Can Capping Prison Sentences Reduce Mass Incarceration?


Last week, Marc Mauer, executive director of the Sentencing Project, in testimony before the Justice Department’s Charles Colson Task Force on Federal Corrections, said that they should cap sentences to 20 years in prison “except for individuals who present an undue risk to public safety.”

The nine-member Colson Task Force is charged with making the federal prison system safer, less costly and more humane, according to its its leader, former U.S. Rep. J.C. Watts Jr., R-Okla. One of the members is Jim Liske, president and CEO of Prison Fellowship Ministries, a group formed by Colson after his imprisonment for obstruction of justice in the Watergate probe, the Religion News Service reported in December.

Here are Marc Mauer’s Full Remarks


After nearly four decades of sustained growth, prison populations in the United States have stabilized in recent years. This has come about in part as a result of declining crime rates, but is also due to changes in policy and practice, particularly at the state level. Notably, these developments have occurred without an adverse impact on public safety.

As promising as these changes may be, we are a long way from solving our national problem of mass incarceration. The dramatic rise in incarceration during the 1980s and 1990s – coinciding with the launch of the modern War on Drugs – produced an overall prison population that remains unprecedented in world history. At the federal level, the growth in the incarceration rate has been even greater and more sustained than in the states. Addressing these problems will require far more than tinkering with sentencing policy for nonviolent drug offenders or revamping prison programming. To achieve a reasonable level of incarceration, we will need to substantially reduce both the number of people admitted to prison and the length of their sentences.

This testimony addresses the second of these two crucial factors: length of sentence. Excessively punitive sentences have contributed substantially to the nation’s scale of incarceration. They have been counterproductive for public safety objectives and have come at great cost.

Therefore, federal sentencing structures should establish an upper limit of no more than 20 years in prison, except for exceptional circumstances. Doing so would reduce the federal prison population considerably, would avert unnecessary costs for excessive incarceration, and would allow for a more appropriate balance of public safety investments. In addition, because sentencing structures are proportional, such a shift would likely reduce excessive sentences even at lower levels of offense severity, and could also serve as a model for state sentencing structures.


In 1980, about 22,000 individuals were incarcerated in the federal prison system. By 2014, the federal prison population had climbed to nearly 220,000. The result is a federal prison system that today is operating at 40% over capacity.1

As the Urban Institute has documented, the federal prison population is driven by the combination of two factors: volume of admissions and length of time served.2 Of these two factors, sentence length, particularly for federal drug offenses, has been the major contributor to federal prison growth in recent years. According to a study by the Urban Institute, between 1998 and 2010, an increase in time served was by far the leading determinant of growth in the federal prison population, accounting for over half of the increase during that time.3

Addressing excessive incarceration will require reforms that address both the “front end” and the “back end” of the federal sentencing regime. However, for reasons both political and practical, strategies that address the length of prison terms after a sentence has been imposed are limited in their ability to produce large and sustained reductions in prison populations.4 No other policy shift would have the impact of reducing lengthy prison sentences.


Sentencing policies in the United States are much more punitive than in other modern democracies. This can be seen most dramatically at the extreme end of the sentencing spectrum. The United States is virtually the only industrialized nation that employs the death penalty, with over 3,000 people on death row.5 Moreover, the nation’s use of life sentences has expanded exponentially in recent decades, with nearly 160,000 people sentenced to life in prison, or one of every nine people in prison.6 Of this group, almost 50,000 are serving life without parole sentences. Such whole-life sentences are exceedingly rare in other countries. For example, in the United Kingdom, only 49 individuals are serving life sentences with no opportunity for release.7

In the federal criminal code, many statutes authorize life imprisonment, for offenses ranging from drug trafficking to racketeering to crimes involving firearms. At least 45 federal statutes require a life sentence as the mandatory minimum penalty.8 According to the United States Sentencing Commission, 42% of individuals who received a federal life sentence in 2013 had been sentenced under the terms of a mandatory minimum penalty.9 Given the breadth by which severe sentences are imposed, sentencing scholar Michael Tonry notes that “modern laws flatly defy conventional notions of proportionality.”10


Because sentencing systems are proportional, extreme penalties exert an upward pressure on offenses at all levels of severity. According to the American Law Institute, the most severe penalty in a criminal sentencing scheme serves as an “anchor point,” defining the scale of the penalties beneath it. 11

