times higher than Italy, five times higher than the U.K., and 15 times
higher than Japan.174
The mass incarceration narrative views America’s high prison population as resulting from the
punitiveness of the American public, and crime rate differences are only mentioned to refute a
potential difficulty in making comparisons.
The United States actually has a crime rate that is lower than the international norm, yet
our incarceration rate is six to 10 times higher than other countries’ around the world.
It’s not crime that makes us more punitive in the United States. It’s the way we respond
to crime and how we view those people who have been labeled criminals.175
These claims do capture a general truth: America does have one of the highest absolute and per
capita prison populations in the world (although it has neither the highest absolute nor the
highest per capita, contrary to widely stated claims).176 Table 6 constructed with data from the
World Prison Brief shows America’s per capita prison and jail (including pretrial prisoners)
population vs. a sample of other countries in recent years. While the inclusion of jailed
offenders inflates the number, the basic picture remains the same even if only offenders
sentenced to prison are examined.177

35

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

Table 6. Per Capita Incarceration Rates in Different Countries
Country178
Incarceration
Rate/100k
residents
El Salvador
1086
Cuba
794
Turkmenistan
576
United States
531
Argentina
254
Poland
201
United Kingdom 146
Italy
102
Canada
88
Sweden
82
Norway
54
But is America’s higher per capita incarceration rate due to exceptional and unjustly punitive
laws or are there other factors at play?
2. Important Factors: Crime, Criminal Justice Effectiveness, Institutionalization,
and Criminal Histories
While American sentencing policies do substantially contribute to America’s higher per
capita incarceration rate compared to many foreign countries, other factors should also be
remembered when making international comparisons.
Crime Rate

The most obvious complicating factor is differing crime rates. Countries with lower per
capita crime rates should have lower per capita incarceration rates, all else equal (of course, all
else is not equal in the real world). For example, America has a homicide rate 6.2 times that of
the European Union, a fact which may contribute to America’s higher rate of prisoners serving
LWOP and other extra-long sentences.179 But contrary to the assumptions of some mass
incarceration critics, crime rates actually do not provide an explanation of America’s higher per
capita incarceration rate compared to Europe because, with the exception of homicide, America
enjoys lower violent and total crime rates than Europe.180 This fact often goes unrecognized in
public perceptions of safety (likely because homicide levels are used—incorrectly—as a heuristic
for total crime). As one study finds:
In 1970 the aggregate crime rate in the seven European countries we consider was 63%
of the corresponding U.S. figure, but by 2007 it was 85% higher than in the United
States. This striking reversal results from a steady increase in the total crime rate in
Europe during the last 40 years, and the decline in the U.S. rate after 1990.181

36

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

But though crime rates do not explain America’s larger prison population compared to
Europe, America’s elevated incarceration levels compared to Europe do help explain why
America enjoyed falling crime rates even as Europe suffered rising ones. As the same study
finds: “back-of-the-envelope calculations based on our estimates indicates that the different
dynamics of the prison populations in Europe and the United States explain 17% of the reversal
of misfortunes for total crime, 33% for property crimes, and 11% for violent crimes.”182
In other words, if American incarceration policy mimicked Europe’s, America would likely
have substantially more crime. Conversely, if Europe had American incarceration policy, it would
likely have significantly less crime. These facts are completely ignored by mass incarceration
activists who often try to paint Europe as a paradisal state of low incarceration and low crime.
To be clear, we do not seek to justify America’s higher incarceration levels on utilitarian grounds
of crime control, but we do feel it is only fair to acknowledge there is some level of tradeoff
between crime and incarceration levels.
Criminal Justice System Effectiveness

But while crime rates alone do not explain much of America’s higher per capita
incarceration rate compared to many countries, criminal justice system effectiveness may. The
mass incarceration narrative sometimes compares America’s incarceration rate with developing
countries around the world with significantly higher crime and lower incarceration. But this is
enormously deceptive. The fact that it may be easier to get away with crimes in other parts of
the world does not mean those countries’ low incarceration levels should be praised. For
example, Mexico’s per capita incarceration rate is less than a third of America’s, but that is
largely because of the ineffectiveness of the Mexican justice system.183 93% of crime goes
unreported in Mexico, and the chance of conviction in case of arrest is also tiny, leading to a
practically non-existent chance of punishment.184 By contrast, over 40% of violent victimizations
and over 30% of property victimizations are reported in the US, and conviction and punishment
rates are also significantly higher than in Mexico.185 Many developing countries like Mexico have
incarceration levels wildly below what they should have given their crime rates. Only a lack of
justice system effectiveness prevents such countries from having higher per capita incarceration
rates than the US. While the U.S. obviously has a more effective justice system than many
developing countries, and this explains much of the U.S.’s higher per capita incarceration rate
compared to them, the mass incarceration narrative also points to European countries with
more developed justice systems.
Unfortunately, precisely comparing the effectiveness of criminal justice systems across
the world, even among developed countries, is difficult—and sometimes impossible—due to
differences in what data is collected and published. Any researcher would desire all countries to
publish victimization surveys, total reported crimes, arrests, convictions, and imprisonments
(ideally with subcategories by offense), but governments around the world (including in the
U.S.) routinely hide or refuse to collect such statistics. But what data exists suggests the U.S.
may have a more effective justice system than at least some European countries. Table 7 shows
data from a BJS report comparing criminal justice systems in the 1990s. Interestingly, the US
seemed to have a more effective justice system than England for most crimes, and a more
effective system than the Netherlands and Sweden for at least some crimes. While the data is
old, the trend in the U.S. since the 1990s has been lower crime and higher justice system
37

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

effectiveness, while the trend in many European countries has been higher crime, making it
even more likely America’s justice system is significantly more effective than at least some
European countries today.186 But without better data and further research, it is hard to quantify
how much this factor affects relative incarceration levels.
Table 7. Convictions per Thousand Recorded Offenders
Convictions per
1000 Recorded
Offenders187
Homicide
Rape
Assault
Robbery
Vehicle Theft
Burglary

U.S.
(1996)

England and
Wales (1995)

Netherlands
(1995)

Sweden
(1995)

554
155
34
24
13
16

555.22
99.7
14.44
6.12
11.66
5.78

455
190
46.6
19.32
112
11.4

673.7
86.1
22.7
65.2
7.6
7.2

Civil Commitment of the Mentally Ill; Criminal Histories

Another overlooked factor that may contribute to prison population differences is
differing rates of institutionalization among the mentally ill and differences in whether mentally
ill offenders are sent to prisons or non-prison mental institutions. It is possible some foreign
countries hide portions of their incarcerated population in non-prison mental institutions.188 We
make no claims that they do, but merely note this is yet another factor that must be explored
when making comparisons between countries.
One potentially important factor completely ignored in international incarceration
comparisons is differing average criminal histories between countries. If one country’s offenders
have more serious criminal histories when sentenced than another’s, one would expect the
country with the more serious offender histories to have a higher per capita incarceration rate,
all else held equal. While data is hard to find, the higher rate of recidivism in the U.S. compared
to some foreign countries suggests the average U.S. offender may have a more severe criminal
history, at least compared to those countries with lower recidivism.189
All these factors are important to consider, but they do not prevent all useful
comparisons of incarceration policies. Nor do they invalidate the claim that many countries have
lower per capita incarceration rates than the U.S. due to different sentencing policies.
3. How Exceptional Are American Sentencing Policies?
How exceptional are American sentencing policies compared to other countries? This
question has two components: How does the chance of a convicted offender receiving a prison
sentence compare? And for those sentenced to prison, how does the time served in prison
compare? Existing data is unideal for answering these questions, but at least some conclusions
can be drawn. As a general statement, the claim that America gives out a higher percentage of
prison sentences and that those prison sentences are longer in time served does appear to be
true when compared to most other Western countries. There are important exceptions for
38

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

certain crimes and certain countries, but while it is unnuanced, the generalization is not false.
However, the mass incarceration narrative often exaggerates the difference with other countries
(and ignores the possible contributions of the many non-sentencing factors discussed in section
II.B.). It also fails to ask whether it might be foreign countries that have problematic punishment
practices.
Chance of Convicted Offenders Receiving a Prison Sentence

Table 8 shows data on the percentage of convicted offenders receiving a custodial
sentence (including jail sentences) in America, England and Wales, Australia, the Netherlands,
and Sweden in 1995/96. More recent data is not readily available, but since America’s
percentage of convicted offenders receiving incarceration has stayed largely the same since the
1990s, and since there have been no massive changes to European incarceration policies (to our
knowledge), the comparison is likely still approximately accurate today.
Table 8. Percentage Receiving Custodial Sentences in Different Countries
% of convicted
U.S.
England and
Australia Netherlands Sweden
receiving custodial (1996) Wales (1995)
(1995)
(1995)
(1995)
sentence190
Homicide
94.5
94.3
96.0
92.0
96.75
191
Rape
75.6
94.5
41.0
63.0
91.67
192
Assault
59.3
27.0
6.0
11.12
29.56
Robbery
75.0
66.8
53.0
70.0
61.17
Vehicle Theft
54.7
29.6
17.0
44.0
24.39
Burglary
54.9
49.9
19.0
66.0
47.54
When it comes to the percentage of convicted offenders incarcerated for a specific
crime, America is unexceptional for certain crimes but an outlier for others. When it comes to
homicide and rape, and to a lesser extent burglary and robbery, America is within the normal
variation among the countries surveyed, a fact seemingly inconsistent with America running an
out-of-control “carceral state.” But America does incarcerate a higher percentage of convicted
assault offenders (59.3%) compared to Sweden (29.5%), the next highest country in the
comparison. It also incarcerates a higher percentage of convicted vehicle theft offenders (54.7%)
compared to the Netherlands (44%), the next highest country in the comparison.
Time Served in America vs. Other Democracies

The mass incarceration narrative often compares average sentences imposed in America
with other countries to prove how much more punitive America is in sentencing. However, it is
time served data that is meaningful. Comparing sentences imposed, which are rarely if ever fully
served, is pointless. Different countries have dramatically different early release policies that
can drastically change the actual punishment meted out to offenders.
While U.S. time served data from 2018 exists, European data is more difficult to come
by. For example, a report on mass incarceration published in December 2022, “Long Sentences:
An International Perspective,” is forced to rely on time served data for comparative purposes
from around 2000.193 Table 9 shows sentences and time served data for different crimes in
39

