Skip navigation
The Habeas Citebook: Prosecutorial Misconduct - Header

Multonomah County Jail Report, Multonomah County DA Office, 2006

Download original document:
Brief thumbnail
This text is machine-read, and may contain errors. Check the original document to verify accuracy.
INDEPENDENT REVIEW OF
POLICIES AND PROCEDURES OF
CORRECTIONAL FACILITIES OPERATED BY
THE MULTNOMAH COUNTY SHERIFF'S OFFICE

By
The Multnomah County District Attorney's Office
2006

TABLE OF CONTENTS
I. SUMMARY OF INVESTIGATION…………...…………………………………...…..1
Authority and Methodology………………………………………………………...1
Summary of Major Concerns…………………………………………………….…2
A. Lack of outside oversight of jail operations………………………………….…2
B. Jail bed and juvenile detention costs……………………………………………2
C. Losses on contracts and agreements to house inmates and juveniles from other
jurisdictions………………………………………………………………….….2
D. Inmate supervision issues……………………………………………………….3
E. Holding inmates and proper identification……………………………………...4
F. Medical emergencies in the jail…………………………………………………4
Specific recommendations…...……………………………………………………...4
II. OUTSIDE OVERSIGHT…………………………………………………………….…7
A. County Commissioners as inspectors of local jails……..………………………8
B. State Department of Corrections oversight and advice…………………………9
C. Corrections Grand Jury investigation……………………………...…………..10
D. Recommendation…………………………………………………...………….10
III. JAIL BED AND JUVENILE DETENTION COSTS……………………..…………...11
A. County sheriff financial reporting practices…………………………….……...12
B. Concerns with MCSO bed cost figures ………………………………………..13
C. Concerns with budgeting practices….…………………………………………15
Recommendation…………………………………………………………………...18
IV. DAILY JAIL BED COSTS…………………………….….…………………………...18
A. MCSO Facilities………………………………………………………………..18
1. Methodology..…………………………………………………………..18
2. Comparison……………………………………………………………..21
3. Reasons for the rise in jail bed costs……………………………………22
4. Workforce costs………………………………………………………...23
a. MCCDA labor contracts………..………………………….…23
b. MCSO workforce management practices…………………….24
i.Sick leave abuse…………………………………………...25
ii.Overtime and Comp-time.………………………………..30
iii.Other leave issues……….……………………………….31
iv. Post relief factor…………………………………………32
v.Closing the least expensive jail in the county…………….34
vi.Medical costs……..………………………………………35
B. Juvenile detention and program costs…………………………………………..37
Detention bed costs…………………………………………………………38
Program bed costs…………………………………………………………..38

V. LOSSES ON CONTRACTS AND AGREEMENTS TO HOUSE
DETAINEES………………………………………………………………………….39
A. Senate Bill 1145 Agreement……..……………………………..……………..40
Recommendation………………………………………………………………….41
B. United States Marshals Service Contract………………………………..……..42
Recommendation………………………………………………………………….44
C. JDH Contract………………………………………………………………….44
Recommendation………………………………………………………………….45
VI. INMATE SUPERVISION ISSUES………………………………………………..…45
A. Direct Supervision, Classification and Double Bunking……………………....45
Recommendations…………………………………………………………………48
B. Video cameras in the jails and criminal prosecution…………………………..49
VII. OTHER ISSUES IN THE JAIL……………………………………………………...50
A. Prosecution of criminal cases in the jail……………………………………….50
B. Command staff must have the full support of the Sheriff……………………..50
C. Earthquake plan………………………………………………………………..51
VIII. INQUIRY INTO SPECIFIC INCIDENTS OCCURING IN THE JAIL…...……….51
A. The murder of Dennis Saban…………………………………………………..51
B. Sexual contact between two inmates in isolation unit…………………………56
C. Identification issues……………………………………………………………58
D. The case of Nick Walter……….………………………………………………60
E. The case of Juan Martinez…….……………………………………………….60
F. In custody death of inmate……………………………………………………..61
CONCLUSION……………………………………………………………………………63
APPENDICES
Appendix A………………………………………………………………………A1
Appendix B………………………………………………………………………A2
Appendix C……………………………………………………………………....A3
Appendix D……………………………………………………………………....A4
Appendix E……………………………………………………………………....A5
Appendix F………………………………………………………………………A6
Appendix G……………………………………………………………………...A7

Introduction
On June 15, 2006 the Multnomah County Board of Commissioners passed Resolution
No. 06-094, which gives the Multnomah County District Attorney’s Office the authority
to conduct an independent review the policies and procedures of the correctional facilities
operated by the Multnomah County Sheriff’s Office. The scope of that review is set forth
in the findings of the resolution. That includes to examine fully: (1) the local facilities,
including, but not limited to the cleanliness of the facility and the health and discipline of
the persons confined; (2) the safety and security policies and procedures within the
correctional facilities and how they affect the health and discipline of the persons
confined therein; and (3) the operating budget and the daily operation and administration
of the jails.
I asked Senior Deputy District Attorney Chuck French and Special Counsel to the
District Attorney John Bradley to conduct the corrections investigation. Combined, Mr.
French and Mr. Bradley have almost 60 years of prosecution and investigative
experience. They have been involved in over 100 homicide investigations; they have
prosecuted most of the racketeering cases ever charged in Multnomah County; they have
conducted internal investigations within this office, and they have participated in the
investigations of both police officers and corrections officers. In addition to their
background in prosecution they both have worked closely with various Corrections
Grand Juries looking into the operations of the jails, prison and juvenile facilities within
Multnomah County. In order to better understand the economic ramifications of the
Sheriff’s budget and projected bed costs, I asked Scott Marcy, my Chief Financial
Supervisor, to assist Mr. French and Mr. Bradley.
In the months following the passage of Resolution No. 06-094 Mr. French and Mr.
Bradley inspected various facilities in Oregon and Washington. They talked to over 100
correctional officers in Oregon and Washington, both administrative and line staff. They
talked to Sheriff Giusto, and former Sheriffs Fred Pearce, Bob Skipper, Don Clark and
Dan Noelle. In addition, they read many jail studies involving all aspects of jail activity,
which were available locally and on the Internet. During this period I was informed of
the progress of the investigation and was provided with updates that I forwarded to the
Board. 1
1

I wish to acknowledge the assistance of numerous officials from a large number of agencies, who spent a
great deal of time meeting with our staff, either in person or by phone conference. Many of those
mentioned assisted this investigation by allowing jail tours and access to their entire staff. Among those
agencies who assisted our staff are the National Institute of Corrections; the National Commission on

On October 1, 2006 Mr. French and Mr. Bradley along with MCSO Captain Bobbi Luna
and Sergeant Phil Anderchuk, who is the President of the Multnomah County Corrections
Deputies Association (MCCDA), went to Orange County (Orlando) and Hillsborough
County (Tampa) Florida. These facilities were chosen because they are considered to be
two of the best-run jails in the country. That trip provided valuable insights into how a
well-managed jail operates. In addition, it was extremely important that Captain Luna
and Sergeant Anderchuk accompanied them because it gave Mr. French and Mr. Bradley
an appreciation of the specific ramifications of certain policies to Multnomah County
jails.
I appreciate the assistance that Sheriff Giusto and his staff have provided in this
investigation. I know it is never easy to have an outside agency review what you do, but
I hope that the Sheriff will consider the suggestions that we have made in this report. I
know that the Sheriff is committed to making his jail and his jail management the best in
the country. I hope this report aids him in that endeavor.
In addition to the operation of the Sheriff’s Office, I hope that this report may prove
valuable for the County Commissioners and all county law enforcement and public
service agencies. Public safety requires an effective jail system, but also an integrated
public service plan throughout the county, which requires informed decisions from
leadership. I hope this report will assist the Commissioners in that effort.

Michael D. Schrunk
District Attorney
Multnomah County

Correctional Health Care; Washington County Sheriff’s Office; Clackamas County Sheriff’s Office; Clark
County, Washington Sheriff’s Office; King County, Washington Department of Adult and Juvenile
Detention; King County, Washington Department of Health; Linn County Sheriff’s Office; Lane County
Sheriff’s Office; Whatcom County, Washington Sheriff’s Office; Hillsborough County, Florida Sheriff’s
Office; Orange County, Florida Sheriff’s Office; San Bernardino County, California Sheriff’s Office.

Foreword
We have spent over four months looking at jails. Our prior involvement with jails has
consisted of conducting Corrections Grand Juries, investigations of corrections officers,
investigations of criminal activity within the jails and going into the jails to interview
witnesses. We have talked to hundreds of witnesses since we were given this assignment.
However, a great deal of our most valuable information came from the men and women
who work in our jails. They provided the information because they are proud of their
profession and they want the jails to be successful.
We hope that the recommendations and observations we make in our investigative report
will help in making our jails better, safer and more economical. Every suggestion we
make has been discussed with experts and various people in corrections. They have
generally agreed with our suggestions. The solutions to these problems will take time to
achieve and will not always be easy. It is important that the men and women who must
find solutions think beyond what has been done in the past. The Multnomah County
Corrections Deputies Association (MCCDA) must be consulted before making changes,
but it is the Sheriff and his command staff who must make the decisions. The answers to
the questions we raise cannot be answered by saying nothing can be done. Necessary
changes must be made. We know that the Sheriff is committed to this process and we will
help him in any manner possible.
Chuck French
Senior Deputy District Attorney
John Bradley
Special Counsel to the District Attorney

I. SUMMARY OF INVESTIGATION

Authority and Methodology
On June 15, 2006 the Multnomah County Board of County Commissioners passed
Resolution 06-094, requesting a broad investigation by the Multnomah County District
Attorney’s Office of local correctional facilities. Beginning in July of 2006, our
investigation was conducted in tandem with the annual Corrections Grand Jury
investigation of the same facilities. While the questions asked in the ordinance are
broader than the statutory requirements of the Corrections Grand Jury, we believed that
because many of the issues were similar it was better to join the two. In the course of the
investigation we heard sworn testimony from over one hundred witnesses in the grand
jury proceedings. In addition to sworn testimony we heard in front of the grand jury, we
spent hundreds of hours speaking to corrections officials throughout the state and
throughout the nation. We have hired an independent financial consultant to advise us on
key fiscal issues. We have visited numerous jail facilities throughout the Pacific
Northwest and have traveled to Florida to visit jails that are nationally recognized as
model direct supervision jails.

-1-

Summary of Major Concerns
A.

Lack of outside oversight of jail operations

State laws require the outside oversight of county jails by the Board of County
Commissioners and by the state Department of Corrections. This oversight has not
occurred as required by law. As a consequence, some of the county commissioners have
failed to acquaint themselves adequately with the operations of local jails, resulting in
uninformed decisions regarding operation and funding of these facilities.
B.

Jail bed and juvenile detention costs

Multnomah County has some of the most expensive jails of any urban area in the
nation. Under our analysis, it costs $157/day to house an inmate in the Multnomah
County jail system. Under the same analysis this compares to $111/day in King County,
Washington, $103/day in Clark County, Washington and $89/day in Washington County,
Oregon. (Page 19) Some of the reasons for this are:
1. Despite the fact that county jail beds have been cut by 20% since 2000 (from
2,104 to 1,690) the Sheriff’s jail budget, adjusted for inflation, has actually been
increased in real terms by over 10%. (Page 22)
2. The two least expensive jails in the system were closed in 2005, forcing inmates
from those facilities to be housed in the remaining two jails, at up to three times
the cost per inmate. (Page 34)
3. Labor contracts, for which neither the Sheriff or Board of County Commissioners
now accepts responsibility, have resulted in wage scales, benefits and work rules
which have raised labor costs, resulting in substantial budget increases to cover
the costs of those contractual obligations. (Page 23)
4. Sheriff’s management practices have allowed the abuse of sick leave (Page 25),
overtime and compensatory time (Page 30) by the workforce.
5. Jail staffing patterns may be inefficient. (Page 33)
6. Extraordinarily high medical costs have contributed to high overall system costs
in the jail (p. 35). Our jail health program, run by the County Health Department,
costs approximately $5 million more annually than would a private plan.
Additionally, Multnomah County has an extremely expensive juvenile detention
facility. It costs $401/day to house a juvenile at that facility. (Page 37)
C.

Losses on contracts and agreements to house inmates and juveniles from
other jurisdictions

Multnomah County has entered into a number of contracts and agreements to
house prisoners from other jurisdictions in our county’s jails and juvenile detention
facility. These agreements result in enormous losses for the county, as they fail by huge
margins to recover the costs of housing these inmates and juveniles. (Page 39) In effect,
Multnomah County is subsidizing other jurisdictions by housing their prisoners while

-2-

matrix releasing its own inmates, who then commit more crimes against the citizens of
Multnomah County. The losses on these agreements are:
1. In housing state inmates under the terms of Senate Bill 1145, the county loses,
by the Sheriff’s estimation, $6.8 million annually, and by our estimation, $7.25
million annually. (Page 40)
2. In housing 150 federal inmates under the terms of a contract with the United
States Marshals Service, the county loses $2.03 million annually. (Page 42)
3. In housing up to 28 juveniles in the juvenile detention facility under the terms of
contracts with Washington and Clackamas counties, the county loses another $2
million annually. (Page 44)
D.

Inmate supervision issues
1. Deterioration of direct supervision model - Both current Multnomah County
jails were designed to operate on a correctional model called direct supervision.
This model is designed to minimize costs by housing as many inmates as possible
in open dormitory-type facilities, with a consequent reduction in supervisory staff.
Primarily because single-inmate isolation cells are not being correctly utilized in
the system, that direct supervision model has been abandoned at the downtown
jail. The result is that predatory inmates have been inappropriately housed in cells
with potential victims, and inmates have been locked in cells for up to 22 hours
each day in the downtown jail. (Page 45)
2. Murder in the jail - On June 16, 2005 an inmate was brutally murdered by
another inmate. Our concerns are:
a. Corrections deputies not at posts for long periods of time where very
dangerous inmates are double bunked.
b. Failure to put information in log books as required.
c. Inmates and staff put at risk because of poor intelligence information.
d. Staff unaware of dangerous inmates.
e. Lack of immediate internal investigation to determine causes and make
changes.
f. Housing very violent inmates with non-violent inmates throughout
MCDC. (Pages 51-56)
3. Sexual relations in isolation cells - On May 1, 2006 an inmate escaped from his
maximum-security segregation cell and was able to enter another segregation cell
where he had sexual relations with a female inmate. Our concerns are:
a. Failure of jail staff to detect that someone had escaped for several hours.
b. Failure of the jail staff to ensure that inmates were in their cells before the
cell doors were closed.
c. Failure of jail staff to thoroughly search cells on a regular basis, thereby
allowing inmates to accumulate contraband.

-3-

d. Failure of staff to supervise and inspect a limited number of cells in the
jail unit that houses the county’s most dangerous inmates. (Pages 56-58)
E.

Holding inmates and proper identification

Oregon law allows police to issue citations to certain offenders, rather than
arresting them. Some of those offenders are brought to jail when the officer is unsure of
their identity. In about 90 % of the cases, the person is identified in the first four hours of
detention. For the remainder it takes more time, but because of jail practices and
procedure they are released. In two separate cases, inmates were released who had
murder warrants in the State of Washington. In one of those cases, after the person was
released he went to Washington and murdered a woman and her daughter. Our concerns
are:
a. We set an arbitrary time limit of 4 hours on holding people to be identified.
b. California and Washington restrict the use of computerized fingerprints.
We must try to persuade them to change their systems and become fully
involved in the Western Identification Network. (Pages 58 –61)
F.

Medical emergencies in the jail

On September 17, 2006 after an altercation with police, a mentally ill man was
brought to the jail with an undiagnosed life-threatening injury. The jail refused to accept
him. He died after police tried to transport him to a hospital some distance from the jail.
Our concerns are:
a. The jail nurse was not informed about the extent of the injuries.
b. The inmate was not taken to the closest hospital when it looked like his
injuries might be life threatening.
c. There was no jail protocol requiring arresting officers to specifically
state the extent of any physical force used upon an inmate.
Specific recommendations
Throughout this document we have made recommendations about things we believe need
to be changed or improved in our jails. The following is a list of specific
recommendations. We place them in a list because we do not want any of our
suggestions to be lost.
While generally we have placed the most important
recommendations first, we believe they are all important.
1. Wapato Jail - We recommend that the Board of County Commissioners negotiate
with the State Department of Corrections to open Wapato Jail by leasing it to the state to
house SB 1145 prisoners now serving time in our jails. This will save the county
approximately $7 million annually, because these inmates can be housed far less
expensively by the state than they are currently housed in our jails. With the money saved
jail beds at Inverness could be kept open to help stop matrix releasing. (Page 41)

-4-

2. United States Marshals Service contract - The county should end the Marshals
Service contract, which we believe to be illegal under state law, and which is subsidizing
the federal government by housing 150 federal prisoners at more than $2 million annually
less than it costs our county to do so. (Page 44)
3. Juvenile detention contracts with Clackamas and Washington counties - The
Board of County Commissioners should insist that these contracts, which currently lose
$2 million annually, be re-negotiated to collect the full cost of housing those counties’
juveniles in our facility. (Page 45)
4. Engaged and informed county leadership - We believe that for some years the
Board of County Commissioners and the Sheriff’s Office have not had a reasonable and
efficient working relationship with one another. County leadership in the Board must
understand the operations and costs of the Sheriff’s Office; and it currently does not, to
the detriment of county taxpayers. Although we can make no specific recommendation
that will dictate better personal relationships, we believe that this must be accomplished,
for the sake of the county.
5. Establishing an oversight committee to report to the Board of County
Commissioners - The Board of County Commissioners should create a permanent jail
oversight committee to assist the Board in its legal oversight functions, which currently
are not being fulfilled. This committee should be supervised by and report periodically to
the Board on the specific concerns noted below in this report, the impact of any labor
negotiations, and concerns noted in annual Corrections Grand Jury reports, in addition to
any other jail issues it deems appropriate. (Page 10)
6. Accounting and budgeting practices - We are greatly concerned with the budgeting
and accounting practices of the Sheriff’s Office, which we find to be misleading and not
serving their stated purpose of informing the Board fully about the financial operations of
the Sheriff’s Office. We recommend the implementation of clear cost accounting
procedures that will establish a consistent model for understanding current and future jail
costs. (Page 18)
7. Labor negotiation - We believe that the labor bargaining practices of the Sheriff’s
Office and the Commissioners have failed to vigorously and diligently present a
management perspective in contract negotiations with the corrections deputies’
association, to the detriment of county taxpayers. We recommend that a process be
instituted which requires both the Board and Sheriff to take responsibility for the results
of labor negotiations. (Page 24)
8. Sick leave abuse - The rampant abuse of sick leave by corrections deputies has
prevented effective workforce management in the jail system and has dramatically
increased costs. We recommend the implementation of a sick leave monitoring
procedure by an independent county agency, with clear enforcement authority, for all

-5-

county employees. We also recommend sick leave policies that will give deputies an
incentive to save sick leave rather than use it indiscriminately. (Page 29)
9. Excessive overtime and compensatory time - The reasons and solution for excessive
overtime and compensatory time usage must be examined. The unchecked use of all
types of leave has made the management of overtime impossible. (Pages 30-31)
10. Jail staffing patterns - We suspect that staffing patterns in the jail are inefficient
and costly. We recommend an independent study of this issue that will consider both the
appropriate staffing levels of the facilities and establish an appropriate post relief factor
that excludes the abuse of leave from its calculations. (Page 33)
11. Private contracting of jail medical services - We recommend that the Board of
County Commissioners consider privatizing jail health care. Private contracting would
save the county approximately $5 million annually. We recommend, should the Board
take this step, that the savings obtained should be returned to the public health system to
treat mental health patients in the community, thereby avoiding much unnecessary and
expensive incarceration of these individuals, who currently represent as much as 30% of
our jail population. (Page 37)
12. Replacement of uniformed officers by civilian staff where possible in the jails Cost-effective jail systems utilize civilian employees instead of uniformed officers as
much as possible. Multnomah County jails do not, and our system must explore how this
can be accomplished. (Page 47)
13. Matrix releases - Matrix releases should not occur, but if they do they should be
after the defendant’s first court appearance, to ensure lower rates of failures to appear in
court.
14. Inmate supervision - The following are recommendations regarding the supervision
of inmates:
Double bunking - Double bunking can be safely retained at the downtown jails
as it is in all the institutions we visited. To safely retain double bunking we
recommend that the current cell doors be replaced with full-visibility glass-panel
cell doors. (Page 49)
Deputy presence - Deputies must be present constantly in housing modules in the
downtown jail. The current practice of leaving modules unattended without
signing in and out of logbooks, against work rules, must be ended, with strict
disciplinary actions if necessary. (Page 53)
Direct supervision - We have noted that the practice of “direct supervision” of
inmates, for which the downtown jail was designed, has been abandoned. We
recommend that the Sheriff’s Office commit itself to that practice once again,
which is both less harsh for inmates and more economical for department
finances. (Page 48)

-6-

Isolation cells - We have found that single-inmate isolation cells are not being
used to isolate violent and disruptive prisoners, as they were designed, but to
house trustees as a reward. This has forced violent inmates into general
population units with disastrous results. We recommend that single-inmate
isolation cells be used to separate dangerous inmates from general population,
making direct supervision of the general population feasible once more. (Page 48)
Investigation of incidents in jails - The Sheriff’s Office has failed to
immediately investigate several recent serious incidents involving the breakdown
of inmate supervision by jail staff. At least one investigation has been delayed for
almost a year. The lack of timely investigations of jail problems has degraded the
quality of any ultimate investigation and has prevented solutions from being
implemented. We recommend a requirement that any serious jail incident be
immediately investigated by the Sheriff’s investigators, unless a corrections
officer is the prime suspect in a criminal matter. In those cases the decision to do
the internal investigations should require the consent of the District Attorney.
(Pages 54-55)
Video cameras - We recommend the use of video cameras throughout the jail
system. This procedure, common in many jails we visited, would reduce
violence, would protect deputies against false allegations and would provide
evidence for criminal prosecutions against inmates who commit crimes. (Page 49)
Prosecution of crimes by inmates - Many crimes committed by inmates are
handled internally and not submitted to the District Attorney’s office. Especially
in cases involving assaults and substantial property damage, we recommend
investigation and referral for criminal prosecution. (Page 50)
Violent inmates - For the safety of staff and inmates, we recommend a system
that clearly identifies violent inmates who have a prior history of violence before
they are admitted into our jails. This is a requirement that has not been enforced.
A mechanism should also be instituted to identify violent inmates to all staff as
these inmates are transferred throughout the jail system. (Page 55)
Identifying inmates - We recommend that no inmate should be released from
jail, except on bail, before that inmate is identified. (Page 60)
Cell searches - We recommend that inmate cells be searched on a regular basis to
prevent the introduction of contraband. (Pages 57-58)
15. Earthquake readiness - We recommend that the Sheriff prepare a detailed plan to
operate jail facilities should a major earthquake occur in the metropolitan area. The
Sheriff should consider housing most inmates who represent security threats to the jails
and the community at the more earthquake-resistant Inverness Jail. (Page 51)
II. OUTSIDE OVERSIGHT
A goal of any well-organized and managed local government is to develop and
promote a clearly defined and integrated mechanism of government services that
responds to the needs of the community. A county jail system is perhaps the essential
gear in that mechanism, because in its starkest terms an effective jail system must be
ultimately available as the final tool of enforcement of community values.

-7-

For a local jail system to function effectively in a network of integrated public
services requires strong collaboration from all levels of government and from the
community itself. Local jails should never be operated independent of outside oversight
and input. First, lack of oversight invites abuse within jails, whose functioning is not
readily open to observation. Second, lack of oversight often leads to the financial
mismanagement of taxpayers’ resources. Finally, lack of oversight and understanding of
jail operations can thwart efforts to streamline county law enforcement activities and
integrate them into general county services.
For these reasons, experts on correctional systems have long understood and
emphasized the necessity for broad outside input into jail operations. In describing the
role outside authorities, such as county commissions, should play in jail operations the
National Institute of Corrections (NIC), a federally funded advisor for local jails,
concludes that such authorities “as much as jail officials, (are) responsible for effective
jail functioning in the local criminal justice system.”2 The NIC recommends a detailed
program of outside oversight by local funding authorities like county commissions that
“includes a mutually agreed-upon process for assessing jail operations and management
and for reviewing jail conditions” and ensures that the outside authority has “an active
and meaningful role in addressing jail liability and cost issues.”3
For years, Oregon law has recognized the need for substantial outside oversight of
local jails, and in fact, state law requires a level of outside oversight well beyond even
that envisioned by the NIC. State law has established at least three interlocking levels of
outside oversight of local jail systems, oversight from differing sectors of government.
County government, state government and the state court system, through the grand jury
process, are all assigned independent jail inspection and advisory functions. In
Multnomah County, however, it appears that the oversight functions of county and state
government officials have not been pursued by those officials4. Only annual grand jury
inspections, as required by state law, have been conducted regularly and in accordance
with law.
A.

County commissioners as inspectors of local jails

Since 1854, Oregon law has assigned to county commissions the task of inspector
of local jails.5 The state law requires the entire commission at least once each term to
visit, inspect and “examine fully into the local correctional facility.” Violations of law
observed by the commission must immediately be reported to the District Attorney. The
2001 Corrections Grand Jury determined that this law was not being followed and that
commissioners were not making attempts to inspect the jail system and familiarize

2

Bowker, G., Jail Resources Issues—What Every Funding Authority Needs to Know, NIC (2002), p. 25
Id., p. 29
4
We note that some of the Commissioners have visited the local jails in varying degrees, although there
does not appear to be any joint or systematic inspection effort by the Board of Commissioners.
5
ORS 169.040
3

-8-

themselves with its operations.6 Three years later the Sheriff’s Office itself, in a report
on compliance with state standards, was troubled with the fact that only one of the
commissioners had even visited the jail.7 To date there is still no evidence that the
commissioners are complying with this legal obligation in any systematic fashion. To
assure effective jail administration that is coordinated with other county services, the NIC
recommends that county commissioners “visit the jail regularly to observe operations, . . .
attend training, . . . read jail-related periodicals, . . . thoroughly review jail related reports
prepared by jail officials and inspection officials, . . . and meet with jail officials
regularly.” In Oregon, this recommendation is actually a state law. The community
benefits by having fully engaged, knowledgeable leadership that understands the mission
and operations of our local jails
B.

State Department of Corrections oversight and advice

In addition to operating state prisons, the Oregon Department of Corrections is
assigned by state law the function of inspecting local correctional facilities and advising
local officials on jail planning and operations. State law requires the Director of the
Oregon Department of Corrections to “designate staff to provide technical assistance to
local government agencies in the planning and operation of local correctional facilities.”8
State law also requires the Department of Corrections to “inspect local correctional
facilities . . . to ensure compliance with (state) standards.” Other state laws require the
Department of Corrections to review the plans for new construction or “major
renovation” of county correctional facilities, and to make recommendations which are not
binding unless they apply to a state jail standard.9 State law also requires the Department
to issue a manual of guidelines for the operation of local correctional facilities.10
In sum, these laws mandate a substantial state oversight and advisory function of
sheriff’s jail operations around the state. The intent of these statutes is consistent with the
concept of broad and overlapping outside oversight policies for correctional facilities
endorsed by the NIC. The Oregon Department of Corrections, however, has by and large
deferred to local officials in this task or has neglected to conduct its functions under law.
This was a concern of the 2001 Corrections Grand Jury, noted in their report.11
No state inspection of local facilities has been conducted since 2003. The
inspections conducted prior to that time appear to be cursory examinations from
boilerplate checklists regarding state standards. The Department has now decided to turn
this inspection function over the Oregon State Sheriffs' Association, although no
agreement on that transfer of responsibility has been finalized. This association conducts
reciprocal inspections of its members’ jail facilities. If the purpose of a law or policy is
6

Report, Conditions and Management of Correctional Facilities within Multnomah County, Oct.-Dec.
2001, p.8.
7
10/31/04 Memo, MCSO Sgt. Charlotte Hasson, p.3.
8
ORS 169.070
9
ORS 169.085
10
ORS 169.090
11
Report, Conditions and Management of Correctional Facilities Within Multnomah County, Oct.-Dec.
2001, p.8.

-9-

to provide outside advice or oversight of an operation from a different level of
government, turning the inspection process back to the sheriffs themselves would seem to
defeat that purpose.12
Regarding other legal requirements discussed above, it would appear that DOC
has not designated staff to perform advisory functions with local jails and that they have
not produced a manual for jail operations, but have relied on that produced by the Oregon
State Sheriff’s Association. In short, their statutory obligations appear to have been
neglected.
C.

Corrections Grand Jury investigation

Oregon law requires a grand jury to be convened each year to “inquire into the
condition and management of every correctional facility and youth correctional
facility…in the county.”13 This is the one level of oversight of local jails that has
occurred regularly. Each year for at least 25 years a Multnomah County grand jury has
conducted inspections of local jails, prisons and juvenile correctional facilities. These
investigations have been extensive, requiring testimony from hundreds of witnesses in
some years. A report is prepared by law and submitted to the local presiding judge,
Sheriff, and County Commissioners. It is distributed locally to law enforcement officials
and the press.
Virtually all of the problems that will be discussed in this current report have been
observed and criticized in past grand jury reports. They include staffing issues, jail costs,
overtime, training, inspections and outside oversight, double bunking, and the
advisability of building Wapato Jail and of closing the Multnomah County Correctional
Facility in Troutdale (MCCF) and the Restitution Center (MCRC). Unfortunately, the
grand jury has no enforcement authority and its recommendations, more often than not,
have been ignored.14
D.

Recommendation

It is apparent that the statutory framework of oversight and collaboration among
different levels of government is not functioning as designed by our legislature.
Organizations often bristle at what is perceived as outside interference in their operations.
We have found, however, that successful and efficient county jail systems gain far more
than they lose by establishing an open and credible relationship with the people who fund
12

While the OSSA guidelines are extensive and Multnomah County usually achieves high marks, there are
some concerns with the rigor with which these guidelines are applied. OSAA OJS C02.02.02 requires
confirmation of inmate security risks before a prisoner is received from another facility. MCDC was
audited in 2004 and found to meet that standard. All evidence, however, indicates that in 2004 none of the
compliance standards for this rule were present. An inmate was admitted to MCDC in May 2005 in
violation of this standard. He later murdered his cellmate in an incident to be discussed later in this report.
13
ORS 132.440
14
Past county sheriffs have followed the practice of filing written responses to Corrections Grand Jury
reports, outlining their position on perceived problems and plans to deal with those problems. The current
Sheriff has not responded formally to any of the grand jury reports during his term.

