CELLS FOR SALE:

UNDERSTANDING PRISON
COSTS & SAVINGS
A Report From

Policy Matters Ohio

Bob Paynter
April, 2011

Author
An investigative reporter and editor for nearly 25 years in Akron and Cleveland, Bob Paynter
took an early retirement buyout from The Plain Dealer in 2008. He continues to investigate,
research and write about matters affecting public policy on a contract basis.

Policy Matters Ohio is a nonprofit, nonpartisan research institute dedicated to
researching an economy that works for all in Ohio. We seek to broaden debate about economic
policy by providing research on issues that matter to Ohio’s working people and their families.
Areas of inquiry for Policy Matters include work, wages and benefits; education; economic
development; energy policy; and tax policy.

Preface to the Report
A few months ago, Policy Matters Ohio asked journalist Bob Paynter to take a
deep look at Ohio’s current record on prison privatization. Though we had no way of
knowing that Governor John Kasich’s first budget would propose selling five publicly
owned prisons, we thought it likely that prison privatization would be considered.
Paynter’s excellent analysis of the cost issues in private prisons follows.
To be clear, Policy Matters believes that there are many reasons to be concerned
about prison privatization and to be skeptical even if cost savings exist. If “savings”
result from higher prisoner-to-guard ratios, more crowding, less psychiatric or medical
care, lower hourly pay for guards, less spending on training, or less spending on prison
education, food, or recreation, they will likely cost us more in the long run. All of these
can result in lower quality supervision, more morale problems, or overall declines in
quality. This in turn can mean less rehabilitation, more recidivism, and possible increases
in tension within the facility.
Second, accountability and transparency are significantly lower in private
companies than in public entities. Incarceration is, by definition, a tremendous incursion
on liberty. The isolation and tarnished reputation of inmates makes them prime targets for
abuse. Thus, this is an area of government where we should want the highest levels of
transparency and accountability, even if that results in slightly higher costs.
Third, private prisons must have room for profit. This can result in perverse
incentives, where the interest of the operator is to maximize the number of inmates. That
is not the same interest as those of Ohioans. In addition, if compensation of guards is
lower, then Ohio communities are receiving less of the return for this state spending,
increasing what economists call “leakage”. In other words, if pay is reduced but profits
for out-of–state operators are increased, the money is not staying and circulating in the
Ohio economy. This reduces the positive impact of state spending, making it less helpful
to Ohio’s beleaguered communities.
The above concerns are all reasonable to raise about publicly-owned, privatelymanaged prisons. The concerns intensify when we consider outright private ownership.
Once prisons are sold or privatized, there is reason to worry that the state will have
reduced bargaining power to change course or switch vendors, giving private operators
more leverage and more ability to garner excessive profits.
For these reasons and more, critics are rightly skeptical about prison
privatization. A demonstration of lower cost is not sufficient reason to give over this most
sensitive government function to private, profit-making companies. This study finds that,
in fact, the cost calculations performed over a number of years by the State of Ohio have
not reliably demonstrated the private-prison savings required under Ohio law. This
history makes it difficult to have confidence in new claims of cost savings. Policy
Matters recommends that policy makers reject the proposal to sell Ohio prisons to private
bidders. We encourage you to read this detailed examination.
-- Amy Hanauer,
Executive Director

Cells for Sale: Understanding Prison Costs & Savings

Policy Matters Ohio

Executive Summary
Ohio taxpayers have purportedly saved more than $45 million over the last decade from the
private operation of two state prisons. Now, with the substantial expansion of prison privatization
recently proposed by Republican Gov. John Kasich, they’re being promised tens of millions more in
savings over the next one.
But if past is prologue, there is little reason to be confident in such promises.
Since the first private prisons were opened here in 2000, Ohio law has required that any
private operator produce savings of at least 5 percent compared to what it would cost the state to run
the same facility. And if you believe past state pronouncements, the savings have exceeded that
threshold by a factor of three.
But a closer look reveals that, as of last month – 10 years into the privatization experiment,
and one day before Kasich’s announced expansion – Ohio officials still hadn’t developed an
accurate, reliable way to compute how much, if anything, they had actually saved.
Their methods for calculating the seemingly robust savings have changed substantially and,
in some case, inexplicably over the last several biennia. A detailed examination of those
calculations shows them not only to be riddled with errors, oversights and omissions of significant
data, but also potentially tainted by controversial accounting assumptions that many experts
consider deeply flawed.
Acknowledging that previous efforts were both inconsistent and imprecise, state officials
more than two years ago started re-tooling those cost-savings calculations, beginning with the 20102011 biennium that’s now coming to a close. And in preparation for the coming biennium, they
started instituting even more fundamental alterations last fall – many of which, had they been
instituted in earlier calculations, would have raised significant red flags about the savings Ohio has
been touting over the years.
Even last month, on the eve of Kasich’s proposal to sell five prisons – which, in a stroke,
would triple the number of Ohio prisoners under private supervision – numbers crunchers at the
Ohio Department of Rehabilitation and Correction (ODRC) were still tinkering with the costsavings calculations in fundamental ways, saying they hoped a new formula could be proposed and
approved by the state budget office sometime this month (April).
Work on a reconstituted formula may well be ongoing. (Officials, promising transparency in
the revision process, said the new methodology would appear on the ODRC website upon
completion and approval. As of April 15, it had not appeared.)
On April 6, the state posted its bid request for the purchase and operation of the five prisons.
And even though the savings formula was still being modified when the sale was proposed,
ODRC’s chief budget analyst, Kevin Stockdale, said he is confident the private-prison savings are
there, “even without all the decimal places double checked.”
But an examination of the calculations the state has relied on to proclaim those savings since
2006 shows that – once apparent errors are corrected and revisions being proposed by ODRC are
made – it’s far from clear that Ohio’s private prisons are producing the savings required by law. Or
that they ever have.
In fact, revised calculations suggest that it may actually have cost taxpayers MORE to
contract with a private vendor for operations at the Lake Erie Correctional Facility in Conneaut for
fiscal years 2006 and 2007 than to have allowed the state to run it. And for both the 2008-09 and the

i

www.policymattersohio.org

Policy Matters Ohio

Cells for Sale: Understanding Prison Costs & Savings

2010-11 biennia, revised calculations for Lake Erie project savings that appear to fall below – and
in some years, well below – the 5 percent minimum required by state law.
If that’s been the case, and the touted savings have been largely illusory, then Ohio’s 10year-experiment with prison privatization becomes considerably less successful than its billing,
leaving little more than blind ideology or political preference to justify the dramatic expansion now
underway. And without a strong fiscal rationale, critics’ concerns about the effectiveness and
security of private prisons – underscored by dramatic events last summer in Arizona, where three
inmates escaped from a private prison near Kingman and allegedly murdered a couple, take on that
much more weight.
Stockdale and his fellow number crunchers at ODRC, working with the state’s office of
budget and management, may still devise and publicize a formula that documents the savings that
state law requires. That they haven’t previously is both telling, given the claims the state has made,
and not terribly surprising.
Nationally, there is little or no consensus about whether private prisons save taxpayers much
of anything. The private corrections industry and its allies claim substantial savings – in the range of
5 to 15 percent a year. Skeptics say there is little credible evidence for that. What independent
research there is tends to be ambiguous.
The hunt for these savings is also politically charged – with Republicans and Democrats and
their supporters tending to land on opposite sides of the debate. And it’s an exceedingly difficult
task to begin with, because it essentially involves comparing the operating costs of a real, privately
operated prison with a make-believe facility, identical in every way to the private one, except that
it’s staffed by state employees and run by the state. That’s what Ohio has been trying to do with its
various calculations.
But a host of assumptions can crop up in such analyses, some of them quite controversial,
which can tip the scale one way or the other depending on how they’re handled. For instance, how
should one account for the corrections department’s central office expenses? And how should the
state’s payments for inmate labor in the private prison be handled – or reimbursements to the
private vendor for outsized medical costs?
In interviews as recently as the day before Kasich’s prison-sale announcement, Stockdale
said these are among the items the state is now revisiting, in addition to trying to clean up the
inconsistencies, oversights and imprecision in past calculations.
The problem is that once past errors in the state calculations are corrected and revisions
made, the private-prison savings computed by the state over the years appear to shrink dramatically.
For the 2006-07 biennium, for instance, when apparently distorted staffing and overhead
estimates are revised in the calculation, the computed savings of $2.4 million for each of the two
years becomes a COST of between $380,000 and $700,000 a year.
For the 2008-09 biennium, potential savings of as much as 21 percent a year shrivel to
between 1.2 percent and -0.3 percent when errors are corrected and state-proposed revisions made,
depending on how state overhead is handled. For 2010, the computed savings could drop from 13.9
percent to 3.6 percent, and for 2011, from 15 percent to 4.7 percent. In each year, once proposed
revisions are made to the state’s calculations, the computed savings could fall below the 5 percent
required by law.
At the very least, that should give taxpayers pause. It may eventually turn out that private
prisons do save money. But after 10 years of claiming it, Ohio has fallen well short of proving it.
www.policymattersohio.org

