Researchers, policymakers, and the public rely on a variety of statistics to measure how society punishes crime.
Among the most common is the imprisonment rate—the number of people in prison per 100,000 residents.
This metric allows for comparisons of prison use over time and across jurisdictions and is widely seen as a proxy
for punishment. States with high imprisonment rates, for example, are considered more punitive than those
with low rates.
A more nuanced assessment of punishment than the ratio of inmates to residents is that of inmates to crime—
what The Pew Charitable Trusts calls the “punishment rate.” This new metric gauges the size of the prison
population relative to the frequency and severity of crime reported in each jurisdiction, putting the imprisonment
rate in a broader context.
Using the punishment rate to examine the U.S. criminal justice system, Pew found that all states became more
punitive from 1983 to 2013, even though they varied widely in the amount of punishment they imposed. The
analysis also shows that the nation as a whole has become more punitive than the imprisonment rate alone
indicates. (See Figure 1.)
How the punishment rate works
Assessing the relationship between crime and punishment is inherently challenging. Many crimes go unreported
and, as a result, unpunished. Reported crimes often do not result in arrest or prosecution, and those that do
frequently do not lead to conviction and imprisonment. As a measure of punitiveness, the imprisonment rate
captures only the small subset of offenders who are apprehended, prosecuted, convicted, and sentenced to
prison—and it reflects only the ratio of those inmates to residents, rather than to overall crime. The punishment
rate, by contrast, calculates the use of prison relative to the frequency and severity of reported crime in each
jurisdiction by combining two data points:
- The imprisonment rate is the count of inmates sentenced to a year or more behind bars per 100,000 residents
of a jurisdiction, as reported by the federal Bureau of Justice Statistics. This is the traditional threshold for
a felony offense. It excludes those who are awaiting trial and those who committed lower-level offenses for
which crime data are limited. (Misdemeanor charges, by contrast, typically result in sentences of probation
or less than a year in local jails.)
- The severity-weighted crime rate captures the frequency and seriousness of reported crime per 100,000
residents, as measured by the seven specific offenses for which reliable, national data are available and that
the FBI classifies as Part I offenses: criminal homicide, rape, robbery, aggravated assault, burglary, larcenytheft,
and motor vehicle theft. To account for differences in the severity of the offenses, Pew constructed a
scale that assigns a weight to each crime according to the average period of imprisonment served by offenders
convicted and sentenced for it. For example, because robbery offenders serve longer prison terms than larceny
offenders, the former is weighted more heavily than the latter.
Although the imprisonment rate counts drug offenders—who make up nearly a fifth of the state prison
population—drug crimes are not counted among the FBI’s Part I offenses, so the severity-weighted crime rate
does not include them.
The punishment rate is not intended to replace the imprisonment rate as an analytical tool, nor should it be
viewed as an assessment of specific corrections policies or practices. Instead, it provides a new lens through
which policymakers, researchers, and the public can view the use of prison relative to reported crime—and the
ways in which that relationship has changed over time.
A more punitive nation
As measured by the punishment rate, the U.S. became 165 percent more punitive from 1983 to 2013. Changes
in the two components of the punishment rate—the imprisonment rate and the severity-weighted crime rate—
explain this trend:
- The imprisonment rate soared over the 30-year period as lawmakers increased criminal penalties, eliminated
parole, and made other policy changes that collectively sent more offenders to prison and kept them there
longer.1 The effects of these changes can be seen in the average amount of time that Part I offenders served,
which was much higher among those released in 2013 than those released in 1983.2 Other practices also
played a significant role in the nation’s prison buildup: Research suggests, for instance, that during this span
prosecutors increasingly pursued felony charges, which led to more offenders being sentenced to at least a
year in prison.3
- Changes in the severity-weighted crime rate also played a
critical role in the nation’s rising punishment rate. Violent and
property crime rates rose sharply from 1983 to 1991, prompting
many of the policy changes referenced above, including the
establishment of mandatory minimum sentences and laws
requiring offenders to spend most of their court-imposed
sentences behind bars.4 But crime rates have dropped by
more than half since 1991, reaching levels not seen since
the 1960s, even as the imprisonment rate has remained far
above historical levels.5 In other words, the United States
has responded to declining crime with the same high level
of imprisonment, leading to a significant increase in the
punishment rate, particularly in recent years.
To assess relative punitiveness across jurisdictional lines, Pew
calculated punishment rates for all 50 states using data from
2013, the most recent year available. Mississippi’s rate of 818, the
highest in the nation, was 71 percent greater than the U.S. average
of 477, while Maine’s rate of 231, the country’s lowest, was less
than half of the U.S. average. (See Table 1.)