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

America and several comparison countries for the years 1995/96 (we use the same years as in
Table 8 to provide a consistent snapshot of incarceration policy—slightly later data reveals the
same general picture). The data shows that while the U.S. indeed had significantly longer
sentences than any of the other studied countries, that is not the case with actual time served,
at least for many comparisons of specific crimes in specific countries. For example, time served
for murder in America was comparable to England and Wales, below Australia, and higher than
the Netherlands and Sweden. The situation for rape is similar: U.S. time served is comparable to
Australia, slightly higher than England and Wales, and noticeably more than the Netherlands
and Sweden. (But, as discussed in the next section, many will think this represents inappropriate
rape sentences in Europe.) When it comes to other crimes such as assault and burglary, America
is comparable to Australia, but does appear to have longer time served than most countries,
although a difference in offenders’ criminal histories might be a contributing factor. While these
results place America on the upper-end of the time served range, they also reveal the
unnuanced nature of the mass incarceration narrative’s foreign comparisons.
Table 9. Sentences and Time Served in Different Countries for Specific Offenses
Sentences & time served (in
months) by offense &
country194
Homicide sentence
Homicide time served
Rape sentence
Rape time served
Assault sentence
Assault time served
Robbery sentence
Robbery time served
Vehicle theft sentence
Vehicle theft time served
Burglary sentence
Burglary time served

America
(1996)

England and
Wales (1995)

250.0
126.2
115.5
59.2
40.4
21.6
76.4
37.4
20.7
10.1
35.0
15.2

229.9
99.8
77.0
44.2
13.7
6.1
40.3
20.5
8.6
3.4
14.9
7

Australia Netherlands
(1995)
(1995)
171.5
129.4
78.2
57.3
34.5195
27.0
73.7
32.8
27.6
8.9
31.0
16.3

109.7
73.1
26.7
19
6.2
6.1
17.1
14.3
8.5
8.1
12.7
11.4

Sweden
(1995)
93.17
46.65
30.8
15.66
5.24
3.35
29.78
15.31
4.14
2.93
11.03
6.58

4. Are the Lenient Sentencing Practices of Foreign Countries Desirable?
If the justice system’s goal is to deliver justice based upon a nuanced assessment of each
offender’s conduct and circumstances, one may wonder whether a thoughtful liberal society
should find the sentencing practices of many foreign countries desirable or even acceptable.
The answer would appear to be no. For example, who in America would consider Sweden’s time
served punishment for murder of less than 4 years acceptable? Should the average punishment
for rape really be 1.5 years spent in prison as in the Netherlands? Should 59% of convicted
rapists avoid prison time as in Australia? It is fair to say that the average member of the
American public across the political spectrum would be shocked by many if not most of the
punishment practices in Table 9, considering them instead to be gross failures of justice.196
Nor do these lenient punishments seem unjust to only the supposedly “exceptionally
punitive” Americans. Despite Europe being held up as a model by the mass incarceration
40

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

narrative, large public majorities in many European countries are dissatisfied with how lenient
sentencing is in their criminal justice systems. For example, a 2023 survey by the House of
Commons Justice Committee found that 71% of respondents in England and Wales thought
criminal sentencing was too lenient, including 38% who thought it was “much too lenient.” By
contrast, only 3% thought sentencing was too tough.197 A study published in 2010 measuring
Swedish public opinion found that 69% of respondents believed sanctions were “too mild” and
only 2% thought they were too severe. A Swedish focus group study came to almost identical
conclusions (67% believing punishments to be too mild). 59% of telephone respondents and
66% of focus group participants were in favor of longer prison sentences.198 A 2022 poll showed
that 65% of French respondents believed criminal punishments were not strict enough, with
majorities from the political right and left wanting stricter punishments.199 A Eurobarometer
survey in 2015 found that 58% of respondents across the EU “totally agreed” with the
statement “criminals should be punished more severely” and another 28% “tended to agree,”
meaning an overwhelming 86% of the European public wanted tougher criminal
punishments.200 As a result, there have been moves to toughen punishments and to build more
prisons in some European countries.201
Clearly, European publics believe something is not right in the supposedly paradisal state
of European incarceration policy. The mass incarceration narrative completely ignores the
failures of foreign, particularly European, sentencing systems. Sentencing policy in Europe is
neither popular nor democratic. It also appears to have done a relatively poor job at dealing
with Europe’s historical crime problems. It is worth asking: if Europeans do not want their own
sentencing policies, should Americans want them? However, the mass incarceration narrative is
not generally interested in providing a punishment policy the public finds acceptable, and so
Europe, a place where anti-democratic sentencing policies have led to chronic public
dissatisfaction, is held up as the goal American sentencing should strive to reach. The facts
suggest that the dispute over incarceration practices is more a dispute between the elites and
the community than a dispute between the U.S. and other democracies.
Recognizing the serious flaws in foreign punishment practices does not mean there is
nothing worth emulating about foreign sentencing or incarceration policies. In fact, the more
creative and widespread use of non-incarcerative sanctions in Europe and other parts of the
world provide a fruitful starting point for considering how the U.S. might find just punishment
alternatives to prison. As the next section shows, we believe America has significant room to
reduce its prison population. The crucial point, however, is that the resulting sentences must be
just, and recognized by the public as such, instead of being seen as acceptable only to a group of
rarified experts and policymakers, many of whom reject the notion of punishment altogether.

III. Aiming Reform at Real Incarceration Problems Rather Than Myths
Some might be inclined to argue the previous sections show American incarceration
policies are in little need of reform and that proponents of lowering prison populations are
utterly wrong. After all, if non-sentencing factors largely explain the rise of America’s per capita
incarceration rate, is there any need to change sentencing and incarceration policies? We
believe this view is mistaken. For all its myths, the mass incarceration narrative is correct that
American incarceration policy is long overdue for an overhaul. Correcting the myths of the mass
41

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

incarceration narrative is a necessary first step in clarifying what parts of incarceration policy
need reform and why, not an excuse for endorsing the status quo.
The central mistake of the mass incarceration narrative is assuming that high prison
populations are a result of the American public’s exceptionally and unjustifiably punitive sense
of justice. If this were true, the natural remedy would be to recommend repressing public
intuitions of justice in favor of reducing prison populations no matter the cost or consequences.
The only constraint acknowledged by the more reasonable wing of the mass incarceration
movement is public safety, which is de facto assumed to be the only legitimate reason why
punishment would be enacted. For example, a Brennan Center report finds that nearly 40% of
the U.S. prison population could be safely released in the interests of ending mass incarceration.
In describing its plan to reduce incarceration and sentence lengths, the report explains: “This
approach is grounded in the premise that the first principle of 21st century sentencing should
be to protect public safety.”202
At first glance, this may sound attractive. But we think it is clearly wrong, for both moral
reasons and effective crime-control reasons. The first principle of any society’s sentencing policy
should be to deliver a just punishment proportional to the offender’s blameworthiness. A
murderer should not escape any punishment simply because he is extremely unlikely to harm
someone again (as the Brennan Center would have it). A shoplifter should not be sentenced to
life in prison because the evidence suggests that he is likely to become a serial killer. Sentencing
policy should not be the playground of social policy, no matter how noble the goals. We have
written at length elsewhere on the importance of the justice system having its first goal being
the doing of justice, operationally defined in terms of delivering punishments based upon a
community’s shared intuitions of justice.203 While a just sentence may serve many goals—
deterrence, incapacitation, or rehabilitation—it should be imposed because it is a just sentence
and not because of its coercive utilitarian benefits. Any attempt to pursue those coercive
utilitarian goals in a way that conflicts with shared community intuitions of justice will
undermine the criminal law’s moral credibility with the community and thereby undermine its
capacity to harness the powerful forces of social and normative influence.204 Thus, even those
who are unconcerned about doing justice and focused strictly on public safety should care
about avoiding serious conflict with the community’s shared intuitions of justice.
Since America’s public does not find most current criminal punishments unjust, this
means that “soft on crime” policies of simply reducing punishment levels generally are not
productive long-term solutions. But the overriding commitment to doing justice, nothing more,
nothing less, opens wide reform possibilities, including a chance to convince the public—
regardless of political persuasion—to support meaningful reform. This is because the public can
support the use of just non-incarcerative sentences for many offenders who would currently
receive prison sentences. We strongly support reducing the use of incarceration where
consonant with desert, and we suspect the American public does too. What we oppose—and
what the American public will always react against—is an attempt to slash incarceration at the
price of foregoing just punishment.
With justice as the priority, there are several promising avenues for reforming
incarceration. First, policymakers must ensure criminal codes accurately reflect the community’s
belief of what is and is not criminally condemnable behavior. Second, policymakers should
abolish unnuanced sentencing policies such as mandatory minimums and three strikes (typically
42

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

passed on utilitarian grounds of coercive crime control) in favor of liability and punishment rules
that track people’s nuanced judgments of justice. Third, policymakers should explore the wider
use of non-incarcerative sanctions that still satisfy public demands for just punishment. Fourth,
policymakers should focus on experimenting to make the prison environment genuinely
rehabilitative.

A. Decriminalizing or Reducing Penalties as Needed to Track Shifting
Community Views
Much, if not most, of the persuasive force behind the mass incarceration narrative
comes from public discontent with the ‘War on Drugs.’ A 2021 poll found that more than 83% of
Americans believe the ‘War on Drugs' is a failure, and 66% support some legalization of
drugs.205 A clear majority of the public supports the legalization of marijuana for medical and
recreational purposes.206 And while a majority of the public may not support legalizing all ‘hard’
drugs, a majority does appear to view the possession of such drugs as constituting no more
than a civil offense on par with a traffic ticket as opposed to a felony.207 It is fair to say the public
views the possession, and perhaps even the sale, of many illicit drugs as far less serious than
current criminal laws suggest. As a result, opposition to America’s drug laws serves as a catalyst
for the mass incarceration narrative, which paints America’s high prison population as resulting
from a mass of drug possessors behind bars. As previously discussed, this myth of prison
composition is false, as drug offenders make up only 15% of the prison population, and most of
these offenders are traffickers and commonly repeat offenders. If America truly wants to
eliminate its prison population of drug offenders, it will have to change the way dealers are
punished, a question on which there has been much less public opinion research.
This is not an article on drug policy, but justice and common sense require lawmakers to
revise criminalization that conflicts with prevailing societal views of what is sufficiently
condemnable conduct to justify criminalization. Public concerns should not be addressed with
haphazard attempts at prosecutorial decriminalization aimed at reducing prison populations.
The worst of all worlds is a law that says one thing (mandating strict punishments for drug
offenses) and a legal system that says another (prosecutors refusing to prosecute drug
crimes).208 Such a disconnect erodes the moral credibility of the law even faster and fuels the
crime economy by encouraging a black market trade whose profits end up in the hands of
gangsters often involved in broader crime. Laws and punishments against drugs should be based
on societal views of what is just, not concerns about prison populations. More broadly,
lawmakers should conduct regular reviews of the entire criminal code to make sure it aligns
with public views and does not lead to what most in society would see as unjust incarceration.