ii

Cells for Sale: Understanding Prison Costs & Savings

Policy Matters Ohio

Cells for Sale: Understanding Prison Costs & Savings
If you believe what you hear from the state of Ohio, then more prison privatization is a fiscal
no-brainer.
For the last decade, the state has consistently claimed annual savings in the 10-to-17 percent
range for the operation of its two privately run prisons – Lake Erie Correctional Institution, in
Conneaut, and North Coast Correctional Treatment Facility, in Grafton. Both have been operated under
contract by Management and Training Corporation (MTC) of Centerville, Utah since 2001 (Lake Erie,
a year earlier). And, if you believe the state, they have combined to generate savings in excess of $45
million for taxpayers compared to what the state would have spent to run them itself. 1
What’s not to like about that?
Even before last fall’s election, when a pre-disposition toward downsizing government was
firmly established in Columbus, ferment for more privatization was building.
Nationally, from 2000 to 2009, the number of state and federal inmates in private prisons had
grown by 48 percent, to nearly 130,000 – about 8 percent of all prisoners in the U.S. 2 With an
established footprint in at least 32 states, private corrections had become a $22.7 billion industry,
Fortune Magazine reported last summer, with an annual growth rate of 4.7 percent and an optimistic
outlook for the future. 3
Citing Ohio’s "successful 10-year experiment" with private management of two facilities, a pair
of state senators introduced legislation last June to privatize at least half of the state’s 31 prisons,
arguing the move could save between 10 and 20 percent a year. 4 The bill didn’t go far.
But with the virtual clean sweep by Republicans at the polls in November, and the $8 billion
hole opened in Ohio’s budget for the next biennium by the state tax cuts approved in 2005 and the
financial meltdown of 2008-2009, a surge in prison privatization seemed all but inevitable.
So, on March 15, when Gov. John Kasich proposed the outright sale of five state-owned
prisons, including the two already operated privately, the only real surprise was in the scope of the
proposal – which in a single stroke would nearly triple the number of Ohio inmates under private
supervision – and even that was mild. The idea just seemed to make sense:
-

On ideological grounds – to those for whom private enterprise is always better and more
efficient than government.

-

On political grounds – to those who find far more support in business circles than in the
public-employee unions most likely to be hurt, and

-

Most importantly, on financial grounds. Not only would the proceeds from the sale –
pegged at $200 million, but estimated to provide only $50 million to the General Revenue
Fund budget – help with immediate budget problems. But, given the requirement in Ohio

1

Ohio Legislative Service Commission (Joseph Rogers, Senior Budget Analyst), “Privately Operated State Prisons save
$5.5 million in FY 2010,” Budget Footnotes, July 2010, page 30.
2
Heather C. West, William J. Sabol and Sarah J. Greenman, “Prisoners in 2009,” Bureau of Justice Statistics, page 34,
December 2010 at http://bjs.ojp.usdoj.gov/content/pub/pdf/p09.pdf
3
D. M. Levine, “What's costlier than a government run prison? A private one,” Fortune Magazine, Aug. 18, 2010 at
http://money.cnn.com/2010/08/17/news/economy/private_prisons_economic_impact.fortune/index.htm
4
Jon Craig, “Privatization Sought for Ohio Prisons,” The Cincinnati Enquirer, June 21, 2010, page B1.
1

www.policymattersohio.org

Policy Matters Ohio

Cells for Sale: Understanding Prison Costs & Savings

law that any private prison must produce a 5 percent savings compared to state operations,
the move would arguably save taxpayers a minimum of $55 million over the next decade. 5
But it’s over this fiscal argument – the only legitimate rationale from a policy perspective – that
deep skepticism is in order. If Ohio is basing the promise of future savings from private prisons on its
evaluation of their past performance, then there is little reason for confidence in that promise.
The state’s method for calculating the hefty savings it has claimed for private prisons does not
stand up well to scrutiny. It has shifted substantially and in some cases mysteriously and inexplicably,
from biennium to biennium. And a detailed examination of the calculations over the last six years
shows them to be riddled with errors, oversights and omissions of significant data and potentially
tainted by controversial accounting assumptions that many experts consider deeply flawed.
Acknowledging that previous efforts were both inconsistent and imprecise, state officials
decided on their own to re-tool the cost-savings calculations for the 2010-2011 biennium that’s now
coming to a close. And last fall, in preparation for the private-prison contract renewal for the coming
biennium, they started instituting even more fundamental alterations in the calculations 6 – changes that
were still being proposed and considered last month, even as Kasich was unveiling his proposed prison
sale.
On March 14, the day before the governor’s announcement, the budget chief for the Ohio
Department of Rehabilitation and Correction (ODRC) disclosed in an interview that a fundamental
proposed change was under review in the way department overhead should be addressed in the
calculation – a change that would substantially reduce the savings projected for the private prisons –
and that a new method for calculating cost savings had neither been finalized nor approved.
Asked if, in an ideal world, the method to calculate privatization savings ought to be in place
before a major expansion in privatization is proposed based on those very savings, ODRC budget chief
Kevin Stockdale replied:
“Well, we don’t live in that world.” 7
Stockdale said he transferred to ODRC from the state budget office in September of 2008 and
immediately saw problems with the way the private-prison savings projections were done for the two
previous biennia – 2006-2007 and 2008-2009. “It wasn’t very precise in the way it was done,”
Stockdale said. “It was not consistent. As an outsider, it sort of shocked me. So we decided to change
it.”
Stockdale said he has no stake in the privatization argument one way or the other. But despite
the continuing flux in the methodology, he said that he and his department superiors, not to mention
the state’s political leaders, are convinced that private-prison savings are real.
“I didn’t come into this with a pre-disposition either way,” Stockdale said. “But after looking at
the numbers, I’m confident that … even without all the decimal places double checked, the savings are
definitely there. At least in my opinion.”
5

Timothy S. Keen, Ohio Office of Budget and Management, “The Savings Book,” State of Ohio Executive Budget, Fiscal
Years 2012 and 2013, page 68 at http://www.scribd.com/doc/50798366/Ohio-2012-13-Budget-Book-Four-The-SavingsBooks
6
Interview with Kevin Stockdale, Chief, Budget Planning and Analysis, Ohio Department of Rehabilitation and Correction,
conducted Feb. 24, 2011.
7
Interview with Kevin Stockdale, Chief, Budget Planning and Analysis, Ohio Department of Rehabilitation and Correction,
conducted March 14, 2011.
www.policymattersohio.org

2

Cells for Sale: Understanding Prison Costs & Savings

Policy Matters Ohio

But if they are, the state has fallen far short of proving it.
For simplicity’s sake, this examination will focus on the calculations as they apply to Lake Erie
Correctional Institution, a medium security prison with similarities to several of Ohio’s state-run
facilities. (The cost-savings calculations follow the same pattern with the North Coast facility in
Grafton. But that facility was designed especially for felony DUI offenders and its unique mission
makes it much more difficult to compare to others.)
In a pair of recent interviews, Stockdale said he has proposed significant changes to virtually
every phase of the calculations for the Lake Erie facility. Several apparent errors have been discovered
as well. But, once the errors are corrected and proposed changes made to the calculations, it’s not at all
clear that Lake Erie is producing the 5 percent savings required by law. Or that it ever has.
In fact, in fiscal years 2006 and 2007, revised calculations suggest that it may actually have
cost taxpayers MORE to contract with MTC for operations at Lake Erie than to have allowed the state
to run it.
If that’s the case, and the touted savings have been largely illusory, then Ohio’s 10-yearexperiment with prison privatization becomes something less than “successful,” leaving little more
than blind ideology, or political preference – or the prospect of the one-time injection of cash from an
initial sale of assets – to justify further ventures in that direction.
And absent a strong fiscal rationale, concerns about the effectiveness and security of private
prisons take on more weight.
National research
The cost question is unresolved in the national discussion as well.
While debate over prison privatization has been heated and divisive, there is little or no
consensus on whether it actually saves money and, if so, how much. Industry leaders typically claim
savings of from 5 to 15 percent, sometimes more, for a private versus public facility, 8 but the research
– such as it is – has been highly ambiguous and often seems more emotional than analytical.
“A lot of the arguments are very hypothetical and are more based on theory than on fact, more
on ideology than anything,” said Dr. Gerry Gaes, director of research at the federal Bureau of Prisons
from 1988 to 2002 and a visiting scientist with the National Institute of Justice, the research,
development and evaluation agency of the U.S. Department of Justice, from 2002 to 2007. 9
Given the stakes and the fervor of the debate, one would expect the issue to have been
thoroughly vetted by now. But Gaes’s own review of the literature last year found very few studies that
compared privately and publicly operated facilities. Many of the studies that were found were “of
questionable value” and often contained widely differing conclusions about what the research actually
shows.
For instance, Gaes noted a triangular dispute over privatization costs several years ago in
Florida that involved the state department of corrections, a private prison contractor and the Office of
Program Policy Analysis and Government Accountability, a research arm of the state legislature. “All
three produced separate and contradictory results regarding the per capita costs of public and private
facilities in the state. The Florida Department of Corrections showed that the private contractors were
8