Many states had similar rankings in their imprisonment and
punishment rates. Michigan, for instance, ranked 19th among
the states in its imprisonment rate and 16th in its punishment
rate in 2013. Nebraska was 39th in imprisonment and 40th in
However, many other states differed by several positions in the
rankings, and 17 states varied by 10 or more slots. Connecticut,
New Jersey, New York, Pennsylvania, South Dakota, Vermont,
Virginia, West Virginia, Wisconsin and Wyoming each ranked
much higher in their punishment rates than in their imprisonment
rates. In other words, these states punished crime significantly
more than their imprisonment rates show. The opposite was
true for Alaska, Arkansas, Georgia, South Carolina, Tennessee,
Nevada, and New Mexico.
The Complex Interplay of
Crime and Imprisonment
Criminologists attribute some
of the U.S. crime decrease
since 1991 to the nation’s
sharp rise in imprisonment.
But they attribute most of the
reduction to other factors,
including better policing, the
aging of the population, and
advances in crime prevention
technology.* Many experts
also argue that the increase in
imprisonment has reached a
point of diminishing or even
negative public safety returns.
The National Research Council,
for example, concluded in a
comprehensive 2014 report
that “statutes mandating
lengthy prison sentences
cannot be justified on the
basis of their effectiveness
in preventing crime.”
At the same time, research
shows that policy choices—
not crime rates—drive
States that imprison a large
share of their residents, for
example, can have a high
crime rate or a low one; the
same is true for states that
lock up a small proportion
of their populations.
To assess how state punishment rates have changed over time, Pew
calculated trends for all 50 states from 1983 to 2013. The rates grew in
all states during this period, with Colorado’s 417 percent spike by far the
largest; North Carolina’s rate rose the least, at 17 percent. (See Table 2.)
The analysis again shows that some states became far more or less
punitive than their imprisonment rates suggest—changes that could be
the result of decreasing crime, increasing penalties, or a combination of
the two. The punishment rates for New York and New Jersey, for example,
dramatically increased, but those states’ imprisonment rates grew
much more slowly. Conversely, in Alabama, Arkansas, and Tennessee,
imprisonment rates rose much more than did punishment rates.
more than their
The long-term rise in U.S. imprisonment is a familiar story. Although the imprisonment rate is an essential
tool in understanding correctional trends, it paints an incomplete picture of the nation’s and individual states’
punitiveness because it does not take crime rates into account. The punishment rate provides a more nuanced
assessment by placing each jurisdiction’s imprisonment rate in the context of the severity and frequency of
Analysis of punishment rates over time and across jurisdictions makes clear that the nation has become more
punitive. What’s more, many states punish crime significantly more—or less—than their imprisonment rates
alone indicate. States with particularly high or low punishment rates and those that experienced significant
increases in their punishment rates over time may benefit from identifying and examining the policy choices
responsible for their rankings and trends.
Pew calculated punishment rates for the 50 states and the nation as a whole with imprisonment rate data from
the annual “Prisoners” series published by the Bureau of Justice Statistics (BJS) and crime rate statistics from
the annual “Uniform Crime Report” series published by the FBI. Both rates are per 100,000 residents. To develop
a scale with which to determine the relative severity of each Part I crime, Pew analyzed time-served data, by
offense, from BJS’ National Corrections Reporting Program. All data are from 1983 to 2013.
The punishment rate counts only the state prison population—not the total incarcerated population, which
includes jail inmates—because the seven crimes tracked in this analysis are more likely to result in prison
sentences than jail terms. In addition, it counts only inmates who were sentenced to more than a year in prison—
the traditional threshold used to define a felony offense—to ensure greater data consistency across all states,
including those with combined prison and jail systems. Further, the punishment rate captures only the FBI’s seven
Part I offenses, because they are most likely to be reported to police and, as a result, to be tracked consistently
in all states. Although Part I offenses omit many crimes that can lead to a term of imprisonment—most notably,
drug-related offenses—they represent the best available measure of crime on an annual state-by-state basis. All
jurisdictional notes listed in BJS’ “Prisoners” series and in the FBI’s “Uniform Crime Report” series apply to this
Pew used data from the National Corrections Reporting Program (NCRP) annual release files from 1983 to 1999
and the term record files from 2000 to 2013 to estimate average time served for each Part I crime and each year.