- 10 -

their operations. That type of relationship does not currently exist in this county, as the
Sheriff himself has stated in his grand jury testimony.
We believe that the following proposal will address that problem and will
promote a better understanding of and integration among law enforcement operations in
this county.
To assist county commissioners in their roles as inspectors of the county
corrections system, we believe that the commission should form a permanent work group
to systematically inspect the jails and report to the commissioners on their operations,
according to the requirements of state law. This work group should include
representatives from the law enforcement agencies in the county, the County Chair, the
County Commissioner responsible for LPSCC, the United States Attorney, the
administrator for the federal facility at Sheridan, people from the business community,
someone from the medical community, the Department of Corrections and the District
Attorney. It should also include an independent financial analyst. Unlike other multiagency law enforcement-related work committees, this work group must focus strictly on
the conditions and operation of the jails. It should be under the supervision of and should
publicly report periodically to the county commissioners. Such reports should be
required at least twice a year on a set schedule; initially it should meet once a month. At
first, this work group should investigate and report on the problems observed in this
report and the efforts and progress made to remedy them. This work group should be
required to examine the conclusions and recommendations of annual Corrections Grand
Juries and report to the commissioners on progress made to rectify problems observed by
that body. Finally, it should report to the Board the ramifications of all union
negotiations, and make appropriate recommendations.
We also recommend that the county make all efforts possible to include the state
Department of Corrections as a full collaborative partner in the operations of the local
jails, as required by law. When contacted in the course of this investigation, Department
of Corrections’ officials have expressed their willingness, if requested, to conduct
thorough inspections and to make their expertise available to the local jail administration.
We do not think it advisable to allow the state Sheriff’s Association to fulfill functions
assigned by law to the state Department of Corrections.
In many ways we believe that re-establishing a healthy relationship between the
county Sheriff and the Board of Commissioners is our most significant proposal. Most of
the other problems identified in this report stem from a breakdown in that relationship.
III. JAIL BED AND JUVENILE DETENTION COSTS
In government, transparent and readily available cost figures are vitally important
to assure good decisions and to maintain public credibility and support. One of the major
efforts of this investigation was an attempt to determine the actual costs of housing adult
and juvenile detainees in our county. Adult inmates are all housed by the Sheriff in the
two Multnomah County jails, MCDC and Inverness. Juvenile detainees are housed in the

- 11 -

Donald E. Long juvenile detention facility under the supervision of the Department of
Community Justice. Medical care at all three facilities is provided by the Corrections
Health division of the county Health Department. As a consequence, we were required to
examine and analyze the budgeting and cost accounting practices of all three county
departments.
To run any type of organization efficiently, a full understanding of operational
costs is essential. The departments we are dealing with here are government
organizations, however, and although both governmental and business organizations
utilize similar accounting practices, the structural differences between the two must be
noted. In private business, accounting practices must ultimately provide a true financial
picture of the operations of the organization. In the private sector, financial selfdeception has a negative impact on management decisions and ultimately on profits.
When a private business spends more money than it takes in it will go out of business and
eventually disappear.
This is not necessarily the case in government organizations, where no profit
impetus exists. The key to the continued viability of government organizations is not as
much their ability to run an efficient operation as it is their ability to obtain funding from
a government funding source, be it the Congress, a state legislature or a county or city
commission. As such, a common tendency of government financial reporting, if it is not
discouraged in some fashion, is to consciously or subconsciously structure their financial
reporting in a manner that tends to justify the continued operation of their organization, at
a greater funding level if possible. We have taken care to consider that fact in assessing
the financial data provided to us in this investigation by the three departments mentioned.
In making system costs assessments, we were very encouraged with the practices
of two of the three departments mentioned. We had little difficulty understanding and
analyzing the budgeting and accounting practices of the Health Department and the
Department of Community Justice. Their fiscal processes were transparent and logical,
and while we have significant concerns about the expenses of their operations, we have
no real disagreement with those departments regarding their cost reporting and fiscal
account structuring. We cannot say the same about the fiscal approaches of the county
Sheriff’s Office. For reasons we will explain, we found so many discrepancies in their
financial procedures we ultimately concluded we could not rely on their financial
conclusions and would need to hire an outside financial analyst to determine the costs
associated with running the corrections division of the Sheriff’s Office.
A.

County Sheriff financial reporting practices

As mentioned above, we came to the conclusion that the figures provided to us
from the county Sheriff’s Office were not sufficiently reliable to continue to use them in
analyzing the costs of that organization. We feel it necessary to detail the reasons we
came to that conclusion.

- 12 -

Initially, several current and former employees at different levels of the county
Sheriff’s Office cautioned us about relying on the figures used by the financial
department of that office. Some spoke to us under a condition of confidentiality. They
discussed questionable, inconsistent and muddled financial reporting practices in that
department. As prosecutors, we are well aware of the potential problems associated with
confidential information. We believe that this type of information needs to be
corroborated before it is considered worthy of belief. After a lengthy review of these
practices we are more than convinced that the information they provided us is correct.
B.

Concerns with MCSO bed cost figures

We have substantial concerns that some of the accounting practices of the
Sheriff’s Office regarding bed costs are “purpose-driven,” in short, prepared in a manner
designed to justify a particular result (specifically, continued funding from some source),
and are not therefore neutral and objective analyses of financial figures. We believe that
a culture of purpose-driven accounting practices has existed for a number of years in the
Sheriff’s Office, pre-dating the current Sheriff. For instance, in 2000, the Sheriff’s Office
reported to that year’s Corrections Grand Jury that the operating cost of the MCCF
facility15 was $54.02/inmate/day, by far the lowest in the system. The next year, as the
Sheriff, in what was then a controversial action, attempted to close that facility, the
Sheriff’s Office reported to the grand jury that the costs of that facility were now far more
expensive--$79.98/inmate/day, making the closing of the facility appear far more
reasonable.
Again, in 2001, in examining jail bed costs, the grand jury noted that “different
figures have emerged year-by-year under what appears to be different accounting
methods.” That year, the Sheriff’s staff estimated the daily jail bed costs at the Wapato
Jail, then in construction, would be $159/day, a figure which prompted that grand jury to
question why the past Sheriff would open a facility at that cost at the same time he was
closing the $79.98/day MCCF facility. Despite this 2001 $159/day Wapato estimate, the
current Sheriff in June 2006 proposed to the county commissioners that he could open
that entire facility, now fully constructed, for just $122/day, and that he could open the
first 150 beds for only $101/day. Additionally, we understand that the Sheriff has
extended a contractual offer to house Washington County prisoners under MCSO
supervision at Wapato Jail for only $95/day. For us, the huge discrepancy between what
the former Sheriff projected in 2001 that it would cost to house prisoners at Wapato when
construction was complete and the modest amount the current Sheriff now claims it will
take to house prisoners at that facility casts into substantial doubt all the figures presented
by the Sheriff's Office regarding that facility and the system as a whole.
Additionally, the recent reporting of jail costs has only served to heighten our
concerns about the reliability of figures presented to us by the Sheriff’s Office. The
following are some instances:

15

MCCF was a jail located in Troutdale designed to house minimum-security inmates. The current sheriff
closed that jail in 2005.

- 13 -

•

•

•

•

•

•
•

•
•

In July 2006 the Multnomah County Auditor analyzed county jail bed costs and
verified 2004 costs at Inverness and MCDC of $91.60/day and $191.17/day,
respectively, which reflect a combined average of $134.36/day. Despite this
auditor’s report to MCSO management, the Sheriff’s Office one month later
reported to the Oregon State Sheriffs' Association bed costs for the current year
of only $117/day.
When the financial department of our office in September of 2006 issued
preliminary investigation figures on bed costs of approximately $157/day
system wide, the Sheriff was asked to respond. His Chief Financial Officer
testified and concluded that while he agreed with our methodology, he
disagreed with some of the expenses we included. Although we do not agree
with those contentions, we recalculated our figures based upon his objections,
and arrived at a figure of $147.52/day. We released these figures in a
preliminary report.
After releasing these figures, the Sheriff's Executive Assistant stated to The
Oregonian that she disagreed with our figures and they had run no estimates
anywhere near those figures. The Sheriff’s Office was at that very time
preparing to release a finding showing the costs to be $120.65, exclusive of
health costs. When the $21/day health costs are added into that figure, the total
bed costs amount to $142/day, or substantially the same as the figures we
arrived at based upon the objections described above.
In examining the spreadsheets used to calculate the $120.65 figure discussed
above, we discovered that the Sheriff’s Office accountants had added the
figures incorrectly. The inability to reconcile columns of figures gives us
substantial concerns about other accounting methods.
In the same Oregonian article discussed above, the Sheriff’s Office stated they
believed they were losing one million dollars each year in their agreement to
house state prisoners under Senate Bill 1145 (to be discussed in detail later in
this report). In testimony before the grand jury on October 20, 2006, the Sheriff
himself stated that his department is losing $6.8 million a year in that
agreement.
On October 31, 2006 we obtained from the Sheriff’s Office yet another estimate
of jail bed costs. This time the figure presented was $139.92/day.
We have discovered numerous instances where the Sheriff’s Office has
understated costs in spreadsheets prepared for various reasons, including for
negotiating the county’s contract to house federal prisoners. Not included in
calculations for the federal contract were costs for transport, temporary hold
facilities, medical and depreciation that are allowed in establishing a price for
the federal contract. This is detailed in Appendix A, attached.
The Sheriff's Office has consistently decided not to include medical costs in its
bed cost analyses, discounting costs by over 10%.
The Sheriff's Office financial officers have continued calculate costs by using
the improper accounting practice of "revenue netting," which deducts specific
revenues from the costs of individual programs within a system. The practice
makes unit costs appear lower than they actually are, and also makes an analysis
of full operating costs impossible.
- 14 -

C.

Concerns with budgeting practices

As we have attempted to assemble financial figures to complete this report, we
have observed that the Multnomah County Sheriff’s Office has utilized budgeting
techniques that have had the effect of obscuring the management practices of the
organization. This has made it difficult for outside observers to determine how the
organization functions and has required our use of independent financial expertise to
unravel the financial outline of the jail system.
This county requires what has been referred to as “priority-based budgeting,”
where each department is required to present a number of “program offer” packages to
the county commissioners for funding. The commissioners then review the packages
individually and decide to “purchase” some and possibly reject funding for others. The
Sheriff’s Office this year has had 66 funding program offers submitted to and approved
for funding by the county commission. The theory behind this priority-based budgeting
is to present an accurate financial picture of various programs so that the commissioners
can make an informed decision on whether or not to fund that particular program. As
such, the commission expects to be presented with figures that accurately reflect the costs
of that program operating alone. This requires an accurate allocation of administrative
and overhead costs to each program in a manner that truly represents their use in that
program.
An example in the jail might be the use of administrative personnel. Payroll
clerks are necessary, for instance, to assemble and process a payroll. The more
employees an organization has, the more payroll clerks will be needed to process
payrolls. The very first payroll clerk should be considered a fixed cost, because the
organization will need this position whether it has hired ten employees or one hundred.
The more employees on the payroll, however, the more payroll clerks will need to be
hired. These additional clerks represent variable expenses, because their number should
fluctuate with the number of employees they serve.
In our estimation, the Sheriff’s Office has “front loaded” an enormous amount of
variable expenses onto its initial MCDC and Inverness Jail program offers. In short, in
presenting its program offers for the first housing units in the two county jails, it has
included virtually all the administrative costs of a fully operating jail. In doing so, it has
labeled these costs as fixed costs when in fact most are variable costs. Labeling variable
costs as fixed costs has the effect of protecting those costs from budget cuts. We disagree
with this practice and our financial consultants concur.
An examination of these budgeting practices is informative. In presenting its
program offers for MCDC, for example, the Sheriff has told the county commission that
in opening the first 46 beds in that facility they need to budget $13,454,488 and employ
103 full-time workers. To supervise these 46 prisoners, the Sheriff has declared that he
needs ten full-time sergeants, three lieutenants and a jail commander. He states he needs
sixteen deputies merely to operate control panels to open doors. He claims to need 51
administrative and core jail support staff. In addition, he states that he also needs

- 15 -

approximately 20 corrections deputies to do the actual supervision. The average jail bed
cost for this unit would figure out to a staggering $801/day. Put another way, no 46-bed
jail would require a staffing pattern such as this. What has happened is that the Sheriff’s
financial office has decided to apply almost all administrative costs of the entire facility
to only one unit.
While we believe that some of these costs may be fixed costs, most are not. Most
are variable costs that are applicable throughout the facility and should fluctuate with the
number of additional beds being used in the jail. This is especially so since the
fluctuation of jail beds in the Multnomah county jail system is an entirely predictable
phenomenon. The jail system in this county operates at a virtual zero percent vacancy
rate. The constant “matrix” release of inmates is evidence of this. Fluctuations in
occupancy rates beyond one or two inmates a day are a result of highly predictable
maintenance or budgetary issues. Entire jail dorms or modules are closed or opened
based on planned budget projections or maintenance determined months in advance. As
a consequence, most administrative costs such as those described above are manageable
as variable costs.
This is important because the budgeting practices described above present an
inaccurate picture of the economic position of the jail system. The practice makes the
initial beds in a jail appear extremely expensive and it makes the last beds in the system
appear deceptively economical, because few administrative costs are assigned to them.
This makes it impossible, without a detailed analysis and reassessment of the
figures, for the public to understand the financial management of the system. It also has
made it difficult for the county commission to knowledgeably evaluate and accept or
reject the Sheriff’s program offers, especially when, as noted, they have not involved
themselves in understanding these matters in any systematic fashion.
We expressed our concerns about this budgeting process to Sheriff's officials.
They were adamant that the budget priorities and costs described above for MCDC were
an accurate method to assign administrative costs for the MCDC facility. They indicated
that this type of front-loading of administrative expenses was a completely legitimate
budgeting exercise.
Given that position, we were concerned when we examined the Sheriff’s June
2006 program offer for the opening of the Wapato Jail, a program offer that was
ultimately rejected by the county commission. In that program offer, the Sheriff sought
to obtain money from the county general fund to open the entire Wapato facility. In
laying out their base or first budget program offer, the opening of the first 150 beds at
that jail, the Sheriff took the exact opposite budgeting strategy than those taken for
program offers for MCDC and Inverness Jail. Instead of front-loading all administrative
costs on the first program offer, as was done for the offers at MCDC and Inverness, the
Sheriff loaded most of the administrative and support functions onto the second program
offer.

- 16 -

A breakdown of this offer is instructive:
•

•

•

The Sheriff chose to apply no “administrative support” costs to his first 150-bed
Wapato program offer. All fiscal support, human resources, technology and
payroll support costs were omitted from the base offer and assigned to the second
75-bed program offer. It is difficult, therefore, to see how computers would
function, how benefits would be calculated for employees and how employees at
Wapato could be paid under those circumstances.
Even more curiously, no transport function was assigned to the 150-bed offer, so
it is difficult to discern how inmates would even be brought to or taken from that
jail under the terms of this program offer. All transport costs were assigned to the
second offer.
The Corrections health budget for the first 150 prisoners was listed as $158,137 or
$5.77/day per inmate (the system wide medical cost is roughly $21/day). The
medical cost, however, for the second program offer for housing 75 prisoners is
$832,473, or $60.81/day/inmate. It is hard to understand how the first set of
Wapato prisoners could be expected to be ten times more healthy than the second
set.

By excluding so many administrative costs from his first program offer, the
Sheriff proposed that he could open the first beds at Wapato Jail for only $101/day,
instead of the $801/day cost of the jail beds from the base MCDC program offer, and the
$269/day cost of jail beds for the base Inverness Jail program offer. However, into the
Sheriff’s second Wapato program offer, designed to open another 75 beds at that facility,
the Sheriff dumped virtually all of the administrative costs. Beds in this second Wapato
program offer would have cost $304/day.
What has happened here appears obvious to us, and it is disturbing. When dealing
with the fully operating facilities at MCDC and Inverness, the Sheriff has adopted a
budgeting procedure designed to portray the first beds opened at that facility as the most
expensive beds, providing a disincentive to the county commission to ever cut the last
beds, because they are pictured in this budgeting scheme as relatively inexpensive. When
attempting to obtain funding to open a new facility at Wapato, however, the Sheriff has
taken the direct opposite budgeting practice. He has portrayed his first beds opened as
the least expensive. Only after those beds are opened would he begin to assign general
administrative costs, and then only to later program offers. Attached as Appendix B are
the three program offers discussed above, demonstrating the exact staffing proposed by
the Sheriff in his budget proposals for these three facilities.
We are concerned that this pattern of budgeting practices presents a distorted
picture of jail functioning. More importantly, the county budgeting process is
specifically designed to be transparent and to allow the county commission to establish
funding priorities. These types of practices serve to obscure from that funding authority
the actual operation of jails and the costs associated with those facilities.

- 17 -

In sum, these accounting and budgeting practices have caused us great concern.
Because of that we concluded that the use of an outside, independent financial consultant
was essential in analyzing the Sheriff’s costs and budgets.
We do not intend to blame the Sheriff’s Office entirely for this situation. As
noted, it is a natural and understandable inclination of a governmental agency to engage
in exactly the type of practices described here. If allowed, the instinct for institutional
preservation will generally produce similar results, especially in an era of financial
uncertainty and economic contraction as has been experienced by this county over the
last five years. We suspect that similar practices are being replicated by governmental
agencies throughout the country. We understand that budget figures publicly distributed
from public agencies may understate costs for political purposes. For this reason, we
ignored publicly stated costs from other jails and did our own calculations directly from
agency budgets.
We feel, as we noted earlier, that the county commission must adopt a more
engaged role in partnership with the Sheriff. When funding authorities such as the
county commissioners fail to understand the operations of departments they fund, such
problems are likely to recur.
Recommendation - The Sheriff’s cost reporting and budgeting practices should be a
major focus of the workgroup we propose in Section II.
IV. DAILY JAIL BED COSTS
A.

MCSO Facilities

A preliminary examination of the unit costs of housing adult prisoners in the
county jail system revealed an average daily per prisoner cost of $157.57. This section
analyzes this unit cost figure on three levels. First, we will explain how we arrived at that
figure. Second, we will compare that figure with the costs of corrections facilities in
other jurisdictions. Finally, we will detail the reasons that appear to have produced this
figure.
1. Methodology - We arrived at a preliminary cost estimate of $157.57/inmate/day by
utilizing budget data for the Sheriff’s Office in its fiscal year 2007 (July 1, 2006 through
June 30, 2007) budget request to the county commissioners. The resultant unit cost,
$157.57/inmate/day, is the projected cost of jail beds in fiscal year 2007 (July 1, 2006June 30, 2007) based on Sheriff’s Office budget figures. Since 2005, all departments of
Multnomah County government are required to submit separate budget proposals for all
the various areas of their operations. These budget “program offers” are designed to
present the county commission with a detailed financial picture of the many different
aspects of a department’s operations, and thereby allow the commissioners to decide
whether to find that particular "program or not.

- 18 -

This “priority-based budgeting” practice forces a department to break down,
assess, and in a budget document, declare the actual costs of each of its various programs.
The 2007 Multnomah County Adopted Budget reflects 66 “program offers” the Sheriff’s
Office submitted to the county commission, and which were then accepted for funding by
that body. Thirty of these were strictly jail-related programs that are under the direct
administration of the Corrections Division of the Sheriff’s Office. We excluded some of
these corrections programs because we felt they were not related to the cost of housing an
inmate. For instance, we excluded from our calculation the costs of mothballing the
unused Wapato Jail facility. We also added a percentage of the costs of the MCSO
executive budget and training program offers,16 a percentage (80%) that was consistent
with the share of corrections spending in the entire MCSO budget. Because health care
for county prisoners is provided by the county health department’s Corrections Health
section we utilized the same process to analyze their budget and assess the total health
care costs for the jail system.
We totaled the cost of all the MCSO and health department jail-related programs
and divided this cost by the total number of annual jail bed days in the system. Based on
information provided for us by the National Institute of Corrections on calculating per
diem jail bed costs, we used the following assumptions to arrive at the total number of
annual jail bed days. We understand these to be standard assumptions used throughout
the nation. We considered a jail bed day to be a unit that is measured by the actual
overnight occupancy by a prisoner in one of the Multnomah County jail facilities. We
did not consider electronic monitoring or any form of “house arrest” to be a jail bed day.
We did not consider the county’s “weekend in jail” program as jail bed days. This
program allows judges to sentence low level offenders to “serve” time from 8:00 a.m. to
4:00 p.m. on Saturdays and Sundays by coming to the downtown courthouse to sit in the
courthouse jail for eight hours and returning home for the evenings. Because inmates do
not sleep overnight in the jail we excluded this program from calculating jail bed days.17
We further assumed an occupancy rate of 100% of the 1,690 jail beds in the
system. We did this because the Multnomah County jail system operates at essentially
full capacity all the time, which is the reason, of course, for constant forced or “matrix”
releases.18 We understand the actual occupancy rate is 99%, because at the instant when
inmate counts are made at midnight there will generally be a few prisoners who have
been bailed out and are actually walking out the door while a “replacement inmate” is
16

The Sheriff’s Office has a small executive office, with eight employees and the training office has seven
employees. They service all sectors of the Sheriff’s Office and their costs were not allocated in the 2007
budget, or “spread,” among the various program offers, as were all other administrative programs. This
required us to allocate a portion of their costs to corrections programs.
17
It should also be noted that because this “sanction” is supervised by the MCSO Court Security detail and
because we excluded the Court Security detail from our calculation of costs, the “weekend in jail” program
has no effect on our jail bed cost calculation. Had we included the program as jail bed days and likewise
included the Courthouse Security detail’s budget in our calculation, it would likely have raised the daily
bed costs of the system.
18
The county jail system operates under statutory maximum population numbers under ORS 169.042-.046.
After the maximum figure is reached, inmates are released according to the terms of a “matrix” which
determines which inmate can be most safely released to the community.

- 19 -

assigned to his bed. The effective occupancy rate, however, is 100%, so we used that
figure. Use of the 99% occupancy rate would actually have produced a higher daily jail
bed cost, adding $1.58 to the daily rate.
After establishing total costs and total annual jail bed days, we divided these costs
by the annual bed days to arrive at our preliminary calculation. The result was the figure
of $157.57 per inmate per day, which represents the full budgeted cost of a jail bed in
Multnomah County. This analysis is attached as Appendix C.
We feel the accuracy of these figures is confirmed in work done by the
Multnomah County Auditor’s Office. In July of this year that office conducted an
independent audit of certain Sheriff’s operations, including a calculation of jail costs.
This audit checked the actual costs of jail beds in fiscal year 2005. Their calculations of
daily jail bed costs in fiscal year 2005 for the two currently existing Multnomah County
jails (MCIJ, $96.50, and MCDC, $191.17) demonstrate the average jail bed cost for those
two jails together was $134.36. Here it must be added that the Auditor’s Office analyzed
the costs of jail beds under Multnomah County’s contract with the federal government to
house 125 federal prisoners. This contract excludes several costs, such as transportation
and work crews, which are not applicable to federal prisoners, and therefore considerably
understates the actual costs of housing prisoners.
This audit was conducted on a two-year-old budget. With a cost of $134/day
established by this audit for jail beds two years ago, therefore, our preliminary estimate of
around $157/day seems not only reasonable, but conservative. When the real increases in
jail budgets over the last two years are added to the costs that are omitted in the federal
contract figures, our preliminary estimate seems accurate. As a consequence, we were
confident that our $157.57 figure accurately reflects the costs of housing prisoners in this
county.
After calculating these figures with information provided by the National Institute
on Corrections and after checking the methodology used by other jail systems, we
presented this figure to the head of the Multnomah County Sheriff’s Office financial
department when he testified before the corrections grand jury. He agreed that the
methodology accurately reflects average jail bed costs. Indeed, this has been the
methodology historically utilized by the Sheriff’s Office in analyzing their own jail
costs.19 It is also the methodology utilized by all the departments we have examined.
The Sheriff's financial officer, however, disagreed with including certain costs in
the calculation, because in his estimation they are not related to the actual costs of
housing prisoners. Specifically, in his testimony before the grand jury he stated that only
70% of booking costs should be allocated to housing prisoners because only 70% of
prisoners booked are actually housed, the remainder being released on bail or on
recognizance before they are ever taken to a jail cell. He also said that most of the cost of
inmate commissary should be excluded because inmates themselves pay for all but
19

A recent study of statewide costs which was compiled by the sheriff specifically states “Average Bed day
costs is calculated by taking the total costs divided by total beds,” the same methodology used here.

- 20 -

$70,413 of that cost. He also felt that only 50% of the costs of transportation should be
allocated to daily jail costs. Although from an accounting perspective we do not
necessarily agree with these minor shifts in allocations, we recalculated total jail costs
after removing those items from the equation. Even allowing for these adjustments, the
average daily jail bed cost in the Multnomah County jail system is $147.52. Attached as
Appendix D. On October 16, 2006 the Sheriff recalculated these figures and concluded
that their daily costs were $120.65, exclusive of corrections health costs. When the
$21/day of corrections health costs are included, the new figure from the Sheriff is about
$142/day. On October 18 of this year, the Sheriff’s Office submitted to us another cost
study for average jail bed costs, now showing the average jail cost to be only $117/day.
After discussing these figures with the Sheriff’s Office we obtained the services
of an independent financial auditor and advisor to analyze all the figures. This audit has
confirmed the accuracy of our initial calculation. See Appendix A. This audit reveals a
full cost of $157.82 per inmate per day, only twenty-five cents from our preliminary
figure. The difference is primarily attributable to a slight discrepancy in allocating
executive and training expenses within the Sheriff’s Office.
We have made substantial attempts to reconcile these figures with the Sheriff’s
Office. They have provided us with at least three different numbers over the last three
weeks. We made the effort to contact a former financial analyst from the Sheriff’s Office
who was intimately familiar with their operations. She was, in fact, recommended to us
by the Sheriff himself. She met with us and with our financial advisor and we discussed
with her our methodology and figures. She prepared her own estimate and arrived at a
figure of $149.51/day.
We understand her analysis and, while at $8.31/day difference, it is quite close to
ours, we believe it violates a fundamental principle of accounting. As described in
Appendix A, the figures reflect the practice of “revenue netting,” where revenue
generated by certain programs is offset against the cost of the program. “Revenue
netting” is considered improper in business accounting and in business management
practice because it obscures the full cost of a system. Without an accurate measure of the
full cost of a system, management of the costs of the system becomes difficult.
We have had no difficulty understanding the analysis that resulted in the $149.51
figure and, after correcting for the improper practice of “revenue netting,” the figure of
this former Sheriff’s analyst becomes $155.48, or substantially the same as ours.
2. Comparison - We have attempted to compare the unit costs of housing prisoners in
the Multnomah County jails with the costs of jails in other counties. We have analyzed
the budgets of those jails, utilizing the same methodology described above. In doing so,
we recognize that not all jail systems are exactly alike, and we also recognize that
managing costs will be somewhat different from program to program. Having
established that, however, there is no other jail system we have examined that comes
even close to duplicating the high cost of the jails in Multnomah County. Using the same
unit cost methodology, we initially calculated that King County, Washington houses

- 21 -

prisoners for about $117/day. An analysis by our financial consultant of the King County
budget figures revised that estimate to $111.18/day.20 Using the same method we
calculate that Clark County, Washington houses inmates for $103/day, and Washington
County for $89/day.
The assertion that Multnomah County operates some of the most expensive jails
in the nation is confirmed by the fact that the federal government’s Marshal Service, in
contracting with our jail system to house approximately 125 inmates each day, pays one
of the highest costs of any contract in a large city in the nation. The federal government
contracts with 180 local jail systems throughout the nation to house prisoners in their
facilities. “As of January 13, 2005 the USMS was managing 180 active (contracts)
totaling about $208 million and guaranteeing 11,203 bed spaces.”21 This works out to an
average bed cost of roughly $51/day throughout the system. This contracting system
utilizes a uniform process to determine costs and thus establishes a nationwide
methodology that allows comparisons between the costs of contracting jail systems.22
The Multnomah County contract with the USMS requires the federal government
to pay $115.90/day to house federal prisoners. Despite the fact that the contract has not
been renegotiated since 1996, the figure of $115.90/day, which has remained the same for
over a decade, is the seventh highest amount paid by the federal government to any jail
system. Only four jails in the New York City area, the jail in Kodiak, Alaska, and the jail
system in San Bernardino County, California, serving the Los Angeles area have higher
costs. All these areas, of course, have dramatically higher costs of living than the
Portland area.23
3. Reasons for the rise in jail bed costs - The rise in jail bed costs in this county could
be seen as the composite result of a number of management and administration decisions
and policies over a number of years. We will attempt later to discuss some of those
separate factors.
Before being lost, however, in the minutiae of management decisions and
practices that have resulted in increased systems costs, it is necessary to understand the
fact and effect of that result. The Multnomah County Sheriff’s Office provides a service
to the community. That service is the incarceration of individuals who are charged with
or convicted of crimes committed in the community. In the past five years that service to
the community has been reduced by 20%, as the number of available jail beds has been
cut from 2,104 to 1,690. At the same time that this service has been drastically slashed,
the budget of the corrections division has paradoxically increased in real terms by over
20

The King County jail system has contracts to house inmates from numerous local municipalities for a
$186 booking fee plus $97/day, or $257/day when housed in specialized medical units. The figures of this
contract align remarkably closely with an average daily rate of $111.18
21
Audit Report 05-28 Office of Inspector General
22
Federal jail contracts exclude costs that are not applicable to federal prisoners, such as outside crews and
transportation. As such, while the figures can be used for cost comparisons between jail systems, they
understate actual jail costs for housing other prisoners in the system.
23
Other comparable West Coast rates for USMS contracts are: Lane County, Oregon--$88.99; Alameda
County, California (serving San Francisco)--$98.00 for Oakland jail and $74.00 for Santa Rita jail.

- 22 -

10%, even after adjusting for inflation.24 While a cut in a public service might be
considered reasonable if a public savings results, a cut in public services that results in
additional spending is not.
The main reason, therefore, for the dramatic increase in jail bed costs is the simple
fact that over the last five years county corrections budgets have been allowed to increase
by over 10% in real terms while services have been decreased by 20%, leading to a
completely predictable increase of over 30% in individual jail bed costs. With that in
mind, we wish to address some of the management decisions and practices that have
resulted in transformed the Multnomah County jails from a moderately expensive
program into one of a handful of the most expensive programs in the country.
Before we begin a discussion of the various “drivers” of jail cost increases, we
wish to stress again the fact that in our estimation this is primarily the result of
breakdown in the partnership between the county commission and the Sheriff’s Office.
Without a reasonable and informed dialogue between a public agency and its funding
source, problems such as these are likely and are likely to be repeated.
4. Workforce costs - The primary component affecting the costs of jailing inmates in the
county jails is labor costs. As noted above, labor costs constitute 82% of the county
Sheriff’s budget, and increases in labor costs are primarily responsible for increases in the
Sheriff’s budget. These increases in labor costs are primarily attributable to two distinct
factors - contract negotiation and workforce management practices. In both of these
areas, significant problems appear to have developed which have allowed labor costs to
rise.
a. MCCDA labor contracts - Corrections officers in the Sheriff’s Office are
represented by the Multnomah County Corrections Deputies Association (MCCDA),
which negotiates periodic contracts with the county regarding compensation and work
conditions and practices. The latest contract between the county and this association was
negotiated and signed this year, resulting in a substantial increase in pay for corrections
deputies. The Corrections Grand Jury heard testimony from the county and from the
Sheriff’s Office regarding the negotiation of this contract. This testimony revealed a
problematic two-way trend in the procedure. The Sheriff testified, as he did before a
previous grand jury, that he considered the negotiation of labor contracts to be primarily a
function of the county commission or their delegates, and that his obligation was merely
to manage the aftermath of such negotiations. The bargaining team for the county, on the
other hand, testified that they considered the Sheriff to be primarily responsible for
negotiating a contract he considered appropriate. What has emerged, therefore, is a
curious “who is minding the store?” situation where the county commission, or their
delegates, believe that they are deferring to the Sheriff regarding key contractual
management points, at the same time that the Sheriff believes that he should actually be
24

In auditing aspects of the sheriff’s correction’s operation in March 2006, the Multnomah County
Auditor’s Office concluded that, adjusted for inflation, the county corrections budget had actually increased
between fiscal year 2001 and fiscal year 2005 by six percent. Since fiscal year 2005 to the present the
corrections budget has again risen dramatically, from $60.3 million to $71.0 million.