Fortune, note 3
Interview with Dr. Gerald G. Gaes, director of research, Federal Bureau of Prisons from 1988 to 2002; visiting scientist at
National Institute of justice, 2002 to 2007, conducted March 23, 2011.
9

3

www.policymattersohio.org

Policy Matters Ohio

Cells for Sale: Understanding Prison Costs & Savings

more expensive than the public operators. The private contractor showed that the privately operated
facilities were less costly than the public counterpart. OPPAGA showed that the costs of the publicly
and privately operated prisons were about the same.”
His conclusion on the state of the research? “Direct comparisons of cost and quality neither
favor the public nor the private sector.” 10
In 2007, researchers at the University of Utah reached essentially the same conclusion: “The
intensity of the debate suggests that a clear, obvious choice about privatizing prisons does not exist.
Eight of the 12 studies … (that were sufficiently rigorous to review) provided information on cost
effectiveness. Half of these eight revealed that privately managed prisons outperformed publicly
managed prisons, with a range of cost effectiveness from 4.6 percent to 15.2 percent. Of the remaining
four studies, two showed that publicly managed prisons were more cost effective than their privately
managed counterparts (10.0 percent and 14.2 percent). The remaining two studies revealed a statistical
tie: that is neither system outperformed the other. Our conclusion is that prison privatization provides
neither a clear advantage nor disadvantage compared to publicly managed prisons. Cost savings from
privatization are not guaranteed and quality of services is not improved.” 11
Because of political and ideological forces that often drive them, government analyses typically
have not been helpful, others have found. Not only have governments generally neglected to develop
independent estimates of what privatization would actually cost before deciding to do it, they also have
typically failed to follow consistent accounting procedures in evaluating the results. 12
If privatization decisions were left to public corrections officials, and if such decisions were not
so politically contentious, Abt Associates concluded in a 2003 privatization study that most officials
could probably be counted on to do the analysis necessary to identify the most cost-effective course to
follow. “But most prison privatization decisions are initiated not by correctional administrators but by
their political masters – the governors or legislatures,” Abt found. “And these decisions are often
surrounded by much politicking by private firms and public employee union representatives who feel
their jobs may be at stake.” 13
Complicating the process further, according to Abt, is the inherent difficulty of trying to
estimate what the government would have spent to operate a facility that’s actually being run by a
private firm – or trying to quantify what Abt describes as “the course not taken.” 14 That’s exactly what
Ohio has tried to do.
An exercise in make believe
Since no state-run facility is directly comparable to Lake Erie Correctional (or the North Coast
facility, for that matter), the state’s calculations essentially involve an exercise in make believe:
Creating a hypothetical prison – identical in every respect to that run by the private company except
that it’s fully staffed and operated by the state, and then calculating what this make-believe prison
10

Gerald G. Gaes, “The Current Status of Prison Privatization Research on American Prisons,” August, 2010 at
http://works.bepress.com/cgi/viewcontent.cgi?article=1000&context=gerald_gaes&seiredir=1#search=%22current+status+of+prison+privatization+research+on%22
11
Lundhal, Brad W., Kunz, Chelsea, Brownell, Cyndi; Harris, Norma; Van Vleet, Russ (2009). "Prison privatization: A
meta-analysis of cost and quality of confinement indicators." Research on Social Work Practice, vol 19(4), July 2009, pp.
383-394, or at http://ucjc.law.utah.edu/wp-content/uploads/86.pdf
12
Douglas McDonald and Carl Patten, Abt Associates Inc., “Governments' Management of Private Prisons,” Sept. 15, 2003
for the National Institute of Justice at http://www.ncjrs.gov/pdffiles1/nij/grants/203968.pdf
13
ibid.
14
ibid.
www.policymattersohio.org

4

Cells for Sale: Understanding Prison Costs & Savings

Policy Matters Ohio

would cost to operate per inmate per day compared to what the state is paying MTC to run the actual
one.
As Abt noted in 2003, and Gaes and others have noted since, such analyses of hypotheticals are
vulnerable to controversial accounting assumptions that can radically skew the results. 15 What, for
instance, is an analyst to do about governmental overhead expenses, which often are figured as high as
10–15 percent of a prison’s direct operating costs? In Ohio’s case, how do you handle Lake Erie’s
share of ODRC’s expenses for central office functions like planning, research and administration, to
name just a few?
Are such expenses actually “saved” by the act of privatization, which is how Ohio has treated
them in past calculations? Or, are they the ongoing costs of running a prison system that taxpayers still
must shoulder whether Lake Erie is public or private and, thus, not really savings at all? As will be
seen, the difference in the two approaches can be huge.
“Whenever you compare a real facility against a hypothetical facility, this becomes an issue,”
said Travis Pratt, an Arizona State University criminologist who has studied privatization questions.
Treating a facility’s share of central office functions as a savings from privatization “is a disingenuous
way of doing the accounting. You fabricate a cost savings that way.” Pratt said the approach is typical
of how private corrections firms “creatively budget to show a savings that is not there. That savings is
not real. The state is not going to be saving money off of that.”
But that’s not to say privatization doesn’t save money, Pratt said. It might, even if there is little
systematic evidence to prove it. But if there are savings, Pratt said, they are likely to come from only
one place: staffing. 16
“If you look at a department of corrections budget, there isn’t a line item that says ‘waste’ that
you can just go ahead and cut. The cuts have to come from somewhere and there are very few places
where you can actually cut costs, whether a public institution or a private. Where you get that savings
is usually going to be in the number of staff, how they are compensated and how extensively they are
trained.”
And as Arizona has experienced recently, Pratt said, “those kinds of cost savings don’t come
without consequences.” 17
Arizona has been a more aggressive – and some might argue, less careful – privatizer than
Ohio. The number of inmates in private prisons there has grown more than six-fold since 2000 and
now accounts for about 22 percent of the state’s prisoners, 18 compared to what would only be about 13
percent in Ohio – even after the contemplated prison sale is completed. 19 Two years ago, the Arizona
legislature passed, and the governor signed, a bill allowing the sale of virtually all of the state’s prisons

15

ibid. Gerry Gaes, “Cost, Performance Studies Look at Prison Privatization,” National Institute of Justice Journal No. 259,
March 2008, or at http://www.nij.gov/journals/259/prison-privatization.htm
16
Interview with Travis Pratt, professor of criminology and criminal justice, Arizona State University, conducted March
23, 2011
17
ibid.
18
Prisoners in 2009, note 2.
19
Ohio Department of Administrative Services, Request for Proposals for Operation, Maintenance and Purchase of
Correctional Facilities, RFP # CSP901412, April 6, 2011 at http://procure.ohio.gov/PDF/462011163521CSP901412.pdf
(the private prisons would have combined capacity of 6,561, about 13 percent of current overall prison population of
50,500).
5

www.policymattersohio.org

Policy Matters Ohio

Cells for Sale: Understanding Prison Costs & Savings

to private vendors willing to pay $100 million in cash upfront. The provision was later stricken from
the budget bill after no suitors came forward. 20
Last July, Arizona officials got a sudden wake-up call when three inmates escaped from a
private prison near Kingman, led authorities on a manhunt throughout the western United States and
allegedly murdered a couple in New Mexico before they were apprehended. Family members of the
slain couple have filed a $40 million claim against Arizona and the prison operator – Management and
Training Corporation (MTC), the same company that operates both of Ohio’s private prisons. 21
Among the culprits identified by state investigators was an inexperienced, poorly trained staff
at the MTC facility and apparently lax monitoring by the state that left that and other serious security
lapses undetected. 22
The day after Ohio’s prison-sale proposal was unveiled last month, Gary C. Mohr – the director
of ODRC and a veteran of both public and private corrections 23 was grilled by skeptical Democratic
lawmakers on precisely that potential vulnerability. Asked where the savings from privatization would
come from, Mohr essentially agreed with Pratt: From personnel. Not only do private firms typically
pay less than the state, Mohr said, but they also allow less vacation and less sick and personal time off,
so they can hire fewer people to cover the same number of required positions.
“We are paying more often for people to be off work than they are,” Mohr said. As a result, for
some security posts, it would take two state employees to staff a position that a private firm could
cover with 1.7. 24
When lawmakers expressed concern about a possible falloff in quality among these less
expensive staffers, Mohr was reassuring. As is currently the case, he said, the private prisons will be
required to meet the same operating standards as the state. And in a change from current practice,
ODRC spokesman Carlo LoParo said the state will require that the private prisons’ corrections officers
be trained by the state – at the private vendor’s expense – at the department’s training academy
alongside their counterparts from state-run institutions.
20

