The Volokh Conspiracy

Mostly law professors | Sometimes contrarian | Often libertarian | Always independent


Is Mass Incarceration Inevitable? Part 4. Show Me the Money

The financial savings of prison and jail reduction may be less than expected.


[The discussion in this post is excerpted from Andrew Leipold, Is Mass Incarceration Inevitable? 56 American Criminal Law Review 1579 (2019).]

There is no argument for reform that is more persistent and comforting than the claim that reducing incarceration rates will save big money.  Running prisons and jail is expensive—states spent more than $43 billion on prisons alone in 2015—any many have assumed that significantly reducing the inmate population will free up money for teachers, parks, highways and pensions.

There are a host of problems with this assumption.  The main problem is that while the average cost of incarcerating a single prison inmate is relatively high (about $100 per day for federal inmates in 2017), the cost savings of releasing a large number of inmates is much smaller.  More specifically, the cost savings of reducing the number of inmates by (say) 10,000 is not $100/day x 10,000 inmates.

The major costs of running a prison are the same as they are for other service businesses: personnel and infrastructure.  The guards' and administrators' salary, benefits, and pension account for about two-thirds of a prison's expenses, and another significant slice of the budget goes to paying for heat, electricity, and maintaining the physical structure, as well as for debt service and legal judgments.

Most of these costs are fixed and lumpy.  Indeed, if a facility reduces its inmate population by, say, 5%, this may not reduce its budget much at all.  The prisons would still need about the same number of guards and administrative staff, and the buildings would still need heat.  For large savings to accrue, there would need to be a sharp enough decline in the population to allow a meaningful drop in personnel costs, which often would not come until an entire cell block was shuttered, or most dramatically, until an entire prison was closed and the inmates consolidated in the state's remaining facilities.  The political barriers to closing and consolidating facilities in small towns with a unionized staff are obvious and large.

But even if some facility consolidation occurs as the inmate population drops, there are other reasons for governments to be cautious about spending the anticipated financial rewards.  Even if we could substantially reduce the size of prisons and jails and thereby the reduce the department of corrections' budgets, this does not mean that the government as a whole would save money.

Former inmates who will now be released earlier (or who would have gone to prison in a pre-reform era but now will not), often have a hard time finding a job, paying rent, and paying health insurance premiums.  This means they will often call on public resources, through welfare benefits, subsidized housing, and subsidized medical care.  This is particularly true for long-term inmates who are released at an advanced age.  Releasing elderly prisoners might safe and even humane, but this will only save "the government" money if the inmate's subsistence, housing, and increasing health care needs, including nursing home care, are met more efficiently by government programs outside of prison than they would have been inside of prison.

There are reasons to be skeptical that this would be true.  It is hard to imagine that medical care and subsistence cost more in prison than on the outside, and so to the extent former inmate still require financial assistance after their release, a lower inmate population may simply result in cost shifting, not cost savings.  Fewer prison inmates might take some pressure off the state budget, but if these costs are simply shifted to local governments or the federal government this will not help the beleaguered tax payer who is footing all of the bills.

A second obstacle to saving money by shrinking prisons and jails is the one hiding in plain sight.  There is now a widespread understanding that serving time in prison is not itself rehabilitative.  We know that substance abuse, mental health problems, dysfunctional families, and poor education and job opportunities contribute significantly to high incarceration rates.  We also know that prison usually does not solve these problems and may exacerbate them, making these untreated problems a big contributor to high levels of recidivism.

Prisons do what they can, but few believe that current levels of programming and treatment inside of prison are adequate, or that community treatment programs for those on probation or supervised release are sufficient to address the problems that inmates carry with them after release.  Criminal justice reform that is serious about reducing the size of prisons, doing it safely, and then sustaining the gains that are made, recognize the need to invest in additional services, including widely-available drug and mental health treatment.  But these programs are expensive, and at least in the short term, can more than offset any financial savings from a reduced inmate population, even if these investments might at some point lead to long-term reductions in crime.