B. Abolishing Unnuanced Punishment Policies in Favor of Rules That Capture
People’s Nuanced Judgments of Justice
In addition to revising criminalization to match public views, policymakers must also
make sure statutory punishments reflect nuanced public views of desert. The mass
incarceration narrative draws force from anecdotes of horribly unjust punishments meted out
43

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

by mandatory minimum sentencing laws in three-strike statutes that ignore factors that can be
highly relevant to an offender’s blameworthiness. As we have argued elsewhere, we support
efforts to repeal such clumsy punishment provisions in favor of nuanced and binding sentencing
guidelines, preferably guidelines that preserve judicial sentencing discretion in unusual cases
(subject to appellate review).209 Such reforms fortunately have already made some progress.
Between 2000 and 2014, at least 29 states and the federal government have moved to modify
or eliminate some of their mandatory minimum sentencing laws.210 For example, California
changed its three strikes laws so as to avoid application to non-violent, non-serious offenses.211
We support reforms that continue this trend. However, reformers should recognize the need to
update statutory punishments and guidelines in both directions of punishment. Overly lenient
punishments can also be unjust, and ignoring their presence in the justice system invites a
punitive backlash. Many of the mandatory minimums passed in the 1990s might never have
been proposed if previous policymakers had made sure punishments reflected shared
community intuitions of justice.

C. Employing More Non-Incarcerative Punishments Consonant with Desert
The reform that could perhaps do the most to reduce America’s prison population is to
replace prison sentences (or portions of prison sentences) with non-incarcerative punishments
that would still do justice as the community sees it. Using more non-incarcerative sentences
should be attractive to policymakers across the political spectrum, and supporters of the mass
incarceration narrative should fully endorse attempting to turn many prison sentences into nonprison alternatives. What differentiates our proposal from those of some existing anti-prison
reformers is that we call for research and tailoring of non-incarcerative sentences to public
views of what amounts to a just punishment, not simply abolishing the punishment as many
mass incarceration reformers would urge.
A creative approach to non-incarcerative sanctions can produce a wide range that can
legitimately be said to provide sufficient punitive severity to substitute for incarceration. Some
non-incarcerative sanctions include home-confinement, Intensive Supervision Programs (ISPs),
weekends in local jail, community service, restrictions on travel, and day fines. Each nonincarcerative sanction may have a slightly different punitive effect, but it is easy enough for
social scientists to reliably assess the different amount of “punishment credit” that an offender
should receive for each kind of non-incarcerative sanction. In other words, one can satisfy the
requirements of just punishment without prison. Studies of lay people show that there is an
intuitive agreement that the right combination of non-incarcerative sanctions can equal the
punitive “bite” of many prison sentences.212 For example, one study found that respondents
perceived a (2023 inflation adjusted) $50,000 fine as being more punitive than a one-year
prison sentence (for certain offenders). Meanwhile, weekends in jail, ISPs, or home confinement
for two years were seen as more punitive than six months in prison. These findings show it is
possible to construct scalable non-incarcerative punishments that would still be seen by the
community as doing justice.213 Implementing such reforms would allow devoting greater
resources to prosecuting and punishing more serious crimes, such as murder or rape, while
allowing many offenders a better chance at societal reintegration than prison would provide.

44

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

Consider a sample range of possible non-incarcerative sanctions, many of which are
already used (including use in other countries):
1. Verbal sanctions, such as public admonitions, reprimands, warnings or
unconditional discharges accompanied by a formal or informal verbal sanction.
2. Conditional discharges (that set out a series of restrictions on the offender
post release, enforced by the threat of reincarceration upon a violation).
3. Status penalties that deny the offender specified rights in the community. Such
a penalty might, for example, prevent someone convicted of fraud from holding a
position of trust as a lawyer or director of a company.
4. Fines are among the most common and effective alternatives to keeping
offenders out of prison.
5. Asset forfeiture in cases where the court has evidence showing that money
found in the possession of the offender is the product of the crime.
6. Restitution to the victim.
7. Community service, which can involve a wide range of required activities.
8. Government work requirements, which would require offenders to engage in
certain work for the government for a certain period of time, such as work on state park
maintenance crews or the like, but without incarceration.
9. Participation in a treatment or training program.
10. Referral to an attendance center, a facility where the offender spends the day,
returning home in the evenings. Attendance centers, also known as day reporting
centers, may provide a centralized location for a host of therapeutic interventions,
training programs, or drug treatment.
11. House arrest.
12. Location monitoring through GPS tracking, sometimes combined with travel
restrictions, such as allowing an offender to be only at a list of locations or traveling
between them.
13. Location monitoring (without location restriction) and contact availability
requirements, which means the offender’s location would be tracked and recorded at all
times and the offender would be obliged to answer his government-issued phone at any
time.
One can also imagine a variety of other possibilities. For many of these and other options, one
could imagine offering offenders one or many of these as an alternative to incarceration that
they could accept or reject. (And those selection decisions might help over time to produce
more accurate equivalency measures among the punishment method alternatives.)
Some possibilities, such as GPS monitoring, have also been shown to reduce recidivism
among likely reoffenders, showing that some non-incarcerative punishments may also preserve
public safety even for moderately risky offenders.214 Given the possibility of altering the length
and intrusiveness of each of these sanctions, such a rich selection of possibilities makes it
possible to construct a non-incarcerative sentence that matches the punitive bite of many
incarcerative sentences.
45

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

Some countries, such a Japan, are well ahead of the U.S. in their reliance on nonincarcerative sanctions. Japan has an incarceration rate 17 times smaller than the U.S. (though
much of this is due to Japan’s lower crime rates and other factors).215 Japan codified its use of
non-incarcerative sanctions in what is now called the Tokyo Rules.216 The rules state that “The
selection of non-custodial measures shall be based on an assessment of established criteria in
respect of both the nature and gravity of the offence and the personality, the background of the
offender, the purposes of sentencing and the rights of victims.”217 Such guidelines are meant to
restrict the use of non-incarcerative punishments to those cases in which the public finds such
sanctions appropriate and satisfactory to justice.
One limitation of non-incarcerative sanctions is their ability to preserve public safety, an
essential factor to consider when choosing a punishment method. This problem can be
addressed by introducing this non-incarcerative sentencing scheme slowly over increasingly
broad categories of offenders and monitoring the recidivism rate to determine which offenders
should be given which non-prison alternatives to guarantee an acceptably low recidivism rate.
Where probation type monitoring may be enough for some offenders, house arrest, GPS
tracking, and phone or video surveillance might be required for other higher-risk felons.218
While such a scheme would likely take decades to fully implement (and fine-tune based on
recidivism data), the possible reduction in the prison population is significant—perhaps
eventually as much as 50%. Importantly, reducing incarceration through using non-incarcerative
sanctions should not be rushed by ignoring public concerns over justice and safety.
We believe much more can be done in the U.S. to create non-incarcerative sentences
that are sufficiently punitive to justly substitute for many prison sentences. Of course, to the
extent mass incarceration activists are not really interested in reducing incarceration, but rather
eliminating punishment, none of these reforms will be of interest. Thus, a person’s willingness
to pursue higher levels of non-incarcerative punishment is a useful test of their true goals in
opposing incarceration.

D. Reforming Prison to Make It Less Destructive to Prisoners
The mass incarceration narrative often rightly points out that prisons are poor
environments for rehabilitating offenders. “Correction” rarely happens at correctional facilities.
Unfortunately, prisons can serve as criminal bootcamps where violence is normalized and
offenders learn the tricks of the criminal trade. A study examining released state prisoners from
2005 to 2014 found that “An estimated 68% of released prisoners were arrested within 3 years,
79% within 6 years, and 83% within 9 years.”219 Recidivism statistics are dismal and should be a
spur to reform.
One downside to the mass incarceration narrative is that it has shifted policymakers’
attention to reducing the use of prison instead of reforming the prison environment to
encourage rehabilitation. For example, mass incarceration activists have repeatedly attempted
to block the construction of new prisons in an attempt to limit incarceration, when the real
result is increased overcrowding and worse prison environments, leading to more recidivism
and more incarceration.220 Improving prison conditions requires more financing, but multiple
studies confirm that better prisons lead to less recidivism, making it a potentially sound
investment of public funds.221 In addition to better funding of prisons, governments should
46

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

experiment with different prison models. Prison bureaus are in a perfect position to run
controlled scientific studies on which prison environments, resources, and programs can
maximize rehabilitation—yet these studies are almost never done. And they are even less likely
to be done when activists focus their energy on simply closing prisons.
Reforming the post-prison environment is also promising, as post-release programs for
prisoners are often currently unavailable or ineffective. Making it easier for released prisoners
to find and hold jobs can also help as the current system actively shunts released felons back
into a life of crime. Instead of attempting to prevent anyone from going to prison in the first
place, anti-incarceration activists should spend more time making sure those entering prison do
so for the last time.

IV. Conclusion
The mass incarceration narrative—the claim that America is suffering from exceptional
and unjustified levels of incarceration due to overly punitive punishment policies—is accepted
by many academics and policymakers without critique or question. Despite hundreds of
researchers publishing on incarceration, practically none have seriously examined the
assumptions of the narrative, a disturbing fact showcasing the dangers of ideological conformity
in academia. This lack of curiosity is damaging since reformers are ill-served by a narrative that
draws so generously upon myths.
Especially troubling is reliance upon the myth that it is draconian U.S. sentencing
changes that are responsible for America’s 300% per capita incarceration rate rise compared to
1960. In fact, non-sentencing factors explain most of that rise, and where punishments have
increased the most, such as for murder and rape, the public is likely to see such changes as
moving the system closer to just sentencing than before. And while foreign countries’ lower
prison populations can offer some useful reform ideas, they do not provide a copyable template
for the American justice system the way the mass incarceration narrative suggests. Many of the
non-sentencing factors that determine current U.S. prison populations – such as crime rates,
deinstitutionalization of the mentally ill, criminal justice system effectiveness, and criminal
histories of sentenced offenders – differ across countries, making meaningful comparisons
difficult. Moreover, many of the lenient sentencing practices in some foreign countries have
inspired chronic and widespread dissatisfaction abroad and would clearly be seen as unjust in
America. Instead of showcasing a draconian America versus a paradisal Europe, differing
punishment policies are often better characterized as the result of elitists wishing to force their
punishment preferences upon the larger community on both sides of the Atlantic, with differing
levels of success.
All that said, America’s incarceration policies do require reform, but that reform must be
guided by doing justice instead of jettisoning just punishment in favor of slashing incarceration
for its own sake. Setting the record straight on the many myths of “mass incarceration” is the
first step toward solving the real problems with America’s use of prison.