- 23 -

deferring to the county commission on the same points. The consequence is that an
effective management perspective is not fully represented in these contract negotiations,
and the ultimate result, as can be seen in the latest contract, is an agreement that makes
effective management and cost control impossible.
This agreement has awarded
MCCDA members an 11.4% cost-of-living increase over the last two years, more than
6% greater than the actual cost of living increase during that period.
We were told that this contract was signed to bring MCSO corrections deputies up
to the pay level of Washington County Sheriff’s Office corrections deputies. We were
told that an analysis of pay structure demonstrated that WCSO deputies were paid over
11% more than corrections deputies here. If that were truly the case, a deeper question
appears never to have been addressed. Jail systems are labor-intensive activities, with
75-85% of budgets resulting from personnel cost. No one disputes the fact that the
Washington County Sheriff’s Office runs a less expensive jail program than the
Multnomah County Sheriff's Office. If the Washington County jails can run a
substantially more economical jail system even when they pay their corrections officers
more than ours are paid, something must be wrong with management practices in our
jails. The productivity of our corrections staff must be substantially lower than
elsewhere, and no one from any side of the table seems to have pondered that point.
In terms of labor costs, negotiating a contract which compensates corrections
officers at the same pay level as their more productive counterparts in other counties
simply ensures that jail bed costs in this county must inevitably be higher than elsewhere.
This is not to say that corrections officers should not be paid the same as deputies
elsewhere. They should, because, by and large, they have the same job and suffer the
same dangers as officers in other facilities. The choices available, however, in
maintaining lower jail costs in a system whose finances are driven almost totally by labor
costs are simple, raise productivity of the labor force through improved management
techniques or avoid increases in the cost of labor. As this report demonstrates, we would
prefer the former, but it is apparent that the latter was not considered in the negotiation of
the last contract. As a result, increased labor costs have driven this year’s 7% increase in
the Sheriff’s budget.
b. MCSO workforce management practices - Aside from the contractual pay scales
and benefits negotiated by management, the day-to-day management of a workforce is
the primary determinant of how cost-effectively a corrections system will operate. By
their nature, jails must operate around-the-clock and many of the posts in the system must
be constantly staffed for obvious security reasons. Because of this, corrections systems
present acute management challenges and an effective operational plan is essential to
effective management of costs. Issues such as staffing patterns and employee leave
administration are critical to a successful plan. We will attempt to comment on several of
the management decisions that have had an impact of jail costs.

- 24 -

i. Sick leave abuse
Attached as Appendix E is a copy of a report on sick time use by MCSO
corrections deputies prepared by Capt. Bruce McCain of the Sheriff’s Office at our
request. The document, prepared by the Sheriff’s own staff, virtually speaks for itself on
this issue of sick leave management and abuse. 77 of the 448 corrections deputies have
used 99% of all their accumulated sick leave. 230 of those deputies have used over 90%
of accumulated sick leave. The mean usage is 91% and the average use is 84%.
The usage of sick leave has increased significantly since the last study was
completed in 1994. In fact, figures for the last year demonstrate that corrections deputies
and sergeants took 55,894 hours of sick leave. This amounts to a total of 27 work years
lost to sick leave in the Sheriff's Office. This figure of hours sick leave taken actually
exceeded the number of hours accrued by corrections deputies that year by over 10,000
hours.
The issue of rampant and unchecked sick leave usage is nothing new to jail
systems. Throughout the nation, corrections authorities consider sick leave abuse (and its
inevitable by-product, excessive overtime budgets) as a management issue that must be
addressed and continually monitored if it is not to run out of control.25 We believe that
the use of sick time in this system has indeed run out of control, with significant
management consequences.
•
•

•
•

•

Because of its essentially unplanned nature, excessive sick time produces a
scheduling nightmare for administrative staff, especially in an operation that
requires round-the-clock staffing for security reasons.
Mismanaged sick time adds enormous personnel cost increases because it
requires replacement of time lost due to abuse. Replacement of this time is
usually accomplished by overtime, adding an additional financial burden.
MCSO records for the corrections division indicate that for fiscal year 2006
the department had to replace
35,190 hours of sick time taken with deputies on overtime.
Unmanaged sick leave permits the manipulation of the system for the benefit
of those who are inclined to use that abuse to increase their paychecks.
Throughout the course of this investigation, we have heard of numerous
methods utilized by corrections deputies to bolster overtime and maintain
benefits, largely based on the virtually unchecked usage of sick leave.
Any attempt to establish an appropriate “post relief factor” (see below for a
discussion of the management tool of post factor) to determine how many
deputies are needed to fill a post is compromised because a post relief factor
which includes sick time abuse simply legitimizes and institutionalizes that
abuse.

25

Reacting to the fiscal impact of sick leave abuse, some jail systems we visited have taken aggressive
management positions on the issue. In Orange County, Fla., the jail system docks deputies “points” for
each unscheduled leave, whether later justified or not, and for every tardy arrival at work. If the deputy
reaches a certain number of points within a pre-determined time-frame, the deputy is fired.

- 25 -

Corrections Deputies Total Sick Hours
Earned Through 8/31/2006

Sick Hours Still
Available

85,175

Sick Hours Used
464,060

Total Hours Earned 549,235

- 26 -

% of Users Using 99% of Earned Sick Hours
20%

18%

17%

16%

2006
14%

12%

10%

9%

8%

1995

6%

4%

2%

3%

1991
% of 99% users, by year
- 27 -

% of Users Using 90% of Sick Time
55%

53%
50%

2006

45%

40%

40%

35%

30%

1995
25%

20%

20%

1991
15%

% of 90% users, by year
- 28 -

•

Inappropriate use of sick leave leaves employees without leave and therefore
financially vulnerable and unprotected against true illnesses, resulting in
unnecessary disability payments and calls for other employees throughout the
county to donate vacation time for these employees.26

In effect, it appears that sick leave is considered by corrections deputies and many
supervisors as an “entitlement” that is the equivalent of vacation time. Indeed, it is
considered a superior form of vacation time because no prior scheduling (and the
potential risk of leave being denied) is required.
Recommendation - To address the overwhelming tide of sick leave usage, there are
several options we believe should be contemplated.
•

•

•

•

•

Consider a system that requires a significant independent observation and
monitoring component. Leaving sick time policing to line supervisors is a
haphazard method whose results are not guaranteed. We consequently believe
that the county should implement a centralized plan of sick time (and perhaps all
type of leave) oversight, managed and monitored by a central county agency. The
county in general would benefit from an accurate accounting of the work patterns
of all its employees, not just sheriff’s deputies.
We believe the Sheriff has not confronted the issue of sick leave abuse and we
recommend that concerted efforts be made to do so within the Sheriff’s Office.
We believe that this issue should be a major topic of review of the oversight
committee discussed above. Should this fail to address the problem we believe the
Board should consider additional actions, including a countywide investigatory
body and county ordinances designed to sanction abuse of leave by employees
that costs the taxpayers millions of dollars.
Consider a restructuring of sick time benefits for corrections officers to allow a
graduated buyout of sick time at the time the employee leaves county service.
This might allow increasing percentage of sick time bought on separation (e.g.payment of 35% for hours over 400 accumulated, 50% of hours over 800, 75% of
hours over 1,200 and 100% of hours over 1,800). This would provide an
incentive for employees to actually accumulate sick leave hours. Considering the
level at which sick time is used and the overtime it generates, the county could
save a lot of money.
If the county Board wishes to avoid the regulation of leave among its employees,
it should consider eliminating sick leave and simply establishing a total leave
figure that is a combination of current vacation, personal holiday and sick leave
(but not the entire 96 hours of sick time currently accumulated).
Negotiate contractual terms where needed, or simply institute procedures where
negotiation is not required, which will put some teeth into the enforcement of sick
leave rules.

26

In the last three years, MCCDA employees have used 1,874 hours of sick leave donated from the
vacation leave accounts of other county employees under county policies that allow employees to donate
vacation time to other employees to be used as sick leave for catastrophic medical conditions.

- 29 -

ii. Overtime and Comp-time
Overtime is a necessary component of any jail system. Because jails operate 24
hours a day, 7 days a week there will always be unexpected situations that give rise to
overtime. To a point it is more economical to staff vacancies with overtime rather than
hiring new staff. However, when an organization is understaffed and positions are not
filled, the resulting overtime can quickly become uneconomical.
For several years the Sheriff’s Office did not hire to their full potential. Past
grand juries had pointed this out and were critical of the practice. Past union presidents
testified that there was too much overtime and it had created an unsafe work
environment. This year the Sheriff testified that he has changed this practice in an effort
to reduce overtime and that there has been some improvement as a result.
Overtime is the result of many things; some of which can be controlled and some
cannot. In looking at the reasons for overtime last year in corrections we found that sick
leave accounts for about one half of the overtime costs. We have discussed above the
reasons sick time is out of control.
In March 2006 the Multnomah County Auditor released her report after reviewing
the sheriff’s overtime. We agree with her recommendations and set them out to reiterate
what her investigation discovered. We appreciate the time the auditor and her staff spent
with us. It gave us a better perspective of the problems and the difficulty of finding
solutions. We were told that what made their jobs so hard was the amount of time that
correction officers were not at their posts. The reasons varied from such things as sick
time and vacation to suicide and medical watches.
The auditor made the following recommendations, which we support:
1.

A review is necessary to see what the effect was on staffing with the cuts in
the number of jail beds in the last several years.

2.

The Sheriff needs to control and limit the use of compensatory time.

3.

A significant effort is needed to manage the high number of absences.

4.

The Sheriff needs to dedicate on-going analysis of staffing trends, conduct a
study of staffing needs, and set a performance measure target for overtime in
agreement with the Board of County Commissioners.

5.

The audit found that MCSO frequently approved vacation, personal holidays,
and comp time at levels above the maximum number of slots allowed. This
practice needed to be curtailed.
At one facility during one shift,
vacation/personal holiday slots were exceeded 170 times during the year.

- 30 -

6.

Training related absences could be managed if the training was not scheduled
during periods of high vacation usage.

7.

The county should evaluate how their existing personnel policies affect the
Sheriff’s 24/7 operation.

8.

The county and the Sheriff should coordinate with each other to establish
long-term strategies and goals.

Compensatory or comp time generally occurs when someone is compensated for
more than a 40-hour week. A corrections officer can elect to be paid for the time worked,
or take comp time.27 Either method will result in time-and-a-half or double time.
Currently each corrections officer can accumulate 80 hours of comp time. Last year’s
Corrections Grand Jury outlined the problems with comp time. In addition, the Auditor’s
report, after studying the practice in great depth, concluded it was the fastest growing
category of absence. Unfortunately when comp time is used the position is filled by
someone who is working overtime and who is getting time-and-half or double time. The
Sheriff acknowledges this has become a big problem and he would like to find a way to
fix it.
Last year, after the Sheriff hired additional officers, the overtime fell slightly.
Unfortunately, the same cannot be said for comp time. Last year’s Correction Grand Jury
was given the comp time hours for the previous 5 years:
Corrections Comp Time Hours for
FY00 through FY05
Fiscal Years

Hours

FY00-01

4,884.25

FY01-02

8,756.50

FY02-03

13,978.75

FY03-04

18,117.50

FY04-05

24,884.25

For FY05-06 the numbers have continued to increase to about 29,000. This is
costing the county a lot of money.
iii. Other leave issues
We heard from our Sheriff and all the other administrators of jails we visited that
one of their greatest problems in dealing with unplanned leave is the Family Medical
Leave Act (FMLA). They understand the importance of family leave for paternity and
27

This is the county policy, although that policy is silent on how much comp time can be accumulated.

- 31 -

maternity leave, catastrophic illness, sick children and other unplanned sicknesses.
However, it has become an instrument that is used by people who want to take advantage
of the system. People who want to abuse the system use it as a shield and as a result in
some cases it becomes almost impossible to investigate sick abuse.
We understand that we cannot change FMLA and that abuses will occur.
However, this problem is not unique to Multnomah County. The best practice is constant
monitoring and management, including employee discipline, to establish an atmosphere
where abuse will not develop. We believe that medical documentation should be
required before approving any county FMLA leave.
Finally, in the last several months we have become aware of a growing problem
that has resulted from the most recent union negotiations. In that contract the county in
order to try and save overtime cost was able to put a clause into the contract that sick time
did not count toward computing the 40-hour week for purposes of overtime. Prior to that
time if an officer reported sick on four days and then worked one day, he would be
eligible for over time if he worked another day. The new contract eliminated this
practice. However, it still allowed vacation time and personal holidays to be considered
in the 40-hour week. As a result of the new provision mandatory overtime and double
time has “skyrocketed” in the words of one command staff. Officers are using vacation
time and then coming in on the 7th day and getting double time.
The new contractual language was narrowly approved and now it appears many
officers are reacting. We were told that officers sometimes are difficult to find and that
sergeants, who should be supervising, are spending most of their shifts on the phone
trying to find people for overtime positions. This is obviously a very bad situation and
something that must be addressed quickly.
iv. Post relief factor
Because the jail is a round-the-clock operation, most posts must be filled all the
time. For a variety of reasons, a person hired for a post will be gone at various times of
the year for various reasons. The reasons are many, including vacation, training, illness,
holiday, a death in the family, etc. To ensure that someone will always be at a post there
must be more people hired than posts. What makes this even harder to understand is that
in a jail people are constantly being transported and certain things, such as meals, must
always be provided. The term “post factor” or a “post relief factor” is used to denote how
many people you need in the system to fill one position. Currently in Multnomah County
the factor is 1.82. That means for every 100 posts (positions) you need to hire 182
people.
Many people looking at a 1.82 post factor would think that was too high. In
practice it may not be high enough. There are reasons to believe that it is too low. If it is
too low it needs to be raised. A post relief factor that is too low will generate too much
overtime.

- 32 -

The formula for coming up with an adequate post relief factor is not difficult to
understand. However, getting the right numbers and then fitting them into the current
staffing patterns is extremely difficult. If the current staffing patterns are wrong and they
are accepted you can build some great inefficiencies into the post relief factor and that is
something that needs to be avoided.
Washington County has done post relief factor studies and they illustrate what
must be calculated. The following were their conclusions in establishing their post relief
factor:

Post Factor Relief
Total Days Per FTE

365

Less Weekends

261

Less Vacation

245.6

Less Holidays

232.6

Less Sick Days

Days Remaining

227.9

Less Other Pay

221.4

Less Other Absences

209.8

Less Vacancies

202.2

Less Training

171.4 Total Available
Days

The number of days available is then divided into the number of days in a year,
because the jail must be opened 365 days a year.
The post factor under these conditions is about 2.1.
One of our primary concerns about using current numbers is that if they are
accepted, any post relief factor will include inefficiencies and abuses. Washington
County factored in about 10 days for sick and FLMA absences. In Multnomah County
because of what we see as sick time abuse we would have to factor in about 15-17 days
for sick time.

- 33 -

In addition, another significant problem with the post relief factor was described
in a recent study by the King County Auditor.28 In a report released on October 9, 2006
the auditor stated: “An inherent weakness of the relief factor method is that it assumes
that all leave taken by officers occurs in a uniform manner throughout the year.” The
Seattle Auditor stated in the report that time taken off is anything but uniform by
corrections officers. The Multnomah County Auditor provided us with overtime taken by
month for the 3 years beginning in FY03 (See attached Appendix F). In January FY05
the number of overtime hours was under 8,000. In June of the same year it rose to over
16,000.
These figures show that unless time taken off during the year is more uniform a
high post relief factor could result in having too many people during the winter and not
enough during the summer.
Recommendation - We recommend that a staffing study be conducted which includes an
analysis of an appropriate post relief factor. The current post relief factor was set some
time ago. It constantly needs to be reviewed because things change. Employee
demography and service produce significant changes to a post factor. For instance, the
longer the average deputy has spent on the job, the more vacation is accrued annually, by
contract. This requires constant re-evaluation of appropriate staffing levels and post
relief factors.
The study should review the post relief factor between various posts. There may
well be a different factor between commanders in a jail and line officers. The study
should examine how many people are needed to fill posts at various times of the day and
of the week. The activity in a jail varies considerably during different times. These
things need to be looked at by an outside, independent expert. The cost of a study by an
independent, outside expert may be high, but the potential savings justify the process.
This study should be used as a basis for continued monitoring by the financial arm of the
oversight committee recommended above.
v. Closing the least expensive jails in the county
One of the key reasons for the increase in jail costs is the fact that the two least
expensive jails in the system have been closed. In 2004 the Sheriff administered four
separate correctional facilities, MCDC, Inverness Jail (MCIJ), the Multnomah County
Correctional Facility (MCCF), and the Multnomah County Restitution Center (MCRC), a
work release center.
In 2005, MCCF and MCRC were permanently closed.
In July 2006 the County Auditor prepared an audit of jail bed costs based on costs
for fiscal year 2005. The report concluded, consistent with common knowledge and
years of testimony before Corrections Grand Juries, that these two facilities were the least
expensive correctional facilities in the county. The figures for fiscal year 2005 were:
28

See Performance Audit Jail Overtime, King County Auditor's Office, Oct. 9, 2006

- 34 -

MCDC--$191.17/day
MCIJ--$95.60/day
MCRC--$84.08/day
MCCF-- $52.10/day
Closing the county’s most economical two jails has had the unavoidable effect of
increasing the overall cost of jail beds in the system. We do not agree with the decisions
to close these two facilities, and it should be pointed out that prior Corrections Grand
Juries have come to the same conclusion,29 as both the previous and current Sheriffs have
at one time or another attempted to close these facilities down.
Responsibility for the decisions to close these facilities is obscured, to some
extent, by the politics of the actions. We understand that political pressure has been
applied for the closure of these facilities from a number of sources outside the Sheriff’s
Office, including in the case of MCCF, from the commissioners’ office. Wherever the
responsibility for the decision lies, from a cost perspective, it was a decision whose
repercussions have been more costs and less jail space.
vi. Medical costs
A major cost of running any jail system is the cost of medical care. 13% of the
daily jail bed costs in our system are health care costs. In this county’s jail system and
juvenile detention, health care is operated by the county Health Department’s Corrections
Health division. We are in complete agreement with the fact that this department
provides an excellent service. It is one of only six county jails in the Pacific Northwest
that is certified by the National Commission on Correctional Health Care (NCCHC), the
gold standard of approval in the business.30
By way of background, corrections facilities, be they prisons or jails, effectively
have three choices for operating health care systems. They can run the operation
themselves, they can have a local public health service run the operation, or they can
contract out health services to a private health care service. As noted, our county has
chosen to have the Health Department run the health service.
In the systems we have studied, using local public health services like our Health
Department is invariably the most expensive option, and the officials of the Health
Department with whom we spoke readily agreed that our Corrections Health program is
expensive. As we noted, the figures for these costs were transparently available to us and
we believe we are in substantial agreement with that department about their costs. We
29

See 2001 Corrections Grand Jury report. Not only was MCCF the least expensive jail in the system, but
it was also the most desired location for inmates, who preferred to be housed there for a number of reasons.
This provided a substantial incentive for compliant inmate behavior, as inmates sought to behave well so
they could be approved for MCCF housing.
30
The others are Washington County, Linn County, King County (Washington), Whatcom County
(Washington) and Ada County (Idaho).

- 35 -

calculate the per inmate costs at $21/day for health care, or over $7,700 per inmate each
year.
We have made a concerted attempt to analyze these costs and compare them with
the health care costs in other jails. At the outset, we determined we would only compare
the costs of our health care to other NCCHC certified facilities. In this manner, we
sought to make sure, first, that we were comparing comparable systems, and second, that
it was understood that we support the philosophy of providing excellent health services
for individuals incarcerated in our jails.
After limiting our examination to NCCHC certified programs we found that the
systems run by public health agencies are roughly 80-90% more expensive than programs
meeting the same health care standards but run by the jails themselves or by private
contractors. Like the high costs for the Sheriff’s staff in the jails, these costs are driven
by high labor expenses and weak management practices. Our Corrections Health
Division was audited by the County Auditor’s Office in September 2005. Many of the
same management weaknesses that plague the Sheriff’s office management staff, such as
leave and overtime management problems, were noted in the audit of the Corrections
Health Division, which “found that weaknesses in the management of staff resources
contributed to less than optimal productivity and higher costs.”31 The Corrections Health
budget was actually cut by the county commissioner’s office for fiscal year 2007 as a
result, we believe, of the management weaknesses uncovered in that audit.
Our contacts with the Health Department have convinced us that they accept and
are actively attempting to address some of the problems described in that document.
Some of the system’s problems, however, such as labor costs, are so structural that we do
not believe costs of a system run by the Health Department could ever be reduced to a
level comparable to that of systems run by a private contractor or the Sheriff’s
department. We believe that officials in the Health Department would agree with that
assessment.
We felt it necessary to explore the potential of utilizing one of the other two
potential jail health care systems—either operated by the Sheriff or by a private health
care contractor. When contacted about this possibility, the Sheriff indicated that he was
strongly opposed to running his own health care system in the county jails.32
We then examined the option presented by private health care contractors. The
Washington County jail contracts with Prison Health Services, Inc. (PHS) to provide
health care for the jail. They provide identical NCCHC certified health services for
$11/inmate/day. Testimony from that county was that they were very satisfied with the
service provided.

31

County Auditor report, September 2005, Corrections Health—Improve management practices for costeffective care, p.1.
32
We visited the Linn County jail in Albany, one of three NCCHC certified county jail health systems in
Oregon. That health department is run by the Linn County Sheriff and costs $11/day.

- 36 -

A huge benefit of the contract signed by PHS and Washington County is that by
the terms of the contract PHS accepts all legal liability for judgments against the county
involving legal actions for inadequate jail medical care.
Recent events have
demonstrated just how significant a contract clause like that might be, as Multnomah
County is now facing the prospect of defending against a probable multi-million dollar
lawsuit involving the death of an arrestee brought to the jail. See the discussion of that
incident below.
We contacted two major private jail health care companies. They analyzed our
jail system and we discussed contractual terms. Based on our substantial discussions, we
believe that the county would save four to five million dollars each year by entering into
a contract for an NCCHC certified medical program in the jails. We doubt that there
would be much dispute about this figure from the Health Department itself, and it is
consistent with the figures for existing private health care providers in jails such as
Washington County.
Recommendation - We believe that the county commission and the Sheriff should
consider the option of privatizing the health care system in the county jails.
We believe, however, that the savings from this action, should it be taken, need to
be returned to the public health system to address the issue of support for mental health
patients in the community. It is beyond dispute that mental health support is underfunded in our community, and it is also beyond dispute that the result of this has been to
overcrowd the jail system with mental health patients. These mental health patients are
often housed for minor criminal actions, in extremely expensive jail cells, when
reasonable treatment in the community would have prevented their incarceration in the
first place. This type of system makes no sense in human terms, or in terms of
correctional philosophy or finances.
B.

JUVENILE DETENTION AND PROGRAM COSTS

The Donald E. Long juvenile custody facility in Northeast Portland was built in
the mid 1990s and has a capacity of 191 beds. The facility is operated by the county
Department of Community Justice (DCJ). In 2005, the average daily population was 88,
according to figures provided to us by DCJ.
When we first analyzed the costs of the juvenile detention beds at this facility we
considered all 88 together. Contacts with DCJ have convinced us that this position was
inaccurate. In fact, there are two distinct groups housed at that facility, one with an
average 2005 daily population of 64 housed in detention modules, and another with an
average daily population of 24 housed in program and treatment modules. The housing
conditions are similar but the funding, treatment, and programs available for the two
groups are radically different.

- 37 -

Detention bed costs - We therefore split the populations and arrived at daily bed costs for
each group separately. Based on the 64 bed average daily population, detention bed costs
are $401.31/day. We arrived at this figure from information provided for us by DCJ on
September 15, 2006.33 We disagree slightly with DCJ on their methodology. In
assigning the daily cost of beds, they divided by the entire 80-bed capacity of the
detention units used to house youths, even though that capacity is not fully utilized. We
are convinced that standard methodology calls for the use of average daily population, in
this case 64, and not the rated capacity of 80, as the divisor. Additionally, DCJ spread
the debt service load over the entire 191 beds of the facility, although only 88 are used.34
We divided the debt service load among the beds actually being used.
By any measure, the figure of $401.31/day for housing delinquents at JDH is
enormous, amounting to over $140,000 per year for each juvenile offender bed. This
expense is over three times the annual tuition at Harvard University.
We compared these costs with juvenile detention facilities in Lane County and
Marion County, using information provided for us from the budgets of those two
counties. Again, no two systems are exactly alike, but our calculations demonstrate a
juvenile detention bed cost of $158.23/day in Marion County and $226.00/day in Lane
County.
Program bed costs - We concluded that the costs for the 24 program beds (one is a sex
offender program and the other drug and alcohol program) are $389.18/day. We arrived
at this figure by totaling the budget program offers for these programs in the fiscal year
2007 budget and dividing by the average daily population of 24. We did not apply any of
the cost of the debt service load to those beds because we understand that these overhead
costs are included in transfers, essentially rent payments, made from the programs to the
facility.
As we have stated on several occasions, one of the major reasons for transparency
in public budgeting is the need to have the community and its leadership make financially
informed decisions that are hopefully aligned with community priorities. One of the
priorities of any community, perhaps its most important priority, is the treatment of its
youth. It is an important exercise in priorities, therefore, to compare the financial support
this community and state provides to these juvenile offenders, all of whom have
committed criminal offenses, to the support provided by our government to those youth
who have committed no criminal offenses. We note, therefore, that while we are
spending approximately $142,000 per offender each year for programs to treat juvenile
offenders with drug and alcohol problems, our state contributes only $3,850/year to
support each student in our state universities and only $8,246 for each student in the
Portland Public Schools.

33

Attached as Appendix G
Using DCJ’s calculations, the cost of juvenile detention beds is $308.60/day. See 9/15/06 memo from
the DCJ Business Services Manager, above.

34

- 38 -

V. LOSSES ON CONTRACTS AND AGREEMENTS TO HOUSE DETAINEES
In Multnomah County, adult and juvenile detention departments have entered into
several separate contracts or agreements with other jurisdictions to house adult and
juvenile detainees in our county’s facilities. They are:
•

•
•
•

An agreement with the State of Oregon, permitted by state law, to house
approximately 325 convicted felons serving sentences of less than twelve
months. Previously these felons would have served their sentences in state
prisons.
A contract with the United States Marshals Service to house federal prisoners.
The number fluctuates, but is currently around 150.
A contract between the Department of Community Justice and Clackamas
County to house up to 14 juveniles at JDH.
An identical agreement between DCJ and Washington County to also house up to
14 juveniles at JDH.

These agreements are extremely problematical for two reasons. First, from a
fiscal standpoint, they lose enormous amounts of money for the county since they house
prisoners from other jurisdictions at substantially below cost, amounting effectively to
vast subsidies for other jurisdictions. Up to $12 million is lost each year because of these
agreements. Second, these contracts (with the exception of the Portland Police contract)
use jail space that could otherwise house our own community’s criminals, rather than
those from elsewhere. Approximately 485 inmates from other jurisdictions are housed in
our jails.
The result is that after excluding from the 1,690 total county jail beds prisoners
awaiting transfer to other jurisdictions, inmates awaiting trials on felony charges, and the
over 400 inmates we are holding for other jurisdictions, our jails are left without space to
sufficiently handle local misdemeanor sentences, required by law to be served in the
county jails.
Because of this lack of jail space to house local misdemeanor offenders, in effect,
judges in our community no longer have the ability to jail offenders convicted of
misdemeanor assault, drunk driving, shoplifting, vandalism or trespassing. If sentenced
to jail for those types of charges, offenders are usually “matrix” released from jail due to
overcrowding. Individuals who have been previously jailed in Multnomah County (and
most arrestees have been) are well aware of the inability of the system to hold them.
Upon being booked into our jails, inmates often ask about the matrix release score for
that night, so they can determine how long they will have wait to be released. As a
consequence, our community’s justice system has lost the credibility it needs to enforce
misdemeanor offenses. One of the reasons our justice system lacks credibility is that our
jails are filled with inmates from other jurisdictions.

- 39 -

A.

Senate Bill 1145 Agreement

In 1996 the Oregon legislature passed Senate Bill 1145 which restructured the
organization of penal sanctions for felony offenders in the state. The stated purpose of
SB 1145 was to allow local communities more control over certain felons from their
community. Prior to SB 1145, all felons who were sentenced to a term of incarceration,
put on probation, or who had been released from prison on parole were under the
jurisdiction of the state government. SB 1145 allowed the various counties, if they chose
to do so, to assume control over those offenders on probation or parole and those
sentenced to incarceration of less than a year. The state government agreed to fund this
effort through direct payments to the counties to cover the costs of these programs they
had assumed from the state. Multnomah County, like all of Oregon’s counties, chose to
enter into the program.35 This county agreed to participate in SB 1145 with a promise to
the community that we would not commit county general fund dollars to finance the
program. That has been far from the case, however, in practice.
Virtually from its inception, at least for Multnomah County, the funding of the SB
1145 program has been extremely problematic. Funding levels from the state to the
counties are set based on a complicated formula. As far as the assumption by the county
of former state prisoners serving terms of less than a year, the formula for direct
payments to the county is computed using a statewide average cost of bed space. The
state payments, based on the statewide average, are currently designed to cover costs of
$79/inmate/day.36 Multnomah County currently houses in its jails approximately 325 SB
1145 prisoners for the state. As noted, these prisoners cost the county $157/day to
house.37
No one questions the fact that housing state SB 1145 prisoners in Multnomah
County’s jails loses this county a great deal of money. A straight-line multiplication of
the $157/day figure yields a cost of $18.7 million to house these prisoners. The Sheriff
only receives $7.4 million to house them. This would point to a loss of over $11 million
each year to this county to house these state prisoners.

35

Once a county decides to accept local control of offenders under SB 1145 it can only withdraw, or “opt
out,” when statewide funding for the program falls below a certain level. The conditions for opting out do
not currently exist, so Multnomah County is effectively stuck with the program.
36
The formula is more complex than a mere monetary payment to the counties based on jail bed days used.
It incorporates a projection of SB 1145 prisoners expected to be either on community supervision or jailed
in each county and then pays the county general fund a lump sum. The lump sum for the projected number
of prisoners is based on a $79/day rate. The county government then decides how much of that money will
go to the jails and how much will go to community supervision. In Multnomah County the jails get
approximately $7.4 million to house SB 1145 prisoners.
37
Housing inmates in prisons is always a less expensive proposition than housing them in county jails,
where the average stay is measured in days instead of years. Administrative, booking, transport, medical
and security costs are much higher for more transient jail populations than they are in prisons.

- 40 -

The Sheriff’s financial department has disputed these figures and has argued that
the complex relationship between fixed and variable costs means that the SB 1145 beds
are somewhat less expensive to operate than the $157/day average figure would indicate.
Our financial consultant disagrees with this analysis, but in the end the dispute is of little
significance. In testimony before the grand jury, the Sheriff stated that in the estimation
of his department he is losing $6.8 million each year on the SB 1145 agreement.
Whether the figure is $11 million each year or $6.8 million each year it is an enormous
sum and represents a huge and unjustifiable subsidy to the state government.38
Recommendation - We believe that county government should attempt to negotiate an
intergovernmental agreement with the state Department of Corrections to allow that
department to use the Wapato Jail facilities, under a nominal lease agreement, to house
SB 1145 prisoners. The state Department of Corrections has a proven track record of
operating prisons inexpensively, even in the Portland area where it currently operates the
Columbia River Correctional Institute for approximately $65/inmate/day.
When SB 1145 was passed, this was an option clearly envisioned by the
legislature, which passed ORS 169.053 to achieve exactly that purpose for counties who
wished to assume control of their offenders in the community on probation and parole,
but did not wish to house state prisoners in their jails.
There are incentives for both sides in entering into such an agreement. The state
Department of Corrections would have the use of a state-of-the-art facility virtually for
free. It would house 325 Multnomah County SB 1145 prisoners, making an additional
200 beds in that facility available to the state Department of Corrections for other state
inmates. A favorable financial agreement could be arranged so that the state could
actually make money on the agreement. Throughout the Oregon prison system, costs are
around $62/day and Multnomah County should be prepared to offer substantially more
than that to obtain an agreement to eliminate housing prisoners in its own jails at
$157/day.
For Multnomah County the benefits are that we will save around $6.8 million
each year. This money can be returned to the jail to finance jail cells for criminals who
are currently being released due to overcrowding. This would effectively increase county
jail population by however many more jail cells the $6.8 million would purchase.
Additionally, the purpose of SB 1145, to keep prisoners serving short sentences close to
their community as a means of aiding their transition from prison, will be retained by
housing them at Wapato Jail.
38

The funding of the SB 1145 program to operate parole and probation supervision of offenders in the
community, operated by the county Department of Community Justice, has proven to be even more
financially problematic than jail funding. While that funding is not the direct subject of this investigation,
it is necessary to understand for perspective when analyzing losses incurred by the sheriff. While
approximately $18.5 million in state funds are used for this community supervision, the county contributes
almost $12 million in additional county general funds to the program. Thus, while the county loses at least
$6.8 million annually in housing state prisoners, it loses $12 million annually supervising state probationers
and parolees. The total yearly bill from county revenues to fund our assumption of state functions is
therefore over $18 million.