Pew’s estimate of average time served is for offenders considered “first releases”—that is, those who met the
following conditions: Their prison sentence was for one year or more; they were sent to prison for a new crime
rather than a supervision violation, categorized in the NCRP data file as a “new court commitment” or “imposed
suspended sentence”; and their release from prison was not due to death, transfer, appeal, or detainer.6 Pew used
average time served by released offenders rather than expected time served by those in prison to account for
variations in state sentencing and release policies that can cause data irregularities.7 Average-time-served data
are also more conservative than estimates of expected time served because they are based on offenders released
from prison and, as a result, over-represent those with short sentences.
All cases in the data files used for this analysis were divided into seven offense categories: criminal homicide,
rape, robbery, aggravated assault, burglary, larceny-theft, and motor vehicle theft, based on the NCRP code books
for the years being evaluated. (A list of offense codes from the 2013 book is available upon request.) Inmates
were placed in an offense category based on the crime for which they received the longest sentence.
Pew calculated total average time served by adding any jail time that was credited toward the current sentence
to the prison time served for that sentence. If prior jail time was missing from the data, Pew estimated it as the
average jail time for that year and offense category.
The NCRP is a voluntary federal data collection program; not all states participate. For the years represented
in this data analysis, 25 to 35 states provided viable time served information. Pew used a centered moving
average—one year before, the current year, and one year after—to account for fluctuations in states’ yearover-
year reporting practices. Pew calculated average time served for 1983 and 2013 using a two-year moving
average (1983 and 1984, and 2012 and 2013, respectively) because data were not available for 1982 or 2014.
The formula used to calculate the punishment rate is:
Imprisonment Rate y / (Murder Rate y * Murder Wt y) + (Rape Rate y * Rape Wt y) +
(Robbery Rate y * Robbery Wt y) + (Assault Rate y * Assault Wt y) + (Burglary Rate y * Burglary Wt y) +
(Larceny Rate y * Larceny Wt y) + (Motor Vehicle Theft Rate y * Motor Vehicle Theft Wt y)
The subscript y in the formula is an indicator for “year” in that the imprisonment and crime rates as well as
weights change each year. The seven weighted crime rates are added together in the denominator of the formula.
As noted above, the formula weights each offense by its relative severity, based on the average time served by
inmates sentenced to prison for each of the seven Part I offenses.
This weighting mechanism is important because it ensures that the punishment rate does not treat all offenses
equally, given important differences in the severity of each crime. Without the weighting mechanism, a ratio of
imprisonment to crime would essentially treat each crime’s potential impact on imprisonment as the same—even
though some offenders serve much longer terms than others.
This study benefited from the insights and expertise of Kristofer “Bret” Bucklen, Ph.D., director of the Bureau
of Planning, Research, and Statistics at the Pennsylvania Department of Corrections; James P. Lynch, Ph.D.,
professor and chair of the Department of Criminology and Criminal Justice at the University of Maryland,
College Park; and Jeremy Travis, Ph.D., president of the John Jay College of Criminal Justice. Pew also recognizes
Mark A.R. Kleiman, Ph.D., professor of public policy at the Marron Institute of Urban Management at New
York University, who provided consultation and whose 2009 book, When Brute Force Fails: How to Have Less Crime
and Less Punishment, provided the inspiration for the punishment rate. Although they have reviewed the analysis
and methodology, neither they nor their organizations necessarily endorse the findings or conclusions.
- National Research Council, The Growth of Incarceration in the United States: Exploring Causes and Consequences (2014),
- Pew analysis of data from the U.S. Bureau of Justice Statistics’ National Corrections Reporting Program, 1983-2013.
- John F. Pfaff, The Causes of Growth in Prison Admissions and Populations (July 2011), http://papers.ssrn.com/sol3/papers.cfm?abstract_ id=1884674.
- For changes in violent and property crime rates from 1983 to 1991, see FBI Uniform Crime Reporting Program,
http://www.ucrdatatool.gov/. For policy changes made in response to rising crime, see National Research Council, The Growth of
Incarceration in the United States.
- Ibid. For historical levels of imprisonment, see Bureau of Justice Statistics, Key Statistics, Estimated Number of Persons Under
Correctional Supervision in the United States, http://www.bjs.gov/index.cfm?ty=kfdetail&iid=488.
- Prison inmates held on detainer are awaiting transfer to another agency or jurisdiction (e.g., federal custody).
- Expected time served is an alternative way to estimate time served by calculating the length of stay for those in prison during a particular
year. For a more detailed explanation of this approach, see The Pew Charitable Trusts, Time Served: The High Cost, Low Return of Longer
Prison Terms (June 2012), 21–22, http://www.pewtrusts.org/en/research-and-analysis/reports/2012/06/06/ time-served-the-high-costlow- return-of-longer-prison-terms.