47

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

NOTES
1

Colin S. Diver Professor of Law, University of Pennsylvania. The authors wish to express their deep appreciation to
Sarah M. Robinson for substantial assistance and to Paul Heaton for his useful criticisms.
2
Bachelor of Arts in Philosophy, Politics, and Economics, 2022, and Master of Science in Behavioral and Decision
Sciences, 2023, University of Pennsylvania.
3
The term “mass incarceration” can be used by a variety of authors to simply refer to the neutral phenomenon of
America’s increased prison population, but the “mass incarceration narrative” critiqued in this article tells a (often
incorrect) story about the causes and consequences of that increase.
4
Kevin M. Carlsmith, John M. Darley & Paul H. Robinson, Why Do We Punish? Deterrence and Just Deserts as
Motives for Punishment. Journal of Personality and Social Psychology, Vol. 83, pp. 284-299, 2002, Available at
SSRN: https://ssrn.com/abstract=678981
5
Note, for example, the American Law Institute’s 2007 revision of the Model Penal Code’s “purposes” section that
sets desert—punishment proportionate to an offender’s moral blameworthiness—as inviolate above all other
distributive principles. See Model Penal Code: Sentencing § 1.02(2)(a) (approved May 16, 2007).
6
See the discussion and citations at note 205 infra.
7
“Mass incarceration on a scale almost unexampled in human history is a fundamental fact of our country today—
perhaps the fundamental fact, as slavery was the fundamental fact of 1850. In truth, there are more black men in
the grip of the criminal-justice system—in prison, on probation, or on parole—than were in slavery then. Over all,
there are now more people under “correctional supervision” in America—more than six million—than were in the
Gulag Archipelago under Stalin at its height. That city of the confined and the controlled, Lockuptown, is now the
second largest in the United States.” Adam Gopnik, The Caging of America, New Yorker, (Jan 22, 2012),
https://www.newyorker.com/magazine/2012/01/30/the-caging-of-america
8
John Pfaff, The War on Drugs and Prison Growth: Limited Importance and Limited Legislative Options, 52 Harv. J.
Legis. 173 (2015), https://ir.lawnet.fordham.edu/cgi/viewcontent.cgi?article=1650&context=faculty_scholarship
(page 173).
9
Jonathan Rothwell, Drug Offenders in American Prisons: The Critical Distinction Between Stock and Flow,
Brookings, Nov 25, 2015, https://www.brookings.edu/articles/drug-offenders-in-american-prisons-the-criticaldistinction-between-stock-and-flow/
10
Lauren-Brooke Eisen & Inimai Chettiar, 39% of Prisoners Should Not Be in Prison, Time Magazine, Dec. 9, 2016,
https://time.com/4596081/incarceration-report/ .
11
Lauren-Brooke Eisen, How Many Americans are unnecessarily Incarcerated, Brennan Center, Dec 9, 2016,
https://www.brennancenter.org/our-work/research-reports/how-many-americans-are-unnecessarily-incarcerated
12
Ross Kleinstuber. 08 May 2018, Mass Incarceration from: Routledge Handbook of Social, Economic, and Criminal
Justice Routledge. Page 331. Accessed on: 30 Oct 2023
https://www.routledgehandbooks.com/doi/10.4324/9781351002707-29
13
Wendy Sawyer & Peter Wagner, Mass Incarceration: The Whole Pie, 2023, March 14, 2023,
https://www.prisonpolicy.org/reports/pie2023.html.
14
The Keys to Our Mass Incarceration Crisis -- and Who Holds them, ACLU Washington, Aug 12, 2022,
https://www.aclu-wa.org/story/keys-our-mass-incarceration-crisis-and-who-holds-them%C2%A0
15
See Part II.D.2.
16
Andrew Cohen, The American ‘Punisher’s Brain’, Brennan Center for Justice, May 17, 2021,
https://www.brennancenter.org/our-work/analysis-opinion/american-punishers-brain.
17
America’s Criminal Legal System is Unduly Harsh. Experts Explain how We Got here and Solutions that will Benefit
Everyone, Brennan Center, https://www.brennancenter.org/series/punitive-excess
18
Perspectives on Punishment: An Interdisciplinary Roundtable on Punitiveness in America, Apr 2015,
https://johnjay.jjay.cuny.edu/punitivenessinamerica/roundtable.asp. As a similar analysis explains, “The
unprecedented rise in incarceration rates can be attributed to an increasingly punitive political climate surrounding
criminal justice policy formed in a period of rising crime and rapid social change.” National Academies of Sciences,
Engineering, and Medicine. 2014. The Growth of Incarceration in the United States: Exploring Causes and
Consequences. Washington, DC: The National Academies Press. https://doi.org/10.17226/18613.

48

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

19

Bettina Muenster & Jennifer Trone, Why is America So Punitive? A Report of the Interdisciplinary Roundtable on
Punitiveness in America (Excerpted), Federal Sentencing Reporter, Vol. 28, No. 5, Changing Punitiveness (June
2016), pp. 340-347, chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://www-jstororg.proxy.library.upenn.edu/stable/pdf/26377016.pdf?refreqid=excelsior%3A0f8ed47682045f9590365e9e92c5941
a&ab_segments=&origin=&initiator=&acceptTC=1.
20
Bettina Muenster & Jennifer Trone, Why is America So Punitive? A Report of the Interdisciplinary Roundtable on
Punitiveness in America (Excerpted), Federal Sentencing Reporter, Vol. 28, No. 5, Changing Punitiveness (June
2016), pp. 340-347, chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://www-jstororg.proxy.library.upenn.edu/stable/pdf/26377016.pdf?refreqid=excelsior%3A0f8ed47682045f9590365e9e92c5941
a&ab_segments=&origin=&initiator=&acceptTC=1.
21
National Academies of Sciences, Engineering, and Medicine. 2014. The Growth of Incarceration in the United
States: Exploring Causes and Consequences. Washington, DC: The National Academies Press.
https://doi.org/10.17226/18613.
22
Ryan King, What David Brooks gets Wrong about Mass Incarceration, Urban Institute, Oct 1, 2015,
https://www.urban.org/urban-wire/what-david-brooks-gets-wrong-about-mass-incarceration.
23
About BJS, https://bjs.ojp.gov/about.
24
Paul H. Robinson, Hugh Rennie, and Clever Earth, Red Codes, Blue Codes? Factors Influencing the Formulation
of Criminal Law Rules, Jan 2024, at 7.
25
Patrick Langan, Americas Soaring Prison Population, March 1991, Science, Vol. 251 at 1571.
26
Shima Baughman, How Effective are Police? The Problem of Clearance Rates and Criminal Accountability,
Alabama Law Review, Vol 72:1:47, 2020, at 106.
27
Bureau of Justice Statistics: Data to Keep our Communities Safe, chromeextension://efaidnbmnnnibpcajpcglclefindmkaj/https://www.amstat.org/docs/default-source/amstatdocuments/pol-bjs-priorities-2021plus.pdf
28
Survey of Inmates in Local Jails, BJS, https://bjs.ojp.gov/data-collection/survey-inmates-local-jails-silj; National
Judicial Reporting Program, BJS, https://bjs.ojp.gov/data-collection/national-judicial-reporting-program-njrp;
Justice Assistance Data Survey, BJS, https://bjs.ojp.gov/data-collection/justice-assistance-data-survey.
29
Recidivism Survey of Felons on Probation, BJS, https://bjs.ojp.gov/data-collection/recidivism-survey-felonsprobation.
30
By way of example, the series “Felony Sentences in the State Courts” has not been published since 2006. And yet
much of the data continues to be collected and is then aggregated and published in other reports. The data that is
used by the BJS is now largely access restricted and despite several efforts to do so, we were never given access to
the data. The data restrictions are not simply related to current information, when trying to access “National
Corrections Reporting Program, 1998 (ICPSR 3029)” the site denies access,
https://www.icpsr.umich.edu/web/ICPSR/studies/3029. We were told that the data was restricted in an effort to
protect sensitive private information but if that were the case, why is already analyzed data, with all identifying
information stripped out, not accessible?
31
Bureau of Justice Statistics: Data to Keep our Communities Safe, chromeextension://efaidnbmnnnibpcajpcglclefindmkaj/https://www.amstat.org/docs/default-source/amstatdocuments/pol-bjs-priorities-2021plus.pdf
32
Paul H. Robinson, Jeffrey Seaman & Muhammad Sarahne, Confronting Failures of Justice: Getting Away with
Murder and Rape, ch. 9A, Rowman and Littlefield, forthcoming, 2024.
33
The per capita prison population peaked a year earlier in 2008 at 506 per 100,000 residents.
34
Of course, the exact starting date of our analysis does not significantly change the overall trends or explanatory
factors.
35
For example: “Our recommendations would reestablish practices that were the norm in America for most of the
previous century, when incarceration rates were a fraction of what they are today.” See James Austin, Unlocking
America: Why and How to Reduce America’s Prison Population, JFA Institute, No 2007, at 2,
https://www.prisonpolicy.org/scans/jfa/UnlockingAmerica.pdf.
36
Population data used to calculate the per 100k rates from 1980-2015 are from the Disaster Center and consistent
with the data in The “United States Crime Rates for Serious Offenses 1960-2021” table. 2021 data is from the U.S.
Census Bureau.
37
1925-1980 incarceration data from: “Prisoners 1925-81,” Bureau of Justice Statistics, 2, 1982.