- 41 -

B.

United States Marshals Service Contract

The next agreement we examined is a contract that the county has entered into
with the United States Marshals Service to house federal prisoners. As noted, this
contract pays the county $115.90 per day to house federal prisoners. This price was
negotiated in 1996 and has never been altered since. We have examined this contract in
detail and we conclude:
•
•
•

This contract at present is a net money loser for the county.
This contract has been subject to abuse and manipulation by federal authorities.
This contract is, and has been from its inception, illegal under Oregon law.

Contract is losing money - That this contract is losing money for the county is not
seriously in dispute. In July of 2006 the Sheriff sought advice from the Multnomah
County Auditor’s Office on this point and the auditor in a “Report to Management”
concluded that “adjustments is (sic) clearly indicated” in the federal contract. Indeed, the
Sheriff’s chief financial officer testified in the grand jury that the contract should be
negotiated to a daily figure in the $130 plus range. The Sheriff himself, in his October
16, 2006 “Response to Corrections Grand Jury Status Reports” concedes “we agree that
the rate charged to the United States Marshal is in need of review.”
Furthermore, as our financial consultants have examined the spreadsheets
prepared by the Sheriff’s Office in negotiating the contract rate, we have discovered
substantial errors and omissions of costs that should be recoverable under the federal
regulations governing this contract, but have not been claimed by the Sheriff in
calculating the contract price.
Federal manipulation of the contract - As we have investigated this contract we have
discovered a pattern of disturbing instances where the federal government has
manipulated this agreement for its own ends, at the expense of Multnomah County
taxpayers. Specifically, we discovered that the federal detention center at Sheridan,
Oregon encountered difficulties contracting with medical providers. To deal with this
problem, we are told that the federal government has engaged in a systematic practice of
shipping prisoners in need of significant medical treatment to the Multnomah County jail
system. Here they are treated by Corrections Health, at great expense, and then shipped
back to Sheridan. Both the Sheriff’s Office and officials from Corrections Health
confirm this practice. We consider this an unconscionable manipulation of this contract,
especially in light of the fact that the federal government is getting below-cost service on
this contract in the first place.
Illegality of contract - Finally, we have examined the legality of this contract. ORS
169.540 states:
The United States shall pay for the support and keeping of prisoners
committed by virtue of legal proceedings issued by or under its

- 42 -

authority, the same charges and allowance provided for the support or
keeping of prisoners committed under the laws of this state.
We believe this statute is clear and requires the county to collect the entire costs
of holding a federal prisoner. A contract negotiated for less than the actual cost of
housing a federal prisoner is illegal under state law.
In his October 16, 2006 response, the Sheriff contends that “the language (of the
statute) does not require the Sheriff to recover the actual cost. Rather, it prevents the
Sheriff from charging the federal government more than the cost of housing state
prisoners.” We see absolutely no legal support for this proposition. The wording of the
statute says nothing about not charging extra for federal prisoners. The statute says the
federal government “shall pay…the same charges and allowances” as state prisoners. It
does not say “federal prisoners shall not pay more than state prisoners.” As such, the
meaning is abundantly clear.
Furthermore, we have checked the appellate history of this statute. Appellate
courts often refine and redefine statutes by appellate decisions, which serve to clarify the
meaning of the law, if necessary. No appellate court has ever issued a ruling on this law.
Consequently, there is no legal basis whatsoever for contending that the clear wording of
the law, that federal prisoners be billed for the same costs as state prisoners, means
anything but what it says.
In addition to the appellate history, however, the legislative history of this statute
is revealing. One of the very first acts of our federal Congress, on September 23, 1789,
was to pass a resolution appealing to the several states “to pass laws making it expressly
the duty of keepers of their jails to receive and safe keep therein all prisoners committed
under the authority of the United States…under the like penalties as in the case of
persons committed under the authority of such states respectively.”
In 1854, the territorial legislature of the Territory of Oregon passed the territory’s
first comprehensive jail and prison legislation. It included sections on accepting federal
prisoners designed to comply with the 1789 congressional resolution. 1855 Oregon
Statutes, Title I, Chapter 1, Section 11 and 12 of An Act Relating to County Prisons and
Their Regulations reads:
Section 11. Every sheriff, jailor, and prison keeper within
the territory, to whom any person shall be sent, or
committed by virtue of legal process, issued under the
authority of the United States, shall be required to receive
such person into his custody, and keep him safely until he
shall be discharged by due course of the laws of the United
States……..

- 43 -

Section 12. The United States shall be liable to pay for the
support and keeping of said prisoners, the same charges and
allowance as are allowed for the support and keeping of
prisoners committed under the laws of the territory.
These statutes have remained in effect with virtually the same wording for 152
years, and their clear import is to accommodate a federal request to house federal
prisoners in local facilities, but to recoup the entire cost of their housing. This is
demonstrated by examining another section of the same chapter of this 1854 Oregon law.
Section 5 requires Oregon counties with jails to accept prisoners from other Oregon
counties that had yet to construct jails, “but the county from which he was taken shall pay
all the expenses of keeping and maintaining him in said jail.” Section 11 simply
reiterates for federal prisoners what Section 5 states for prisoners from other counties—
that the jurisdiction utilizing another’s facility will be required to pay all the costs. The
Sheriff’s contention that ORS 169.540 allows him to charge less than the entire costs of
housing a prisoner is without support in a plain reading of the law, in any appellate
decisions on the statute, or in an examination of the history of the law.
A review of the costs listed by the Sheriff in this contract reveals, as we have
noted, that the Sheriff has not claimed many of the actual expenses of housing prisoners.
The Sheriff has not claimed these costs for two reasons. First, as the Sheriff notes, there
are certain costs of keeping prisoners that federal regulations prevent a contracting jail
system from recouping. This means that the state statute requirement of full
reimbursement and the federal regulation forbidding full reimbursement are
incompatible. Under these circumstances no contract can be signed without violating
either one law or the other. Second, there are a number of costs that the federal
government allows to be claimed that the Sheriff has chosen not to include in his contract
bid. We do not know why he has done this, but it has the effect of making the contract
illegal under Oregon law.
Finally, we do not understand the Sheriff’s position either politically or
economically. Given the current prospects for a loss in county revenues over the next
two years, Multnomah County should not be subsidizing the federal government, even if
it can. We believe the Sheriff’s legal position is tenuous but just because he might feel he
has the legal right to sign a document that will result in millions of dollars in losses for
the county, does not mean that he should do so.
Recommendation - Eliminate this contractual obligation unless the federal government
is willing to waive their regulations prohibiting full-cost recovery, and then pay the full
costs.
C.

JDH Contract

The Department of Community Justice currently contracts with Washington
County and Clackamas County to house up to 14 juveniles from each county in the
Donald E. Long juvenile detention facility. The contract price is $132.19/day for the first

- 44 -

ten beds and $191.00/day for the next four. The average daily population for those
contracts in 2005 was 12 for Washington County and 13 for Clackamas County.
As we discussed earlier, the actual average cost for housing juveniles for these
two counties is $401.31/day. Our county loses over two million dollars each year on this
contract. In effect, taxpayers in this county are paying over $200/day to house juvenile
detainees from other counties. We can discern no fiscally responsible reason to continue
this subsidy for neighboring counties.
Recommendation. Re-negotiate this contract to charge actual costs.
VI. INMATE SUPERVISION ISSUES
A.

Direct Supervision, Classification and Double Bunking

Over the course of the last two years, several troubling incidents have occurred in
Multnomah County jail facilities, including the murder of an inmate by another inmate.
These incidents will be discussed in detail later. The reaction to these incidents within
the jail system management and staff has generally been to blame these problems on the
double bunking of inmates in cells at MCDC originally designed for single cell
occupancy. Their claim is that the double bunking of inmates has put pressure on the
system that has made violent incidents inevitable. We disagree with this claim and
believe that double bunking is not related in any way to disciplinary and security issues in
the jail system. Instead, we believe that these problems have been generated by
management decisions that have resulted in the steady erosion and compromising of the
“direct supervision” function for which both county jails were designed.
Here, a review of the philosophy of our county jail system is in order. The two
currently operating Multnomah County jails are designed as “direct supervision” jails.
Direct supervision systems originated in the 1980s as a measure to minimize operating
costs in correctional facilities, where labor costs constitute 75% or more of operating
costs. The essential concept of a direct supervision system is that costs can be contained
and reduced by housing as many inmates as possible in dormitory or barracks type
facilities with open areas, supervised by only one corrections officer who circulates
among the dormitory. It has now been widely accepted by experts on direct supervision
jail systems that as many as 90 inmates can be supervised effectively by only one
corrections officer, if those inmates are properly identified as suitable for that type of
open housing. This represents a vast reduction of staffing ratios from the “linear” type of
jail designs popular until the 1980s.
MCDC was designed in the early 1980s and was based on these emerging direct
supervision principles. At that time, county designers traveled to other jails, including
Contra Costa County, California where the concept was being pioneered, to adapt their
designs for our new facility. In that era, jail designers were unsure of just how many
inmates could safely be directly supervised by one officer, so the modules in MCDC

- 45 -

were conservatively designed to hold only 32 inmates, well below what has been proven
feasible twenty years later.
MCIJ was also designed as a direct supervision jail. This facility was planned in
several phases beginning the late 1980s. Original direct supervision dormitories at that
jail were to hold 55 inmates, reflecting the developing understanding of the increased
capacity of these housing units. Direct supervision dormitories constructed at MCIJ in
the 1990s were designed to hold 75 inmates, again reflecting the evolving nationwide
appreciation of the increased inmate capacity of well-managed direct supervision units.
In 1996 230 single cells at MCDC were converted to double bunked cells, again
reflecting the developing understanding that direct supervision principles could support
that number of inmates in one housing unit. Direct supervision modules that previously
held 32 inmates now held 63 inmates. In theory, this number is well within the
boundaries of the supervision capacity of corrections deputies under direct supervision.
Without the rigorous application of the basic management principles of direct
supervision, however, such an effort is destined for failure, and failure is what has
happened as the Sheriff’s Office over the course of several years has lost sight of the
fundamentals of this process, at least as it is applied in MCDC.
There are three essential requirements that must be fulfilled to effectively operate
a direct supervision facility. First, there must be a sufficient number of single inmate
confinement cells to isolate inmates who are inappropriate for direct supervision.
Second, there must be a classification system that adequately identifies those inmates
who must be jailed in single cells because they are unsuited for open housing. Third,
there must a system-wide commitment to the direct supervision model. This commitment
must ensure that in actual direct supervision housing units, corrections officers actively
and vigilantly circulate among the inmates and aggressively supervise those inmates,
without resorting to the natural inclination to retreat to places of isolation from the
prisoner population.
Confinement cells—the 10% rule - Experts on direct supervision jail systems recognize
the need to isolate disruptive inmates in single inmate confinement cells. Violent or
disruptive inmates simply cannot be housed in open units with a population of otherwise
compliant prisoners. Experts who run successful direct supervision systems state
unequivocally that 10% of total jail beds in a direct supervision system must be dedicated
to single inmate confinement cells. Put another way, 90% of prisoner populations, if
properly identified, can appropriately be housed in open housing units and 10% must be
locked in single cells. A successful direct supervision system must therefore have that
critical 10% confinement cell number to operate effectively.
In the Multnomah County jail system, well over 10% of its total beds are in single
prisoner confinement cells. Consequently, a direct supervision system should be able to
readily function for the remainder of the prisoner population. In fact, well over 200 beds
in this system are currently in single inmate cells. Unfortunately, these single-inmate
cells are being misused and mismanaged in a manner that has seriously compromised the

- 46 -

direct supervision model. At MCIJ, a full 75-inmate unit of single cells is being used to
house trustees, the best inmates in the system, as a type of reward for good behavior and
because it is easier to get trustees to different locations throughout the institution at
different times. Also at MCIJ another 75-inmate single cell unit is being used to house
mental health prisoners, rather than to isolate troublemakers. We are told that many of
these inmates do not need single cell confinement, and, in fact, the unit is often populated
by inmates without mental health issues simply to fill up the beds.
As noted, in an effective direct supervision system 10% of total beds must be used
for the isolation of potential troublemakers. As a consequence of the management
practices described above, 150 single inmate isolation cells have been removed from their
best use, the confinement of these problem inmates. This has totally compromised the
direct supervision model by forcing into the open housing at MCDC a population of
inmates who do not belong in that environment.
Classification - Critical to the functioning of an effective direct supervision jail system is
the ability to analyze the custodial behavioral characteristics of inmates so they can be
appropriately assigned either to single cells or to open housing. Multnomah County’s jail
system has such a system that can reasonably predict the behavior of inmates for this
purpose. There are several different models for this type of classification system and we
believe that ours is as effective as others.39
A good classification system, however, cannot overcome deficiencies in a flawed
housing configuration. In this case, while the classification system may be able to
identify inmates who are inappropriate for open housing, if there are insufficient single
cells assigned to isolate those inmates, the classification system is essentially worthless to
the effective functioning of a direct supervision jail system.
By removing 150 available single inmate confinement cells from their proper
function of isolating troublemakers, MCSO management has forced the classification unit
to assign inmates it knows to be inappropriate for direct supervision into what were
supposed to be direct supervision modules at MCDC. Line officers supervising those
units have quickly understood that with the addition of violent and disruptive inmates to
these modules, the units could no longer function as direct supervision units. Step by
step, therefore, these officers have taken measures to manage these MCDC modules as
semi-solitary confinement units, where prisoners are celled in for up to 22 hours each
day.
We need at this point to reiterate and stress that we do not believe that the
problems at MCDC are related in any way to double bunking. These problems are the
consequence of the forced classification (actually, forced misclassification) of disruptive
inmates to MCDC units that are not designed for them. With this type of disruptive

39

Whether our classification system is as cost-effective as others should be examined. MCSO uses a
classification staff of 18 sworn deputies. Other leading direct supervision systems perform this function
much more economically with a civilian staff.

- 47 -

inmate eliminated from MCDC modules, and the classification unit has a proven ability
to identify them, these units could operate as full direct supervision modules.
Commitment to direct supervision - As we have noted, both county corrections
facilities were designed as direct supervision jails, but only MCIJ continues to operate on
that model. As a system-wide model, direct supervision has disappeared and has been
replaced by a system where 40% of the total inmate population is housed in what
amounts to semi-isolation cells. This happened as managers have reacted to any number
of issues that have arisen over the years without asking the fundamental question of how
their decision might affect the general direct supervision model. Some examples:
•
•
•
•
•
•

Corrections officers are allowed to place MCDC inmates into their cells and leave
the module unsupervised.
Whole modules at MCDC are staffed to operate without any officer on duty on
night shift, except to enter the module periodically for safety rounds.
Lines are drawn on module floors on areas around officer workstations to separate
the inmates from the officer on duty in that module.
Inmates are permitted limited time out of cells in MCDC so that officers need not
supervise more than sixteen at a time out on the floor.
As noted, classifications deputies are required to place disruptive inmates in what
should be open housing at MCDC.
Again as noted, isolation cells are used for functions other than removing inmates
from open housing.

The cumulative effect of these decisions, and undoubtedly many more, has been
to eliminate the possibility of direct supervision at MCDC. In a system where multimillion dollar jails have been specifically constructed to function under a particular
model, every operational decision made by the management of the system should be
analyzed, first and foremost, on how that decision will work in the general model. This
has not occurred; it appears, because of a lack of understanding of, or commitment to, the
fundamental operational principles of the system.
Recommendations
1. Re-establish single inmate cells as isolation cells for inmate troublemakers
throughout the system. To establish the required number of single inmate cells to
use to isolate inmates unsuited for open housing, we believe that the Sheriff
should use the 150 single inmate cells at Inverness Jail for that purpose. This will
require moving trustees out of one 75 cell unit and moving 75 mental health
inmates our of another 75 cell unit and back to a 32 unit single cell module in
MCDC. One of the 63 inmate modules in MCDC should be reduced to 32 single
cells to isolate prisoners.
2. Re-establish direct supervision at MCDC’s remaining modules. To re-establish
direct supervision at MCDC, we recommend that all wooden doors in the facility

- 48 -

be replaced with glass doors to improve visibility40. All of the facilities that we
visited that were double bunked had glass doors. We were told at these facilities
that the doors provided more protection for the inmates because it was easy for
the officers to see what was going on. We believe that with the new federal law
mandating jails to reduce the incidents of rape, glass doors are an important
component in achieving that goal. We understand that the original doors have
reached the end of their effective lifespan and are slated to be replaced within
months. We believe that they should be replaced with glass paneled doors. In
isolation cell units, these doors should be equipped with food ports. The
remaining single cell in each 63-inmate module should be double bunk, raising
the count in those modules to 64.
The effect of these changes would be to reduce the total system-wide inmate
count by 19 (loss of 31 beds by single bunking one MCDC module counteracted by the
addition of 12 beds by fully bunking the remaining modules at MCDC.
B.

Video cameras in the jails and criminal prosecution

Some of the literature we reviewed maintains that it is important to have videos
that record activity in the jail. One of the most important things it tends to do is to reduce
violence in the areas that are being videoed. As we toured other facilities we observed
that some had video cameras at most locations. The images were stored for a proscribed
period of time. We were told by the administrators at those locations that they have
become a very important tool for reducing violence and eliminating false claims by
inmates. In addition, they provide and excellent tool for the prosecution for crimes
occurring in the facilities.
We have been told that in MCDC most of the assaults by inmates do not occur in
the double bunked cells, but rather in the general housing. Virtually no cases of assault
by one inmate on another are prosecuted unless there is a death or serious physical injury.
The use of video cameras would greatly aid the prosecution of these cases.
We were told by the union president that he would probably not oppose the use of
video cameras and the recordation of activity in the jail but only if sound were used.
Unfortunately, this would probably cost considerably more and the chances of picking up
conversations in an open area where many people were present might be very difficult.
We believe that the jail should experiment with installing video cameras even if there is
no sound. We believe that once the benefits are demonstrated it will become an
important tool for the officers, the inmates and criminal prosecution.
We know there is some reluctance to use video cameras by some officers, who
believe that the cameras are there to spy on them. We did hear some disturbing
information that at Inverness Jail, where some of the dorms do have cameras, sometimes
in the morning the sergeants have had to remove paper placed in front of camera lenses.
40

No one disputes that glass doors will provide more cell visibility and improve module security. We have
observed this type of door in most of the facilities we visited.

- 49 -

We were told that corrections deputies are ones who placing the paper there to obscure
surveillance. Obviously, this kind of conduct cannot be tolerated, and when it happens it
needs to be investigated.
VII.

OTHER ISSUES IN THE JAIL

A.

Prosecution of criminal cases in the jail

During our investigation it became apparent that most criminal conduct in the jail
is neither investigated nor prosecuted adequately. There is a perception that the District
Attorney’s Office will not prosecute crimes that occur in jail. We were told that most of
the assaults on corrections officers were rejected by our office. As a result, we ran a
check on a case tracking system and found that only about 10% of these cases are
rejected, which is a much lower rejection rate than most every other kind of case
processed by the office.
As a result of deputies’ misperceptions about lack of prosecution of crimes
committed in the jails, most criminal matters in the jail are handled administratively.
There is seldom any criminal investigation into these matters. We were told that when
police officers from the enforcement division were called they were slow to respond, if
they responded at all, and generally had a negative attitude about any criminal
investigation. We certainly understand that a fight between two inmates might be
difficult to prove, but we heard of one instance where an unprovoked fight broke out,
witnessed by a corrections officer, which went unreported. This is an area where video
cameras would be critical.
In King County these kinds of cases are vigorously prosecuted and there is a
belief that it is important to reduce violence and vandalism in the jail. There is an
agreement with the prosecutor’s office to take these cases and the arrangement has
worked well.
We think the Sheriff’s Office should be more aggressive in prosecuting these
kinds of cases. We think more of these cases need to be investigated and referred to the
District Attorney's Office for prosecution. In some of the cases it would be important for
an enforcement officer to do the investigation, but there are cases that a trained
corrections officer could adequately investigate and document.
B.

Command staff must have the full support of the Sheriff

Currently there is a perception that the Sheriff will not back his command staff on
the issue of workplace discipline and work rule enforcement. There is a belief among
some that he will undercut any recommendation they make. Obviously, this is not a good
situation. The Sheriff must show his support for recommendations of his staff or they
will lose their ability to command. The Sheriff must consider what an arbitrator may do
if there is a dispute about discipline, but in any close question we believe he should side
with his commanders.

- 50 -

C.

Earthquake plan

The Sheriff’s Office has no comprehensive earthquake plan. We know from last
year’s Corrections Grand Jury that the construction of MCDC renders it susceptible to a
great deal of damage from a strong earthquake. It simply was not built under today’s
strong codes. We heard testimony, however, that Inverness is built to withstand
earthquakes because of the way its walls are constructed.
If we continue to house our worst inmates at MCDC, a substantial earthquake will
create a major problem. Under some scenarios, most inmates would be released from
that facility, and only the very worst could be kept. In the event of a major earthquake,
the community would expect that, if any inmates must be released from jail because of a
compromised jail structure, only the least dangerous should be released into the
community. By placing all of the most dangerous inmates in the MCDC facility, the
weaker structure, we are risking the opposite.
We therefore recommend, as we have done above, that the isolation cells at
Inverness be continually used to house as many of these violent inmates as possible, as
part of a comprehensive earthquake plan. Housing these people in this facility will aid
the community if we have a catastrophic earthquake.
VIII. INQUIRY INTO SPECIFIC INCIDENTS OCCURRING IN THE JAIL
A. The murder of Dennis Saban on June 16, 2005 by Thomas Gordon
Facts
On June 16, 2005 some time around 7:30 a.m., Corrections Officer John Plock
was at his officer station in MCDC’s module 5-A, which houses 62 males. It is a double
tier dorm with 16 cells on the top and 16 cells on the bottom. Officer Plock described it
as a typical morning. He had served breakfast, rounded up people for court, did a
security check and then left the dorm for a short break. He told investigators he left the
unit around 8:30 a.m. and returned about 8:45 a.m. There was no officer in the room
while he left the room. When he entered the dorm the phone was ringing. He answered
it and had a short conversation. Thomas Gordon had activated the cell light, he started
kicking on the door and yelling that he wanted Officer Plock to open his door. When
Officer Plock got to the door, he looked inside and saw Dennis Saban laying on the floor
and not moving. The cell was covered with blood. Officer Plock called for other officers
before entering the cell. It took a minute or two before anyone entered the cell.
Mr. Gordon was housed in Module 5-A-19 with Dennis Saban. Mr. Saban had
come into the jail after he failed to appear several times on Possession of a Controlled
Substance in Schedule I (Heroin) and Possession of a Controlled Substance in Schedule
II (Cocaine) charges, both from the same incident. He was taken into custody on June

- 51 -

13, 2005. He had numerous felony and misdemeanor convictions. He was on parole at
the time he was arrested.
Mr. Gordon came into the jail on May 25, 2005. He was extradited from Walla
Walla Correctional Institute. He had been convicted in Vancouver, Washington of killing
an acquaintance. The Oregonian on May 30, 2005 described Mr. Gordon’s crime as an
“execution style killing.” In a very cold and calculating manner, he shot an acquaintance
in the back of the head while they sat in a car. While in custody he admitted to killing his
mother’s boyfriend, again in a very calculated manner. He was being housed on this
charge at the time he killed Mr. Saban.
When Mr. Gordon came to MCDC he was housed in a two-person cell. The
Oregonian became aware of his history and ran a story on how dangerous he was. As a
result of the story, Mr. Gordon was moved to the isolation unit. On June 9, 2005 Mr.
Gordon complained that he wanted to be housed in 2x general housing. Ultimately,
Sergeant Scott Johnson transferred Mr. Gordon to general housing on June 15, 2005.
Sergeant Scott Johnson testified before the grand jury that after The Oregonian
article, because he had not received any documentation from Washington State
Corrections, he assumed that the information in the article related to Mr. Gordon’s
problems in the Clark County Jail in 2002. This is consistent with an email that Sergeant
Johnson sent on June 17, 2005 a day after the killing in regard to Mr. Gordon’s conduct.
In that email he stated, “After 14 days in ad-seg the only concerning info we received
was that while pending trial in Clark, County in 2001/2002 he had a number of disp.
Violation. I waited for further relevant documentation from Wash. State DOC and none
arrived. At that point I assumed that the issues referenced in the Oregonian article were
from his Clark. County custody in 2002.”
What is concerning is that two weeks before the murder, on June 1, 2005 a
“Safety Alert” was sent out to the staff in regard to Mr. Gordon. Although the “Alert”
came from someone else’s computer, it was Sergeant Johnson who wrote it. That “Alert”
contained the following information:
“Gordon was the subject of an article in the Oregonian on Monday May 30th. In
the article it was noted that the sentencing Judge in his Washington State murder case
stated it was his opinion that Gordon is extremely violent with a propensity for violent
outbursts both in and out of custody. As a result of this article Gordon was immediately
moved to 4C. Washington state DOC was contacted based on this article and we
discovered that Gordon does indeed have a recent history of both Staff and inmate
assaults. Extreme caution should be used when in contact with this inmate.” This is
inconsistent with Sergeant Johnson’s testimony.
One other concerning piece of information is that Sergeant Johnson asked his
intelligence officer, Daniel Nordstrom, to find out what information the State of
Washington had on Gordon. Officer Nordstrom had no telephone number to call, so he
looked up numbers on the Internet. He only made one substantive call to Washington,

- 52 -

and that was not even to the institution where Mr. Gordon had been housed. He was
“passed around” a lot, but ultimately talked to a woman who found Mr. Gordon’s
information on the computer. She verbally told him the information that was contained
in the safety alert. When Officer Nordstrom asked her to send him the information, she
told him she was new and did not know how to send it. He asked her to try. Officer
Nordstrom testified he made no further attempts to follow up on the information he had
been given. Sergeant Johnson testified that Officer Nordstrom told him he had called
three times attempting to get the information.
Officer Nordstrom told Sergeant Johnson what information he had received,
including the fact that Mr. Gordon had been in the I.M.U. (Intensive Management Unit)
at the Washington State Penitentiary just prior to being transported to Oregon. The
I.M.U. is known to be the most restrictive confinement in a prison system. It is generally
reserved for the inmates who show the worst behavior, unless there is a need for
protective custody. Sergeant Johnson had received no information indicating that Mr.
Gordon was in protective custody.
In spite of the information Sergeant Johnson received, because he did not have
“documented evidence” he released Mr. Gordon into the 2x general population after
being on good behavior for 15 days. Mr. Gordon killed Mr. Saban the next morning.
Concerns
Why was the officer not at his post during the beating of Mr. Saban, which
lasted between 20-30 minutes? Had the officer been at his post Mr. Gordon would not
have killed Mr. Saban. The beating took an inordinate amount of time. All the other
inmates in the area could hear it. Some commented on how the victim pleaded for his
life. At the time there was no way that anyone could have summoned help. When the
officer returned to the module and discovered what had happened it took another minute
or two for him to call and get help.
It was our impression from past grand juries that officers were always at their
posts. In the past we heard complaints that it was sometimes difficult for officers to even
go to the bathroom. That is no longer the case. In talking to witnesses, it appears that in
the morning it is not unusual for the officer to be gone for half an hour. During the noon
hour officers can be gone up to an hour and a half, and then in the afternoon they may be
gone another half hour.
Why was Mr. Gordon, given his violent past, transferred from maximum
security to a general housing cell that was occupied by a non-violent inmate? Mr.
Gordon behaved himself for two weeks and was then transferred into the general double
cell population. We were told by every jurisdiction we visited that given what
information we had, no other jurisdiction would have put Mr. Gordon in the general
population after just two weeks. Some said they never would have put him in general
population and the remainder said if they had, they would have waited longer. In King
County, Washington we were told that, had they considered such a move, they would
have done it on a graduated basis. They would have held him in segregation for a time

- 53 -

(more than two weeks). If he had shown good behavior they would have transferred him
to a single cell, but then let him have time with the other inmates in that section.
Eventually, if he was compliant and non-violent he might have been transferred to the
general population.
Unfortunately, in Multnomah County there is no such system in place. Although
there are an adequate number of single cells to run such a program, neither MCDC nor
Inverness have sufficient isolation cells to do so. This is based upon a philosophy not to
house maximum-security inmates at Inverness. Therefore, when Mr. Gordon came into
our system there were only 16 isolation cells to house the most dangerous people in the
system. There was no intermediate system to transfer him to and as a result he was able
to kill Mr. Saban.
There is also a question raised as to why a very violent criminal was being housed
with someone who had no history of violence. In Orange County, Florida we were told
that classification levels should not be mixed because the non-violent inmates are “more
vulnerable.” There is a three-fold problem with doing this: (1) dangerous inmates are
placed throughout the institution with no direct notification to the officer, (2) all of the
beds at MCDC are maximum security so that everyone must considered the equivalent of
Mr. Gordon, even though at any given time there are over 100 inmates waiting to be
transferred to Inverness and could be out of their cells in direct supervision, and (3) nonviolent inmates are put at risk and given no notice that the person sharing the cell with
them is a killer and subject to attack them. We were told that Multnomah County in the
past generally put violent inmates with other violent inmates so that they could and would
protect themselves. This was changed and one of the results was the Saban murder. The
new procedure is certainly easier for the classification unit, but we think it is a bad
system. We think that people like Mr. Gordon should be singled celled. Not by
eliminating double bunking but rather by reconfiguring all the current single cells in the
system. This has been discussed in the section on direct supervision.
Why did the officer who was on duty have no information about the
dangerousness of Mr. Gordon? Two weeks before the killing, a safety alert was issued
in regard to Mr. Gordon. His picture was part of the alert. In that picture he had long
hair, a mustache and a beard. When he was transferred to the general population he had
little facial hair, other than a mustache. His appearance had changed substantially.
Officer Plock had no idea that Mr. Gordon was the subject of the earlier safety alert.
There was nothing to alert the officer about Mr. Gordon’s past violent behavior. We
were told that safety alerts do not follow the inmate, but rather remain in a lieutenant’s
office. Most of the other jurisdictions we visited would have alerted the dorm officer.
Why did it take over a year to start the internal investigation of this matter
within the Sheriff’s Office? It appears that the death of Mr. Saban was the result of a
number of mistakes by the Sheriff’s Office; the fact that the county settled the lawsuit
quickly and the amount of the settlement demonstrates that mistakes were made.
However, the Internal Affairs Unit did not immediately conduct an investigation. We
were told that this was at the request of the County Attorney’s Office, but we heard

- 54 -

testimony that it was not. Whatever the reason, we question why it has taken so long to
look into the matter. Had there been potential criminal charges against an officer we
could understand why an internal investigation might be postponed. But that was never a
possibility in this case. If the Internal Affairs Unit is to look at things when there is a
potential breakdown in the system, this is a prime case. If it does not look at incidents
because there is a lawsuit or potential for a lawsuit, it is questionable whether it should
ever look at anything.
Here, there was a substantial breakdown in the system and at the time the
Corrections Grand Jury heard testimony there had been no changes made to the system.
We were told that no policies were changed as a result. There may be changes ultimately,
but taking more than a year to decide what to do is simply too long.
Why was there a failure to obtain relevant information on Mr. Gordon’s past
conduct? Calling the State of Washington one time and eliciting information that should
have been followed up on and then failing to do so is not acceptable. The person who
provided the information told Officer Nordstrom she was new and did not know how to
send the information. When that information did not come immediately there should
have been follow-up phone calls; but there weren’t any. Failure to act upon critical
information is a recipe for disaster. There were no jail commanders we spoke to who
would have released Mr. Gordon into the general population. To quote one
administrator, in these circumstances you need to “err on the side of caution.”
By failing to obtain vital information, the Sheriff’s Office put both inmates and
staff at risk when they put Mr. Gordon in the general population. Mr. Gordon could have
just as easily attacked the officer and the officer would have had no information
indicating the dangerous propensities of Mr. Gordon.
Why was Mr. Gordon transported and taken into MCDC when there was no
paper work as required by OSAA Jail Standards? Oregon Jail Standards C02.02.02
require inquiry and documentation of potential security issues presented by inmates being
transferred to our jails. Despite the fact that an Oregon State Sheriffs' Association
inspection of MCDC in October of 2004 found MCDC in "full compliance" with that
section, sheriff's deputies testified that this procedure was never followed until after the
Gordon incident. This is troubling on two levels. It is troubling that a reasonable
procedure was never followed, and it is also troubling that an audit could find full
compliance with a standard where compliance was absent.
Why did the officer fail to record in the log his absence from the module?
Each dorm has a log. The policy is that every time an officer leaves a dorm or comes
back into a dorm they must sign the log. In this case this procedure was not followed so
it is impossible to know exactly how long the officer was out of the unit. We are left to
guess based upon the time estimation of the beating by other inmates. We have been told
although the policy requires log entries it is not something that is routinely done.
Officers are reminded of this task at roll call, but many do not take it seriously. While we
can see that this could be a problem at Inverness because the officer is watching so many

- 55 -

people, it should not be a problem at MCDC. We believe that oftentimes it is not done
because officers do not want a log indicating when they were not with the inmates. This
is something that should be a requirement at MCDC and there should be consequences
when there is any kind of pattern that it is not being done.
Was excessive overtime usage a factor in this situation? We have noted that
both the sergeant who made the decision to house Mr. Gordon in general population and
the deputy who was absent from the module at the time of the killing were among the top
overtime workers in the system. One accumulated over 900 hours of overtime last year,
the most in the department. We believe that the Sheriff should examine the effect of
excessive overtime on workplace efficiency.
B. Sexual contact between two inmates in isolation unit
Facts
On May 1, 2006 DeRay Willis was in administrative segregation because he had
attempted to escape by putting together a homemade mannequin. He had been in
segregation for about a month. The unit consists of five isolation cells that are all in a
straight row on the same floor. There are two locations where the officers can observe
the inmates. One location is next to the controls that open and close doors to the cells.
The other location is some distance from the controls. If someone were to walk from the
second location to the controls they would lose sight of the unit for a period of time.
At 8:05 a.m. Officer Kessinger let Mr. Willis out for a 15-minute walk. He saw
Mr. Willis return to his cell. Officer Kessinger then got up and walked over to the
control panel, temporarily losing sight of Mr. Willis. When he got to the control panel he
closed Mr. Willis’s door. He then let out Inmate Quinn, who also took a 15-minute walk.
After Inmate Quinn took his walk and was put back in his cell, Officer Kessinger made a
check of each cell. To him it appeared that all the inmates were in their cells.
Unbeknownst to him, Mr. Willis had put extra linen in the bed to resemble a body and
was hiding in shower area. The shower area is located at the far end of the complex.
Upon entry into the shower, there is an area where a person can hide behind a wall.
Apparently this is what Mr. Willis did.
At 8:35 a.m. Officer Kessinger let Ms. Misha Cooper out of her cell. She took a
short walk and then returned to her cell. Officer Kessinger closed her cell upon her
return. At 9:05 a.m. Sergeant Pina did a security and welfare check; nothing unusual was
noted in the cellblock. At about 11:00 a.m. there was a call from Ms. Cooper’s cell.
Instead of Ms. Cooper talking it was Mr. Willis who said “…I was just kicking it with
Ms. Cooper and I am ready to go back to my cell.” Officer Kessinger went to Ms.
Cooper’s cell where he saw Mr. Willis and Ms. Cooper. He then went to Mr. Willis’ cell
and noted that there appeared to be a body lying in bed with its head covered. When the
officers went into Mr. Willis’s cell they noted that he had stockpiled extra linen items and
had fashioned a human shape under the blankets.