Casey Newton, Ginger Rough and JJ Hensley, “Arizona inmate escape puts spotlight on state private prisons,” The
Arizona Republic, Aug. 22, 2010 at http://www.azcentral.com/news/articles/2010/08/22/20100822arizona-privateprisons.html
21
JJ Hensley and Ginger Rough, “Kingman prison still under scrutiny,” The Arizona Republic, Jan. 30, 2011 at
http://www.azcentral.com/news/articles/2011/01/30/20110130kingman-prison-still-under-scrutiny0130.html
22
ibid.
23
Mohr is a former consultant and managing director for Corrections Corporation of America, the Nashville-based
corrections giant. Mohr moved to the private sector in 2005 after an extensive career in Ohio's prison system. He was an
ODRC deputy director under two different administrations and a warden at three Ohio prisons. Mohr was a managing
director for CCA from 2007 to 2009. The company also was a client of his consulting firm, Mohr Correctional Insight,
before and after his CCA employment. CCA would be eligible to bid on the state prison sales, but Mohr has promised to
remove himself from all consideration of those bids. Joe Guillen, “Private corrections company with ties to government
officials will not get special treatment while Ohio sells five prisons, director says,” The Cleveland Plain Dealer, March 21,
2011 at http://www.cleveland.com/open/index.ssf/2011/03/private_corrections_company_wi.html. On March 30, according
to The Columbus Dispatch, the interim executive director of the Ohio Ethics Commission said Mohr does not have a
conflict of interest with regard to the state’s privatization plans.
24

Laura A. Bischoff, “Ohio wants 5 prisons sold by August,” The Dayton Daily News, March 22,
2011 at http://www.daytondailynews.com/news/dayton-news/ohio-wants-5-prisons-sold-by-august1114455.html ;
Alan Johnson, “Sale of prisons, fewer inmates part of budget plan,” The Columbus Dispatch, March
22, 2011 at http://www.dispatchpolitics.com/live/content/local_news/stories/2011/03/22/copy/sale-ofprisons-fewer-inmates-part-of-plan.html?sid=101;
Gongwer News Service, “Administration pitches prison sales, sentencing changes to House panel,” March 22, 2011
www.policymattersohio.org

6

Cells for Sale: Understanding Prison Costs & Savings

Policy Matters Ohio

Currently, MTC, although required to meet ODRC standards, trains its own staff. LoParo said
the change is designed to ensure a “uniform system” where not only are procedures identical in public
and private facilities, but staff training is also. He and Mohr emphasized that, as is currently the case
with Lake Erie and North Coast, state-paid monitors would be on hand at all times to oversee
compliance with state requirements. But as was discovered last year at MTC’s facility in Kingman,
Arizona, the presence of state-paid monitors isn’t always a panacea.
In the beginning
Ohio was a relative late-comer to the era of private prisons, which dawned in the U.S. in 1983
with the founding of Corrections Corporation of America (CCA) and the opening in Texas the
following year of its first privately designed, built and operated facility. CCA still has the largest
private-corrections footprint in the nation, according to Gaes, with 17,000 employees and 75,000
inmates housed in 60 facilities in 19 states, 44 of which are company owned. 25 A close second is The
GEO Group, formerly Wackenhut Corrections Corporation, which, following a merger last year with
former competitor Cornell Corrections, now claims some 80,000 prison beds under contract
worldwide, including operations in Canada, South Africa, Australia and the United Kingdom. 26 The
corrections division of Wackenhut was formed in 1984. MTC started in 1981 as a job training
organization operating U.S. Department of Labor Job Corps centers. The company entered private
corrections in 1987 and now operates 20 state and federal facilities with 25,000 beds in seven states,
including Arizona and Ohio. 27
The first private prison on Ohio soil was the Northeast Ohio Correctional Facility, initially a
troubled facility for federal inmates that was opened by CCA in Youngstown in 1997.28 By the time
the state of Ohio first opened its own privately operated prisons three years later, nearly 70,000
inmates were already under private supervision in 29 other states, along with more than 15,500 federal
prisoners. 29
By the end of 2009, the number of federal inmates in private prisons had more than doubled
and those under state authority had increased by a third. With about 4 percent of its prisoners under
private supervision at the end of 2009, Ohio ranked 20th among the states in the percentage of
privatized beds (behind New Mexico, with 43.3 percent; Montana, Alaska and Vermont, with between
30 and 40 percent; and Hawaii, Idaho, Mississippi, Oklahoma, Colorado and Arizona, all with 20
percent or more). The federal government, with 34,087, had the largest number of privately held
prisoners at the end of 2009. Ohio had roughly 2,200, ranking it 15th in the number of privately held
inmates, behind Texas – far and away the national leader with more than 19,000 – and Florida,
Arizona, Oklahoma, Mississippi, Georgia and Tennessee, all with 5,000 or more. 30

25

Gaes, note 10; Corrections Corporation of America website at http://www.cca.com/about/
The GEO Group Inc. website at http://www.thegeogroupinc.com/about.asp
27
Management & Training Corporation website at http://www.mtctrains.com/about-mtc/overview
28
Michael Hallett and Amy Hanauer, “Selective Celling: Inmate Population in Ohio’s Private Prison,” Policy Matters Ohio,
May 2001 at http://www.policymattersohio.org/pris.html
29
Prisoners in 2009, note 2
30
ibid.
26

7

www.policymattersohio.org

Policy Matters Ohio

Cells for Sale: Understanding Prison Costs & Savings

If the Kasich plan is realized 31 , Ohio’s private prisons would have a combined population of
about 6,560. Assuming no change elsewhere, that would put Ohio in the top five in the number of
privatized beds and roughly 12th among the states in percentage of private prisoners.
Ohio’s prison population and corrections budget have exploded since the 1970s just as they
have in virtually every other jurisdiction. But unlike Arizona and most of the other private-prison
states, Ohio appears to have ventured into private corrections almost exclusively as an experiment in
cost control. In a survey conducted by Abt Associates, of the 23 states (plus the federal government,
District of Columbia and the Commonwealth of Puerto Rico) that reported having prison contracts
with private firms at the end of 1997, only eight reported that cost savings were of primary importance
in the privatization decisions. For the rest, privatization was seen as a means to reduce overcrowding
and to acquire prison beds quickly. 32 For Ohio, private prisons were seen as a way to save money. In
fact, the requirement that any private prison firm operate at least 5 percent below ODRC’s estimated
costs for a comparable state institution was inserted into law from the outset. 33
But the process for assuring that that goal is met has been both mysterious and something short
of rigorous.
In 2001, after CiviGenics – the first operator of the North Coast Correctional Treatment
Facility – ran afoul of state officials for contract violations and other issues there, ODRC offered to
pay potential replacements significantly more than CiviGenics was getting per inmate per day to
operate the facility. But when an insufficient number of bids met the 5 percent savings goal, ODRC
simply raised its estimate of what the state would have spent to run it. That allowed would-be operators
to ask for as much as $62.88 per inmate per day – 13.5 percent more than CiviGenics had been
receiving – and still meet the 5 percent “savings” requirement. 34
That contract was eventually awarded to MTC, the original contractor hired (without a
competitive bid and to the chagrin of competitors) to operate Lake Erie Correctional Institution, a
1380-bed institution that opened in Conneaut in April of 2000, about a month after North Coast
Correctional. 35
For Lake Erie, ODRC estimated that MTC would save 12 percent annually over what the state
would incur to run the facility, based on the contract per diem of $36.47 awarded to the company and
the department’s estimate that it would have cost the state $41.47 per inmate per day to run it. ODRC
used a hypothetical model to derive this estimated cost and released limited information about the
variables that went into that model. 36
Mystery has shrouded the state’s savings calculations almost ever since.
In fact, officials of the Ohio Civil Service Employees Association, the public-employees union
whose members’ jobs are most threatened by the push toward privatization, are deeply skeptical about
the savings the department has claimed. But they say they have never seen the actual calculations that
produced those savings projections. 37
31

The state issued its request for proposals (RFP) for the prison sale on April 6. All proposals are due by June 16, with an
estimated contract-award date of Aug. 31 and transfer of operations by Dec. 31. ODRC director Mohr has said he expects
five to seven bidders.
32
Abt, note 12
33
Selective Celling note 28
34
ibid.
35
ibid.
36
ibid.
37
Interview with Tim Roberts, vice president of the Ohio Civil Service Employees Association/AFSCME Local 11,
conducted on Feb. 16, 2010
www.policymattersohio.org