This is not to say that reducing the size of prisons won't save any money—between 2010 and 2015 several states, in particular, New York and New Jersey substantially cut their prison population and their prison spending.  (Although in a comparable number of states during this same period,  prison spending increased while the inmate population decreased.) But those who think that reducing our current levels of incarceration will lead to near-term, significant economic savings should temper their enthusiasm, and base their justification for prison and sentencing reform on some other source.



NEXT: "An Introduction to Constitutional Law" is back in Stock on Amazon

Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Report abuses.

  1. Just as a counterpoint to the OP, cutting incarceration rates does see some quite measurable immediate drop off in overtime expenses. It is quite difficult to hire and retain corrections officers, and you end up paying a lot of overtime and having a large vacancy rate for your positions inventory.

    But he is right that there are fixed costs that don't go down, and you never see a long term permanent savings in personnel costs unless a "pod" (as they are called, jails/prisons being modular these days) gets closed.

    1. Like he says, unless you close entire blocks or entire prisons, the number of guards doesn't change.

      Furthermore, he does not mention overcrowding. News reports used to be full of stories on how crowded prisons were. Unless new prisons have been built and none are now overcrowded, you'd have to release a lot of prisoners just to get to "normal" levels, and there won't be any block or prison closings until you get way below the "normal" level.

      If prisons are 50% overcrowded now, you'd have to release 1/3 of the prisoners just to get to normal. At 20% overcrowded, you'd have to release 1/6 of the prisoners. And so on.

      1. Yes, I know he says that, I acknowledge it, but you do see an immediate decrease in overtime. These guards are all unionized, and it's a lot of money up front.

        It's entirely possible, to your point, that pods/wings never close because they just go from overcrowded to normal and it takes a lot of reduction to get to the point of closing a wing.

        1. How can there be any reduction in hours worked? They assign guards by the block or prison or cell or even square feet, not by prisoner count. They don't have guards trailing prisoners shackled together in groups of ten.

          A warehouse guard doesn't work more or fewer hours depending on how much is in the warehouse.

          Cops patrol streets regardless of how many houses are unoccupied.

          1. A warehouse is a bad analogy, because more boxes on shelves don't cause more problems other than a lack of room.

            I'm not an expert on this and I work with only a county jail, but the number of guards required to be present depends on the number of prisoners. More prisoners means more guards on a ratio basis, also because less crowding means less problems (violence, write-ups, inspections, etc.). The ratio is not fixed, but depends on the type of prisoner being held and other factors. If you reduce prisoner populations you don't need as many guards present, and you can reduce overtime.

          2. Edit: I work *with* a county jail. I don't work in the jail itself.

  2. I think the idea behind financial savings is that entire prisons would be closed, not just prison populations reduced with the same set of prisons and the same staffing.

  3. What the additional expenses created by felons who commit new crimes after their early release?

    1. A lot, but unfortunately it's unmeasurable, as is accurately predicting recidivism, as discussed in previous parts of this series. It's not like you can make a trade-off decision.

      There isn't really a way I can think of to say: $100 per day to keep this guy locked up for ten years ($365k), but with a 35% change he will re-offend, that means that the $200,000 (a SWAG) in man hours we will spend to prosecute and convict to send him back means that there is a chance we will come out ahead.

      1. Perhaps the decision would become easier if we allowed the third obvious possibility to be considered - reduce the cost of imprisonment from $100 a day to $20 a day.

        I'm very confident that, say, Ukraine could happily make a profit from a $20 per prisoner per day fee, what with lower labor costs. Moreover prisoners would hardly ever bother to try escaping since they would get about 50 feet before their lack of fluency in Ukrainian would sink their escape.

  4. Prof. Leipold, you have a typo in your headline:

    In Mass Incarceration Inevitable? ....

    Should read:

    Is Mass Incarceration Inevitable? ...."

  5. Medical care in prisons is tremendously expensive.
    Not only do you have to pay high salaries to get medical personnel under awful working conditions (I remember a quotation years ago in an article by a prison physician that "my patients try to bite me as often as a vet's do"), you also have correctional officers that must be assigned to the patients.

Please to post comments