- 56 -

Mr. Willis was then returned to his cell. Although neither Mr. Willis nor Ms.
Cooper were very cooperative with the investigation, Ms. Cooper did say “look, this is
the first time I had sex in a long time, and I’m tired of all these people tryin’ to talk to me
about it.”
When Officer Kessinger testified before the Corrections Grand Jury months after
this incident, he said that he was unaware of any change in procedures as a result of this
incident.
Concerns
Why did the officers in the unit fail to notice that an inmate had escaped
from his cell when there was information available that he was an escape risk? We
were told by some officers that the first rule of checking an inmate is that you either see
skin, you see the person moving or you see the person breathing before you are satisfied
the person is in their cell. We were told by other officers that this is difficult to do in
some cells because of the lighting.
Officers need to do more than just look into cells during routine checks. It is
important to determine whether someone is in the cell and that the person is showing
some kind of movement or breathing. That was obviously not done in this case.
Corrections policy requires that there be checks every half hour. At about 8:35
a.m. Officer Kessinger checked the cells. At 9:05 a.m. Sergeant Pina did a safety check.
From the time Mr. Willis left his cell and the time his escape was discovered, there
should have been at least five checks. If there had been, it would have been noticed that
Mr. Willis had not moved in over 2 hours. That should have sent out warning signals.
However, it unclear whether there were more than two checks during this period. Officer
Kessinger assumed that when the escorts came at 9:32 a.m. to pick up Mr. Quinn, they
did a check. There is nothing in Officer Kessinger’s report that indicates he made any
check after the escorts left. In this case there was added reason to watch the inmate
because he was an escape risk. However, that never appeared to be a factor with the
officer.
If it is difficult to understand how, in what were supposedly several safety checks,
a dummy could not have been discovered in Mr. Willis’ cell, it is more difficult to
understand how the presence of two inmates, one male and the other female, perhaps
undressed, went undetected Ms. Cooper’s single-inmate segregation cell during the same
safety checks.
Why have there been no procedural changes implemented after this
incident? It was obvious to us that this matter needed to be examined closely and
changes made. When the officer testified at the grand jury, he was asked if any
procedural changes had been made as a result of this incident. He was unaware of any.
One suggestion is that every time someone does a check they log it at the time. We were
told that within corrections institutions there is a tendency not to do the checks. In one

- 57 -

jurisdiction an officer doing checks has to swipe a card on a computer based tracker
every time he or she does a safety check. This ensures that the check is done.
Why was an inmate able to hide in the shower and the officer was unable to
see him? Someone needs to examine why there is any space in the shower where
someone can hide any portion of his or her body. Had the inmate been intent on
overpowering the officer and trying to escape the unit, he might have been able to do it.
This is a safety issue for staff and other inmates. It is obvious that someone can hide in
the shower because Mr. Willis did so. With the inmates in these segregation units, line of
sight is important. This unit is not big. It only has 5 cells in it. Inmates should have no
place to hide.
How was an inmate in segregation able to accumulate enough linen to
disguise the fact that he was no longer in the cell? We were told that this is the result
of a combination of sloppy work and not searching cells on a regular basis. In addition,
this was an inmate who was known for collecting contraband and trying to escape.
Why are the officers given no information about the inmates they are
watching? Officer Kessinger testified that he did not know why Mr. Willis was in
isolation. In the Gordon case, that officer had no knowledge that Mr. Gordon had been
violent to staff and other inmates. It is unimaginable that corrections officers who are
responsible for some of the most dangerous people in society are not given more
information about the people they are supervising. In the case of Mr. Gordon it does not
do much good to have the information in a Lieutenant’s office and not with the officer
guarding the inmates. In a number of the institutions we visited there were various ways
the officers were informed who they were dealing with. In many jurisdictions security
information is readily available to the officers. In some jurisdictions they place notice on
the cell door of the inmate, and in others they put the information on computers. We feel
that was missing at MCDC.
C. Two people were brought into MCDC to be identified; both had murder
warrants from other states. They were never accurately identified and both were
released. One of them returned to the State of Washington and killed two people.
Facts
In order to best understand what happened here, it is important to understand the
concept of “cite in lieu,” also known as “book and release".
In Multnomah County, because of the lack of jail space, we do not arrest and
book a lot of people. The Sheriff, in order to maintain the population of the jail, has
guidelines as to who he will book in the front door and place in detention. For most
crimes in Multnomah County, a criminal is given a citation by the police and given a
court date. It is similar to a traffic ticket. In many cases people given citations do not
appear and then a bench warrant is issued.

- 58 -

The Sheriff has agreed that in those cases where the police officer is not satisfied
with the identification the defendant has given, the officer can bring the person to the jail
to be identified only. There is never any intent that the person will be booked unless
other more serious warrants of arrest are outstanding. The Sheriff’s Office has agreed
that they will not hold anyone longer than 4 hours to be identified. The process is called
“cite in lieu” or “book and release.”
The Sheriff’s Office contracts with the City of Portland to do the identification of
the people brought in by the police. One of the reasons that the Sheriff contracts with the
city is because the city has the actual fingerprints of most people who have previously
been arrested in Multnomah County. While there are several modern computer
techniques for comparing fingerprints, it is important to look at the actual fingerprint of a
suspect. Often times the computer systems will give the police a number of possible
people based upon a mathematical formula of the print. Unfortunately, because of budget
cuts in the past there are only a limited number of people who can search prints.
Each day, the city must identify between 100 and 120 people. Some of these
people are “cites in lieu” and others are people who are arrested. In about 88% of the
cases, the city will know who the person is because they have that person’s fingerprint
from a previous occasion. It is estimated by the city that of the 12% who are not
identified, only about 2-3% have never been booked anywhere before. That leaves 910% of the people in the jail who will not be identified within the first 4 hours.
When someone comes in to the facility to be identified, if they do not have prints
on file there are several options the city has to identify the person. Unfortunately,
sometimes the process takes more than four hours and the person is released.
When a person is booked, an electronic fingerprint is also taken of the suspect. If
the Portland Police are unable to identify the person, that fingerprint is sent electronically
to WIN (Western Identification Network). There are nine western states that participate
in WIN. Unfortunately, California and Washington, who are part of the system, do not
participate fully. Neither state allows a general search in their database unless there is a
specific request for a print search. Because these requests take some time, unless it is
known that a suspect has some relation to either California or Washington, there are
simply not enough resources to do a search. Simple data entry takes 10-15 minutes to do
one request. It can take up to 45 minutes to process a search request on one inmate.
Washington and California are apparently hesitant to become full members of the
WIN system because of privacy fears. It is all right to search each state individually, but
they prohibit their database from being used in the overall WIN system.
The prints can also be sent to the FBI. However, this has to be done through the
Oregon State Police. If everything is working correctly it will take 8-10 hours to get the
information. If it is a holiday or weekend it may take 2-3 days to get the information.
Obviously, if a person is in on a “cite in lieu” the person will be released in 4 hours,
which is generally hours or days before the city gets back the information.

- 59 -

D. The case of Nick Walter
Nick Walter was brought into the MCDC on a low level charge. The officers
could only book and release Mr. Walter under the guidelines in place at the time. Mr.
Walter had identified himself as Mr. Metcalf. He was mugged and printed. He had no
local record so the information went into the WIN system. Although there was a warrant
for murder outstanding from the State of Washington, that information was not provided
to the WIN system. Mr. Walter was released after four hours. Less than seven hours
after he was brought in, Mr. Walter’s murder indictment was found by the FBI. That
information was provided to the city, but too late to do any good.
E. The case of Juan Martinez
Mr. Martinez, like Mr. Walter, was brought into MCDC on a low level offense to
be identified. He gave a false name and was not identified within four hours. In fact
there was a warrant from the state of Washington charging him with a double murder.
After being released Mr. Martinez returned to Benton County, Washington where he
killed a woman by cutting her throat and then killed her daughter. Later Mr. Martinez
committed suicide in Vancouver, Washington.
Concerns
Why do we only hold people four hours on “cite in lieu” and then let them go
if we have not adequately identified them? No one we spoke to was aware of why we
have a four-hour rule on “cite in lieu” cases. It may have had more relevance when we
had closed booking, in which people were put in small confined areas for long periods of
time. Today we have open booking, which is much better and much easier on both the
inmates and the staff. We were told that the number of people who have not been
identified is not large. Therefore if the four-hour rule was changed, it should not result in
increased numbers. Historically, booking has not been counted in the overall number for
the jails.
Why did we not learn about the murder warrants in the two cases? When a
warrant of arrest is issued against a person, especially for murder, the rights of society
must be balanced against those of the individual. We believe California and Washington
should provide information to other states in a meaningful manner. Mr. Walters and Mr.
Martinez, two people charged with murder, were able to elude authorities because of
procedures that we believe need to be re-examined. In one case two people were
violently murdered because Washington does not participate fully in the WIN system. It
is a tragedy that two victims died for a defendant’s rights that are already adequately
protected.
Is there something that can be done to prevent this from happening again? It
is our suggestion that our Sheriff, District Attorney and County Commissioners jointly
ask our governor and attorney general to request that the governors and attorney generals

- 60 -

of California and Washington change their rules. As a practical matter it is much more
likely that on the West coast people travel north and south, rather than east to west.
Oregon, California and Washington should all be participating in a computer network to
identify people who have serious warrants outstanding. The failure to do so will result in
future crimes of violence against innocent victims.
Is there something we can do in the meantime? Unless we expand the ability
to always search databases from California and Washington when they cannot otherwise
identify someone, this will remain a problem. In addition, we need to hold people until
we know who they are. The Sheriff should reconsider his four-hour rule for people who
remain unidentified. The Portland Police Bureau stated that it would not result in holding
that many additional people. The benefits are that (1) we are able to correctly identify the
person, (2) we will correct misinformation so that innocent people will not be arrested
because the defendant used their identification and (3) we will protect the public.
F. An inmate was taken to the jail after a physical altercation with the police.
Because of the physical condition of the inmate, the nurse on duty told the police to
take the inmate to the hospital. The police took the inmate to the hospital. The
inmate died shortly after arriving at the hospital.
Facts
On September 17, 2006, starting sometime around 5:20 p.m., law enforcement
observed James Chasse, a 42 year old man with a history of mental health problems,
acting in a strange manner. After they made contact with him there was a struggle.
Because our investigation is dealing with the corrections component of this case and a
separate grand jury looked into the conduct of the officers, we are not going to address
the stop or the ensuing struggle.
After the struggle, because the officers observed that Mr. Chasse appeared to have
lost consciousness, an ambulance was called and given a “code 1,” which requires
immediate response. There was some question as to whether Mr. Chasse had stopped
breathing. When Mr. Chasse re-gained consciousness, the call was changed to a code 3.
The medical response team arrived and did a brief examination of Mr. Chasse. His vital
signs were such that they did not believe he should be transported to the hospital in an
ambulance.
Mr. Chasse was transported to MCDC. The officers stopped momentarily to fill
out a booking form. Mr. Chasse’s physical condition was described as appearing
malnourished and unclean. The officers thought that he might have been on drugs. He
was brought into the jail in restraints and a spit sock. He was placed in an isolation cell.
Once inside the cell, his restraints were removed.
At some point while he is in the isolation cell, someone thought he was having
problems breathing. A nurse was called. After seeing Mr. Chasse, the nurse told the
police to take him to a hospital. Mr. Chasse had to be put back into restraints. The

- 61 -

officers took him outside and placed him in their car and then drove toward Portland
Adventist Hospital. On the way to the hospital one of the officers noticed that Mr.
Chasse was losing color and was motionless. The police got off the freeway at 33rd
Street. They examined Mr. Chasse and observed that he was not breathing and did not
have a pulse. They called for an ambulance, which arrived shortly thereafter. Mr.
Chasse was transported to Providence Hospital where he subsequently died.
Patricia Gayman was the nurse on duty when Mr. Chasse came in around 6:15
p.m. She had been told that he was in the booking area and that he might not be
breathing. Ms. Gayman got to the cell and saw Mr. Chasse on the floor. She noted he
had a spit sock on his head. She knew that there had been a struggle, but had not been
given any of the specifics. She was unaware that one of the officers had attempted to use
a taser charge on him.
Ms. Gayman’s watched Mr. Chasse from outside the cell. She noted that he was
dirty and had sores around his feet. One of her first impressions was that he was on
drugs. She observed that he appeared to have a little seizure for about five seconds. She
immediately told the police that Mr. Chasse should be taken to the hospital. She did not
suggest he be taken in an ambulance because there was no reason to think he would not
make it to the hospital.
Although Ms. Gayman observed Mr. Chasse from a distance, this is normal
practice in the jail when someone is in a segregation cell without restraints. She noted
that they generally send inmates to Portland Adventist, which is located at S.E. 102nd and
S.E. Market. There are several other hospitals that are closer, but Portland Adventist has
the contract for holding inmates.
Concerns
Why wasn’t the nurse informed about the extent of the altercation? The
nurse thought her involvement in this matter was about two minutes. She never asked
anyone to be specific about the circumstances that brought Mr. Chasse to jail. She was
aware of the fact that he had been cleared by the medical response team on the street.
She stated that it is not unusual for a medical response team to arrive at the scene of an
arrest that involves a fight.
In hindsight Ms. Gayman said that, had she known the extent of the fight, she
would have called an ambulance. In the past she seldom knew what happened on the
street. As a result of this incident she wants to be better informed.
In at least one jurisdiction we visited, before the jail takes a prisoner the police
must provide any information relevant to the use of force, including the use of chemical
agents or tasers. The information can be provided with a quick form that will alert the
jail of potential problems so that the medical staff can make better informed decisions.

- 62 -

When the police knew that someone might be dying, why was he not taken to
the closest hospital? When the police realized that Mr. Chasse might be dying in the
car, they should have driven to Providence Hospital, which was less than two miles from
where they pulled off the freeway. Mr. Chasse would have gotten to the hospital faster
than waiting for an ambulance to arrive.
There must be some uniform written procedures given to the police, corrections
officers and corrections health workers for when to call an ambulance and when to get
the person to the closest hospital.
Considering Mr. Chasse’s mental health history was there a better
alternative to arresting him and taking him to jail? Unfortunately, with the lack of
mental health resources, including secure housing this will continue to be a problem. Too
often arrest and jail are the only immediate solutions for dealing with mental health
problems, and this case is an excellent example of why arrests and jail should be used as
the last resort. It is also an excellent example of the police being asked to make the city
more livable by addressing quality of life crimes and the crisis they face when they run
directly into the lack of mental health resources. The result is that the jail has become the
biggest secure mental health facility in the county.
CONCLUSION
We have spent almost four months gathering information on how this county’s
jail system functions and documenting the problems we feel we have observed in that
operation. We do not believe that our findings and recommendations will be universally
accepted by the parties involved in this process, and nor should they be. We would hope
and even expect that the leaders of this community will use this report as a starting point
in a general discussion of how the justice system operates in this county.
While we have observed a number of flaws in this system, and given more time
could probably find more, many of the concerns we have mentioned in this report are the
byproducts of a more significant structural problem. This problem is the lack of
communication within the leadership of this community, and even the lack of a desire to
communicate. We were as startled by how little the Board of County Commissioners
knew of the operations of the sheriff’s office as we were disturbed by the manner with
which the sheriff’s office provided county leadership access to information regarding its
operations.
Accessible information and frank, open and knowledgeable debate are the
foundations for good decisions. Unfortunately, in this county, on the cornerstone issue of
our jails, good information has been hard to come by, and debate has hardly been open
and certainly not knowledgeable.
The sheriff himself has told us that it is time to move past personal issues among
county leadership. We applaud that sentiment and hope that new leadership in county
government will work with the sheriff on that goal.

- 63 -

Appendix A

5, 2006

Mr. Michael D. Schrunk
Multnomah County District Attorney
1021 S.W. 4th Avenue
Portland, Oregon 97204
Dear Mr. Schrunk:
EXPERT REPORT TO MULTNOMAH COUNTY CORRECTIONS GRAND JURY
Our opinions are offered in reply to the questions that you posed as summarized in this
letter.
We reserve the right to update this report and our resultant opinions as necessary as
additional or new data becomes available.
ˆ Based on the Multnomah County Sheriff’s Office Fiscal 2007 budget, as adopted
by Multnomah County, what is the average budgeted full cost per jail bed, per
day?
ˆ How does this average budgeted full cost per jail bed per day compare to other
jurisdictions in general and King County, Washington in particular?
This letter comprises our work to date in connection with the above referenced matter.
Best professional regards,

Darrell D. Dorrell, CPA/ABV, MBA, ASA, CVA, CMA, DABFA, CMC
Principal
financialforensics®

Thomas S. Brown, CPA/ABV, CVA
Director
financialforensics®
†

THE NATURE OF OUR PROFESSIONAL RESPONSIBILITIES AND ETHICS
We practice as forensic accountants within the profession of certified public accountants,
CPAs. Therefore, we are compelled to meet a wide variety of ethical, legal, professional
and technical standards. These requirements reflect our fiduciary capacity and comprise
a set of standards exceeding perhaps any other profession.

Page 1 of 33

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006

Our core discipline of forensic accounting distinguishes us within and among the
financial profession and differentiates us from mere accountants, appraisers, business
valuators, CPAs, bankers, brokers, economists, financial analysts, investment analysts
and others claiming capabilities in the forensic accounting, litigation or valuation realm.
We hold our ethical, legal, professional and technical practice responsibilities in the
highest regard, particularly the hallmark of our profession - independence. It is through
independence that we serve the public’s interest in providing unbiased, objective and
technically correct opinions on financial matters, whether in court or in public or private
transactions.
The authorities governing our ethical, legal, professional and technical practice actions
include the following non-inclusive bodies:
‰

AICPA – American Institute of Certified Public Accountants. This national body,
in connection with the Financial Accounting Standards Board, FASB and related
entities establishes financial reporting requirements commonly known as
“Generally Accepted Accounting Principles,” GAAP. The AICPA Rules of
Professional Conduct contains twelve (12) Rules organized into the numbered
categories below:
¾
¾
¾
¾
¾

100 – Independence, Integrity and Objectivity
200 – General and Technical
300 – Responsibilities to Clients
400 – Responsibilities to Colleagues (superseded, i.e. no longer in force)
500 – Other Responsibilities and Practices

‰

OSBA – Oregon State Board of Accountancy.1 This body, authorized by ORS
673.410 has as its primary purpose to protect the public interest. It is the
principal government state agency which issues CPA licenses and certificates.
The Oregon laws governing certified public accountants generally follow the
Uniform Accountancy Act, UAA, but where conflicts arise Oregon law prevails.
The Board rules specifically refer to breach of fiduciary duty as a part of general
prohibitions against dishonesty and fraud under OAR 801-030-0020.

‰

OSCPA – Oregon Society of Certified Public Accountants. This body furthers the
interests of CPA-members, and coordinates professional requirements with the
AICPA through the “joint ethical enforcement program” (JEEP).

‰

SEC – Securities and Exchange Commission. This US law enforcement
government body, in coordination with the Public Companies Accounting
Oversight Board (PCAOB) governs financial reporting requirements for all
publicly held and certain privately held businesses and transactions.

1

A partner of our firm presently serves on the OSBA’s Complaints Committee, with the purpose:
“Provide technical assistance to the Board Investigator in the investigation of complaints received
by the Board. The Committee reviews investigation reports and makes recommendations to the
Board regarding disciplinary action or other resolution of complaint investigations.”

Page 2 of 33

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006
‰

IRS – Internal Revenue Service. This US Treasury body is responsible for
administering the laws and collections processes concerning revenue for the
federal government.

‰

Jurisdictional and Regulatory Authorities – Depending upon the venue of our
practice we may be compelled to meet additional criteria. For example, in many
damages matters the composition of the damages is defined by statute and/or
precedent. Also, in many valuation matters the standard of value is defined by
statute, precedent or prevailing practice.

‰

Other Professional Bodies - The unique nature of our forensic accounting
practice compels us to apply additional professional and ethical standards from
the following:
¾
¾
¾
¾
¾
¾
¾

ACFE – American College of Forensic Examiners
ACFE – Association of Certified Fraud Examiners
ASA – American Society of Appraisers
IBA – Institute of Business Appraisers
IMA – Institute of Management Accountants
IMC – Institute of Management Consultants
NACVA – National Association of Certified Valuation Analysts

We strive to meet and exceed the preceding standards through our firm’s guidance as
summarized below:
OUR MISSION STATEMENT:
To spearhead forensic accounting innovation in civil, criminal and combat matters.2
THE DEFINITION OF FORENSIC ACCOUNTING IS:
The art and science of applying forensic accounting techniques to matters of law.3

2
3

(financialforensics® October 2002).
(financialforensics® Newsletter, September 1993).

Page 3 of 33

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006

SUMMARY OF FINDINGS
WHAT WERE WE ASKED TO DO?
Michael D. Schrunk, Multnomah County District Attorney asked us to answer the
following questions:
ˆ Based on the Multnomah County Sheriff’s Office Fiscal 2007 budget, as adopted
by Multnomah County, what is the average budgeted full cost4 per jail bed, per
day?
ˆ How does this average budgeted full cost per jail bed per day compare to other
jurisdictions in general and King County, Washington in particular?
You also inquired regarding whether the methodology for calculating the average
budgeted full cost per jail bed per day, hereinafter also referred to as “jail bed cost”, used
by the Multnomah County District Attorney’s office is correct and properly includes all the
costs of operating the County jail facilities for purposes of evaluating the Federal and
State Jail Bed contracts against the budgeted cost of operating theses facilities.
WHAT INFORMATION WAS USED TO CONDUCT OUR INVESTIGATION?
We used four categories of information in conducting our investigation.
ˆ Information obtained through the Multnomah County District Attorney’s Office
including Scott Marcy’s calculation of average budgeted full cost per jail bed per
day and information from the Sheriff’s office concerning jail bed costs.
ˆ Information that we obtained directly from the Multnomah County Budget Office
ˆ Information obtained from the Multnomah County Sheriff’s Office representatives
including Bernie Giusto, Larry Aab, Markley Drake and Sharie Lewis.
ˆ Information that we obtained through our own independent research and
investigation.
See the “Information Sources” attachment to this report for details.
WHAT ACTIONS WERE NECESSARY TO COMPLETE OUR ASSIGNMENT?
We completed the following actions.
ˆ We reviewed all the information described in the four categories listed in the
preceding question. Refer to the “Information Sources” section of this report.
4

Full Cost - The sum of all costs required by a cost object including the costs of activities
performed by other entities regardless of funding sources. See Glossary of Terms at the
end of this report.

Page 4 of 33

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006

ˆ We conducted various work sessions and discussions with Scott Marcy, John
Bradley and Charles French of the Multnomah County District Attorney’s office.
We also met with Christian Elkin of the County Budget Office and Sharie Lewis,
formerly of the Sheriff’s Office.
ˆ We obtained a working knowledge of the budgeting process through documents
obtained from the County Budget Office. This is explained briefly in the next
section of or report.
ˆ We reviewed calculations of average budgeted full cost per jail bed per day
together with supporting schedules prepared by Scott Marcy.
ˆ Two (2) CPAs invested more than 100 hours in conducting the analysis, technical
research and preparing the calculations

UNIQUE FACTS AND CIRCUMSTANCES
Every consulting assignment contains unique aspects that have an impact on the
investigation approach, and techniques considered and applied. In this assignment the
following unique facts and circumstances are specifically identified in order to apprise
the reader of their application.
ˆ

In order to prepare for the sunset of the ITAX5, Multnomah County moved
to Priority-Based Budgeting (PBB) in FY 2005. PBB has required a
significant investment of staff time and resources and the Board has
indicated its commitment to future use of this process. The County
Budget Manager explains PBB as follows:
¾

Priority-Based Budgeting is a concept developed by Peter Hutchinson and
David Osborne from the Public Strategies Group (PSG). It provides a proven
and innovative approach for agencies to identify funding priorities that show
quantifiable results that support the agency’s overall goals, and that are most
highly valued by the community served. This approach helped the Budget
Office shift its focus from line-item budgeting and constraint (across-theboard cuts) that weaken infrastructure, to a funding methodology that
considers measurable outcomes that best support the County’s strategic
objectives. The work performed in each step is brought to the Board for its
approval and for any necessary course correction. It is a highly transparent
process—to the Board, other elected officials, department heads, county
employees, and citizens.6

5

The Multnomah County personal income tax (ITAX) was created when voters approved Ballot
Measure 26-48 in May 2003. It is a three-year measure that raises local funds to prevent further
cuts to Multnomah County’s public schools and programs that help the County’s most vulnerable
residents. It is intended to be a temporary, local answer to recent state budget cuts and a poor
economy.
6
Multnomah County Budget Manager’s Message, FY 2007 Adopted Budget, page 6

Page 5 of 33

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006
¾

ˆ

The Priority-Based Budgeting Process was implemented to answer the
following questions:7
□

1. How much money do we want to spend? The formulation of the
budget must be based on the premise that the County cannot
spend more than it receives in revenue. Note: This question
suggests that a full-cost analysis is essential.

□

2. What do we want to accomplish? The budget must prioritize the
services that most efficiently achieve the desired results.

□

3. What is the most effective way to accomplish our priorities with
available funds? As part of the Priority-Based Budgeting Process,
every department is asked to find ways to work more efficiently
and to leverage scarce resources.

Program offers form the basis for the Price of Government budgeting
process. They constitute a department’s budget request and narrative
budget, and provide the information that the Board of County
Commissioners uses to select programs that will best address the
County’s priorities.8
¾

Program Offer Description. In Multnomah County's budget, a “program” is an
organizational unit that provides services to the public or to other County
departments or divisions. Its description should briefly and clearly explain the
activities encompassed in the program offer. If the offer represents a change
in current practice (a reorganization or change in staffing levels, service
hours, etc.), it should briefly describe the nature of that change. Indeed,
Multnomah County’s FY 2007 Priority Based Budgeting description9 indicates
that the Price of Government must be considered in totality, i.e. comprising
“Fees, Taxes and Revenue.”

¾

Projected Program Costs. Program offers should show the “full cost” to the
County of providing the program, including the program’s share of
administrative, support, or other shared costs, as well as the revenues
generated by the program for its support.

¾

Avoiding a “full cost” analysis distorts measurements that lead to flawed
decision making. For example, the fuel efficiency of an automobile can be
determined by calculating the miles per gallon, i.e. “total miles driven” divided
by “total gallons consumed.” However, selectively calculating miles per
gallon for highway-driven mileage only distorts and misstates overall fuel
efficiency.

¾

Types of Program Offers. Program offers were categorized based on the

7

Ibid
Information regarding “Program Offers” is taken from the Readers Guide – Volume 2, FY 2007
Adopted Budget
9
FY 2006 Priority Based Budgeting, MULTNOMAH COUNTY OREGON, page 3 of 8.
8

Page 6 of 33

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006

“type” of services they deliver. Program offers fell into one of the following
groups:
• Administration
• Support
• Operating Program
• New Program
• Program Alternative/Reconstruction
• Internal Service
• Revenue
¾

The following are definitions of some of the above listed types of program
offers:
□

Administration: Department- or division-level management and
related expenses (i.e. office space, supplies, telephones, etc.)
Direct program supervision is considered to be a part of the
operating program (NOT administration), and should be included
in the operating program’s offer.

□

Support: An activity in a department that provides services
directly to some or all operating programs within a department.
Examples include the Health Department’s pharmacy, which
supports the various health clinics; the Library’s Technical
Services, which maintains the Library’s materials and catalog
system-wide; or the District Attorney’s Human Resources unit.

□

Operating: An “on the ground” activity of the County. Includes
front-line
supervisors/program
managers
with
specific
responsibilities for particular operating programs. An example
would be the Department of Schools and Community Partnerships
program for Runaway Youth or the Health Departments Primary
Care Clinics.

ˆ

The program offers display each program’s proportionate share of
administrative and support costs. This information is intended to display
the full cost of operating the program or providing the service.
Departments determine how the total amount of department administration
and support is divided among program offers. Some programs may be
more administration-intensive than others, or may use support services
that others do not. Note that administrative and support program offers
also are included in each department’s budget. These are included in order
to display the particulars of each department’s administrative and support
functions.