8

Cells for Sale: Understanding Prison Costs & Savings

Policy Matters Ohio

Last December, union officials submitted a public records request to ODRC for “all documents
that explain the process whereby the 5 percent savings that private prisons must demonstrate was
calculated.” In return, they received a document entitled “Overview of Privately-Operated Prison Cost
Comparison” that described the components of the calculations, but contained no calculations.
In response to its own public records request, Policy Matters Ohio in January received
electronic copies of the actual spreadsheets that ODRC officials have used to calculate the projected
savings from private prisons in each of the last three biennia – 2006-07; 2008-09 and 2010- 2011.
The three spreadsheets contain significant detail on how the recent calculations have been
made. However, it’s clear they were not designed to determine an appropriate per diem for the private
prisons, one that would ensure that the required 5 percent savings is realized. Rather, the per diem is
negotiated with the private provider first. The calculation is then used to determine that the per diem
would generate the requisite savings.
The process has always been the same, Stockdale said in an interview: Negotiate a rate
acceptable to the private vendor, then run a series of calculations to show it saves 5 percent. 38
Focusing, as mentioned earlier, on Lake Erie Correctional, and based on a detailed examination
of the recent spreadsheets, here’s how the process has worked.
In order to calculate Lake Erie’s projected savings, one needs to imagine a hypothetical
institution that looks exactly like Lake Erie, but is operated by the state, and then ask how much it
would cost:
„ to staff such a facility with state employees;
„ to stock, supply and equip such a facility and to provide an appropriate menu of services;
and
„ to provide the facility with heat, electricity, water and sewer services.
Utilities appear to be the smallest of these cost components and the most straightforward to
calculate. State officials have simply assumed that utilities would cost the same whether the state or a
private firm was in charge. So Lake Erie’s actual utility costs for previous years, adjusted for inflation
and divvied up per inmate per day, are simply inserted into the per diem cost calculation for its
hypothetical twin. 39
Since there is no state-run facility that works just like Lake Erie, calculating costs for
equipment, supplies and contract services for the hypothetical is much trickier. So, the ODRC has tried
to create a composite facility of sorts by selecting five prisons that share some of Lake Erie’s
characteristics, averaging their non-utility, non-personnel expenses, dividing the total by the average of
their inmate populations per day and inserting the result into the hypothetical’s per diem calculation. 40
Staffing requirements – the number of posts that need to be manned per shift – are established
for Lake Erie by contract. But because private firms have had different training regimens than the state,
offer less time off and tend to pay less overall, real cost differences may exist here. To measure that,
state officials determine how many state employees would be needed to fill the required number of
positions and how much those employees would receive in state pay. That total is also divided by the
number of inmates per day and added to the previous two components to produce a
payroll+costs+utilities per diem for the hypothetical.
38

Stockdale interview, Feb. 24, 2011, note 6.
ibid.
40
ibid.
39

9

www.policymattersohio.org

Policy Matters Ohio

Cells for Sale: Understanding Prison Costs & Savings

Then, the spreadsheets show, an estimate or calculation has been made of the hypothetical’s
share of “indirect costs,” which essentially are all ODRC costs associated with running prisons – as
opposed to functions like community-based corrections and parole operations. These, too, are divided
by the number of inmates per day and added to the hypothetical’s per diem.
Under the private-prison contract, ODRC-paid staffers must be on hand at the private prison to
maintain correctional records and to monitor adherence to contractual obligations. Salaries and benefits
for those staffers are totaled, divided by the number of inmates per day and added to Lake Erie’s
negotiated per diem – because these are additional costs associated with privatization, as opposed to
savings. 41
Finally, the hypothetical’s per diem costs for payroll, supplies and equipment, utilities and
indirect expenses are adjusted for inflation where appropriate and added together to create a total per
diem for the facility if it were state run. That, in turn, is compared to Lake Erie’s negotiated per diem,
plus the daily per-inmate cost for contract monitors and records staff. If the Lake Erie figure is at least
5 percent lower than the hypothetical’s, then the statutory savings requirement has been met.
But a close examination of the ODRC spreadsheets shows that errors, omissions and
inconsistencies have dramatically distorted the savings projected by the state.
The Spreadsheets
Take the 2006-07 biennium, for instance. ODRC officials calculated the savings for Lake Erie
over its hypothetical, state-run counterpart for those fiscal years at 10.2 percent, or about $2.4 million
for each year of the biennium (Table 1, Column A). But a detailed examination of the state’s
calculations shows that they include an apparently bloated staffing estimate for the state-run
hypothetical – a figure nearly 20 percent higher than the actual number of employees at Lake Erie and
approximately 12 percent higher than what the state would have needed to operate it.
The calculation also assesses the hypothetical for central office “indirect costs” at the rate of
about 10 percent of its payroll+costs+utility per diem. As was mentioned earlier, it is controversial to
include such central office “savings” at all. But granting for a moment that it should be included, based
on the size of the adjustments now being considered by the state, that assessment was at least six times
too high. Once those figures are adjusted, the picture of Lake Erie as a clear bargain for taxpayers is
tarnished considerably.
In fact, with more realistic treatment of overhead costs, the savings projected by the state’s
calculation for a privatized Lake Erie could slip to as low as 1.1 percent to 2.5 per cent, well below the
minimum required by law. (Table 1, Columns B and C) And when staffing levels for the state-run
hypothetical are cut to their proper size, the privatized Lake Erie actually appears to have cost between
1.8 percent and 3.4 percent MORE to operate that its state-run twin – a loss for taxpayers of between
$380,000 and $700,000 for each of those fiscal years. (Table 1, Columns D and E)

41

ibid.
www.policymattersohio.org

10

Cells for Sale: Understanding Prison Costs & Savings

Policy Matters Ohio

Table 1
Ohio's Cost Savings Calculations for 2006-2007 biennium including revisions
for hypothetical state-run prison identical to Lake Erie Correctional Institution, both with 1360 inmates
Column A
Column B
Column C
Column D
Column E
With corWith corWith no
With adjusPer diem*
rected staff
rected staff
central
ted central
costs and
levels; no
levels and
office costs
office costs
savings
central
central
computed by per new state as some
office costs
recommend office costs
state officials revision
Utilities
$2.34
$2.34
$2.34
$2.34
$2.34
Costs**
$6.94
$6.94
$6.94
$6.94
$6.94
Payroll
$33.68
$33.68
$33.68
$31.81
$31.81
Payroll+costs+utilities per diem
Central Office costs ***
Hypothetical's per diem cost

$42.96
$4.36
$47.32

$42.96
$0.64
$43.60

$42.96
$0.00
$42.96

$41.08
$0.64
$41.72

$41.08
$0.00
$41.08

Negotiated per diem for Lake Erie
State monitor; other charges
Lake Erie's per diem

$42.27
$0.22
$42.49

$42.27
$0.22
$42.49

$42.27
$0.22
$42.49

$42.27
$0.22
$42.49

$42.27
$0.22
$42.49

Lake Erie's savings vs hypothetical
Percent savings per year
Dollar savings per year
(per diem savings*365*1360)

$4.83
10.2

$1.11
2.5

$0.47
1.1

-$0.77
-1.8

-$1.41
-3.4

$2,397,580

$551,575

$233,879

-$380,420

-$698,116

Source: Ohio Department of Rehabilitation and Correction; author’s calculations
* amounts per inmate, per day
** equipment, supplies, other expenses
*** about 10% of payroll+costs+utilities

For the 2008-09 biennium, officials calculated the projected savings for Lake Erie over its
hypothetical, state-run counterpart at about 6 percent per year, or roughly $1.4 million for each year of
the biennium. (Table 2, Column A) Actually, had state analysts been consistent with the previous
biennium, those projected savings would have been dramatically higher – a whopping 21 percent, or
nearly $6 million a year for Lake Erie alone. 42 (Table 2, Column B)

42

Here’s why the projected savings reported by the state were so much lower. A detailed examination of the calculations
reveals two apparent oversights by state officials that, if corrected, would have produced the much more dramatic savings
figures. As with the previous biennium, the spreadsheet contains the calculation of a percentage assessment for “indirect
costs” (although for the 2008-09 biennium, the assessment was for about 13.9 percent of the payroll+costs+utilities per
diem instead of the 10 percent used previously). For whatever reason, however, officials either neglected or declined to
apply the assessment to the state-run facility as they had in the previous biennium and would again in the next one. Had
they applied the assessment, the projected annual savings for Lake Erie would have been 17.6 percent – or more than $4.6
million each year. The spreadsheet also reveals that, after estimating the position-by-position wages and salaries that state
employees would earn in fiscal 2008 if they were staffing the Lake Erie facility, officials inadvertently used an earlier
year’s pay estimates in the final cost calculations – seemingly understating the payroll costs. If both of those errors are
corrected, the purported savings shoot to 21 percent per year.
11

www.policymattersohio.org

Policy Matters Ohio

Cells for Sale: Understanding Prison Costs & Savings

Table 2
Ohio's Cost Savings Calculations for 2008-2009 biennium, including revisions
for hypothetical state-run prison identical to Lake Erie Correctional Institution, both with 1380 inmates
Column A
Column B
Column C
Column D
Column E
Column F
With
inmate pay,
revised
staffing
central
office costs