49

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

38

This number includes only new court commitments and violators returned to prison. It does not include other
admissions such as escapees returned to prison, special court orders, or administrative transfers. See “Prisoners in
State and Federal Institutions” https://www.ojp.gov/pdffiles1/Digitization/2172NCJRS.pdf page 12, table 5. A more
comparable number to post 1978 data (which includes all admission categories) would be 1961’s total new
admissions which totaled 175,824. For the 1961 data, see table 2 of
https://www.ojp.gov/pdffiles1/Digitization/2160NCJRS.pdf.
39
Prisoners in State and Federal Institutions on December 31, 1975, Dept of Justice, Law Enforcement Assistance
Administration, Feb 1977, at Table 4, https://www.ojp.gov/pdffiles1/Digitization/39194NCJRS.pdf.
40
1980-2015 new imprisonments data from: “Prison Admissions, 1978-2019,” Bureau of Justice Statistics,
https://csat.bjs.ojp.gov/freq-requested-charts.
41
1990-2000 incarceration data from: Allen Beck and Paige Harrison, “Prisoners in 2000,” Bureau of Justice
Statistics, 1, 2001.
42
Ann Carson & William Sabol, Prisoners in 2011, BJS, Dec 2012, tables 1 and 6,
https://bjs.ojp.gov/content/pub/pdf/p11.pdf.
43
Ann Carson & William Sabol, Prisoners in 2011, BJS, Dec 2012, tables 1 and 6,
https://bjs.ojp.gov/content/pub/pdf/p11.pdf.
44
2011-2021 data from: Ann Carson, “Prisoners in 2021 – Statistical Tables,” Bureau of Justice Statistics, 1, 2022.
45
Ann Carson, “Prisoners in 2021 – Statistical Tables,” Bureau of Justice Statistics, 17, 2022.
46
Wendy Sawyer & Peter Wagner, Mass Incarceration: The Whole Pie, 2023, March 14, 2023,
https://www.prisonpolicy.org/reports/pie2023.html.
47
Wendy Sawyer & Peter Wagner, Mass Incarceration: The Whole Pie, 2023, March 14, 2023,
https://www.prisonpolicy.org/reports/pie2023.html.
48
John Dilulio, The Numbers don’t Lie; It’s the Hard Core doing Hard Time,” Brookings institute, March 17, 1996,
https://www.brookings.edu/opinions/the-numbers-dont-lie-its-the-hard-core-doing-hard-time/; Matt DeLisi & John
Paul Wright, Mass Incarceration Hysteria, City Journal, Spring 2022, https://www.city-journal.org/article/massincarceration-hysteria.
49
See Part II.B.5.
50
John Dilulio, The Numbers don’t Lie; It’s the Hard Core doing Hard Time,” Brookings institute, March 17, 1996,
https://www.brookings.edu/opinions/the-numbers-dont-lie-its-the-hard-core-doing-hard-time/
51
Data for 1960 and 1974 comes from Margaret Callahan, Historical Corrections Statistics in the United States,
1850- 1984, BJS, Dec 1986, table 3-20. Data for 1980 and 1992 comes from table 14 of “Prisoners in 1994” by Allen
Beck and Darrell Gilliard. Data for 2001 comes from Paige Harrison, Prisoners in 2001, BJS, July 2002. Data for 2010
comes from Ann Carson, Prisoners in 2011 – Statistical Tables, BJS, 2012
51
Ann Carson, Prisoners in 2021 – Statistical Tables, BJS, 2022. Data for 2021 comes from Ann Carson, Prisoners in
2021 – Statistical Tables, BJS, 2022.
52
We recognize one possible exception covered in this section’s discussion of new or tightened criminalization. The
tightening of criminalization against the distribution and possession of illicit drugs could be considered an
unjustified increase in punitiveness, though even there, the picture is more complicated than often assumed.
53
Historical Population Change Data (1910-2020), US Census Bureau, Apr 26, 2021,
https://www.census.gov/data/tables/time-series/dec/popchange-data-text.html.
54
See, for example, this tweet on mass incarceration that garnered hundreds of retweets: “sometimes i forget that
our hyperincarceration crisis happened so recently (less than fifty years ago) and so quickly (prison population grew
700% in 40 years). it makes me sick and it makes me hopeful. this country made this in less than one lifetime, we
can unmake it in ours.” Rena, @ReenNahMean, Twitter (June 15, 2021, 7:01 PM),
https://twitter.com/ReenNahMean/status/1404937052324392961. While technically correct, statements like this
one add no context either about population increases or any other factors such as crime increases.
55
United Staes Crime Rates 1960-2019, Disaster Center, https://www.disastercenter.com/crime/uscrime.htm.
56
The violent crime rate rose to 395.5 crimes per 100,000 in 2021.
57
The increase in severe violent crime was even greater, with advances in medical care turning what would have
been homicides in 1960 into non-fatal assaults today. As one study found: “Murder rates would be up to five times
higher than they are but for medical developments over the past 40 years.” See Dobson R. Medical advances mask
epidemic of violence by cutting murder rate. BMJ. 2002 Sep 21;325(7365):615. doi: 10.1136/bmj.325.7365.615/a.
PMID: 12242167; PMCID: PMC1124155.

50

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

58

United States Crime Rates 1960-2019, Disaster Center, https://www.disastercenter.com/crime/uscrime.htm.
Data for 1960 comes from Margaret Callahan, Historical Corrections Statistics in the United States, 1850- 1984,
BJS, Dec 1986, table 3-20.
60
This is because 0.55 (1.22) + 0.4 (2.35) + 0.05 = 1.66 or a 66% increase over the previous incarceration rate.
61
Gfroerer, Joseph, and Marc Brodsky. "The incidence of illicit drug use in the United States, 1962–1989." British
Journal of Addiction 87, no. 9 (1992): 1345-1351.
62
Drug Abuse Statistics, National Center for Drug Abuse Statistics, https://drugabusestatistics.org/.
63
That is because 0.05 x 5 = 0.25 and 0.05 x 10 = 0.5.
64
Of course, incarceration did not neatly track crime increases, mainly due to changes in criminal justice system
effectiveness (the rate at which crime is punished). In other words, when crime went up, system effectiveness went
down, and when crime went down, system effectiveness went up. As a result, increases in crime are less important
when modelling incarceration increases for specific slices of the 1960-2021 period, and this is one reason increased
crime is overlooked in the literature as an explanatory factor. However, higher levels of crime today clearly do
contribute to the higher levels of incarceration since criminal justice system effectiveness has caught up and
exceeded its 1960 benchmarks. Since police make a similar percentage of arrests compared to reported crime
today as they did in 1960, and prosecutors convert a higher percentage of arrests into convictions, it would be
impossible for the higher crime rate not to contribute to a higher per capita incarceration rate.
65
Deinstitutionalization resulted from a confluence of factors: the discovery of effective psychiatric drugs, an
advocacy campaign to raise awareness of poor conditions in mental institutions, new laws designed to move the
mentally ill out of institutions into the community, and Supreme Court decisions making it harder to civilly commit
mentally ill individuals. Megan Testa & Sara G. West, Civil Commitment in the United States, National Center for
Biotechnology Information - National Library of Medicine, (Oct. 7, 2010),
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3392176/.
66
Anna Swanson, A shocking number of mentally ill Americans end up in prison instead of treatment, Wash. Post,
(Apr. 30, 2015), https://www.washingtonpost.com/news/wonk/wp/2015/04/30/a-shocking-number-of-mentally-illamericans-end-up-in-prisons-instead-of-psychiatric-hospitals/
67
Bernard E. Harcourt, “An Institutionalization Effect: The Impact of Mental Hospitalization and Imprisonment on
Homicide in the United States, 1934-2001,” 40 J. LEGAL STUD. 39, 41-46, 2011.
68
Steven Raphael and Michael Stoll, “Assessing the Contribution of the Deinstitutionalization of the Mentally Ill to
Growth in the U.S. Incarceration Rate,” The Journal of Legal Studies vol. 42 No. 1, 190-219, 2013.
69
Laura M. Maruschak, Indicators of Mental Health Problems Reported by Prisoners, U.S. Department of Justice Bureau of Justice Statistics, (June 2021), https://bjs.ojp.gov/library/publications/indicators-mental-healthproblems-reported-prisoners-survey-prison-inmates
70
When considering deinstitutionalization’s effect on increasing the per capita incarceration rate from 1960s levels,
it is important to avoid double counting between it and the effect of generic crime increases which it contributed
to. Deinstitutionalization both meant more crime (thus increasing incarceration) and a higher likelihood that a
mentally ill offender who committed a crime would end up in prison as opposed to a mental hospital (thus also
increasing incarceration).
71
Bernard E. Harcourt, “An Institutionalization Effect: The Impact of Mental Hospitalization and Imprisonment on
Homicide in the United States, 1934-2001,” 40 J. LEGAL STUD. 39, 42, 2011.
72
In 1960, there were approximately 175,000 new prison admissions, but many of these were prison transfers or
other admissions not representative of punishment for a new crime. The number of prison admissions where a
crime was punished can be estimated since there were 88,575 new court commitments (new convictions resulting
in prison) and 13,586 parole violators returned to prison, around half of whom (6,793) were likely returned for a
new criminal offense. For new court commitments, see Langan “Race of Prisoners 1926-1986”. For number of
parole violators in 1960, see https://www.ojp.gov/pdffiles1/Digitization/2160NCJRS.pdf page 1 for the parole
violators in 1960. We estimate 50% of these parole violators were returned for criminal violations based on a
reasonable guess at the current national rate. See Schuman, Jacob, Criminal Violations (February 15, 2022). Virginia
Law Review, Vol. 108, p. 1817, 2022, Penn State Law Research Paper No. 11-2022, Available at SSRN:
https://ssrn.com/abstract=4034991 which based on U.S. Sentencing Commission Data estimated that over 50% of
supervised release revocations were “criminal violations.” The actual rate of criminal vs. technical violations differs
by jurisdiction (and likely somewhat across time), but 50% seems a reasonable guess at the national number across
time.
59