“Arguably, the entire scale of authorized sanctions has some tendency to be stretched upward, or compressed downward, depending on where the absolute maximum is located.”12

As law professor Jonathan Simon has written, extreme sentences — particularly life sentences — have had an “inflationary effect” on the entire structure of punishment and the overall scale of criminal penalties.13 The “high price” for murder (i.e. often death or a life sentence) eases the way for harsher sentences for all manner of less serious offenses such as burglary, larceny, and drug selling.14 Indeed, during the 1980s and 1990s, as the average time served for murder increased by 238%, trends in time served for many other crime types also showed growth.15 It follows that if the most extreme punishments are less severe, punishments for less serious offenses will be less severe as well. Doing so would place the United States closer to the sentencing severity range of comparable nations since “…the United States generally imposes longer sentences on persons sentenced to incarceration than other industrialized nations.”16

These issues are particularly prevalent in the federal system, due to the combination of the presumption of incarceration in the sentencing guidelines structure, the use of mandatory penalties for many drug and gun offenses, the 85% truth in sentencing requirement, and the imposition of life prison terms, even for nonviolent offenses. Each of these features pushes the length of federal prison terms higher.


For a number of reasons, sentences of more than 20 years are largely counterproductive and are extremely costly.

  • Most offenders “age out” of crime. Research shows that after peaking in the mid-to-late teenage years, offending begins to decline as individuals are in their 20s and drops sharply as they reach their 30s and 40s.17 As a result, for each successive year of incarceration there are diminishing returns for crime control. As the National Research Council concluded, “Because recidivism rates decline markedly with age, lengthy prison sentences, unless they specifically target very high-rate or extremely dangerous offenders, are an inefficient approach to preventing crime by incapacitation.”18
  • False promise of long sentences for public safety. Support for extremely long sentences, such as life and life without parole, is premised in part on the assumption that individuals receiving such sentences will reoffend if released. However, an analysis by The Sentencing Project found that individuals released from life sentences were less than one third as likely to be rearrested within three years compared to other formerly incarcerated individuals.19
  • Caring for aging prisoners is extremely expensive. The cost of housing aging prisoners rises substantially, due in large part to increased healthcare costs. Because Medicaid and Medicare funds may not be used to provide treatment to prisoners, healthcare costs generally must be borne by correctional facilities.20
  • Opportunity costs. Long sentences also incur opportunity costs, as they divert resources and attention from other public safety measures. For example, public funds spent on prisons are not available to invest in crime prevention initiatives such as early childhood education, therapeutic interventions for at-risk youth, and treatment for substance abuse and mental illness.


Therefore, federal sentencing structures should generally employ an upper limit of no more than 20 years in prison, except for exceptional circumstances. This upper limit incorporating appropriate exceptions could come about through a variety of mechanisms.

First, Congress could act to reduce statutory penalties of greater than 20 years. Second, short of abolishing mandatory minimum penalties, which drive long federal sentences while undermining justice, fairness, and effectiveness, Congress could prohibit the practice of “stacking” such penalties to achieve sentences exceeding 20 years. The stacking of mandatory minimum penalties is the means by which an individual can be sentenced to 55 years of imprisonment for possessing firearms in connection with selling small amounts of marijuana.21

Next, Congress could enact legislation prohibiting any federal sentence longer than 20 years unless a court has determined that the individual presents an undue risk to public safety. Congress could establish a mechanism for making this determination, with appropriate safeguards to ensure due process. For example, in the event that a public safety exception is found to require a sentence longer than 20 years, the defendant could have an opportunity to present evidence and arguments regarding mitigating factors.

In addition, the United States Sentencing Commission could restructure its sentencing guidelines to create systems that encompass within them a maximum sentence of 20 years for any offense, regardless of the formal maximum statutory limit. All lesser penalties could be reduced to ensure that penalties in the guidelines are proportional.

Finally, all of these measures could be applied retroactively to individuals serving federal sentences. In lieu of retroactive application, the President could exercise the executive clemency power, commuting any federal sentence in excess of 20 years.