With
inmate pay,
revised
staffing, no
central
office costs

$2.87
$7.94
$37.46

$2.87
$7.94
$33.11

$2.87
$7.94
$33.11

$48.27
$0.68
$48.95

$48.27
$0.00
$48.27

$43.93
$0.68
$44.61

$43.93
$0.00
$43.93

$43.25
0
$0.10
$43.35

$43.25
0
$0.10
$43.35

$43.25
0
$0.10
$43.35

$43.25
$0.71
$0.10
$44.06

$43.25
$0.71
$0.10
$44.06

$2.81
6.1

$11.63
21.2

$5.60
11.4

$4.92
10.2

$0.55
1.2

-$0.13
-0.3

$1,416,721

$5,857,496

$2,820,185

$2,477,669

$276,039

-$66,477

Per diem*
costs and
savings
computed
by state
officials****

With
oversights
restored
and payroll
errors
corrected

With
adjusted
central
office costs
per state's
revision

Utilities
Costs **
Payroll

$2.87
$7.94
$35.25

$2.87
$7.94
$37.46

$2.87
$7.94
$37.46

Payroll+costs+utilities per diem
Central Office costs ***
Hypothetical's per diem cost

$46.06
$6.42
$46.06

$48.27
$6.71
$54.98

Negotiated per diem for Lake Erie
Inmate pay (columns E and F)
State monitor; other required charges
Lake Erie's per diem

$43.25
0
$0.10
$43.25

Lake Erie's savings vs hypothetical
Percent savings per year
Dollar savings per year
(per diem savings*365*1380)

With no
central
office costs
as some
recommend

* amounts per inmate, per day
** equipment, supplies, other expenses
*** originally (Columns A and B) about 13.9 % of payroll+costs+utilities
**** the central office costs for the hypothetical, and the state monitor costs for Lake Erie apparently were overlooked and
the wrong year's pay estimates used in this calculation

Numbers like that make the privatization experiment look even more like a slam dunk.
But, as with the 2006-07 figures before them, the 2008-09 calculations for the hypothetical are
burdened by dubious overhead assumptions and even more bloated staffing estimates (16 percent more
employees than the state would actually have needed) which make the mythical state-run institution
appear more expensive than it actually would be.
When those figures are cut to size in the calculation, and when previously ignored state costs
(like inmate pay) are inserted, the projected savings virtually disappear – shrinking to between 1.2
percent and -0.3 percent, depending on how central office costs are handled, well below the 5 percent
savings required by law. (Table 2, Columns E and F)
It should be mentioned that Stockdale was not with ODRC when those calculations were made
and can’t speak directly to how they were constructed. Arriving from the state budget office, Stockdale
became chief of budget planning and analysis for ODRC in the fall of 2008. The following year, as he
and his staff began preparing for the renewal of the private-prison contracts for Lake Erie and its sister
institution, Stockdale said that several inconsistencies and problems with the state’s earlier calculations
became apparent and he set out to fix them.

www.policymattersohio.org

12

Cells for Sale: Understanding Prison Costs & Savings

Policy Matters Ohio

Some of the more obvious concerns were rectified in the 2010-11 calculations, including the
use of more precise salary figures for various positions in the state-run hypothetical and a more
realistic estimate of its staffing levels. Stockdale also began to refine the calculations for “indirect
costs,” using actual figures for the department’s prison-related central office expenses as a starting
point instead of the shifting percentage estimation (about 10 percent for 2006-07, 13.9 percent in 200809) used previously.
“That is one of the things where it’s inconsistent from year to year,” Stockdale said in an
interview. “Instead of trying to calculate the cost of central office, they applied a flat percentage. That
is one of the things that bothered me.”
One result of Stockdale’s revision is what appears to be a much cleaner analysis for 2010-11,
one that still shows an attractive projected savings for Lake Erie of 13.9 percent ($3.8 million) in fiscal
2010 and an even more attractive 15 percent ($4.2 million) for 2011. 43 (Table 3, next page, Columns
C and D)
But there are ample reasons to be wary about these savings figures as well. In their ongoing
review of the cost-saving calculations, Stockdale and his superiors at ODRC have discovered a number
of other problems with the methodology – problems with significant implications not only for the
savings attributed to the private facilities in the past, but also for projected savings in the future.
Multiple changes to the methodology – proposed by ODRC and awaiting approval by the
Office of Budget and Management – were expected to be in place by this month (April), Stockdale
said, before the department was scheduled to renew contracts for Lake Erie and North Coast. In the
interest of transparency, he said the new methodology would appear on the ODRC website upon
completion and approval. As of April 15, it had not appeared. Stockdale disclosed several of those
proposed revisions during an interview in February.
They include probable changes in the complement of comparison facilities used to calculate
cost estimates for equipment and supplies for the state-run hypothetical. While similar in size to Lake
Erie, Grafton Correctional Institution (one of the five prisons, including the two already under private
supervision, to be sold) and Allen Correctional Institution don’t compare well otherwise, ODRC has
concluded. Both have mental-health residential components within or attached to them that make them
considerably more expensive to operate and the inmate populations of both are generally older and less
healthy than at Lake Erie. 44

43

Actually, the projected savings figures the state had been using for 2010-2011 are somewhat lower because, in the
original calculation, Stockdale’s office mistakenly used Lake Erie’s actual staffing figures for the state-run hypothetical,
instead of the higher number the state would have needed to fill the same number of posts. As a result, the original
calculations for the current biennium understated the personnel costs for the state-run hypothetical by about $1.36 per
inmate per day, which amounts to about $725,000 per year. (Table 3, Columns A and B)
44
ibid.
13

www.policymattersohio.org

Policy Matters Ohio

Cells for Sale: Understanding Prison Costs & Savings

Table 3
Ohio's Cost Savings Calculations for the 2010-2011 biennium, including revisions
For hypothetical state-run prison identical to Lake Erie, both with projected average inmate population of 1450
Column A

Column B

Column C

Column D

Column E

Column F

2010 per
diem costs
and savings
computed
by state*

2011 per
diem costs
and savings
computed
by state*

2010
calculation
with hypothetical's staff
level
corrected

2011
calculation
with hypothetical's staff
level
corrected

2010 calculation,
adjusted to
reflect state
revisions
***

2011 calculation,
adjusted to
reflect state
revisions
***

Column G
Revised
2010 calculation
(Column E),
no central
office costs
***
$3.06
$8.65
$36.86

Column H
Revised
2011 calculation
(Column F),
no central
office costs
***
$3.15
$8.90
$37.07

Utilities
$3.06
$3.15
$3.06
$3.15
$3.06
$3.15
Costs
$8.75
$9.02
$8.75
$9.02
$8.65
$8.90
Payroll
$35.50
$35.70
$36.86
$37.07
$36.86
$37.07
State monitor; records
($0.56)
($0.57)
($0.56)
($0.57)
($0.56)
($0.57)
($0.56)
($0.57)
staff for Lake Erie
Payroll+costs+ utilities
$46.74
$47.29
$48.10
$48.67
$48.00
$48.55
$48.01
$48.55
per diem
Central office costs**
$3.94
$4.06
$3.94
$4.06
$0.72
$0.75
$$Hypothetical's per
$50.68
$51.35
$52.04
$52.72
$48.72
$49.30
$48.01
$48.55
diem
Negotiated per diem
44.25
44.25
44.25
44.25
44.25
44.25
44.25
44.25
State monitor; records
$0.56
$0.57
$0.56
$0.57
$0.56
$0.57
$0.56
$0.57
staff
Inmate pay and overthe-cap medical
$1.44
$1.44
$1.44
$1.44
(Actual 2010 figures)
Lake Erie's per diem
$44.81
$44.82
$44.81
$44.82
$46.25
$46.26
$46.25
$46.26
Lake Erie's savings vs
$5.87
$6.53
$7.23
$7.91
$2.46
$3.04
$1.75
$2.30
hypothetical
Percent savings per
11.6
12.7
13.9
15.0
5.1
6.2
3.6
4.7
year
Dollar savings per year $3,104,102 $3,458,641
$3,825,454
$4,184,278
$1,303,602
$1,609,635
$926,199
$1,214,713
* officials mistakenly used actual Lake Erie staffing levels instead of higher number of employees that the state would need to hire
** share of actual prison-related department functions
*** includes costs for remaining comparison facilities, actual 2010 inmate pay and over-the-cap medical payments and revised central office costs based
on proposed state adjustment