51

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

73

We use 2018 FBI UCR data because of changes in 2019 to reporting rules that resulted in participation changes.
There are limitations to this comparison, of course. FBI UCR data only counts certain types of crime (while new
imprisonments result from all types of crime), and the 1960 FBI UCR data is unideal and may undercount crimes
more than the 2018 data, but this does not prevent an approximate instructive comparison. The data for new
imprisonments in 2018 comes from https://bjs.ojp.gov/content/pub/pdf/p18.pdf (prisoners 2018 table 8). In 2018,
there were 410,867 new court commitments and 169,663 supervised release revocations. We use the previous
50% guess at new imprisonments from parole caused by criminal as opposed to technical violations.
74
Felony Sentences in State Courts 1986 (page 2, table 2) and Felony Sentences in State Courts 2006 (table 1.2). In
1986, 46% of convicted felony defendants in state courts received a prison sentence, 21% received a jail sentence,
and 33% received non-incarcerative punishments (usually probation). In 2006, 41% of convicted felony defendants
in state courts received prison sentences, 28% received jail sentences, and 31% received non-incarcerative
sentences.
75
There were 145,895 new state court commitments in 1986 and 63,830 parolee/probation violators returned to
state prison, of which we assume half (31,915) were returned for criminal violations. There were also 25,185
federal prison admissions. See National Corrections Reporting Program 1986
https://bjs.ojp.gov/content/pub/pdf/ncrp86.pdf
76
In 2006, New court commitments totaled 491,810 and there were 246,781 parole violators, of which we assume
half (123,390) were returned for criminal violations. See https://bjs.ojp.gov/content/pub/pdf/pim07.pdf at 19.
77
This is the most plausible interpretation of the data of which we are aware. It represents an oversimplification of
many complex factors, but the general trend appears sound.
78
Shima Baughman, How Effective are Police? The Problem of Clearance Rates and Criminal Accountability,
Alabama Law Review, Vol 72:1:47, 2020, at 82. The use of crime victimization data as opposed to reported crime
does not complicate the picture as the reporting rate was about the same in 1990 and 2006.
79
How Effective Are Police? The Problem of Clearance Rates and
Criminal Accountability (Shima Baughman). In 1990, 15.98% of all reported crimes resulted in arrest, while in 2006
the number was 21.16%, a 32.4% increase in the arrest rate of reported crimes. An increase in prosecution
effectiveness also must have occurred to achieve the 57.2% increase in “true” conviction rate.
80
John F. Pfaff, The Causes of Growth in Prison Admissions and Populations (January 23, 2012). Available at SSRN:
https://ssrn.com/abstract=1990508 or http://dx.doi.org/10.2139/ssrn.1990508 .
81
This allows for the possibility that problems in the underlying data comparison (such as the higher proportion of
prison admissions caused by drug crimes in later years) makes 110% an overstatement.
82
Because 1.85 x 1.67 = 3.09.
83
Consider one anti-incarceration source: “Law enforcement lie to secure the arrest and prosecution of innocent
persons; prosecutors suppress mitigating evidence, knowingly use perjured testimony, and fabricate evidence to
secure criminal convictions…It is estimated that anywhere from 2 to 10 percent of the nearly 2 million people
incarcerated in U.S. prisons are innocent. That means there could be as many as 200,000 innocent people are
victims of the nation’s prison industrial complex.” See https://www.unincarcerated.com/blog/human-cost-ofprison-industrial-complex While these estimates do exist, they are wildly exaggerated.
84
The true rate of wrongful convictions is likely between 0.016% and 0.062%. See Cassell, Paul G., 60 Ariz. L. Rev.
815 (2018).
85
For example, the conviction rate for murder was 36% in 2006, the last year data was made available, and it has
almost certainly decreased since. See Sean Rosenmerkel, Matthew Durose & Donald Farole, "Felony Sentences in
State Courts, 2006 – Statistical Tables," Department of Justice, December 2009,
https://bjs.ojp.gov/library/publications/felony-sentences-state-courts-2006-statistical-tables-standard-errortables#:~:text=In%202006%20an%20estimated%2069,and%2011%20months%20in%202006. Also, around 92% of
aggravated assaults and 97% of rapes and sexual assaults end in no conviction. For assault data, see Rachel Morgan
& Jennifer Truman, Criminal Victimization, 2019, U.S. Dept. Of Justice, Sept. 2020, at table 1, chromeextension://efaidnbmnnnibpcajpcglclefindmkaj/https://bjs.ojp.gov/content/pub/pdf/cv19.pdf. For sexual assault
data, see The Criminal Justice System: Statistics, RAINN, https://www.rainn.org/statistics/criminal-justicesystem;Victims of sexual Violence: Statistics, RAINN, https://www.rainn.org/statistics/victims-sexualviolence#:~:text=Sexual%20Violence%20Affects%20Millions%20of,year%20in%20the%20United%20States

52

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

86

Andrew Von Hirsch, Past or Future Crimes: Deservedness and Dangerousness in the Sentencing of Criminals, 7885 (1985).
87
This can be demonstrated looking at the data from the BJS series Felony Defendants in large Urban Counties, for
example: Gerard Rainville & Brian Reaves, Felony Defendants in Large Urban Counties, 2000, BJS, chromeextension://efaidnbmnnnibpcajpcglclefindmkaj/https://bjs.ojp.gov/content/pub/pdf/fdluc00.pdf.
88
This can be demonstrated looking at the data from the BJS series Felony Defendants in large Urban Counties, for
example: Gerard Rainville & Brian Reaves, Felony Defendants in Large Urban Counties, 2000, BJS, chromeextension://efaidnbmnnnibpcajpcglclefindmkaj/https://bjs.ojp.gov/content/pub/pdf/fdluc00.pdf.
89
Brian Reaves, Felony Defendants in Large Urban Counties, 2009 – Statistical Tables, Dept of Justice, Dec 2013,
chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://bjs.ojp.gov/content/pub/pdf/fdluc09.pdf
And also see https://bjs.ojp.gov/content/pub/pdf/p90.pdf for 1990 data.
90
Some might argue the increase in criminal records did not come from offender recidivism worsening but rather
police or prosecutors being more effective than in the past. Increased criminal justice system effectiveness may
indeed have contributed to a worsening in offenders’ official criminal records, but it is impossible to say how much.
91
For this extrapolation to be inaccurate, the criminal records of offenders would have had to improve significantly
between 1960 and 1990, a virtual impossibility given the increase in crime over that period and the increasing
chance of an arrest resulting in imprisonment over that period. See Langan, Patrick A. "America's soaring prison
population." Science 251, no. 5001 (1991): 1568-1573.
92
Sentencing Drug Offenders: The Incarceration Addiction, DEA.gov, https://www.dea.gov/sites/default/files/201805/Early%20Years%20p%2012-29.pdf at 22.
93
Lawrence Greenfeld, Trends in Prison Population 1926-1984, at table 12.
94
Sentencing Drug Offenders: The Incarceration Addiction, DEA.gov, https://www.dea.gov/sites/default/files/201805/Early%20Years%20p%2012-29.pdf at 25
95
That is because 0.55 (1.22) + 0.4 (2.35) + 0.05 (5) = 1.861 and 0.25/1.861 = 0.13. Again, this is quite rough and
assumes all other factors (such as criminal justice system effectiveness) being held equal for all crimes. Still, it gives
an idea for how much the drug offender prison population might have changed if the government had not made
any punitive changes to drug policy post-1960 but had simply attempted to keep the level of enforcement
consistent even as drug crimes proliferated.
96
Single convention of Narcotic Drugs, 1961, UN, https://www.unodc.org/pdf/convention_1961_en.pdf.7.
97
Facing Addiction in America, Surgeon General’s Report on Alcohol, Drugs, and Health, Nov 2016,
https://store.samhsa.gov/product/facing-addiction-america-surgeon-generals-report-alcohol-drugs-and-health-fullreport/sma16-4991.
98
We of course do not deny that the motives of some individual policymakers in tightening drug criminalization
may have been to target groups they were biased against, but taken wholistically, greater drug criminalization
would not have been accepted if it did not reflect prevailing public concerns—including among minority
communities.
99
Jennifer Robinson, Decades of Drug Use: Data fro the ‘60s and ‘70s, Gallup, Jul 2, 2002,
https://news.gallup.com/poll/6331/decades-drug-use-data-from-60s-70s.aspx
100
Some argue that drug criminalization is a contributing factor to the worsening of criminal offender histories, as
without drug criminalization, some offenders would have fewer total prior convictions. As a result, drug
criminalization may indirectly explain a bit more than 15% of the current prison population, but without better
data, it is impossible to say how much more—though it is not likely to be particularly large.
101
National Center for Drug Abuse Statistics, Drug Related Crime Statistics, https://drugabusestatistics.org/drugrelated-crime-statistics/.
102
Offenders tend not to stick to just one crime type. Consider one report which found that “More than threequarters (77%) of released drug offenders were arrested for a non-drug crime within 9 years.” See Mariel Alper,
2018 Update on Prisoner Recidivism: A 9-year Follow-up Period (2005-2014).
103
Lara Geer Farley, “The Adam Walsh Act: The Scarlet Letter of the Twenty-First Century.” 47 Washburn L.J. 471
(2008).
104
Pre-internet, the occurrence of child sexual abuse appeared to be stable across multiple decades and a 1997
study found it was 14.5% for females and 7.2% for males (heartbreakingly high numbers). See Kevin M. Gorey,
Donald R. Leslie, The prevalence of child sexual abuse: Integrative review adjustment for potential response and
measurement biases, Child Abuse & Neglect, Volume 21, Issue 4, 1997, Pages 391-398, ISSN 0145-2134,

53

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

https://doi.org/10.1016/S0145-2134(96)00180-9.
(https://www.sciencedirect.com/science/article/pii/S0145213496001809
105
Maria Blackburn, Incarcerating child sexual abusers topped $5.4 billion in 2021 according to study, Hub, Apr. 5,
2022, https://hub.jhu.edu/2022/04/05/incarcerating-child-sexual-abusers-5billion/#:~:text=At%20the%20federal%20level%2C%20the,of%20%24508%20million%20in%20spending.
106
Human Trafficking: Key Legislation, DOJ, https://www.justice.gov/humantrafficking/key-legislation
107
Human Trafficking Data Collection
Activities, DOJ, 2022, https://bjs.ojp.gov/sites/g/files/xyckuh236/files/media/document/htdca22.pdf
108
2023 Trafficking in Persons Report: United States, US Dept of State,
https://www.state.gov/reports/2023-trafficking-in-persons-report/united-states
109
David M. Uhlmann, Environmental Crime Comes of Age: The Evolution of Criminal Enforcement in the
Environmental Regulatory Scheme. Utah L. Rev. 2009, no. 4 (2009): 1223-52,
https://repository.law.umich.edu/cgi/viewcontent.cgi?article=1786&context=articles
110
https://www.justice.gov/criminal/criminal-fraud/identity-theft/identity-theft-and-identity-fraud
111
William Pryor et al. “Mandatory Minimum Penalties for Identity Theft Offenses in the Federal Criminal Justice
System,” United States Sentencing Commission, 33, 2018. However, there is significant room for greater state and
federal enforcement of statutes against fraud and identity theft since as “as many as 700,000 people fall victim to
identity theft and other forms of Internet fraud every year.” Dan Verton, Criminals Using High-tech Methods for
Old-style Crimes, Computerworld, Feb, 2003.
112
History of Anti-money Laundering Laws, US Treasury, https://www.fincen.gov/history-anti-money-launderinglaws.
113
Joseph Fawbush, How Much time will I Serve for Fraud and Related Crimes?, Jun 7, 2021,
https://www.lawinfo.com/resources/criminal-defense/sentencing/sentencing-statistics/how-much-time-will-iserve-for-fraud-and-related-crimes.html.
114
Kyla Bishop, A Reflection on the History of Sexual Assault Laws in the United States, Arkansas Journal of Social
Change and Public Service, Apr 15, 2018, https://ualr.edu/socialchange/2018/04/15/reflection-history-sexualassault-laws-united-states/
115
Kyla Bishop, A Reflection on the History of Sexual Assault Laws in the United States, Arkansas Journal of Social
Change and Public Service, Apr 15, 2018, https://ualr.edu/socialchange/2018/04/15/reflection-history-sexualassault-laws-united-states/
116
Kyla Bishop, A Reflection on the History of Sexual Assault Laws in the United States, Arkansas Journal of Social
Change and Public Service, Apr 15, 2018, https://ualr.edu/socialchange/2018/04/15/reflection-history-sexualassault-laws-united-states/. The shift to take all forms of sexual assault more seriously is exemplified by the FBI’s
2013 decision to change the definition of rape to include previously excluded crimes. The previous definition of
“carnal knowledge of a female forcibly and against her will” was changed to “Penetration, no matter how slight, of
the vagina or anus with any body part or object, or oral penetration by a sex organ of another person, without the
consent of the victim.” Rape, FBI:UCR, https://ucr.fbi.gov/crime-in-the-u.s/2013/crime-in-the-u.s.-2013/violentcrime/rape#:~:text=The%20revised%20UCR%20definition%20of,rape%20and%20incest%20are%20excluded.
117
Andrew Van Dam. “Analysis | Less than 1% of Rapes Lead to Felony Convictions. At Least 89% of Victims Face
Emotional and Physical Consequences.” The Washington Post, October 6, 2018.
https://www.washingtonpost.com/business/2018/10/06/less-than-percent-rapes-lead-felony-convictions-leastpercent-victims-face-emotional-physical-consequences/.
118
German Lopez, The US Incarcerates too many People. But Comparisons with Europe are Flawed, VOX, Apr 7,
2015, https://www.vox.com/2015/4/7/8364263/us-europe-mass-incarceration.
119
Jeffrey Fagan, The Criminalization of Domestic Violence: Promises and Limits, National Institute of Justice, 1995,
at 6.
120
Bureau of Justice Statistics. “Bureau of Justice Statistics Family Violence Statistics Including Statistics on
Strangers and Acquaintances,” 2005. https://bjs.ojp.gov/content/pub/pdf/fvs08.pdf.
121
32% of these incarceration sentences were prison sentences. See Bureau of Justice Statistics. “Bureau of Justice
Statistics Family Violence Statistics Including Statistics on Strangers and Acquaintances,” 2005.
https://bjs.ojp.gov/content/pub/pdf/fvs08.pdf.
122
Bureau of Justice Statistics. “Bureau of Justice Statistics Family Violence Statistics Including Statistics on
Strangers and Acquaintances,” 2005. https://bjs.ojp.gov/content/pub/pdf/fvs08.pdf.