Such a structure is not as “radical” as one might think. In many democratic nations, sentences of greater than 20 years are quite rare. For example, in Norway, the maximum sentence for any offense is 21 years, with the possibility of civil commitment for public safety reasons following that. Nations such as Germany, France, and Italy have found life sentences to be unconstitutional.22


The notion that punishment should be no greater than necessary to achieve a legitimate public purpose has been one of the central conceptions of justice shared by Western nations since the Enlightenment. But in recent decades, the concept of parsimony in punishment has been lost. Rather than treating our criminal justice system as an institution whose purpose is to ensure social norms, it has become a system of punishment operating grossly disproportionately and with little regard for promoting effective public safety strategies or individual liberties.

It is time to recalibrate our system of punishment, and there is nowhere better to start than with the federal sentencing regime. By employing an upper limit of no more than 20 years in prison, except for exceptional circumstances, we could establish a new anchor point for proportional sentencing. Such an evidence-based approach would mitigate the counterproductive effects of lengthy sentencing, help control costs, and bring the United States more in line with other industrialized nations. Setting such an upper limit in the federal system could serve as a model for state consideration as well, and therefore contribute to a serious reconsideration of sentencing policy nationally.

About The Author

David Greenwald is the founder, editor, and executive director of the Davis Vanguard. He founded the Vanguard in 2006. David Greenwald moved to Davis in 1996 to attend Graduate School at UC Davis in Political Science. He lives in South Davis with his wife Cecilia Escamilla Greenwald and three children.

Related posts


  1. TrueBlueDevil

    This sure sounds like a one-sided report without some touch points to refer to.

    According to PBS / Frontline, here are the average prison times for various crimes:

    Sexual assault:      35 months
    Homicide:              71 months
    Rape:                    65 months
    All violent crimes: 43 months

    So “life sentences” doesn’t mean much when the average time served is 71 months. Sentences for drug trafficking are stiff – imagine how many lives can be impacted by one drug dealer? Imagine how many young teenagers get hooked on drugs for life, in part due to the role of the drug salesman.

    This speech / report also soft pedals the substantial reduction we have had in violent crime the past 3 decades, and also bypasses the huge increase in violent gang activities.

    There are many parts of the world that are notoriously soft on crime, like Canada, which up until recently had the “age of consent” for sexual relations at 14 years old. It is now 16 after a high-profile case of 31-year-old American Dale Eric Beckham legally shacked up with a 14-year old boy in a motel that he had met on the Internet. Canadian officials could do nothing as the child said it was consensual, but Canadian officials tipped off the American police, they got a warrant, searched his Texas computer, and found loads of child pornography.

    I’m all for reason, and think we should revisit the idea of work camps so that prisoners can help pay for their incarceration and also gain some job skills. If there are really this many non-violent people serving time, have them clean graffiti, build highways, and fight fires.

    1. Davis Progressive

      “So “life sentences” doesn’t mean much when the average time served is 71 months.”

      that’s a dumb comment.  the only way you could get 71 months for a homicide is if you are collapsing involuntary manslaughter in with first and second degree murder – both of which are life sentences in most states.  that’s not a helpful comparison.

      EDIT: I see, you’re pulling from a 1998 article. wow. you might want to look at things like this. so if that’s a 1998, you’re probably looking at 1996 data, which means for instance, three strikes in california was less than two years old and therefore not even really factoring into that data.

  2. Miwok

    I think the real reason the sentencing climbed is the crying victims, mothers and fathers testifying in front of juries and Congress. This is what the Justice system was supposed to avoid, emotional sentencing.

    the problems with letting these convicts out is the recidivism. Like politicians, they cannot promise to NOT do something, like uphold the law, they go right back to doing what they did to get them in prison. If the prison system did their jobs at keeping them incarcerated, they failed at rehabilitation.

    Would that the Prisons would get help for the people they incarcerate, the families and victims of this person, then then it would be a Justice System. Instead you have people who only want to live in peace, yet know there are people who argue every day to let out the person who caused them pain and loss, and the families who either were a part of the crime, or live in the same fear the victims do.

    When they say “you can never go home again”, I wish that would apply to any parolee from even the County Lockup, and now with Prop 42, it should be that way. I don’t think sentences can be capped, I think they need to have that flexibility. But the convicts should have more of a stake in their release. Some are there because it is better than working for a living. That is why I like the comment by TBD,

Leave a Reply

X Close

Newsletter Sign-Up

X Close

Monthly Subscriber Sign-Up

Enter the maximum amount you want to pay each month
Sign up for