www.policymattersohio.org

14

Cells for Sale: Understanding Prison Costs & Savings

Policy Matters Ohio

By way of illustration: in 2009, 7.4 times more prescriptions were written per 100 inmates at
Allen than at Lake Erie, according to state data, and nearly 3 times more at Grafton. 45 As of the middle
of last year, about 12 percent of the combined inmate population of Allen and Grafton were considered
seriously mentally ill, compared to just about 4 percent at Lake Erie. 46 According to ODRC data, both
Allen and Grafton were also carrying about 35 percent more staffers per inmate at the beginning of this
year than Lake Erie and the three other comparison prisons. 47
Stockdale said analysts would probably retain the more similar, albeit larger, facilities now
used as comparables – North Central Correctional Institution, in Marion, (which also is to be sold
under the Kasich plan); Noble Correctional Institution and Belmont Correctional Institution. Allen and
Grafton, he said, would probably by replaced by Southeastern Correctional Institution, in Lancaster,
which is roughly the same size and has a younger, healthier inmate population more similar to Lake
Erie’s.
Changes to the calculations are also likely to include assurances, in Stockdale’s words, “that all
of the costs that should be assigned to private prisons are assigned.” Because it’s a projection for the
coming fiscal year, the per diem negotiated for Lake Erie (and North Coast) doesn’t include some of
the actual costs incurred by the state for the facility during the course of a year’s operations.
For instance, the state typically reimburses Lake Erie for payments made to inmates for work
performed in the prison (at the estimated rate of $18 per inmate, per month) and for the amounts they
are given for initial expenses upon their release. Those costs have not been included in the
department’s previous cost-savings calculations, Stockdale said. 48
A more significant example is what Stockdale described as “over-the-cap medical payments.”
Beginning with the 2007-08 biennium, according to an amendment to Lake Erie’s contract, the state
agreed to reimburse MTC for all hospitalization costs beyond $20,000 for any inmate per admission.
Such reimbursements, which totaled $484,000 in fiscal year 2010, should be added to Lake Erie’s per
diem in the savings calculation, but haven’t been to date.
Stockdale acknowledged that those changes would tend to make the hypothetical state-run
facility look more competitive with Lake Erie in the savings analysis. 49
But one of the biggest changes the department is proposing – a dramatic alteration in the way
the department’s “central office” overhead expenses are handled in future calculations, wasn’t
disclosed until Stockdale was specifically asked about the issue in a follow-up interview on March 14
– the day before the announcement of Kasich’s intention to sell five prisons. 50
The significance of this “central office” issue is dramatically illustrated by a pair of in-depth
privatization studies by private research firms that were commissioned by the federal government and
completed in 2005. Both were designed to compare the cost to the government of contracting with
what was then the Wackenhut Corrections Corporation (now The GEO Group) to run a low-security
prison in Taft, California, from 1998 through 2002 with what it would have cost the federal bureau of
prisons to run it.

45

Gregory T. Geisler, “The Cost of Correctional Health Care,” Ohio Correctional Institution Inspection Committee,
December 15, 2010 at http://www.ciic.state.oh.us/health-care-or-medical-services/view-category.html page 11, Table 1
46
CIIC Progress Report and Staff Briefing, Ohio Correctional Institution Inspection Committee, June 29, 2010 at
http://www.ciic.state.oh.us/199-ciic-progress-report-and-staff-briefing-2010.html pages 7 and 8, Tables 3 and 4
47
Ohio Department of Rehabilitation and Correction website at http://www.drc.ohio.gov/web/prisprog.htm
48
Stockdale interview, February 24, 2011
49
ibid.
50
Stockdale interview, March 14, 2011
15

www.policymattersohio.org

Policy Matters Ohio

Cells for Sale: Understanding Prison Costs & Savings

Both studies – one by Abt Associates, the other by The CNA Corporation – used the same three
federally operated prisons for cost comparisons. And yet they came to dramatically different
conclusions which, according to Gaes, should hammer home two key lessons for would-be privatizers
and their foes:
-

Public-private cost comparisons are “deceivingly complex,” and
Be careful with overhead. 51

The “problem of overhead costing is so pervasive and difficult,” Gaes wrote last year, “that …
this calculation alone can tip the balance (often erroneously) in favor of one sector over the other.”
Unless considerable effort is made to determine which overhead expenses represent actual “savings”
for taxpayers, “large errors in the cost comparisons between private and public sector comparisons can
occur.” 52
In the studies at issue, Abt found that Taft, the privately operated prison, was much less
expensive to operate than the public facilities – by an average of about 14 percent in fiscal years 2001
and 2002. CNA, on the other hand, found a statistical dead heat: Taft appeared cheaper, but only by 1
percent. The much bigger cost difference found by Abt between the public and private facilities was,
almost dollar for dollar, equivalent to government overhead expenses it calculated for the public
facilities but did not charge against the private one.
Abt concluded that Taft’s share of “government overhead” – which it calculated for the public
facilities at 12 percent of their salary, wages and fringe benefits – should be counted as a savings from
privatization even while conceding that taxpayers didn’t actually save that money because the Bureau
of Prisons’ overhead didn’t actually change. 53
By contrast, CNA reasoned that if the government doesn’t actually avoid an expense as a result
of privatization – if the taxpayers don’t experience an actual savings – than the expense should not be
included in the calculation. CNA analyst Julianne Nelson concluded what Abt conceded – that most
Bureau of Prison (BOP) overhead costs associated with Taft would continue to be incurred by the
government even though a private company was operating the prison.54
“Many of these support expenditures – such as central office costs—benefit all BOP facilities,
both publicly- and privately-managed,” Nelson wrote. “As a result, these costs are essentially
independent of outsourcing – they are the same whether GEO runs Taft as a BOP contractor or the
BOP runs Taft directly on its own behalf.” According to CNA’s Nelson, “central office expenditures
are unavoidable and can thus be ignored when evaluating the merits of outsourcing.” An alternative to
ignoring them completely, Nelson wrote, would be to add them to both the public and the private
scenarios, which would have roughly the same effect.
When asked about Nelson’s argument on March 14, the day before Kasich’s budget was
released, Stockdale acknowledged that state officials were re-evaluating the overhead charges even in
the refined 2010-2011 calculation and were proposing significant reductions. Instead of charging the
hypothetical state-run institution a per-inmate share of $67.9 million in central office functions –
essentially for all prison-related activity in the department’s Columbus headquarters – he said what
51

Gaes, NIJ journal, note 15
Gaes, research review, note 10
53
Douglas C. McDonald and Kenneth Carlson, Abt Associates, Inc., “Contracting for Imprisonment in the Federal Prison
System: Cost and Performance of the Privately Operated Taft Correctional Institution,” November 2005 for National
Institute of Justice at http://www.ncjrs.gov/pdffiles1/nij/grants/211990.pdf
54
Julianne Nelson, The CNA Corporation, “Competition in Corrections: Comparing Public and Private Sector Operations,”
December 2005 at http://www.bop.gov/news/research_projects/published_reports/pub_vs_priv/cnanelson.pdf
52

www.policymattersohio.org

16

Cells for Sale: Understanding Prison Costs & Savings

Policy Matters Ohio

“we are going toward” is a function-by-function breakdown of what should be included and what
shouldn’t.
For instance, the department’s office of prisons oversees facility operations whether they are
public or private, Stockdale said. By excluding that office, the central office figure becomes
“significantly smaller,” he said, somewhere in the $10 million to $15 million range. 55 Taking the midpoint, that’s a cut of more than 80 percent in the “central office” costs attributed in past calculations to
the state-run hypothetical. When that adjustment is inserted in the calculation along with the other
changes already discussed, the projected 2010 savings for Lake Erie suddenly shrinks by nearly two
thirds – from 13.9 percent to just 5.1 percent, just a hair above the minimum allowed by state law (or
from $3.8 million to about $1.3 million). (Table 3, Column E)
But even those changes may not be enough.
The central office calculation could still include Lake Erie’s share of such functions as the
department director’s office, public affairs, legal services, human resources, planning, research,
budgeting, accounting, purchasing and other department-wide operations, most of which would have to
be performed from Columbus whether the facility is publicly or privately operated.
Following Nelson’s argument – ignoring such expenses or charging them to both the public and
private scenarios in the 2010-2011 calculations – would drop the projected savings for Lake Erie to
about 3.6 percent for 2010 and 4.7 percent for 2011, (Table 3, Columns G and H) below the legal
requirement for privatized facilities.
Stockdale said he and other ODRC officials are aware of the argument Nelson puts forward.
“It’s something that we kick around too,” he said.
“You could make that argument – I’m not saying whether it’s appropriate or not. I am not
judging either way. I want to be as accurate as possible. I don’t want to put my finger on the scale
either way, because all this affects people and I recognize that. So I just try to take as dispassionate a
stance as possible and that’s why we wanted to re-look at this.”
A race to the bottom?
Whether or nor the Lake Erie facility meets Ohio’s 5 percent savings requirement, it’s clear
from ODRC calculations that personnel costs – which account for about 70 percent of facilities’
operating expenses – are the likeliest place a private vendor would look to for cuts. One question, from
a public policy point of view, is how far can such expenses be trimmed before effectiveness and safety
are compromised?
According to Gaes, it’s an area of privatization that has received even less research attention –
and may be even harder to quantify – than cost. “What is the impact of lower labor costs on staff
turnover or staff performance?” he asked in his research review last year. “How do private companies
develop training for workers with fewer skills? Do private companies, in fact, hire lower skilled
workers? Have private companies re-engineered the prison employee’s job?” Gaes fears what he calls
a “McDonaldization” of prison labor.56
Once you introduce the profit motive into prisons, says Arizona State University’s Travis Pratt,
there’s always a danger that cost cutting in the most expensive areas – personnel and programming –
can impact quality and security. “It’s difficult to keep good people,” Pratt said, “when they are
55
56