54

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

123

Muller, Why We need to take Stalking More Seriously
Meyers, R. “Anti-Stalking Statutes | Office of Justice Programs.” www.ojp.gov, 1998.
https://www.ojp.gov/ncjrs/virtual-library/abstracts/anti-stalking-statutes
125
Katrina Baum, “Stalking Victimization in the United States,” Bureau of Justice Statistics, 15, 2009.
126
Sarah Gray, Here’s a Timeline of the Major Gun Control Laws in America, Time, Apr 30, 2019,
https://time.com/5169210/us-gun-control-laws-history-timeline/.
127
Lawrence Greenfeld & Marianne Zawitz, Weapons Offenses and Offenders, BJS, Nov 1995,
https://bjs.ojp.gov/content/pub/pdf/woofccj.pdf.
128
Mariel Alper & Lauren Glaze, Source and use of Firearms Involved in Crime: Survey of Prison Inmates, 2016, BJS,
Jan 2019, https://bjs.ojp.gov/content/pub/pdf/suficspi16.pdf.
129
David Olson, Illegal Firearm Possession: A Reflection on Policies and Practices that may Miss the Mark and
Exacerbate Racial Disparity in the Justice System, Duke Center for Firearms Law, Jan 19, 2022,
https://firearmslaw.duke.edu/2022/01/illegal-firearm-possession-a-reflection-on-policies-and-practices-that-maymiss-the-mark-and-exacerbate-racial-disparity-in-the-justice-system/#_ftn12
130
Wendy Sawyer & Peter Wagner, Mass Incarceration: The Whole Pie, 2023, March 14, 2023,
https://www.prisonpolicy.org/reports/pie2023.html.
131
U.S. Life Expectancy 1950-2024, Macrotrends, https://www.macrotrends.net/countries/USA/united-states/lifeexpectancy.
132
Life Expectancy for Men at the age of 65 in the U.S. from 1960 to 2021, Statista,
https://www.statista.com/statistics/266657/us-life-expectancy-for-men-aat-the-age-of-65-years-since-1960/
133
That is because 13586/88575 = 0.15
134
That is because 169,663/410867 = 0.41
135
Incarceration researcher John Pfaff concludes the role of parole in incarceration increases is much smaller than
often assumed precisely because it is a downstream effect of incarceration. See Pfaff, John F., The Causes of Growth
in Prison Admissions and Populations (January 23, 2012). Available at SSRN: https://ssrn.com/abstract=1990508 or
http://dx.doi.org/10.2139/ssrn.1990508 (page 36).
136
Nor would this finding be unique in the literature. John Pfaff’s extensive empirical work on incarceration has
concluded that non-sentencing factors are largely responsible for America’s incarceration rate increase. See for
example Pfaff, John. 2008. “The Empirics of Prison Growth: A Critical Review and Path Journal of Criminal Law and
Criminology 98: 101–179. Pfaff, John. 2010. “The Durability of Prison Populations.” University of Chicago Legal
Forum 2010: 73 – 115. Pfaff, John. 2011. “The Myths and Realities of Correctional Severity: Evidence from the
National Corrections Reporting Program.” American Law and Economics Review. Patrick Langan also cast doubt on
the idea that changes to sentencing were a main driver of incarceration increases. See Langan, Patrick A. "America's
soaring prison population." Science 251, no. 5001 (1991): 1568-1573.
137
Langan, Patrick A. "America's soaring prison population." Science 251, no. 5001 (1991): 1568-1573.
138
Langan seems to cast doubt on these explanations in his 1991 study, citing prison surveys. However, such
surveys do not necessarily capture all changes in the pool of convicted defendants, nor do they consider how plea
bargaining might obscure a rise in serious crime admittances while still impacting incarceration chances.
139
Felony Sentences in State Courts 1986 (page 2, table 2).
140
Felony Sentences in State Courts 2006 (table 1.2).
141
Danielle Kaeble NCJ 252205 “Time Served in State Prison,” U.S. Department of Justice Bureau of Justice
Statistics, November 2018, 1.
142
Danielle Kaeble NCJ 252205 “Time Served in State Prison,” U.S. Department of Justice Bureau of Justice
Statistics, November 2018, 1.
143
“Most mandatory minimum penalty offenses result in a negotiated disposition whereby the defendant pleads
guilty to a lesser charge or other offense not subject to a mandatory minimum penalty.” Mandatory Minimum
Sentences, Prison Policy Initiative, at 5,
https://www.prisonpolicy.org/scans/cga/mandatory_minimum_sentences_finding_and_recommendations.pdf.
144
For both mean and median data, “first release” refers to the first release for a particular sentence, not first-time
offenders. It excludes time served between first and second releases, which occurs when a parolee is returned to
prison after violating the conditions of their first release. The data also counts deaths as a first release, so it
includes life sentences and death sentences.
124

55

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

145

1923-1960 data from Margaret Werner Cahalan, “Historical Corrections Statistics in the United States, 19501984,” Bureau of Justice Statistics, 52, 1986, at Table 3.23.
146
1923, 1935, and 1992 data includes manslaughter offenses
147
M means that the data is for males only.
148
Felony Sentences in State Courts, 1986, BJS, at table 4.
149
Paula Ditton & Doris Wilson, Truth in Sentencing in State Prisons, Jan 1999, U.S. Dept of Justice, at Table 7,
chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://bjs.ojp.gov/content/pub/pdf/tssp.pdf.
150
Lawrence Greenfeld, “Prison Sentences and Time Served for Violence,” Bureau of Justice Statistics, 1, 1995.
151
Supra note 81.
152
Matthew Durose, Felony Sentences in State Courts, 2000, Bureau of Justice Statistics, at table 4.
153
Danielle Kaeble, “Time Served in State Prison, 2018,” Bureau of Justice Statistics, 2, 2021.
154
1926-1982 data from: Stephanie Minor-Harper and Lawrence Greenfeld, “Prison Admissions and Releases,
1982,” Bureau of Justice Statistics, Table 13, 1985, https://bjs.ojp.gov/content/pub/pdf/par82.pdf.
155
Margaret Cahalan, Historical Corrections Statistics in the United States, 1850-1984, Dept of Justice, Dec 1986, at
table 3-23.
156
Margaret Cahalan, Historical Corrections Statistics in the United States, 1850-1984, Dept of Justice, Dec 1986, at
table 3-25.
157
Margaret Cahalan, Historical Corrections Statistics in the United States, 1850-1984, Dept of Justice, Dec 1986, at
table 3-23.
158
“National Corrections Reporting Program, 1986,” Bureau of Justice Statistics, Table 2-6,
https://bjs.ojp.gov/content/pub/pdf/ncrp86.pdf.
159
1993-2009 data from: National Corrections Reporting Program: Time served in state prison, by offense, release
type, sex, and race, Bureau of Justice Statistics, 2011, https://bjs.ojp.gov/library/publications/national-correctionsreporting-program-time-served-state-prison-offense.
160
Sourcebook of Criminal Justice Statistics online, 2009, Table 6,
http://www.albany.edu/sourcebook/pdf/t600272009.pdf.
161
Danielle Kaeble, “Time Served in State Prison, 2018,” Bureau of Justice Statistics, Table 1, 2021.
162
2018 data does not distinguish between aggravated assault and simple assault.
163
See discussion and citations at note 91 supra.
164
Richard P. Conaboy, Survey of Public Opinion on Sentencing Federal Crimes, Mar 14, 1997, United States
Sentencing Commission, https://www.ussc.gov/research/research-reports/survey-public-opinion-sentencingfederal-crimes.
165
P.H. Rossi, Guidelines and Sample Sentences Compared, National Sample Survey Public Opinion and Sentencing,
Ch 4, 1997, https://www.ussc.gov/sites/default/files/pdf/research-and-publications/research-projects-andsurveys/surveys/19970314-public-opinion-on-sentencing/JP_CH4.pdf.
166
John Gramlich, U.S. Public Divided over whether People convicted of Crimes Spend too much or too little Time in
Prison, Pew Research Center, Dec 6, 2021, https://www.pewresearch.org/short-reads/2021/12/06/u-s-publicdivided-over-whether-people-convicted-of-crimes-spend-too-much-or-too-little-time-in-prison/.
167
Of course, many in the mass incarceration movement believe their elite notions of a “just” sentence should take
precedence over the public’s. We do not seek to change their private views of justice (or lack thereof); we merely
note that the justice system is meant to do justice as society sees it or it loses its legitimacy among the very people
it is supposed to satisfy and protect. See discussion and citations at note 205 infra.
168
See Rummel v. Estelle (1980).
169
Peter French, (2001). The virtues of vengeance. Lawrence: University of
Kansas Press, at 207; Barbara H. Fried, Beyond Blame, BOS. REV. (June 28, 2013),
http://bostonreview.net/forum/barbarafried-beyond-blame-moral-responsibility-philosophy-law; VICTOR
TADROS, THE ENDS OF HARM: THE MORAL FOUNDATIONS OF CRIMINAL LAW 60 (2011) at 61 (“Until the recent
revival of retributivism, the common view was that retributivism is barbaric in treating the suffering of human
beings as good.”).
170
Executed Inmate Summary – William George Bonin, CDC, https://www.cdcr.ca.gov/capitalpunishment/inmates-executed-1978-to-present/william-george-bonin/
171
See Rummel v. Estelle (1980).