ˆ

Many County programs can deliver their services at varying levels. For
example, an Alcohol and Drug Treatment Program may be able to offer 100
beds, 150 beds, or 200 beds. Departments were asked, when appropriate,

Page 7 of 33

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006

to provide decision-makers with the option to choose from among those
levels. The term used most often for this is “scaled” or “scalability”.
Looking at scaled program offers, one should consider:
¾

The Base Level of Service - Program “A” is the base level of service that
can be provided while still delivering the expected results. In the example
above, the base level of service would be 100 beds of A&D treatment. To see
an illustration in the program offers please see the Department of Community
Justice’s Program 50052A-C Addiction Services.

¾

Increments of Service - Program “B and beyond” represent incremental
levels of service that buy additional outcomes. In the A&D treatment example,
one increment could be 50 additional beds, and another, separate increment
could be 100 additional beds.

SITUATIONAL ASSESSMENT
The FY 2007 Adopted Budget for the Multnomah County Sheriff’s Office is shown in
summary form below.10
Sheriff's Office - Budget for FY 2007
FY 2006
Adopted
Budget
Category
Professional Services
$ 77,830,685
Contractual Services
1,423,407
Materials & Supplies
19,767,964
Capital Outlay
194,066
Total Costs
$ 99,216,122

FY 2007
Adopted
Budget
$ 85,446,231
1,523,153
18,847,350
209,816
$ 106,026,550

Percent
Change
9.8%
7.0%
-4.7%
8.1%
6.9%

Source: Multnomah County Budget Office FY 2007 Adopted Budget

10

¾

We noted a 6.9% increase in the overall FY 2007 budget over the prior
budget for FY 2006.

¾

For purposes of determining average budgeted cost per jail bed per day,
program offers in the Sheriff’s FY 2007 Adopted budget may be separated
into two primary components – Jail Related Program Offers and Other
Program Offers. These two major groupings are illustrated below:

Source: http://www.co.multnomah.or.us/dbcs/budget/web_page_07_adt_vol_two/mcso_adt_07_two.pdf

Page 8 of 33

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006
Sheriff's Office - Budget for FY 2007
Jail Related Program Offers
FY 2007
Adopted
Budget
$ 84,683,943
21,342,607
$ 106,026,550

Category
Jail Related Program Offers
Other Program Offers
Total Costs

Percent
Total
79.9%
20.1%
100.0%

Source: Multnomah County Budget Office FY 2007 Adopted Budget

¾

In addition to the Sheriff’s FY 2007 Jail Related Program Offers, the FY 2007
Budgeted cost of health services for inmates, referred to as “Corrections
Health” developed by the County Health Department, of $13,185,762 must
also be included in a calculation of jail bed costs. This becomes the total jail
related costs as illustrated below:
Sheriff's Office - Budget for FY 2007
Total Jail Related Costs

Category
Jail Related Program Offers
Corrections Health Jail Program Offers
Total Costs

FY 2007
Adopted
Budget
$ 84,683,943
13,185,762
$ 97,869,705

Percent
Total
86.5%
13.5%
100.0%

Source: Multnomah County Budget Office FY 2007 Adopted Budget

The Corrections Health portion of the above total jail costs is taken directly from the
Health Department budget for FY 2007.11

COMPUTATION OF AVERAGE BUDGETED COST PER JAIL BED PER DAY
Using the total jail related costs presented earlier, we developed an average budgeted
cost per jail bed per day of $157.82 as illustrated in the following schedule.

11

Source: Corrections relate program offers per Health Department FY 2007 Summary by
Program Offer, http://www.co.multnomah.or.us/dbcs/budget/2007narrativeguide.shtml

Page 9 of 33

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006

Average FY 2007 Budgeted Jail Bed Cost
M ultnomah Co. Detention Center (MCDC)
Beds/Annual Bed Days

676

Percent of Total Beds

40%

Program Offer

MCDC

Inverness Jail (MCIJ)

246,740

1,014

System-Wide Total
370,110

1,690

60%
Per Bed Per
Day

Program Offer

Inverness

616,850

100%
Per Bed
Per Day

Totals

Per Bed Per
Day

Percent
Total

Program Cost Allocations
Program
Allocable
Cost
%
Base

%

MCDC

%

MCIJ

Federal Allowed Costs

Dormitory costs
60021A

$13,252,077

60022A

60021B

$3,238,647

60022B

$15,592,448
$2,463,430

60021C

$1,292,119

60022C

$2,244,079

60021D

$4,505,829

60022D

$617,307

60021E

$1,084,687

60022E

$4,007,358

60021F

$3,245,500

60022F

$1,349,876

60021G

$569,397

60022G

$3,159,968

60021H

$2,977,624

60022H

$1,549,182

60021I

$569,397

60022I

sub-total

$30,735,277

$124.57

$1,473,810
$32,457,458

$87.70

$63,192,735

$102.44

64.9%

Executive/Prof. Standards
60001

$586,782

60001

$723,755

$1,610,290

81%

$1,310,537

45%

$586,782

55%

60002

$424,247

60002

$523,278

$1,164,248

81%

$947,525

45%

$424,247

55%

sub-total

$1,011,029

$4.10

$278,893

$1.13

$1,247,033

$3.37

$2,258,062

$3.66

$650,749

$1.76

$929,642

$1.51

$4,158,708

$11.24

$7,530,369

$12.21

$2,020,928

$5.46

$3,659,390

$81,429

$0.22

2.3%

$2,774,538

$2,258,062

$1,011,029

$723,755
$523,278
$1,247,033

Offender Transport (Fed.Portion)
60015

60015

$2,601,539

36%

$929,642

30%

$278,893

70%

$650,749

7.7%

$7,530,369 100%

$7,530,369

45%

$3,371,661

55%

$4,158,708

$5.93

3.8%

$3,659,390 100%

$3,659,390

45%

$1,638,462

55%

$2,020,928

$147,447

$0.24

0.2%

$147,447 100%

$147,447

45%

$66,018

55%

$81,429

64%

$1,671,897

30%

$501,569

70%

$1,170,328

$2,458,912 100%

$2,458,912

45%

$1,100,958

55%

$1,357,954

Booking-Release
60016A

$3,371,661

$13.66

60016A
Classification

60016B

$1,638,462

$6.64

60016B
Gresham Temp

60016C

$66,018

$0.27

60016C
Corrections Health Program

Federal Allowed Costs

$6,132,446

$24.85

$7,053,316

$19.06

$13,185,762

$21.38

13.5%

$43,233,786

$175.22

$47,669,621

$128.80

$90,903,407

$147.37

93.4%

$501,569

$2.03

$43,735,355

$177.25

$1,100,958

$4.46

$44,836,313

$181.71

Offender Transport
Other Costs

60015
sub-total

60015

$1,170,328

$3.16

$1,671,897

$2.71

$48,839,949

$131.96

$92,575,304

$150.08

$2,601,539
95.1%

Inmate Commissary
60019
sub-total

60019

$1,357,954

$3.67

$2,458,912

$3.99

2.5%

$50,197,903

$135.63

$95,034,216

$154.06

97.6%

Corrections W ork Crews
NA

$2,319,013

$6.27

$2,319,013

$3.76

2.4%

Other Costs

$1,602,527

$6.49

60025

$4,847,295

$13.10

$6,449,822

$10.46

6.6%

Total Jail Costs

$44,836,313

$181.71

$52,516,916

$141.90

$97,353,229

$157.82

100.0%

Page 10 of 33

I
Several points of explanation regarding the preceding schedule follow.
1. The schedule has separate columns for the Multnomah County Detention Center
(green), the Multnomah County Inverness Jail (tan) and total system-wide
columns (yellow). Costs are listed by program offer for each facility and totaled.
2. The right side of the schedule shows how certain costs that were not already split
between the two facilities were allocated. The allocation of these costs and
alternative allocation methods considered but rejected is shown below.
Certain costs are allocated based on the following assumptions:
Program Offer
Allocation Base
Ratio of corrections portion of Sheriff's budget excluding Administration and Support to budget total, then
Exec./Prof. Standards:
based on the number of beds at each facility. See calculation below.
Administration and Support Program Offers
60003A
$1,187,180
60004
$1,011,262
60005
$781,483
60006
$3,695,147
60007
$941,295
60009
$2,600,961
60010
$2,124,088
60011
$3,379,894
60012
$1,999,841
60013
$333,627
60014A
$1,827,531
60014B
$2,386,870
60017
$2,641,253
60034
$1,835,499
60035
$1,634,116
sub-total
$28,380,047
Total MCSO Budget
$106,026,550
Costs excluding Administration and Support
$77,646,503
Dormitory Costs (includes allocated Administration and Support Costs)
$63,192,735
Percent of Dormitory Costs to Costs Excluding Administration and Support
81.4%
Offender Transport

Per USM-243 Basic Guidelines, only the portion benefiting federal prisoners is allowable. Per Sheriff's
office worksheet "Jail Bed Costs Oct 11 061.xls" $929,642 or 35.73% allowable.
60015
$2,601,539
Portion deemed allowable by MSCO
$929,642
35.7%

Facilities Allocation Methods
Beds
Number of Beds

MCDC
676

40%

MCIJ
1,014

60%

TOTAL
1,690

100%

$30,735,277
$20,497,800

48.6%
44.8%

$32,457,458
$25,282,598

51.4%
55.2%

$63,192,735
$45,780,398

100.0%
100.0%

Direct Costs
Dorm. Costs
Dorm. Costs no Admin & Sup.

FTE's (Excluding any allocated FTE categories)
Dormintory FTE's
158.57
Corrections Work Crews
Corrections Health
33.10
Totals
191.67
46.1%

169.98
15.20
38.47
223.65

53.9%

328.55
15.20
71.57
415.32

100.0%

Program Offer
Allocation Base
Booking-Release
Based on the direct costs of each facility.
Gresham Temporay Holding FBased on the direct costs of each facility.
Offender Transport
Estimate provided by Scott Marcy based on the nature and location of each facility . All offenders are
processed through the MCDC facility. Inmates requiring higher level of security are not likely to be
transported to Inverness.
Classification
Based on the direct costs of each facility.
Inmate Commissary
Based on the direct costs of each facility.

Page 11 of 33

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006

3. At total of 81.4% of the Executive and Professional Standards program offers
were allocated to the jail system based on the ratio of jail costs to the total MCSO
budget.
4. The schedule is also divided horizontally into two sections. The top section
contains costs that would be allowable according to Federal Marshal program
cost guidelines.12
5. Based on figures obtained from the Sheriff’s Office13, approximately 36% of total
Offender Transport Program Costs is includable in the determination of jail bed
costs per day for federal prisoner contract purposes. The remainder, 64% is
shown below in “Other Costs” and included in the Average Budgeted Cost per
Jail Bed per Day calculation.
6. As shown in the schedule, Average Federal Allowable Budgeted Costs per jail
bed per day is $147.37.
7. How certain programs were allocated among the two facilities is summarized at
the bottom of the schedule above. As indicated, most of the costs were allocated
among the facilities based on direct costs which represent the total dormitory
costs less administration and support costs allocated to the dormitory program
offers.
8. Corrections Health costs were allocated using the same allocation base for
consistency. Refer to the Appendix for a schedule showing the Corrections
Health Budgeted costs. The chart below depicts the two major components of
our jail bed per diem calculation.

MCSO FY 07 System-Wide Program Cost Components

$ 3.76

$ 3.99
$ 157.82

$ 10.46

$ 2.71

Federal Allowed Costs

Offender Transport

Inmate Commissary

Corrections Work Crews

12

Guidelines for determining costs includable in the contract with the U.S. Marshal’s Office are
taken from the USM-243 Form Guidelines and OMB Circular A-87. These documents explain
which costs may be included in the computation of costs related to federal prisoners. Any cost
that benefits federal prisoners, unless specifically disallowed, should be included in the USM
Contract Cost.
13
No explanation was forthcoming from MCSO as to why only 36% of Offender Transport
Program costs would be includable in the U.S. Marshal Service contract allowable costs.

Page 12 of 33

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006

As the chart indicates, the Average Budgeted Cost per Jail Bed per Day is $157.82, of
which $10.46 in total represents costs not includable in the federal contract rate.
The chart below illustrates the various components of Average Budgeted Costs per Jail
Bed per Day.

FY07 System-Wide Jail Bed Cost - $157.82/day
Correct ions Work
Crews, $3.76, 2%
Inmat e Commissary,
$3.99, 3%
Off ender Transport,
$2.71, 2%
Corrections Health
Program, $21.38, 14%
Gresham Temp, $0.24,
0%
Classificat ion, $5.93,
4%
Booking-Release,
$12.21, 8%
Of fender Transport
(Fed.Port ion), $1.51, 1%

Dormitory costs,
$102.44, 64%

Execut ive/Prof .
St andards , $3.66, 2%

The two largest components are Dormitory Costs (64%) and Corrections Health
Costs (14%).

Page 13 of 33

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006

The second question that we were asked to address was:
ˆ “How does this average budgeted cost per jail bed per day compare to other
jurisdictions in general and King County, Washington in particular?.”
The Multnomah County District Attorney’s office determined that the King County,
Washington Jail System was sufficiently comparable to the Multnomah County Jail
System to make a comparison of the two systems meaningful.
A Comparison of jail bed costs/day was made between Multnomah County and King
County, Washington. Regarding the latter, the following excerpt was taken from a recent
King County Auditor’s Report: “Jail operations are complex, and in King County this
complexity is heightened by the fact that the department operates two large urban jails
that are separately located, have different designs and sizes of inmate housing units,
and employ different approaches to inmate supervision.”14
A schedule prepared by Scott Marcy of the Multnomah County District Attorney’s Office
showing King County Jail Cost follows. Note that the Average Budgeted Cost per jail
bed per day at full or near full capacity would be $85.05. At present capacity of
approximately 80%, the average daily rate is $111.18.

14

“Performance Audit – Jail Overtime” King County, Cheryle A. Broom, King County Auditor,
October 9, 2006, page 5

Page 14 of 33

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006

King County Jail Bed Costs
KCCF (1697 beds)

Per Day

RJC-DF (1388 beds)

Per Day

Expenses
Description
Salaries & W ages
Overtim e Adult
Overtim e Juvenile
Personal Benefits
Supplies
Services & Other Charges
Intragovernm ental service
Capital Outlay
Debt Service
Intra County contributions
Special Budget Account
Contingencies
Contra Expenditures

Budget
$ 24,040,264
1,573,624
0
9,335,180
538,268
712,911
60,417
0
0
0
0
0
-368,699

Total
allocated adm in.
Total
percentage of adult corrections direct
allocated corr. Health by percentage

Totals

Description
Salaries & W ages
Overtim e Adult
Overtim e Juvenile
Personal Benefits
Supplies
Services & Other Charges
Intragovernm ental service
Capital Outlay
Debt Service
Intra County contributions
Special Budget Account
Contingencies
Contra Expenditures

Budget
$ 15,834,746
1,461,382
0
6,307,618
2,892,438
167,027
32,144
0
0
0
0
0
-244,523

$ 35,891,965

Total

$ 26,450,832

62,342,797

6,155,223
$ 42,047,188

allocated adm in.

4,536,134
$ 30,986,966

10,691,358
73,034,155

42.43%
9,647,002

100%
22,737,321

Total
percentage of adult corrections
allocated corr. Health by percentag

57.57%
13,090,319

Total cost for facility
$ 55,137,508
79.4% occupancy- avg. cost per bed day

Per Day

%

Expenses

$ 89.02
$112.11

Total annual cost for facility
$ 40,633,968
73% occupancy- avg cost per bed day

39,875,010
3,035,006
0
15,642,798
3,430,706
879,938
92,561
0
0
0
0
0
-613,222

$ 80.21 $
$109.87

95,771,476

* note jail population alert data found on the Adult and Juvenile Dention web pages shows an overall adult capacity of 2,973 beds.
(This differs from the facilitiy capacity 3,085). I am not sure why.
** jail population alert inform ation indicates that the system is running at approx. 80% occupancy rates currently.
(See link http://www.m etrokc.gov/dad/population/Population-Alert.htm l)
Per notes from Chunck French visit average daily population was 1347/day and 1013/ day respectively. This form s the basis of m y occupancy rate.

Page 15 of 33

24%
$ 85.05 100%
$111.18

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006

The two jail systems both operate two jail facilities. The King County Jails capacity is
nearly 50% larger than that of Multnomah County.
Chuck French of the Multnomah County District Attorney’s Office, who toured these
facilities and spoke with representatives from King County, identified the following key
differences between Multnomah County and King County jail systems:
1. “King County should be expected to have lower transport costs, since there
are court buildings attached to both of the jails. Inter-jail transport might be
lower also because both Mult. Co jails act as feeders to our one courthouse,
while the King County jails feed, by and large, into their own courthouses,
where crimes are processed based on (the) geographic location where they
were committed.”
2. “King County has no federal contract.”
3. “King County apparently gets to use the county hospital free of charge for
inmates who are (in) need (of) outside treatment.”
4. “King County has a substantially higher transportation charge for transporting
and guarding prisoners in the county hospital since they apparently send out
more prisoners for hospital treatment (probably because it is free to them).”
5. “King County has contracts with approximately 20 municipalities to house
inmates for municipal ordinance violations from that city. My guess is that
dealing with so many different jurisdictions produces a somewhat higher
administrative cost.”
6. “King County jails are administered by only one department, the Department
of Adult and Juvenile Detention. This means that costs are more transparent,
since 100% of all overhead and administrative costs are assigned to the
department.”
It would appear to us that the overall impact of the above differences on jail bed costs
would be relatively small. However, we did not attempt to quantify in dollar terms the
impact that these differences between the two systems would have on jail bed costs.
ADDITIONAL FINDINGS AND OBSERVATIONS
During the course of conducting our analysis, we also noted the following additional
items.
1. The first program offers for both the Detention Center and the Inverness Jail
contain the bulk of the Administration and Support Costs allocation for the
dormitories. The schedule below and chart that follows shows this clearly.

Page 16 of 33

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006

MCSO FY 2007 Budgeted Costs - Dorm itory Adm in. & Support Allocations
Multnomah Co. Detention Center (MCDC)
Beds
Program Offer
Dormitory costs
60021A
60021B
60021C
60021D
60021E
60021F
60021G
60021H
60021I
sub-total
percent of total

676
MCDC
$13,252,077
$3,238,647
$1,292,119
$4,505,829
$1,084,687
$3,245,500
$569,397
$2,977,624
$569,397
$30,735,277

System-Wide
T otal

Inverness Jail (MCIJ)
1,014

MCDC
Adm in/Support
$6,573,927
$605,753
$367,768
$842,207
$203,112
$768,663
$106,487
$663,073
$106,487
$10,237,477

%T
49.6%
18.7%
28.5%
18.7%
18.7%
23.7%
18.7%
22.3%
18.7%
33.3%

Direct Costs Program Offer
$6,678,150
$2,632,894
$924,351
$3,663,622
$881,575
$2,476,837
$462,910
$2,314,551
$462,910
$20,497,800
44.8%

60022A
60022B
60022C
60022D
60022E
60022F
60022G
60022H
60022I

Inverness
$15,592,448
$2,463,430
$2,244,079
$617,307
$4,007,358
$1,349,876
$3,159,968
$1,549,182
$1,473,810
$32,457,458

1,690
MCIJ
Adm in/Support
$3,181,717
$456,938
$489,406
$115,970
$1,122,743
$254,678
$985,141
$291,002
$277,265
$7,174,860

%T
20.4%
18.5%
21.8%
18.8%
28.0%
18.9%
31.2%
18.8%
18.8%
22.1%

total adm in/support per FY 07 budget
total adm in/support-dorm itory
percent of total

Page 17 of 33

Direct Costs
$12,410,731
$2,006,492
$1,754,673
$501,337
$2,884,615
$1,095,198
$2,174,827
$1,258,180
$1,196,545
$25,282,598
55.2%

Total

$45,780,398
100.0%
$28,380,047
$17,412,337
61.4%

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006

2. Note the significant difference in the Admin/Support amount allocated by MCSO
to the first dormitory program offer for each facility compared with subsequent
program offers. This can be seen more clearly in the following chart.

MCSO FY 2007 Budgeted Admin. and Support Costs

$18,000,000
$16,000,000
$14,000,000
$12,000,000
$10,000,000
$8,000,000
$6,000,000
$4,000,000
$2,000,000
$0
First Program Offers

All Program Offers

The first program offer for MCDC represents 46 of 676 beds. The first program
offer for MCIJ represents 160 of 1014 beds. It is unlikely that the true cost of
operating these first few beds in each facility is anywhere near the total program
offer cost listed. In fact, the practice of “frontloading” costs to the first dormitory
program offer for each facility distorts the cost of the remaining program offers
and defeats the scalability desired by the County. In essence, the Sheriff is
treating some variable costs as fixed costs. 15
The same frontloading of
admin/support costs exist in the Corrections Health Program Offers 40025A
(MCDC) and 40028A (MCIJ). Note that such practice violates the “Price of
Government” principle foundational to the budgeting methodology adopted by
Multnomah County.
3. In his Response to Corrections Grand Jury Status Reports dated October 16,
2006, the Sheriff indicates that jail bed costs are $120.65. The supporting
15

In general, variable costs are those costs that increase with the number of jail beds occupied.
Fixed costs generally do not increase with the level of jail beds occupied. See the glossary of
terms at the end of this report.

Page 18 of 33

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006

calculations provided by MCSO (Jail Bed Costs Oct 11 061.xls) come up to
$124.26/day. Whichever figure the Sheriff’s office intends to use, neither figure
represents “full jail bed costs”. The Sheriff omits a number of budgeted cost
components and includes only portions of others, without explanation. Note that
such practice violates the “Price of Government” principle foundational to the
budgeting methodology adopted by Multnomah County. The chart below
compares the full cost per jail bed to the figures received from the Sheriff’s Office
and King County, Washington.
MCSO Full Cost, King County Cost and Sheriff's Cost Figures

$160.00
$140.00
$120.00
$100.00
$80.00
$60.00
$40.00
$20.00
$Jail Bed Cost

Full Cost

King County

MCSO-Sheriff

MCSO-Drake

$157.82

$111.18

$120.65

$124.26

4. We have seen nothing in the USM-243 instructions or OMB Circular A-87 that
specifically prohibits including the following indirect costs16 in determining the
rate to be charged for housing federal prisoners:
a. Offender Transport - Cost of transporting prisoners from booking at MCDC to
MCIJ. (Costs for transporting prisoners to local court proceedings from either
facility are clearly not allowed, however.)
b. Booking and Release – The process by which a criminal on the street
becomes an inmate in custody.
c. Classification - Cost to sort detainees by facility to the extent federal
prisoners go through the classification process.

16

Indirect Cost - A cost that cannot be identified specifically with or traced to a given
cost object in an economically feasible way.

Page 19 of 33

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006

d. Gresham Temporary Hold Facility – While excluded from the County
Auditor’s cost per bed per day calculations, if federal prisoners are being held
at this facility, this cost should be included in the USM-243 process.
e. Commissary - The net cost of commissary operations (admittedly a small
number, but recoverable under USM-243 guidelines nonetheless).
5. The Multnomah County Audit report (July 2006) acknowledges the Sheriff’s use
of costs allowed for the U.S. Marshal contract only in calculating costs per
bed/day and reports a combined MCDC and MCIJ cost per day of $134.37 for
the FY06 budget year. The Sheriff’s budget increased from FY06 to FY07 by
6.9% indicating that an even higher system-wide cost per day would be expected
for the FY 2007 budget year.
6. Federal and state prisoners are not segregated from other prisoners in either
facility. Hence, the all-inclusive system-wide full cost per day ($157.82) should
be the yardstick used to measure the real impact of housing federal and state
prisoners, not the U.S. Marshal Contract costs per day. The chart below shows
the full cost of jail beds in Multnomah County compared with the Federal and
State contract rates.

MCSO System-Wide Full Costs
Federal and State Contract Payments
$157.82
$170.00
$150.00
$130.00
$110.00
$90.00
$70.00
$50.00
$30.00
$10.00
$(10.00)

$115.90
$79.94

Full Costs

Federal Marshall
Contract

State SB1145 Contract

SB1145 based on FY07 budget revenues $7,440,145 divided by the
FY06 Avg. 1145 beds/day of 255 (eSWIS data)

7. Revenues received from the federal and state contracts fall short of covering full
jail bed costs by substantial margins. This amounts to a subsidy “by default”,

Page 20 of 33

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006

instead of a conscious decision by policy makers. The chart below illustrates the
amount of the shortfall in full cost recovery.

MCSO System-Wide Full Costs
Federal and State Contract Payment Shortfall
$2,030,111
$7,249,248

$5,612,405
$7,440,145

Shortfall
@ Contract
@ Cost

$7,642,517

$14,689,393

Federal Marshall Contract

State SB1145 Contract

Data Sources: FY06 Avg. Daily Population-Fed: 132.67/day per Larry Aab,
"Information Request" 10/4/06. State: eSWIS data

8. USM-243 Basic Guidelines state that accurate prisoner population data must be
submitted for the same accounting period as the cost data submitted. However,
the form (Section III) makes no provision for multiple facilities. Sheriff’s Office
response #5 to Information Request dated October 4, 2006 indicates that actual
prisoner count data by facility is not available. Larry Aab states “I am unable to
identify a data source that identifies USM prisoners housed by facility”. The
same is indicated in the Department of Justice audit report dated July 28, 2006.
9. Assertions that some beds cost less than others or that the last bed purchased is
the cheapest are irrelevant to the calculation of the full cost of jail beds. In
simplest terms, as defined earlier, system-wide, all-inclusive per diem costs –
“full costs” is determined by dividing the total corrections related program offers
by the total number of bed days per year.

Page 21 of 33

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006

The netting of program revenues (e.g. Commissary, Workcrews etc.) against
program costs does not yield a full and accurate picture of the budgeted costs of
the program offers in question. Netting revenues against costs may give the
uninformed reader the appearance of “lower” cost. In reality, netting revenues
against costs merely serves to obscure the expense side of the MCSO budget
and hampers expense analysis, management planning and cost control. This is
why revenues are shown separately in the budget and not just netted against the
program costs. The same process is followed in the private sector. Revenues
are shown separately. Expenses follow. And net income or loss is the bottom
line. Note that such practice violates the “Price of Government” principle
foundational to the budgeting methodology adopted by Multnomah County.
10. It is our understanding that the federal contract rate of $115.90/day for
Multnomah County – a rate set in 1996, is presently the seventh (7th) highest in
the nation and yet, this amount does not cover the County’s cost. This is
alarming and should be prompting discussion as to why it costs so much to
house prisoners in the County jail system. Or better yet, how might costs be
reduced without eliminating jail beds? Instead, the discussion seems to be
focused on which costs (that the County is actually paying by the way) should or
should not be included in the County’s jail bed cost per diem rate. The point is –
cost must be reduced without eliminating jail beds. And, the full cost of housing a
prisoner in the County jail system must be the starting point for decision makers.
11. We would recommend a detailed jail system cost study focused primarily on
payroll and benefit costs, particularly costs related to uniformed officers including
a detailed comparison with other jurisdictions. We suspect that the number of
individuals on the payroll, their rate of pay, benefits, sick leave and overtime
compensation may explain, in large measure, the relatively high cost of operating
current facilities compared with other systems such as King County, Washington.
12. Consideration should also be given to privatizing components of jail operations
where feasible, as we understand other jurisdictions have done. This would no
doubt result in a substantial savings to the County over the present system.
13. The following observations were developed from our review of the Department of
Justice, U.S. Marshal Service Audit report dated July 28, 2006 covering the years
2002 and 2004.
a. A portion of the corrections health expense was disallowed. Medical
services provided to federal inmates offsite at local hospitals or clinics are
billed directly to the USMS by the hospital or clinic. “Thus, the excluded
medical costs are allowable to the County but are not allocable to the
daily jail rate because they do not benefit USMS prisoners.” If these
costs are billed directly to the USMS by the health care providers, how
are they a cost to the County and why were they included in the MSCO
costs per the USM-243 form to begin with? Does the FY07 MSCO
budget include these costs? If so, should they be eliminated from the
Corrections Health portion of the jail bed rate?

Page 22 of 33

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006

b. “Capital Depreciation” of equipment as referenced in the DOJ audit
report, was computed using a 15 year life. However, the actual useful life
on a GAAP basis could be considerably less than 15 years – especially
for EDP equipment. OMB Circ. A-87, beginning on page 23, discusses
two acceptable methods that may be employed for the cost of equipment
– 1) “usage allowance” or, 2) “depreciation”. Depreciation must be
computed on a straight-line basis unless it can be shown that
consumption of the asset will be significantly greater in the early years. If
depreciation is not used, a usage allowance of 6.67%/year (15 years)
may be used instead. The auditor added usage allowance amounts to
the Sheriff’s allowable costs for equipment at both facilities which appears
to have been significantly understated. If depreciation were used instead
of the much lower usage allowance, the understatement would have been
even greater.
c. Certain assets were expensed that should have been capitalized and
depreciated at both facilities in FY2004. This indicates that Generally
Accepted Accounting Principles “GAAP” is not being followed in these
cases as required in the budgeting process.
14. We noted the following Items for discussion and follow up during our review of
Sheriff’s Response to Grand Jury Status Report, October 16, 2006.
a. The Sheriff states that “In the early years, Sheriff’s office used to employ a
“fully loaded” methodology to determine jail bed costs.” (See page 2,
paragraph 4.) “This methodology, while easy to calculate, quickly broke down
when budget cuts forced jail bed closures and average bed day costs did not
provide sufficient clarity to support policy decisions by the agency or the
Board of County Commissioners.” Please explain what is meant by the
phrase “average bed day costs did not provide sufficient clarity to support
policy decisions”. Also, please define the term, “fully loaded”.
b. Quoting the Sheriff, “Program budgeting forces transparency of costs and
many of the authors cited above caution policy makers to be prepared for
“sticker shock” as they begin seeing the “true cost” of programs based on
collective resource consumption rather than general ledger budget
submissions.” How does exclusion of some cost elements or netting
revenues against cost in the jail bed calculation on the part of the Sheriff’s
office further the goals of “transparency” and understanding of “true cost”?
What is the rational for excluding these cost items from the jail bed
calculation and how were the exclusions determined? The schedule below
was prepared from a spreadsheet prepared by Markley Drake of the Sheriff’s
office to illustrate the program offer components excluded from their
calculation of jail bed costs17.

17

Developed from Excel file “jail bed cost Oct 11 061.xls” prepared by Markley Drake, Senior
Research and Evaluation Analyst, MCSO.

Page 23 of 33

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006
Program
Excluded or Reduced Program Items
Offer
Transport
$ 2,601,539
Booking
7,530,369
Court Services
5,082,321
Commissary
2,458,912
W orkcrews
2,319,013
Gresham Temp Hold
147,447
$ 20,139,601

$

Included
929,642
4,992,635

$

5,922,277

Difference
1,671,897
2,537,734
5,082,321
2,458,912
2,319,013
147,447
$ 14,217,324
$

c. Please define the terms “collective resource consumption” and “general
ledger budget” and explain why the “true cost” of a jail bed would be different
under one or the other. (Page 3, paragraph 1).
d. What “fixed and overhead costs” were redistributed to MCDC and MCIJ as a
result of the closure of MCCF and MCRC? Define the terms “fixed” and
“overhead”. (Page 4 paragraph 2).
e. Explain the huge discrepancy between the ratio of uniformed staff – 3.6 per
MCDA, 6.1 (DC) and 8.5 (IJ) per MSCO. (Page 4, paragraph 4)
f.

The Sheriff seems to be indicating the SB1145 revenues allocated to
corrections was a policy decision. “The local policy direction was to provide
more funding for jail beds than provided in the SB 1145 rate, but to not cover
the total systems cost of the jail bed.” (Page 5, paragraph 4) Please explain.

g. City of Portland purchased 57 beds from the County at $62.48/day. “This
policy decision put 57 beds into use that were not going to be funded
otherwise.” (Page 6, paragraph 3) Please explain how receiving less than
half the full cost per bed from the City of Portland is beneficial to the County?
If there were no other non-financial reasons for doing so, wouldn’t it be better
to recover the full cost?
h. “Recovery of fixed and overhead costs within our rates has a positive impact
on the general fund. It is also true that policy decisions other than cost have
played an important role in determining that rate.” (Page 8, paragraph 4) If
recovery of fixed and overhead costs has a positive impact on the general
fund, would not recovery of the full cost for federal and state prisoners be an
even better outcome? What policy decisions is the Sheriff referring to
specifically, that result in the County subsidizing a significant portion of the
cost of housing federal and state prisoners? Note that even if genuine “fixed
costs” existed in MCSO it is universally acknowledged that variable costs
must be covered FIRST before any “fixed cost” benefit is gained.