Stockdale interview March 14, 2011
Gaes, research review, note 10
17

www.policymattersohio.org

Policy Matters Ohio

Cells for Sale: Understanding Prison Costs & Savings

understaffed and not compensated very well. If the cost saving is coming in the form of staffing, if it’s
coming in the form of encouraging turnover, than there is a safety risk at that point.” 57
In a 2009 letter to the governor protesting Arizona’s later-scuttled initiative to put for-sale signs
on virtually all of the state’s prisons, the state’s corrections chief issued a prophetic warning:
“Undoubtedly, a private company would pay its employees significantly lower wages and provide
them lesser training to realize cost savings,” department director Charles Ryan wrote. “This would lead
to higher staff turnover, low morale and place public safety at risk.”
A year later, that prophecy was seemingly fulfilled with the escape from the MTC facility near
Kingman and its tragic aftermath.
State investigators found that turnover at the facility was high, leading to a lack of training, the
Arizona Republic reported; some officers struggled to load and use their weapons during a security
review. The warden reported that nearly 80 percent of the staff was new or newly promoted and
investigators found that many showed a lack of experience and “command presence.” One investigator
estimated that a third of the security employees had less than three months on the job. Inmates at the
facility were found to be unkempt, unruly and poorly controlled. And false alarms were so frequent
that the staff simply ignored them. During a 16-hour review period before and after the escape, 89
alarms sounded and only two were legitimate. The rest were false. 58
After the investigation, MTC was warned that if it didn’t comply with revised safety standards
it could lose its contract to run the facility. Some of the prison’s administrators resigned; others were
re-assigned and the state increased the number of state employees on hand to monitor MTC’s
operations. 59
As Ohio readies itself to triple the number of inmates under private supervision, the debacle at
Kingman Arizona last year stands as a cautionary tale. Ohio officials say they have experienced no
such security lapses with MTC in its 10-year operation of Lake Erie and North Coast. ODRC’s LoParo
said there have been no prisoner walk-aways from either facility on MTC’s watch. Records provided
by LoParo show that over the last three fiscal years combined, Lake Erie has experienced inmate-oninmate assaults at only about half the rate of the state’s designated comparison facilities (three per 100
inmates, compared to 5.8 on average at the five comparables.) Lake Erie has experienced about the
same level of inmate-on-staff assaults as the comparables over the period (just over 3 per 100
inmates). 60
As for staff stability, average corrections-officer turnover for the last two years has been higher
at the MTC facilities than at state-run prisons (12.5 percent at Lake Erie and 10.2 percent at North
Coast, compared to 8.7 percent for ODRC as a whole), according to figures provided by LoParo. But
the private-prison turnover rate is still below what LoParo described as the maximum permissible rate
of about 20 percent. “Anything less than that is acceptable,” he said.
Still, MTC’s problems in Arizona raise questions. It’s not as if the company was a newcomer to
corrections last year and didn’t know better. MTC had been running prisons for nearly 25 years at the
time and had operated the Kingman facility for six. Surely, experience had taught company officials
57

Travis Pratt interview, March 23, 2011
JJ Hensley, “State rips private prison, begins making changes,” The Arizona Republic, Aug. 20, 2010 at
http://www.azcentral.com/news/articles/2010/08/19/20100819arizona-prison-escape-report-brk19-ON.html ; Arizona
Department of Corrections, Security Review of MTC Facility at Kingman, Aug. 4-6, 2010 at
http://www.azcorrections.gov/adc/news/2010/kingman_sec_review/Kingman_Assessment_1.pdf
59
ibid, Arizona Republic, note
60
Ohio Department of Rehabilitation and Correction, Inmate on Staff Assault Report – Fiscal Years 2008, 2009, 2010, and
Inmate on Inmate Assault Report – Fiscal Years 2008, 2009, 2010.
58

www.policymattersohio.org

18

Cells for Sale: Understanding Prison Costs & Savings

Policy Matters Ohio

how to maintain a stable, well-trained staff. So, one is entitled to ask, were the staff deficiencies at
Kingman a result of corner-cutting to boost the bottom line?
In 1998, after six prisoners, including five murderers, escaped from the CCA-operated federal
prison in Youngstown, many of the guards were found to have had little or no experience in
corrections and conditions at the prison were described as generally chaotic. 61 And yet, CCA had been
in the corrections business for a decade and a half at that point and presumably knew better.
Escapes can happen. Prisons tend to be volatile and unpredictable places. So, when seasoned
prison operators staff their facilities in one location with inexperienced, insufficiently trained
employees, one is entitled to ask what motives are at work and whether similar problems could crop up
again elsewhere.
Whatever the answers to such questions, one thing is probably safe to assume: Once the
commitment is made to aggressively go private – for whatever reason – it’s a decision that’s unlikely
to be reversed.
Many commentators have warned that once a jurisdiction delegates its authority to manage and
operate a correctional facility to a private corporation, the jurisdiction also runs the risk of becoming
overly dependent upon that corporation for services. 62 The limited number of private corrections firms
makes it difficult for states to shop around for replacements if they are unhappy with the services –
especially if the vendor owns the facility. In its prison-sale RFP, Ohio has asked would-be purchasers
to outline a procedure for transferring the facility operations and assets to the state or a new vendor if
their contract is terminated.
If the vendors happen to become political players, the dynamic can be that much more
complicated. Last August, the Arizona Republic cited a 2006 report from the National Institute on
Money in State Politics stating that the private-prison industry had contributed to the campaigns of 29
of 42 Arizona lawmakers who heard a 2003 proposal to increase state private-prison beds there.
Between 2001 and 2004, according to the report, the industry contributed $77,267 to Arizona's
legislative and gubernatorial candidates, most through lobbyists representing company interests at the
legislature. 63
Last year, before the Kingman episode, Arizona lawmakers approved 5,000 more privateprison beds, an expansion that was slowed by the outcry over the escape last summer and fall. But
those beds have since gone out for bid 64 , while multiple bills to increase state oversight of for-profit
prisons have languished – without so much as a hearing – in the legislature. 65
Meanwhile it’s not at all clear that Arizona’s private prisons are saving taxpayers anything.
In fact, when the comparison is apples-to-apples, the evidence shows the opposite. In a study
completed last year, the state department of corrections concluded that, overall, it costs taxpayers about
$2.45 less per inmate per day to house medium- and minimum-security prisoners in private facilities.
But that comparison is skewed by the fact that contract provisions allow some operators to screen
inmates based on medical conditions. As a result, private prisons spend about 35 percent less per
prisoner on health care than the state-run facilities do. Once the playing field is leveled, the tables are

61

Hallett and Hanauer, note 28
ibid.
63
Newton, Rough and Hensley, Arizona Republic, note 20
64
Hensley and Rough, Arizona Republic, note 21
65
Mary K. Reinhard, “Arizona private prison oversight bills die,” The Arizona Republic, Feb. 12, 2011 at
http://www.azcentral.com/news/articles/2011/02/15/20110215arizona-private-prison-bills.html
62

19

www.policymattersohio.org

Policy Matters Ohio

Cells for Sale: Understanding Prison Costs & Savings

turned. Setting medical costs aside, Arizona’s state-run facilities cost $1.78 per inmate LESS per day
to operate than its private facilities – or about $5.8 million less per year. 66
The bottom line?
“I don’t want to be ideological about this,” said Gaes, who acknowledges deep skepticism on
the wisdom of private prisons.
“There are going to be occasions where you are going to save money,” he said in an interview.
“You can begin to squeeze money out of the system. Maybe you can squeeze a half a percent out, who
knows? But it’s not as if these systems are overfunded to begin with. And at some point, you start to
lose quality. And because quality is very difficult to measure in prisons, I’m just worried that you’re
getting into a race to the bottom.”

66

Arizona Department of Corrections, “Revised FY 2009 Operating Per Capita Cost Report,” Sept. 28, 2010 at
http://www.azcorrections.gov/adc/reports/ADC_FY2009_PerCapitaRep_revised.pdf
www.policymattersohio.org

20

Policy Matters Ohio is a non-profit, non-partisan research
institute dedicated to researching an economy that works for all in
Ohio. We seek to broaden debate about economic policy by
providing research on issues that matter to Ohio’s working people and
their families. Areas of inquiry for Policy Matters include work,
wages, and benefits; education; economic development; energy policy;
and tax policy. To those who want a more prosperous, equitable,
sustainable and inclusive economy… Policy Matters.
Main Office: 3631 Perkins Avenue, Suite 4C-E • Cleveland, Ohio 44114
Columbus: 300 E. Broad Street, Suite 490 • Columbus, Ohio 43215
http://www.policymattersohio.org/

© 2011 Policy Matters Ohio. Permission to reproduce this report is granted provided that credit is given to Policy
Matters Ohio. All rights reserved.