56

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

172

Alexandria Ocasio-Cortez on her Catholic faith and the urgency of criminal justice reform, Ocasio Cortez House,
https://ocasio-cortez.house.gov/media/in-the-news/alexandria-ocasio-cortez-her-catholic-faith-and-urgencycriminal-justice-reform
173
Emily Widra & Tiana Herring, States of Incarceration: The Global Context 2021, Prison Policy Initiative, Sept.
2021, https://www.prisonpolicy.org/global/2021.html
174
James Cullen, The History of Mass Incarceration, Brennan Center for Justice, July 20, 2018,
https://www.brennancenter.org/our-work/analysis-opinion/history-mass-incarceration
175
Sarah Childress, Michelle Alexander: “A System of Racial and Social Control,” Frontline, Apr. 29, 2014,
https://www.pbs.org/wgbh/frontline/article/michelle-alexander-a-system-of-racial-and-social-control/
176
World Prison Brief data cited in the table shows America having the sixth highest incarceration rate. China has
the highest absolute prison population in the world. See https://www.statista.com/statistics/262961/countrieswith-the-most-prisoners/
177
World Prison Brief data shows that even excluding jailed populations, America still does not have the highest per
capita incarceration rate, but it remains close to the top.
178
https://www.prisonstudies.org/highest-to-lowest/prison_population_rate?field_region_taxonomy_tid=All
179
A clear factor is greater availability of guns in America. America’s gun homicide rate is 22 times that of the EU.
See Warren Fisk, Fact Check: Is Gun Violence 23 Times Higher in the U.S. than in the EU?, WRAL News, Jan 10,
2020, https://www.wral.com/story/fact-check-is-gun-violence-23-times-higher-in-the-u-s-than-the-eu/20665571/.
180
“Europe” refers to a representative basket of countries that contain most of the European population—there are
of course plenty of individual European countries that have much lower crime rates than the U.S., and for these
countries, their lower crime rates do contribute to their lower per capita incarceration rates.
181
Paolo Buonanno, Francesco Drago, Roberto Galbiati, Giulio Zanella. Crime in Europe and the United
States: dissecting the ‘reversal of misfortunes’. Economic Policy, 2011, p. 349. 10.1111/j.14680327.2011.00267.x. hal-03461525
182
Paolo Buonanno, Francesco Drago, Roberto Galbiati, Giulio Zanella. Crime in Europe and the United
States: dissecting the ‘reversal of misfortunes’. Economic Policy, 2011, p. 375. 10.1111/j.14680327.2011.00267.x. hal-03461525
183
https://www.prisonstudies.org/country/mexico Number is from 2020.
184
https://qz.com/610369/only-1-of-crimes-are-punished-in-mexico
185
https://www.pewresearch.org/short-reads/2017/03/01/most-violent-and-property-crimes-in-the-u-s-gounsolved/
186
For trends in American justice system effectiveness, see the previous discussion at II.B.4. For the European crime
trend, see note 182.
187
Laurence Greenfeld, Cross-National Studies in Crime and Justice, Sep 2004, DOJ.
188
There is some evidence that hints at this. “Cross National Studies in Crime and Justice” felt it necessary to
include prisoners sent to mental institutions in some European countries when calculating incarcerated offenders
(see page 168). However, the effect of this factor is likely quite small.
189
While measures of recidivism can differ, making data comparison difficult, data suggests the US has a higher rate
of recidivism than many other Western countries. See Yukhnenko, Denis, Leen Farouki, and Seena Fazel. "Criminal
recidivism rates globally: A 6-year systematic review update." Journal of Criminal Justice 88 (2023): 102115.
190
Laurence Greenfeld, Cross-National Studies in Crime and Justice, DOJ, Sept 2004,
https://bjs.ojp.gov/content/pub/pdf/cnscj.pdf.
191
This number includes only adult offenders, whereas other countries do not make this distinction.
192
This includes only crimes categorized as “serious assault,” which despite its name, does not include many
assaults causing grievous injury. As a result, it may not be comparable to other forms of assault in the table.
193
Kazemian, L. (2022). Long sentences: An international perspective. Council on Criminal
Justice. https://counciloncj.foleon.com/tfls/long-sentences-by-the-numbers/an-internationalperspective
194
Laurence Greenfeld, Cross-National Studies in Crime and Justice, DOJ, Sept 2004,
https://bjs.ojp.gov/content/pub/pdf/cnscj.pdf.
195
These numbers are for a subcategory of assault labeled “serious assault.”
196
See section II.b.6 for a discussion of American punishment levels and the views of ordinary people.

57

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

197

Sentencing Poll – House of Commons Justice Committee – March 27, 2023, Savanta,
https://savanta.com/knowledge-centre/published-polls/sentencing-poll-house-of-commons-justice-committee-27march-2023/.
198
Kristina Jerre, Public’s Sense of Justice in Sweden – a Smorgasbord of Opinions, Stockholm University, 2013,
http://su.diva-portal.org/smash/get/diva2:643158/FULLTEXT01.pdf.
199
Public Opinion on the Severity of Sentences in France in 2022, by Political Preference, Statista,
https://www.statista.com/statistics/1364050/opinion-severity-justice-penalties-france-political-preference/.
200
To What Extent do you Agree with the Following Statement: “Criminals should be Punished more Severely”,
Statista, https://www.statista.com/statistics/549254/public-opinion-about-punishment-of-criminals-in-thenetherlands/./
201
Jacqueline Beard, What is the Government doing to Reduce Pressure on Prison Capacity?, UK Parliament, Oct
19, 2023, ommonslibrary.parliament.uk/what-is-the-government-doing-to-reduce-pressure-on-prisoncapacity/#:~:text=The%20Government%20is%20undertaking%20a,5%2C400%20places%20had%20been%20deliver
ed.
202
Lauren-Brooke Eisen, How Many Americans are Unnecessarily Incarcerated, Brennan Center, Dec 9, 2016,
https://www.brennancenter.org/our-work/research-reports/how-many-americans-are-unnecessarily-incarcerated.
203
Robinson, supra note 32, ch. 3.
204
Paul Robinson, INTUITIONS OF JUSTICE & THE UTILITY OF DESERT (Oxford 2013); DISTRIBUTIVE PRINCIPLES OF
CRIMINAL LAW: WHO SHOULD BE PUNISHED HOW MUCH? (Oxford 2008); The Criminogenic Effects of Damaging
Criminal Law’s Moral Credibility, 31 Southern California Interdisciplinary Law Journal 277-327 (2022) (with Lindsay
Holcomb); The Disutility of Injustice, 85 New York University Law Review 1940-2033 (2010) (with Geoff Goodwin,
Penn Psych, and Michael Reisig, ASU Criminology).
205
Aila Slisco, Two-Thirds of American Voters Support Decriminalizing All Drugs: Poll, Newsweek, June 10, 2021,
https://www.newsweek.com/two-thirds-american-voters-support-decriminalizing-all-drugs-poll-1599645.
206
Ted Van Green, Americans Overwhelmingly say Marijuana should be Legal for Medical or Recreational Use, Pew
Research, Nov 22, 2022, https://www.pewresearch.org/short-reads/2022/11/22/americans-overwhelmingly-saymarijuana-should-be-legal-for-medical-or-recreational-use/.
207
Emily Ekins, Poll: 55% of Americans Favor Decriminalizing Drugs, Cato, Oct 2, 2019,
https://www.cato.org/blog/poll-55-americans-favor-decriminalizing-drugs.
208
Paul H Robinson and Sarah M Robinson, American Criminal Law: It's People, Principles, and Evolution (2023), at
302 (discussing the problem of having a criminal offense that is rarely if ever enforced).
209
Robinson, supra note 32, ch. 11.
210
Ram Subramanian & Ruth Delaney, Playbook for Change? States Reconsider Mandatory Minimum Sentences,
Feb. 2014, at 8
211
In-Prison Credit-Earning Opportunities, California Department of Corrections and Rehabilitation,
https://www.cdcr.ca.gov/proposition57/.
212
Id. at 71 & 86.
213
Id. at 76.
214
Barry Latzer, It’s Possible to Reduce Imprisonment and Crime Rates” Wall St. J., Oct. 27, 2022,
https://www.wsj.com/articles/its-possible-to-reduce-incarceration-and-crime-gps-ecarceration-paroleeprobationer-prison-ankle-bracelet-privacy-offender-recidivism-11666896237
215
David McNeill, Life in Japan: The softer corners and harder edges of the Japanese justice system, The Mainichi,
Apr. 3, 2022, https://mainichi.jp/english/articles/20220402/p2a/00m/0op/017000c; Mari Kita, Proxy Punishment:
Consequences of Informal Sanctions among Families of Offenders In Japan, 6(1) Qualitative Criminology, 2018.
216
NCJRS 147416, Commentary on the United Nations Standard Minimum Rules for Non-Custodial Measures (The
Tokyo Rules), National Institute of Justice, 1993, at 1–39.
217
United Nations Office on Drugs and Crimes, Handbook of basic principles and promising practices on
alternatives to imprisonment, 2007, at 41.
218
Modern surveillance technology allows for greater control of non-imprisoned offenders today than at any time
in the past while still granting such offenders more liberty than they would have in prison. Of course, some
offenders would simply never be given non-incarcerative punishments—where the exact line should be drawn
requires public research and experimentation.
219
Mariel Alper, 2018 Update on Prisoner Recidivism: A 9-year Follow-up Period (2005-2014).

58

Electronic copy available at: https://ssrn.com/abstract=4698071

Robinson & Seaman

1 15 24

220

Ezra Kleinbaum, Activists Unite to Protest against a new Women’s Prison, The Cypress, Apr 10, 2023,
https://thecypressonline.com/49153/news/activists-unite-to-protest-against-a-new-womens-prison/; Activist try to
Hancuff Delano Prison, AP, 2016, https://threestrikes.org/activists-try-to-handcuff-delano-prison/; Henry K. Lee,
Bay Area Activists File Lawsuit to Stop Construction of Delano Prison, SFGate, Jul 11, 2000,
https://www.sfgate.com/bayarea/article/Bay-Area-Activists-File-Lawsuit-to-Stop-2749372.php.
221
See M. Keith Chen & Jesse M. Shapiro, Do Harsher Prison Conditions Reduce Recidivism? A Discontinuity-based
Approach, Apr. 16, 2007, https://www.brown.edu/Research/Shapiro/pdfs/prison041607_web.pdf;
https://thecrimereport.org/2020/04/06/want-to-reduce-recidivism-stop-overcrowding-prisons-researcher/.

59

Electronic copy available at: https://ssrn.com/abstract=4698071