Page 24 of 33

$ 23.05

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006

CLOSING
We sincerely appreciate the trust placed in us in connection with providing information in
this significant matter.
This letter comprises our work to date in connection with the above referenced matter.
We reserve the right to update, amend or modify this document based upon available
data, and solely at our discretion.
Best professional regards,

Darrell D. Dorrell, CPA/ABV, MBA, ASA, CVA, CMA, DABFA, CMC
Principal
financialforensics®

Thomas S. Brown, CPA/ABV, CVA
Director
financialforensics®
†

The Appendix is an integral of this report.

Page 25 of 33

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006

Corrections Health FY 2007 Budgeted Costs
Mult. Co. Detention Center (MCDC)
Beds
676
Program Offer
Dormitory costs
40025A
40025B
40025C
40025D
40025E
40025F
40025G
40025H
40025I
sub-total

MCDC

Inverness Jail (MCIJ)
1,014
Program Offer

$2,220,912
$710,393
$277,749
$432,387
$763,806
$266,580
$266,580
$281,776
$281,776
$5,501,959

40028A
40028B
40028C
40028D
40028E
40028F
40028G
40028H
40028I

Inverness

System-Wide
Total
1,690
Totals

$2,131,177
$277,745
$937,102
$177,602
$998,758
$998,766
$301,803
$301,803
$150,899
$6,275,655

$11,777,614

Program Cost Allocations
Allocable
Program Cost %
Base

%

MCDC

%

MCIJ

Reception
40026

$364,785

40026

$449,937

$814,722

$814,722 100%

$814,722 45%

$364,785 55%

$449,937

Mental Health
40059

$713,444

40059

$879,982

$1,593,426

$1,593,426 100%

$1,593,426 45%

$713,444 55%

$879,982

-$895,484
$447,742
-$447,742

40905

-$1,104,516
$552,258
-$552,258

-$2,000,000
$1,000,000
-$1,000,000

-$2,000,000 100%
$1,000,000 100%
-$1,000,000 100%

-$2,000,000 45%
$1,000,000 45%
-$1,000,000 45%

-$895,484 55%
$447,742 55%
-$447,742 55%

-$1,104,516
$552,258
-$552,258

Budget cut
40905
Contingency

Total Allocated Costs

$6,132,446

$7,053,316

$13,185,762

Per Bed Per Day

$24.85

$19.06

$21.38

I

Page 26 of 33

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006

INFORMATION SOURCES
The following sources and the footnote references in our report comprise the primary
information sources upon which we relied, cumulative knowledge notwithstanding.
Budget Office Documents
ˆ
ˆ
ˆ
ˆ
ˆ
ˆ
ˆ
ˆ

Chief Executive Budget Message 5/4/2006
Budget Manager’s Message
Priority-Based Budgeting
FY 07 Budget Readers Guide – Volume 2
Sheriff’s Office Budget for FY 2007
MCSO Corrections (Jail) related Program Offers
Health Department Budget for FY 2007
Corrections Health Program Offers

Newspaper Articles
ˆ
ˆ

“Report says prisoner contracts costly”, The Oregonian, October 10, 2006
“DA to end its probe of Multnomah County Sheriff’s Office”, Portland Tribune,
October 10, 2006

USM-243 Cost Sheet for Detention Services, USM-243 Instructions and Form
OMB Circular A-87 (Revised 5/10/04)
U. S. Department of Justice, U.S. Marshals Service Intergovernmental Service
Agreement For Detention Facilities With The Multnomah County Sheriff’s Office
Portland, Oregon, U.S. Department of Justice, Office of the Inspector General, Audit
Division, Audit Report GR-90-06-007, July 28, 2006, FY 2004 and FY 2002.
Adjustments to USM Rates for FY03/04 Audits – Prior to Report
Report to Management, MCSO, FY05 SEA Reporting Process, Suzanne Flynn,
Multnomah County Auditor and Excel file with FY 04 and FY 05 Actual Cost Allocations
Multnomah County District Attorney’s Office calculation of jail bed costs and supporting
worksheets.
Memorandum from Larry Aab, MCSO to John Bradley dated 10/4/06 “Information
Request”
Local Control Average Daily Population based on Beds Used by Jail, State SB1145 Bed
Day Monthly Data
MCSO Facilities Budget – FY 2007 - Adopted
Sheriff’s Response to Grand Jury Status Report dated 10/16/06.

Page 27 of 33

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006

MSCO Excel file, Jail Bed Costs Oct 11 061.xls – Supporting calculations for Sheriff’s jail
bed cost figure cited in his Response to Grand Jury Status Report dated 10/16/06.
King County Jail Costs
ˆ
ˆ
ˆ
ˆ
ˆ
ˆ

King County Corrections Budget and jail bed cost data prepared by Scott
Marcy, Multnomah County District Attorney’s Office
Adult Population Alert Notification,
www.metrokc.gov/dad/population/Populations-Alert.html
Key Differences between King County and Multnomah County Jail Systems,
Chuck French, Multnomah County District Attorney’s Office
Performance Audit Jail Overtime, Cheryle A. Broom, King County Auditor,
October 9, 2006
Adult Detention Jail Costs Follow-up Study, Cheryle A. Broom, King County
Auditor, July 23, 2003
Adult Detention Jail Costs Audit, Cheryle A. Broom, King County Auditor,
January 28, 2003.

Statement #4/ Managerial Cost Accounting Concepts and Standards for the Federal
Government, Executive Office of the President, Office of Management and Budget,
"Managerial Cost Accounting Concepts and Standards for the Federal Government"
Statement of Federal Financial Accounting Standards Number 4, July 31, 1995

Page 28 of 33

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006
18

GLOSSARY OF TERMS

Activity - The actual work task or step performed in producing and delivering products
and services. An aggregation of actions performed within an organization
that is useful for purposes of activity-based costing.
Activity Analysis - The identification and description of activities in an organization.
Activity analysis involves determining what activities are done within a much time they
spend performing the activities, what resources are required to perform the activities,
what operational data best reflect the performance of the activities, and what customer
value the activity has for the organization. Activity analysis is accomplished with
interviews, questionnaires, observation, and review of physical records of work.
It is the foundation for agency process value analysis, which is key to overall review of
program delivery.
Activity-Based Costing - A cost accounting method that measures the cost and
performance of process related activities and cost objects. It assigns cost to cost
objects, such as products or customers, based on their use of activities. It recognizes the
causal relationship of cost drivers to activities.
Actual Cost - An amount determined on the basis of cost incurred including standard
cost properly adjusted for applicable variance.
Avoidable Cost - A cost associated with an activity that would not be incurred if the
activity were not performed.
Common Cost - The cost of resources employed jointly in the production of two or more
outputs and the cost cannot be directly traced to any one of those outputs.
Common Data Source - All of the financial and programmatic information available for
the budgetary, cost, and financial accounting processes. It includes all financial and
much non-financial data, such as environmental data, that are necessary for budgeting
and financial reporting as well as evaluation and decision information developed as a
result of prior reporting and feedback.
Controllable Cost - A cost that can be influenced by the action of the responsible
manager. The term always refers to a specified manager since all costs are controllable
by someone.
Cost - The monetary value of resources used or sacrificed or liabilities incurred to
achieve an objective, such as to acquire or produce a good or to perform an activity or
service.

18

Statement #4/Managerial Cost Accounting Concepts and Standards for the Federal
Government, Executive Office of the President, Office of Management and Budget, "Managerial
Cost Accounting Concepts and Standards for the Federal Government"
Statement of Federal Financial Accounting Standards Number 4, July 31, 1995

Page 29 of 33

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006

Cost Accounting Practice - Any disclosed or established accounting method or technique
which is used for measurement of cost, assignment of cost to accounting periods, and
assignment of cost to cost objects.
Cost Allocation - A method of assigning costs to activities, outputs, or other cost objects.
The allocation base used to assign a cost to objects is not necessarily the cause of the
cost. For example, assigning the cost of power to machine activities by machine hours is
an allocation because machine hours are an indirect measure of power consumption.
Cost Assignment - A process that identifies costs with activities, outputs, or other cost
objects. In a broad sense, costs can be assigned to processes, activities, organizational
divisions, products, and services. There are three methods of cost assignment: (a)
directly tracing costs wherever economically feasible, (b) cause-and-effect, and (c)
allocating costs on a reasonable and consistent basis.
Cost Driver - Any factor that causes a change in the cost of an activity or output. For
example, the quality of parts received by an activity, or the degree of complexity of tax
returns to be reviewed by the IRS.
Cost Finding - Cost finding techniques produce cost data by analytical or sampling
methods. Cost finding techniques are appropriate for certain kinds of costs, such as
indirect costs, items with costs below set thresholds within programs, or for some
programs in their entirety. Cost finding techniques support the overall managerial cost
accounting process and can represent non-recurring analysis of specific costs.
Cost Object (also referred to as Cost Objective) – An activity, output, or item whose cost
is to be measured. In a broad sense, a cost object can be an organizational division, a
function, task, product, service, or a customer.
Differential Cost - The cost difference expected if one course of action is adopted
instead of others.
Direct Cost - The cost of resources directly consumed by an activity. Direct costs are
assigned to activities by direct tracing of units of resources consumed by individual
activities. A cost that is specifically identified with a single cost object.
Estimated Cost - The process of projecting a future result in terms of cost, based on
information available at the time. Estimated costs, rather than actual costs, are
sometimes the basis for credits to work-inprocess accounts and debits to finished goods
inventory.
Expense - Outflow or other using up of resources or incurring liabilities (or a combination
of both), the benefits from which apply to an entity's operations for the current
accounting period, but do not extend to future periods.
Fixed Cost - A cost that does not vary in the short term with the volume of activity. Fixed
cost information is useful for cost savings by adjusting existing capacity, or by
eliminating idle facilities. Also called Non-Variable Cost or Constant Cost.

Page 30 of 33

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006

Full-Absorption Costing - A method of costing that assigns (absorbs) all labor, material,
and service/manufacturing facilities and support costs to products or other cost objects.
The costs assigned include those that do and do not vary with the level of activity
performed.
Full Cost - The sum of all costs required by a cost object including the costs of activities
performed by other entities regardless of funding sources.
Incremental Cost - The increase or decrease in total costs that would result from a
decision to increase or decrease output level, to add a service or task, or to change any
portion of operations. This information helps in making decisions such as to contract
work out, undertake a project, or increase, decrease, modify, or eliminate an activity or
product.
Indirect Cost - A cost that cannot be identified specifically with or traced to a given cost
object in an economically feasible way.
Inter-Entity - A term meaning between or among different federal reporting entities. It
commonly refers to activities or costs between two or more agencies, departments, or
bureaus.
Job Order Costing - A method of cost accounting that accumulates costs for individual
jobs or lots. A job may be a service or manufactured item, such as the repair of
equipment or the treatment of a patient in a hospital.
Managerial Cost Accounting System - The organization and procedures, whether
automated or not, and whether part of the general ledger or stand-alone, that information
and performance data from various agency feeder systems. The accumulated and
reported data enable management and other interested parties to measure and make
decisions about the agency's/segment's ability to improve operations, safeguard assets,
control its resources, and determine if mission objectives are being met.
Opportunity Cost - The value of the alternatives foregone by adopting a particular
strategy or employing resources in a specific manner. Also called Alternative Cost or
Economic Cost.
Outcome - The results of a program activity compared to its intended purposes. Program
results may be evaluated in terms of service or product quantity and quality, customer
satisfaction, and effectiveness.
Outputs - Any product or service generated from the consumption of resources. It can
include information or paper work generated by the completion of the tasks of an activity.
Performance Measurement - A means of evaluating efficiency, effectiveness, and
results. A balanced performance measurement scorecard includes financial and
nonfinancial measures focusing on quality, cycle time, and cost. Performance
measurement should include program accomplishments in terms of outputs (quantity of
products or services provided, e.g., how many items efficiently produced?) and
outcomes (results of providing outputs, e.g., are outputs effectively meeting intended
agency mission objectives?). See

Page 31 of 33

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006

Statement of Federal Financial Accounting Concepts No. 1, Objectives of Federal
Financial Reporting, page 65.
Process - The organized method of converting inputs (people, equipment, methods,
materials, and environment), to outputs (products or services). The natural aggregation
of work activities and tasks performed for program delivery.
Process Costing - A method of cost accounting that first collects costs by processes and
then allocates the total costs of each process equally to each unit of output flowing
through it during an accounting period.
Process Value Analysis - Tools and techniques for studying processes through customer
value analysis. Its objective is to identify opportunities for lasting improvement in the
performance of an organization. It provides an in-depth review of work activities and
tasks, through activity analysis, which aggregate to form processes for agency program
delivery. In addition to activity-based costing, quality and cycle time factors are studied
for a complete analysis of performance measurement. Each activity within the process is
analyzed, including whether or not the activity adds value for the customer.
Product - Any discrete, traceable, or measurable good or service provided to a
customer. Often goods are referred to as tangible products, and services are referred to
as intangible products. A good or service is the product of a process resulting from the
consumption of resources.
Responsibility Center - An organizational unit headed by a manager or a group of
managers who are responsible for its activities. Responsibility centers can be measured
as revenue centers (accountable for revenue/sales only), cost centers (accountable for
costs/expenses only), profit centers (accountable for revenues and costs), or investment
centers (accountable for investments, revenues, and costs).
Responsibility Segment - A significant organizational, operational, functional, or process
component which has the following characteristics: (a) its manager reports to the entity's
top management; (b) it is responsible for carrying out a mission, performing a line of
activities or services, or producing one or a group of products; and (c) for financial
reporting and cost management purposes, its resources and results of operations can be
clearly distinguished, physically and operationally, from those of other segments of the
entity.
Service - An intangible product or task rendered directly to a customer.
Standard Costing - A costing method that attaches costs to cost objects based on
reasonable estimates or cost studies and by means of budgeted rates rather than
according to actual costs incurred. The anticipated cost of producing a unit of output. A
predetermined cost to be assigned to products produced. Standard cost implies a norm,
or what costs should be. Standard costing may be based on either absorption or direct
costing principles, and may apply either to all or some cost elements.
Support Costs - Costs of activities not directly associated with production. Typical
examples are the costs of automation support, communications, postage, process
engineering, and purchasing.

Page 32 of 33

MULTNOMAH COUNTY DISTRICT ATTORNEY
Expert Report
October 25, 2006

Traceability - The ability to assign a cost directly to a specific activity or cost object by
identifying or observing specific resources consumed by the activity or cost object.
Uncontrollable Cost - The cost over which a responsible manager has no influence.
Unit Cost - The cost of a selected unit of a good or service. Examples include dollar cost
per ton, machine hour, labor hour, or department hour.
Value-Added Activity - An activity that is judged to contribute to customer value or satisfy
an organizational need. The attribute "value-added" reflects a belief that the activity
cannot be eliminated without reducing the quantity, responsiveness, or quality of output
required by a customer or organization. Value-added activities should physically change
the product or service in a manner that meets customer expectations.
Variable Cost - A cost that varies with changes in the level of an activity, when other
factors are held constant. The cost of material handling to an activity, for example, varies
according to the number of material deliveries and pickups to and from that activity.
Variance - The amount, rate, extent, or degree of change, or the divergence from a
desired characteristic or state.

Page 33 of 33

Appendix B

MCDC Jail Purchasing Plan

Multnomah
County Sheriff’s
Office

FY 06/07 Budget Offerings

PROGRAM GROUP
Total Offer 60021
Total Beds - 670
Total FTE - 220.87
Total Budget - $31,383,420

Offers are subject to change

Blocks in White are Scalable
Left to right to purchase, and
right to left to scale down.

OFFER SET FIVE
Offer 60021 H & I

Jail Support

Beds 156
FTE 24.84
Cost $3,636,290

Jail
Operations

Corrections
Counselors
1.00 FTE

Dorm 8B
Disciplinary

Dorm 8C
General

Dorm 8D
General

Female

Male

Female

Male

Single Cells

Escort
5.46 FTE

Equipment/
Property Tech
FTE 1.0

3.64 FTE

Floor
Control
5.46 FTE

Records
Techs
FTE 1.0

Purchasing Order

Dorm 8A
General

Jail Support

Beds 156
FTE 26.66
Cost $3,911,605

Jail
Operations

Corrections
Counselors
1.00 FTE

Escort
5.46 FTE

Equipment/
Property Tech
FTE 1.0
Records
Techs
FTE 1.0

Purchasing Order

Dorm 7A
General/
Suicide Watch

Floor
Control
5.46 FTE

3.64 FTE

HUD 1.82

62 Offenders

Offer 60021F 78 Beds FTE 23.02

OFFER SET THREE
Offer 60021 D & E
Beds 156
FTE 37.47
Cost $5,735,496

Records
Techs
FTE 1.0

Sergeants
(Sgt)
8.17 FTE

Dorm 7C
Mental
Health

Single Cells

Single Cells

16 Offenders

Female
3.64 FTE

62 Offenders

16 Offenders

Offer 60021G 78 Beds 3.64 FTE
$584,275

$3,327,330

Dorm 6B
Disciplinary

Dorm 6C
General

Male

Female

Floor
Control
5.46 FTE

3.64 FTE

3.64 FTE

HUD 1.82

62 Offenders

16 Offenders

Offer 60021D 78 Beds FTE 30.19

Dorm 7D
General

3.64 FTE

Dorm 6A
General

62 Offenders

Offer 60021I Beds 78 FTE 3.64
$584,275

Dorm 7B
Mental
Health

Escort
5.46 FTE

Corrections
Counselors
1.00 FTE

Jail
Administration

Purchasing Order

Jail
Operations

Jail Support

16 Offenders

16 Offenders

62 Offenders

Offer 60021H 78 Beds FTE 21.20 $3,052,015

OFFER SET FOUR
Offer 60021 F & G

3.64 FTE

3.64 FTE

Dorm 6D
General

3.64 FTE

3.64 FTE

16 Offenders

62 Offenders

Offer 60021E 78 Beds 7.28 FTE
$1,114,443

$4,621,053

Purchasing Order
OFFER SET TWO
Offers 60021 B & C

Jail
Operations

Beds 156
FTE 28.08
Cost $4,645,541

Dorm 5A
General

Escort
5.46 FTE

Recreation
10 A,B & C

Dorm 5C
Disciplinary

Male

Female

Single Cells

Single Cells

3.64 FTE

Floor
Control
5.46 FTE

2.60

Dorm 5B
Disciplinary

62 Offenders

Jail
Operations

Beds 46
FTE 103.82
Cost $13,454,488

Floor
Control
5.46 FTE

Escort
5.46 FTE

39.4%
39.4%

Dorm 4A
Medical
Single Cells

3.64 FTE
10 Offenders

Dorm 4B

Dorm 4C

Administrative
Segregation

Administrative
Segregation

Single Cells

Single Cells

8 Offenders

Core Jail
Operations

3.64 FTE

16 Offenders

16 Offenders

Master Jail
Control
10.92 FTE

Medical Clinic
1.30 FTE

Core Jail Administration

Manager Correction
Counselors
1.0 FTE

Lieutenant (Lt)
L
Executive
1.0 FTE
OA Sr
1.0 FTE
Sergeants (Sgt)
Supervisors of
units. 9.90 FTE

Administrative
Sgt 1.0 FTE

Dorm 4E
Disciplinary

Dorm 4F
Disciplinary

Single Cells

Single Cells

5 Offenders

10 Offenders

Corrections
Counselors
5.00 FTE
Chaplain
1.0 FTE

Maintenance
1.30 FTE

Transport

Court
Escort/ Holding
2.60

Manager
Correction Records
1.0 FTE

Administrative
Support

Records Techs
FTE 2.0
OA 2
Records
FTE 1.0
Facility Security
Officers
FTE 14.3

Manager Aux
Services Unit
1.0 FTE

HR
2.0 FTE

Laundry Supervisor
1.0 FTE

Payroll
2.0 FTE

Equipment/
Property Techs
FTE 12.0
Buyer
1.0 FTE

Sewing
Specialist
FTE 1.0
OA Sr
1.0 FTE

Fiscal
2.0 FTE

Laundry
ServicesIT

Designates 100 % External Funding
Source(s)

24 x 7 Twenty-four hours/seven days a week

75%

Designates 50 % up to 100% External
Funding Source(s)

8 x 5 Eight hours/five days a week

35%

Designates under 50 % External Funding
Source(s)

16 x 7 Sixteen hours/seven days a week

75%

35%

MCDC Process Maps Feb 15 06 Budget.vsd
OA Senior
1.0 FTE

5 Offenders

Arraignment
Court

100%

100%

Costs include
FTE, M&S,
and ISRs

Single Cells

3.64 FTE

Core Jail/Corrections Support
Corrections
Chief Deputy
1.0 FTE

Captain –
cJail Commander
1.0 FTE

LLieutenant (Lt)
Shift Supervisors
3.0 FTE

Dorm 4D
Mental Health

8 Offenders

Medical
Transport
1.30 FTE

62 Offenders

Offer 60021C 78 Beds FTE 7.28
$1,321,875

Core Jail Services
Medical
Clinic

3.64 FTE

3.64 FTE

Offer 60021B 78 Beds FTE 20.80 $3,323,666

BASE OFFER
OFFER SET ONE
Offer 60021A

Dorm 5D
Facility
Workers

2.0 FTE

Printed 3/1/2006

Inverness Jail Purchasing Plan

Program Group
60022 Totals
Total Beds 1,014
Total FTE 217.13*
Total Costs $33,098,119*

Multnomah
County Sheriff’s
Office

FY 06/07 Budget Offerings
Offers are subject to change

Blocks in White are Scalable
Left to right to purchase, and right to left to scale down.

Purchasing Order
OFFER SET FIVE
Offer 60022- I

Dorm 1
General

Dorm 2
General

5.46 FTE

3.64 FTE

Beds - 114
FTE 9.10
Cost $1,521,309

57 Offenders

60022 I Beds 114 FTE 9.10
$1,521,309

OFFER SET FOUR
Offer 60022- G & H
Beds - 171
FTE 30.66*
Cost $4,837,868*

East Jail
Administration

Inmate Workcrews
Offer 60025
Corrections
Workcrews

Sgt
1.00 FTE

Jail Support

14.20 FTE

68.5%

Dorm 3
General

Dorm 4
General

Dorm 5
General

5.46 FTE

3.64 FTE

3.64 FTE

Escorts
5.46 FTE
East Jail
Control
5.46 FTE

Corrections
Counselor
4.00 FTE

1.0 FTE

$2,339,208

East Jail
Operations

Records
Techs
FTE 2.0

Lieutenant
L
(Lt)

57 Offenders

57 Offenders
57 Offenders

57 Offenders

100%
100%

68.5%

60022 G Beds 57 FTE 16.10

OFFER SET THREE
Offer 60022- E & F

60022 H Beds 114 FTE 9.10
$1,596,681

$3,241,187

Jail Support

Beds - 232
FTE 35.12
Costs $5,509,231

Corrections
Counselors
FTE 3.0
Records
Techs
FTE 2.0
Office
Assistants
FTE 1.0

Jail
Administration
Sergeants
(Sgt)
5.46 FTE

Dorm 6
General

Dorm 7
General

Dorm 8
General

Dorm 9
General

5.46 FTE

3.64 FTE

5.46 FTE

3.64 FTE

Jail
Operations
Processing
Control
5.46 FTE

60022 E Beds 116

FTE 26.02

58 Offenders

58 Offenders

58 Offenders

58 Offenders

60022 F Beds 116 FTE 9.10
$1,397,375

$4,111,856

Purchasing Order

Purchasing Order
OFFER SET TWO
Offers 60022- B - D
Beds - 337
FTE* 31.94
Cost $5,486,312

Dorm 14
0.59%

0.59%

Dorm 12
General

Dorm 13
Mental Health

Female

Mental Health
Single Cells

9.10 FTE

9.10 FTE

75 Offenders

65 Offenders

65 Offenders

78 Offenders

31 Offenders

23 Offenders

Jail Control
10.92 FTE

47.3%

Visiting
3.64 FTE

Maintenance
1.30 FTE

3.64 FTE

3.64 FTE

75 Offenders

75 Offenders

10 Offenders

Designates 100 % External Funding Source(s)

Medical
Transport
1.30 FTE
Transport

Core Jail Support
Operations Supervisor
Aux Services
1.0 FTE

Corrections
Counselors
FTE 7.0

Equipment/
Property Techs
FTE 5.0

Corrections
Techs
FTE 2.0

OA 2s

* Does not include FTE or
costs for Workcrews,
Program Offer 60025

MCIJ Process Budget Map Feb 15 06

5.46 FTE

Medical
Clinic
1.30 FTE

Manager
Corrections Counselors
1.0 FTE

Sergeants (Sgt)
Supervisors various work units.
12.48 FTE

100%

Dorm 18
Medical

Medical Clinic

Kitchen
Services

cCaptain –
Jail Commander
1.0 FTE
Lieutenant (Lt)
L
Shift Supervisors
3.0 FTE

100%

Dorm 11
General

Core Jail Services
Kitchen
3.64 FTE

Core Jail Administration

Costs include FTE,
M&S, and ISRs

60022 D Beds 54 FTE 3.64
$636,307

Dorm 10
General

Escorts
16.38 FTE

75%

(Records/Counselor)

FTE 2.0

Designates 50 % up to 100% External Funding
Source(s)

Administrative
Support

Chaplain
1.0 FTE

24 x 7 Twenty -four hours/seven days a week
8 x 5 Eight hours/five days a week
16 x 7 Sixteen hours/seven days a week

75%

Single Cells

1.82 FTE

47.3%

Processing
Control
5.46 FTE

Female

Single Cells

1.82 FTE

60022 C Beds 143 FTE 13.74
$2,310,578

Core Jail Operations

Dorm 17
Disciplinary

Male
3.64 FTE

Base Offer
OFFER SET ONE
Offer 60022 – A
Beds - 160
FTE 110.31
Cost $15,743,399

Dorm 16
Disciplinary

Single Cells

MH Counselor
1.00 FTE

5.46 FTE

60022 B Beds 140 FTE 14.56
$2,539,427

Dorm 15
General

35%

Records Techs
FTE 4.00

Payroll
2.0 FTE

Facility Security
Officers
FTE 7.15

Fiscal
2.0 FTE

HR
2.0 FTE
IT
1.0 FTE

Dorm 3 in Red is purchased in part by City of
Portland System wide contribution

Laundry35%
Services

Designates under 50 % External Funding Source(s)

Printed 2/28/2006

Program Group
Total Beds - 525
Total FTE - 123.40
Total MCSO Budget $ 9,921,174.24
Totals with Corrections Health
FTE – 137.75
Costs - $11,740,200

Wapato Jail Purchasing Plan

Multnomah
County Sheriff’s
Office

FY 06/07 Budget Offerings
Offers are subject to change

Budget is for 6 months starting January 01, 2007
Blocks in White are Scalable
Left to right to purchase, and right to left to scale down.

Floor 3
OFFER SET FIVE
Offer 60023E

Dorm 5
General

Corrections Health

Jail
Operations

Jail Support

Beds 100
FTE 16.56
Costs $1,198,044

FTE 0.35
Costs $108,788
Corrections
Counselor
1.00 FTE

Totals with Corrections Health
FTE – 16.91
Costs - $1,306,832

Records
Techs
1.00 FTE

5.46 FTE

Dorm 6
General

5.46 FTE

3.64 FTE

50 Offenders

50 Offenders

Purchasing Order
Floor 2
OFFER SET FOUR
Offer 60023 D

Corrections Health

Beds 100
FTE 11.10
Costs $ 888,541

Dorm 3
General

Dorm 4
General

5.46 FTE

3.64 FTE

50 Offenders

50 Offenders

Jail Support

FTE 2.25
Costs $277,227

Records
Techs
1.00 FTE

Corrections
Counselor
1.00 FTE

Totals with Corrections Health
FTE – 13.35
Costs - $1,165,768

Floor 1
OFFER SET THREE
Offer 60023 C

Corrections Health

Beds 100
FTE 11.10
Costs $887,791

Dorm 2
General

5.46 FTE

3.64 FTE

50 Offenders

50 Offenders

Jail Support

FTE 3.75
Costs $442,401

Records
Techs
1.00 FTE

Corrections
Counselor
1.00 FTE

Totals with Corrections Health
FTE – 14.85
Costs - $1,330,192

Purchasing Order

Dorm 1
General

Purchasing Order
OFFER SET TWO
Offer 60023 B
Beds 75
FTE 46.06
Costs $ 4,167,663

Dorm 9
General

Corrections Health
FTE 6.85
Costs $832,473

Totals with Corrections Health
FTE – 52.91
Costs - $5,000,136

Core Jail
Operations

Core Jail Services
Medical
Clinic

Processing
5.46 FTE
Escort
5.46 FTE

Core Jail Support

Lieutenant (Lt)
L
Shift Supervisors
2.0 FTE

Sergeants
Lt Relief
1.68 FTE

Admin
Sergeant
1.00 FTE

Shift
Sergeants
5.46 FTE

Manager
Correction
Counselors
1.0 FTE
Corrections
Tech
2.00 FTE

Corrections
Counselor
1.00 FTE
Office
Assistants
1.00 FTE

5.46 FTE

Maintenance
1.30 FTE

Transport

Core Jail Administration
Captain
–
c
Jail Commander
1.0 FTE

Medical Clinic
1.30 FTE
Transport
1.30 FTE

75 Offenders

Administrative
Support

Facility
Security
Officers
FTE 3.64

HR
1.0 FTE

Records
Techs
FTE 2.0

Payroll
1.0 FTE

Fiscal
1.0 FTE
IT
1.0 FTE

Purchasing Order

Base Offer One
OFFER SET ONE
Offer 60023 A
Beds 150
FTE 38.58
Costs $ 2,779,135

Corrections Health
FTE 1.15
Costs $158,137

Totals with Corrections Health
FTE 39.73
Costs $2,937,272

Dorm 8

General

General

5.46 FTE

75 Offenders

Core Jail Support

Core Jail Administration
Lieutenant (Lt)
L Commander
Jail
1.0 FTE

Dorm 7

Shift
Sergeants (Sgt)
5.46 FTE

Facility Security
Officers
FTE 3.64

Corrections
Counselor
2.00 FTE

Costs include: FTE,
M&S, ISRs and
Admin & Support

3.64 FTE
75 Offenders

Core Jail Operations

Equipment/
Property Techs
1.00 FTE

Central Jail
Control
5.46 FTE

Escort
10.92 FTE

24 x 7 Twenty -four hours/seven days a week
8 x 5 eight hours/five days a week
16 x 7 sixteen hours/seven days a week

Wapato Process Costing Feb 16 06.vsd

Not Reconciled with PCP

Printed 2/27/2006

Appendix C

Summary Corrections FY07
MCSO Jail Program
MCSO Dorm Costs
MCSO Exec/PS
Offender Transport
Booking-Release/Classification/temp hold
Inmate Commissary
MCSO Corrections Work Crews

MCDC (676 beds)
$30,735,277.00
$980,991.00
$780,461.00
$4,534,881.00
$983,564.00
$0.00
$38,015,174.00

Inverness (1014 beds)
$32,457,458.00
$1,238,638.00
$1,821,077.00
$6,802,323.00
$1,475,347.00
$2,319,013.00
$46,113,856.00

$5,501,959.00
$325,888.00
$637,370.00
-$400,000.00
$6,065,217.00

$6,205,665.00
$488,833.00
$956,055.00
-$600,000.00
$7,050,553.00

Total Corrections Cost

$44,080,391.00

$53,164,409.00

Court Services

Courthouse
$3,899,276.00

Justice Center
$1,183,045.00

Total Cost of Corrections/Court Services

$102,327,121.00

Corrections Facility Bed Days

246,740

370,110

Cost per bed day
Average daily bed cost system wide

MCDC
$178.65
$157.65

MCIJ
$143.64

Misc. other corrections costs
Wapato Moth ball costs
Adult Recog (DCJ)

$315,929.00
$1,342,173.00

Corrections Health Jail Program
HD Dorm Costs
HD CH Reception
CH Mental Health
CH budget reduction (net)

comments
MCSO shows $8,112,688 SB1145,
2,388,499 commissary revenue,
1.2 million from Portland,
864,252 work crews revenue

does not include Court Services or Wapato

MCSO Corrections Cost FY07
Dormitory costs
Program Offer
60021A
60021B
60021C
60021D
60021E
60021F
60021G
60021H
60021I
sub-total

MCDC
$13,252,077.00
$3,238,647.00
$1,292,119.00
$4,505,829.00
$1,084,687.00
$3,245,500.00
$569,397.00
$2,977,624.00
$569,397.00
$30,735,277.00

Program Offer
60022A
60022B
60022C
60022D
60022E
60022F
60022G
60022H
60022I

Inverness
$15,592,448.00
$2,463,430.00
$2,244,079.00
$617,307.00
$4,007,358.00
$1,349,876.00
$3,159,968.00
$1,549,182.00
$1,473,810.00
$32,457,458.00

Program Offer
60001
60002
sub-total

MCDC
$515,292.00
$465,699.00
$980,991.00

Program Offer
60001
60002

Inverness
$772,939.00
$465,699.00
$1,238,638.00

Offender Transport
Program Offer
60015

MCDC
$780,461.00

Program Offer
60015

Inverness
$1,821,077.00

MCDC
$3,012,147.00
$1,463,756.00
$58,978.00
$4,534,881.00

Program Offer
60016A Booking-Release
60016B Classification
60016C Gresham Temp hold

Inverness
$4,518,221.00
$2,195,634.00
$88,468.00
$6,802,323.00

Court Services
Program Offer
60018A

Courthouse
$3,899,276.00

Program Offer
60018B

Justice Center
$1,183,045.00

Inmate Commissary
Program Offer
60019

MCDC
$983,564.00

Program Offer
60019

Inverness
$1,475,347.00

MCSO Corrections Work Crews

MCDC
$0.00

MCIJ
$2,319,013.00

* includes $8,112,688 SB 1145 revenue

Executive/Professional Standards

Booking-Release/Classification
Program Offer
60016A Booking-Release
60016B Classification
60016C Gresham Temp hold

* corrections 80%, then MCIJ 60%, MCDC 40%
* corrections 80%, then MCIJ 50%, MCDC 50%

MCIJ 70%, MCDS 30%

MCIJ 60%, MCDC 40%
MCIJ 60%, MCDC 40%
MCIJ 60%, MCDC 40%

MCIJ 60%, MCDC 40%
97% self paid thru sales

Corrections Health FY07
Dormitory costs
Program Offer
40025A
40025B
40025C
40025D
40025E
40025F
40025G
40025H
40025I
sub-total

MCDC
$2,220,912.00
$710,393.00
$277,749.00
$432,387.00
$763,806.00
$266,580.00
$266,580.00
$281,776.00
$281,776.00
$5,501,959.00

Program Offer
40028A
40028B
40028C
40028D
40028E
40028F
40028G
40028H
40028I

Inverness
$2,131,177.00
$277,745.00
$937,102.00
$107,602.00
$998,758.00
$998,776.00
$301,803.00
$301,803.00
$150,899.00
$6,205,665.00

Detention Center
Program Offer
40026 reception

MCDC
$325,888.00

Program Offer
40026

Inverness
$488,833.00

CH Mental Health
Program Offer
40059

MCDC
$637,370.00

Program Offer
40059

Inverness
$956,055.00

MCDC
-$800,000.00
$400,000.00
-$400,000.00

Program Offer
40905

Inverness
-$1,200,000.00
$600,000.00
-$600,000.00

budget cut
Program Offer
40905
Contingency

, MCDC 40% located at MCDC

MCIJ 60%, MCDC 40%

MCIJ 60%, MCDC 40%
budget note

Appendix D

Summary Corrections FY07
Adjusted based on MCSO Grand Jury Tesitmony
MCSO Jail Program

MCDC (676 beds)

Inverness (1014 beds)

comments

$30,735,277.00
$980,991.00
$390,230.50
$3,510,751.00
$28,166.00
$0.00
$35,645,415.50

$32,457,458.00
$1,238,638.00
$910,538.50
$5,266,127.00
$42,247.00
$2,319,013.00
$42,234,021.50

MCSO shows $8,112,688 SB1145,
2,388,499 commissary revenue,
1.2 million from Portland,
864,252 work crews revenue

$5,501,959.00
$325,888.00
$637,370.00
-$400,000.00
$6,065,217.00

$6,205,665.00
$488,833.00
$956,055.00
-$600,000.00
$7,050,553.00

Total Corrections Cost

$41,710,632.50

$49,284,574.50

Court Services

Courthouse
$3,899,276.00

Justice Center
$1,183,045.00

Total Cost of Corrections/Court Services

$96,077,528.00

Corrections Facility Bed Days

246,740

370,110

Cost per bed day
Average daily bed cost system wide

MCDC
$169.05
$147.52

MCIJ
$133.16

Misc. other corrections costs
Wapato Moth ball costs
Adult Recog (DCJ)

$315,929.00
$1,342,173.00

MCSO Dorm Costs
MCSO Exec/PS
Offender Transport
Booking-Release/Classification/temp hold
Inmate Commissary
MCSO Corrections Work Crews

Corrections Health Jail Program
HD Dorm Costs
HD CH Reception
CH Mental Health
CH budget reduction (net)

adjust transport 1/2
*adjust booking 1/3
* adjust to 70,413

does not include Court Services or Wapato

Appendix E

MULTNOMAH COUNTY SHERIFF’S OFFICE
501 SE HAWTHORNE BLVD., SUITE 350 • PORTLAND, OR 97214

Exemplary service for a safe, livable community

BERNIE GIUSTO
SHERIFF
503 988-4300 PHONE
503 988-4500 TTY
www.sheriff-mcso.org

MEMORANDUM

TO:

JOHN BRADLEY, SPECIAL COUNSEL TO THE DISTRICT ATTORNEY
CHUCK FRENCH, SENIOR DEPUTY DISTRICT ATTORNEY

Cc:

BERNIE GIUSTO, Sheriff

FROM:

CAPTAIN BRUCE MCCAIN, Executive Office

DATE:

October 11, 2006

SUBJECT:

MCCDA Career Sick Time Usage

As part of your inquiry into the condition and management of Multnomah County’s jails, you
asked me to replicate two studies I conducted in 1991 and 1995. The focus of those studies,
as well as the 2006 study, is the career usage of sick time by Corrections Sergeants and
Corrections Deputies. Before presenting some preliminary findings, I offer a brief explanation
of the background and methodology used, as well as a few caveats regarding conclusions that
may or may not be drawn from this data.
Background
The job classifications used in this study, Corrections Sergeants and Corrections Deputies, are
members of the Multnomah County Corrections Deputy Association (MCCDA). The accrual
and use of sick leave by these employees is governed by Article 10 of the current collective
bargaining agreement (CBA) between the county and MCCDA (see Appendix).
Unlike vacation leave that accrues at a faster rate with more seniority, sick leave is accrued by
all county employees at the same, flat rate from the date of hire until severance. All county
employees, including MCCDA members, accrue 96 hours (12 eight-hour days) of sick leave
per year. In most instances, those hours are accrued at the rate of 4.0 hours per pay period.
With 24 pay periods per calendar year, that results in 96 hours of annual sick leave accrual.
Caveats on Drawing Conclusions
This study merely reveals the number and percentage of career sick leave hours used by
corrections employees. It offers or suggests no opinion as to the cause of such usage by
individual employees or the group as a whole. However, compared to the prior studies, the
usage of career sick leave has increased substantially in the past 10-15 years.

MCCDA Sick Leave Usage
October 11, 2006
Page 2 of 15
Methodology
The 2006 study, as well as the earlier studies, used a spreadsheet formula to calculate the
estimated number of sick leave hours an employee has accrued during their career. That result
is compared against the actual number of sick leave hours the employee has retained as of a
given date ─ in this case, all data used was current as of August 31, 2006.
The formula used to estimate career accrual is as follows:
1. August 31, 2006 – [county seniority date] = total number of days employed, including
weekends and holidays.
2. Total days employed ÷ 15.208* = Total number of pay periods employed [*365 days ÷
24 = 15.208]
3. Total Pay Periods x 4.0 = Career Sick Hours Accrued.
EXAMPLE: Deputy John Doe was hired on July 4, 1990.
1. August 31, 2006 – July 4, 1990 = 5,902 days employed
2. 5,902 ÷ 15.208 = 388.08 pay periods
3. 388.08 x 4.0 = 1552.34 career hours accrued
Summary of Findings
The table below summarizes the results of the 2006 study compared to those done in 1991
and 1995 using similar formulas. The spreadsheet data used to calculate the 2006 numbers is
enclosed herewith, with employees listed by number, county seniority date, career hours
accrued, actual hours available as of 08/31/2006, percentage available and percentage used.
08/31/2006 11/24/1995 09/09/1991
Number of Records
448
351
326
Average Sick Leave Use
84.49%
76.87%
70.74%
Average Career Accrual
1226
828
633
Average Available
190
192
185
Number of 99% Users
77
33
10
Number of 90% Users
236
140
65
Total Hours Accrued
549,235
290,637
206,268
Total Hours Available
85,175
67,257
60,347

MCCDA Sick Leave Usage
October 11, 2006
Page 3 of 15
MCCDA Career Sick Leave Usage
County
Date
7/26/1999
9/11/1995
4/8/2002
3/9/1998
2/17/1987
11/27/2000
6/7/1994
11/29/2004
4/20/1998
6/1/1998
4/26/1982
4/20/1998
9/16/1996
8/23/1993
9/19/1988
7/11/1988
12/15/1986
7/1/1991
1/29/1996
11/28/1988
1/3/1978
6/6/1988
11/27/1989
4/29/1991
3/4/1996
11/9/1994
1/10/1994
7/2/1990
8/28/1995
5/28/1991
12/8/1997
8/29/1994
3/4/1996
3/19/1996
7/13/1987
8/23/1975
7/22/1991
2/6/1984
10/16/1984
3/18/1977
8/19/1985
5/2/1988
6/6/1988
8/11/1986

Accrued
682.01
1053.92
422.41
814.57
1876.64
553.13
1175.17
168.33
803.52
792.48
2339.03
803.52
956.34
1250.92
1724.09
1742.50
1893.48
1457.13
1017.10
1705.68
2753.02
1751.71
1609.94
1473.70
1007.89
1134.40
1214.10
1552.87
1057.60
1466.07
838.51
1153.34
1007.89
1003.95
1838.24
2980.27
1451.60
2167.81
2101.26
2829.56
2020.52
1760.92
1751.71
1926.62

Available

Pct Avl

0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.01
0.02
0.03
0.04
0.04
0.04
0.07
0.14
0.10
0.10
0.10
0.07
0.09
0.11
0.18
0.13
0.22
0.15
0.23
0.22
0.24
0.44
4.09
2.71
4.09
4.21
5.69
4.12
4.17
4.18
5.01

0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.00%
0.01%
0.01%
0.01%
0.01%
0.01%
0.01%
0.01%
0.01%
0.01%
0.02%
0.02%
0.02%
0.02%
0.02%
0.02%
0.14%
0.19%
0.19%
0.20%
0.20%
0.20%
0.24%
0.24%
0.26%

Pct
Used
100.00%
100.00%
100.00%
100.00%
100.00%
100.00%
100.00%
100.00%
100.00%
100.00%
100.00%
100.00%
100.00%
100.00%
100.00%
100.00%
100.00%
100.00%
100.00%
100.00%
99.99%
99.99%
99.99%
99.99%
99.99%
99.99%
99.99%
99.99%
99.99%
99.98%
99.98%
99.98%
99.98%
99.98%
99.98%
99.86%
99.81%
99.81%
99.80%
99.80%
99.80%
99.76%
99.76%
99.74%

MCCDA Sick Leave Usage
October 11, 2006
Page 4 of 15
1/29/1990
12/16/1991
12/9/1991
9/30/1996
11/27/1992
8/23/1993
5/9/1994
8/24/1988
11/17/1986
11/3/1987
1/26/1988
3/9/1998
10/6/1997
12/18/1989
5/18/1998
6/1/1998
10/15/1990
8/24/1998
1/6/1986
3/22/1999
4/19/1999
3/1/1993
8/23/1993
9/20/1993
4/12/1978
4/30/1984
4/5/1977
10/6/1995
1/29/1996
1/17/1983
2/17/1987
11/17/1997
3/23/1998
7/27/1998
2/26/1990
8/9/1999
4/26/1982
1/6/1986
6/17/1991
7/31/1989
10/29/1984
6/10/1987
7/13/1998
1/29/1990
6/2/1986
11/17/1997
6/25/1979
11/16/1997
7/13/1987

1593.37
1412.94
1414.78
952.66
1321.67
1250.92
1182.80
1730.93
1900.84
1808.52
1786.43
814.57
855.08
1604.42
796.16
792.48
1525.25
770.38
1983.69
715.15
707.79
1296.95
1250.92
1243.56
2726.99
2145.71
2824.83
1047.34
1017.10
2269.07
1876.64
844.03
810.89
777.75
1586.01
678.33
2339.03
1983.69
1460.81
1641.24
2097.84
1846.92
781.43
1593.37
1945.03
844.03
2611.52
844.29
1838.24

4.18
4.18
4.21
3.00
4.18
4.01
4.19
7.15
8.22
8.13
8.09
3.70
4.00
8.06
4.00
4.00
7.76
4.00
10.53
4.00
4.00
8.23
8.01
8.33
20.13
16.21
21.50
8.13
8.03
20.14
17.94
8.25
8.00
8.00
18.05
8.00
28.04
24.21
17.84
20.15
26.51
23.39
10.08
20.93
26.16
11.50
37.26
12.22
27.06

0.26%
0.30%
0.30%
0.31%
0.32%
0.32%
0.35%
0.41%
0.43%
0.45%
0.45%
0.45%
0.47%
0.50%
0.50%
0.50%
0.51%
0.52%
0.53%
0.56%
0.57%
0.63%
0.64%
0.67%
0.74%
0.76%
0.76%
0.78%
0.79%
0.89%
0.96%
0.98%
0.99%
1.03%
1.14%
1.18%
1.20%
1.22%
1.22%
1.23%
1.26%
1.27%
1.29%
1.31%
1.34%
1.36%
1.43%
1.45%
1.47%

99.74%
99.70%
99.70%
99.69%
99.68%
99.68%
99.65%
99.59%
99.57%
99.55%
99.55%
99.55%
99.53%
99.50%
99.50%
99.50%
99.49%
99.48%
99.47%
99.44%
99.43%
99.37%
99.36%
99.33%
99.26%
99.24%
99.24%
99.22%
99.21%
99.11%
99.04%
99.02%
99.01%
98.97%
98.86%
98.82%
98.80%
98.78%
98.78%
98.77%
98.74%
98.73%
98.71%
98.69%
98.66%
98.64%
98.57%
98.55%
98.53%

MCCDA Sick Leave Usage
October 11, 2006
Page 5 of 15
5/2/1988
5/16/1977
6/4/1998
5/28/1991
8/19/1991
3/22/1999
5/12/1986
10/26/1998
5/28/1991
1/3/1995
1/20/2002
8/22/1988
12/14/1998
10/21/1991
2/28/1995
5/18/1998
6/19/2000
3/26/1979
7/21/1982
7/2/1990
2/1/1988
4/29/1991
1/24/2000
7/9/1990
3/27/1995
8/13/1984
4/26/1982
10/11/1999
5/10/1988
4/29/1991
5/3/1999
11/17/1997
4/19/2002
7/26/1999
1/26/1998
9/27/1999
7/28/1997
10/1/1984
4/25/2005
12/17/1990
9/28/1998
4/29/1991
1/13/1998
1/5/1998
1/9/1989
5/26/1989
7/13/1987
2/22/1999
8/26/1996

1760.92
2814.05
791.69
1466.07
1444.24
715.15
1950.55
753.81
1466.07
1119.94
442.92
1731.46
740.93
1427.67
1105.21
796.16
595.48
2635.46
2316.41
1552.87
1784.85
1473.70
634.14
1551.03
1098.11
2118.10
2339.03
661.76
1758.81
1473.70
704.10
844.03
419.52
682.01
825.62
665.44
873.49
2105.21
129.67
1508.68
761.18
1473.70
829.04
831.14
1694.63
1658.60
1838.24
722.51
961.86

27.42
45.03
13.00
24.22
24.08
12.00
33.01
13.00
26.22
20.19
8.00
33.21
14.50
28.01
21.89
16.00
12.00
53.38
47.32
32.09
37.18
32.05
14.00
35.81
26.22
50.72
59.01
16.70
47.54
40.10
20.00
24.04
12.00
20.00
24.50
20.00
26.31
64.60
4.00
46.56
24.00
46.60
26.78
27.00
55.83
54.73
61.06
24.00
32.00

1.56%
1.60%
1.64%
1.65%
1.67%
1.68%
1.69%
1.72%
1.79%
1.80%
1.81%
1.92%
1.96%
1.96%
1.98%
2.01%
2.02%
2.03%
2.04%
2.07%
2.08%
2.17%
2.21%
2.31%
2.39%
2.39%
2.52%
2.52%
2.70%
2.72%
2.84%
2.85%
2.86%
2.93%
2.97%
3.01%
3.01%
3.07%
3.08%
3.09%
3.15%
3.16%
3.23%
3.25%
3.29%
3.30%
3.32%
3.32%
3.33%

98.44%
98.40%
98.36%
98.35%
98.33%
98.32%
98.31%
98.28%
98.21%
98.20%
98.19%
98.08%
98.04%
98.04%
98.02%
97.99%
97.98%
97.97%
97.96%
97.93%
97.92%
97.83%
97.79%
97.69%
97.61%
97.61%
97.48%
97.48%
97.30%
97.28%
97.16%
97.15%
97.14%
97.07%
97.03%
96.99%
96.99%
96.93%
96.92%
96.91%
96.85%
96.84%
96.77%
96.75%
96.71%
96.70%
96.68%
96.68%
96.67%

MCCDA Sick Leave Usage
October 11, 2006
Page 6 of 15
3/23/1998
9/27/1999
11/28/1988
5/15/1989
3/31/1986
4/29/1996
12/8/1997
3/6/2000
2/27/1995
8/24/1987
8/6/1990
7/22/1991
3/4/1996
9/15/1997
8/4/1997
8/25/1997
4/24/1995
2/27/1995
11/3/1986
1/5/1998
5/29/1990
1/10/1994
7/26/1999
8/25/1997
6/25/1996
6/14/2001
6/15/1998
2/19/1991
12/15/1986
10/12/1984
6/29/1998
4/25/1994
12/2/1996
4/20/1998
12/14/1998
9/27/1999
10/11/1999
2/14/1994
2/19/1991
9/14/1998
7/31/1989
5/18/1998
2/27/1995
1/28/1991
12/8/1997
10/2/1989
11/1/1993
10/24/1994
5/27/1986

810.89
665.44
1705.68
1661.49
1961.60
993.16
838.51
623.09
1105.47
1827.20
1543.66
1451.60
1007.89
860.60
871.65
866.12
1090.74
1105.47
1904.52
831.14
1561.81
1214.10
682.01
866.12
978.17
500.79
788.80
1491.85
1893.48
2102.31
785.11
1186.48
936.09
803.52
740.93
665.44
661.76
1204.89
1491.85
764.86
1641.24
796.16
1105.47
1497.63
838.51
1624.67
1232.51
1138.61
1946.61

27.00
22.25
60.82
60.04
71.19
36.21
30.75
24.00
42.60
72.74
63.21
60.21
41.97
36.12
37.11
37.75
47.59
49.10
85.16
38.50
72.78
56.86
32.00
41.00
46.32
24.00
38.00
72.11
92.95
104.12
39.00
60.04
47.94
42.00
38.86
35.50
35.50
65.13
81.86
42.00
90.65
44.00
61.10
82.84
48.06
94.07
71.56
66.31
114.72

3.33%
3.34%
3.57%
3.61%
3.63%
3.65%
3.67%
3.85%
3.85%
3.98%
4.09%
4.15%
4.16%
4.20%
4.26%
4.36%
4.36%
4.44%
4.47%
4.63%
4.66%
4.68%
4.69%
4.73%
4.74%
4.79%
4.82%
4.83%
4.91%
4.95%
4.97%
5.06%
5.12%
5.23%
5.24%
5.33%
5.36%
5.41%
5.49%
5.49%
5.52%
5.53%
5.53%
5.53%
5.73%
5.79%
5.81%
5.82%
5.89%

96.67%
96.66%
96.43%
96.39%
96.37%
96.35%
96.33%
96.15%
96.15%
96.02%
95.91%
95.85%
95.84%
95.80%
95.74%
95.64%
95.64%
95.56%
95.53%
95.37%
95.34%
95.32%
95.31%
95.27%
95.26%
95.21%
95.18%
95.17%
95.09%
95.05%
95.03%
94.94%
94.88%
94.77%
94.76%
94.67%
94.64%
94.59%
94.51%
94.51%
94.48%
94.47%
94.47%
94.47%
94.27%
94.21%
94.19%
94.18%
94.11%

MCCDA Sick Leave Usage
October 11, 2006
Page 7 of 15
3/23/1998
1/5/1998
1/27/1992
3/19/1984
8/23/1999
12/2/1996
6/1/1998
11/16/1991
11/15/1999
8/22/1988
7/2/1990
7/22/1991
5/15/1989
3/9/1983
11/28/1994
6/17/1991
12/9/1991
9/13/1999
4/6/1998
12/20/1994
8/9/1999
6/1/1998
9/26/1988
10/9/1995
11/17/1997
11/27/1989
12/18/1989
12/17/1990
7/11/1994
3/8/1999
10/1/1983
11/15/1999
2/19/1991
10/12/1987
11/17/1997
6/15/1998
3/13/1985
4/10/1987
1/26/1998
8/29/1994
4/24/1995
8/19/1991
10/12/1987
9/13/1999
5/4/1998
1/26/1998
10/25/1999
1/27/1992
7/31/1989

810.89
831.14
1401.89
2156.76
674.64
936.09
792.48
1420.83
652.55
1731.46
1552.87
1451.60
1661.49
2255.65
1129.41
1460.81
1414.78
669.12
807.21
1123.62
678.33
792.48
1722.25
1046.55
844.03
1609.94
1604.42
1508.68
1166.23
718.83
2201.47
652.55
1491.85
1814.31
844.03
788.80
2062.34
1862.97
825.62
1153.34
1090.74
1444.24
1814.31
669.12
799.84
825.62
658.07
1401.89
1641.24

48.00
49.50
83.55
130.59
41.00
56.98
48.50
87.01
40.00
107.40
98.59
94.71
110.29
150.21
76.24
99.34
101.18
48.25
58.75
84.61
51.75
60.50
132.61
81.06
65.50
125.19
128.25
127.56
100.64
63.50
195.82
59.25
136.05
166.58
78.00
73.50
193.92
176.51
80.25
113.23
107.09
141.83
178.58
66.50
79.50
83.75
67.00
146.05
171.15

5.92%
5.96%
5.96%
6.05%
6.08%
6.09%
6.12%
6.12%
6.13%
6.20%
6.35%
6.52%
6.64%
6.66%
6.75%
6.80%
7.15%
7.21%
7.28%
7.53%
7.63%
7.63%
7.70%
7.75%
7.76%
7.78%
7.99%
8.46%
8.63%
8.83%
8.89%
9.08%
9.12%
9.18%
9.24%
9.32%
9.40%
9.47%
9.72%
9.82%
9.82%
9.82%
9.84%
9.94%
9.94%
10.14%
10.18%
10.42%
10.43%

94.08%
94.04%
94.04%
93.95%
93.92%
93.91%
93.88%
93.88%
93.87%
93.80%
93.65%
93.48%
93.36%
93.34%
93.25%
93.20%
92.85%
92.79%
92.72%
92.47%
92.37%
92.37%
92.30%
92.25%
92.24%
92.22%
92.01%
91.54%
91.37%
91.17%
91.11%
90.92%
90.88%
90.82%
90.76%
90.68%
90.60%
90.53%
90.28%
90.18%
90.18%
90.18%
90.16%
90.06%
90.06%
89.86%
89.82%
89.58%
89.57%

MCCDA Sick Leave Usage
October 11, 2006
Page 8 of 15
1/28/1991
8/19/1991
1/27/1992
4/11/1994
3/9/1998
7/15/1996
2/23/1998
1/26/1998
7/8/1985
5/6/1985
3/23/1998
6/19/2000
6/17/1991
4/19/1999
2/17/1987
7/10/1995
5/17/1999
8/23/1993
10/26/1998
1/29/1996
7/12/1999
1/24/2000
7/26/1999
6/1/1992
4/13/1987
11/30/1998
7/27/1998
9/16/1996
5/4/1998
4/13/1987
3/22/1999
7/27/1998
12/9/1991
2/1/1994
3/9/1998
11/17/1997
3/15/1995
6/14/2001
2/27/1995
10/9/1995
3/23/1998
11/12/1985
6/6/1988
2/18/1986
11/30/1998
10/11/1999
10/17/1988
12/18/1995
3/22/1999

1497.63
1444.24
1401.89
1190.16
814.57
972.91
818.25
825.62
2031.56
2048.13
810.89
595.48
1460.81
707.79
1876.64
1070.49
700.42
1250.92
753.81
1017.10
685.69
634.14
682.01
1368.75
1862.18
744.61
777.75
956.34
799.84
1862.18
715.15
777.75
1414.78
1208.31
814.57
844.03
1101.26
500.79
1105.47
1046.55
810.89
1998.16
1751.71
1972.38
744.61
661.76
1716.73
1028.14
715.15

156.74
151.73
148.55
126.22
87.00
104.00
87.50
89.00
220.07
222.57
88.25
65.00
160.20
78.00
208.88
122.20
80.00
143.01
87.25
117.79
80.00
76.00
84.00
170.34
236.71
95.00
100.72
127.00
106.75
248.96
96.00
105.00
198.21
171.32
119.00
124.75
162.81
75.00
165.85
157.06
124.00
308.59
273.85
315.35
120.00
108.00
280.39
170.80
119.50

10.47%
10.51%
10.60%
10.61%
10.68%
10.69%
10.69%
10.78%
10.83%
10.87%
10.88%
10.92%
10.97%
11.02%
11.13%
11.42%
11.42%
11.43%
11.57%
11.58%
11.67%
11.98%
12.32%
12.44%
12.71%
12.76%
12.95%
13.28%
13.35%
13.37%
13.42%
13.50%
14.01%
14.18%
14.61%
14.78%
14.78%
14.98%
15.00%
15.01%
15.29%
15.44%
15.63%
15.99%
16.12%
16.32%
16.33%
16.61%
16.71%

89.53%
89.49%
89.40%
89.39%
89.32%
89.31%
89.31%
89.22%
89.17%
89.13%
89.12%
89.08%
89.03%
88.98%
88.87%
88.58%
88.58%
88.57%
88.43%
88.42%
88.33%
88.02%
87.68%
87.56%
87.29%
87.24%
87.05%
86.72%
86.65%
86.63%
86.58%
86.50%
85.99%
85.82%
85.39%
85.22%
85.22%
85.02%
85.00%
84.99%
84.71%
84.56%
84.37%
84.01%
83.88%
83.68%
83.67%
83.39%
83.29%

MCCDA Sick Leave Usage
October 11, 2006
Page 9 of 15
5/18/1987
8/20/1993
7/8/1985
6/24/1996
9/13/1999
9/14/1987
3/23/1998
10/27/1997
7/10/1995
2/13/2002
3/9/1998
6/1/1998
7/13/1998
11/19/1984
3/14/1983
5/3/1999
10/28/1996
2/27/1995
1/11/1999
12/8/1997
11/20/1990
11/17/1997
1/30/1984
11/17/1997
2/13/2002
7/7/1986
5/18/1998
2/5/1990
4/29/1996
8/25/1997
3/9/1998
6/19/2000
8/14/1989
12/19/1994
10/25/1999
6/5/1989
9/27/1999
8/27/1984
10/28/1996
8/24/1998
9/30/1996
2/9/1998
7/28/1994
8/7/2000
11/7/2003
1/29/1990
2/28/2000
8/19/1985
10/6/1997

1852.97
1251.71
2031.56
978.43
669.12
1821.67
810.89
849.55
1070.49
436.61
814.57
792.48
781.43
2092.32
2254.34
704.10
945.29
1105.47
733.56
838.51
1515.78
844.03
2169.65
844.03
436.61
1935.82
796.16
1591.53
993.16
866.12
814.57
595.48
1637.56
1123.88
658.07
1655.97
665.44
2114.41
945.29
770.38
952.66
821.94
1161.76
582.59
270.38
1593.37
624.93
2020.52
855.08

322.96
219.01
360.44
173.82
119.25
325.21
146.00
153.50
195.51
80.00
149.50
145.50
143.50
384.39
419.75
132.50
180.63
212.60
144.00
165.25
302.19
172.25
448.24
174.50
90.50
403.62
166.50
335.58
209.71
183.00
173.00
128.25
355.96
251.55
147.50
374.41
151.05
490.22
220.00
182.25
226.50
198.25
280.82
141.50
66.00
389.68
153.00
499.07
213.50

17.43%
17.50%
17.74%
17.77%
17.82%
17.85%
18.00%
18.07%
18.26%
18.32%
18.35%
18.36%
18.36%
18.37%
18.62%
18.82%
19.11%
19.23%
19.63%
19.71%
19.94%
20.41%
20.66%
20.67%
20.73%
20.85%
20.91%
21.09%
21.12%
21.13%
21.24%
21.54%
21.74%
22.38%
22.41%
22.61%
22.70%
23.18%
23.27%
23.66%
23.78%
24.12%
24.17%
24.29%
24.41%
24.46%
24.48%
24.70%
24.97%

82.57%
82.50%
82.26%
82.23%
82.18%
82.15%
82.00%
81.93%
81.74%
81.68%
81.65%
81.64%
81.64%
81.63%
81.38%
81.18%
80.89%
80.77%
80.37%
80.29%
80.06%
79.59%
79.34%
79.33%
79.27%
79.15%
79.09%
78.91%
78.88%
78.87%
78.76%
78.46%
78.26%
77.62%
77.59%
77.39%
77.30%
76.82%
76.73%
76.34%
76.22%
75.88%
75.83%
75.71%
75.59%
75.54%
75.52%
75.30%
75.03%

MCCDA Sick Leave Usage
October 11, 2006
Page 10 of 15
9/18/2000
11/27/1989
5/20/1985
11/15/1999
5/28/1991
5/15/1989
10/25/1999
5/17/1999
3/6/1984
11/28/1995
6/15/1987
11/30/1987
1/25/1999
10/30/1995
3/7/1988
5/27/1986
8/23/1999
2/7/1995
3/5/2001
8/16/1980
10/9/1989
1/25/1999
7/10/1995
4/12/1999
7/26/1999
11/17/1997
11/27/1995
3/8/1999
12/1/1980
7/16/1993
10/17/1988
4/12/1999
9/14/1987
3/23/1998
1/29/1990
1/26/1998
10/12/1987
4/24/1995
8/4/1997
3/8/1999
12/17/1990
11/25/1996
12/15/1986
4/8/1996
6/15/1998
4/7/1980
7/28/1983
4/1/1986
9/14/1987

571.54
1609.94
2044.45
652.55
1466.07
1661.49
658.07
700.42
2160.18
1033.40
1845.61
1801.42
729.88
1041.03
1775.64
1946.61
674.64
1110.73
527.35
2501.58
1622.83
729.88
1070.49
709.63
682.01
844.03
1033.67
718.83
2473.44
1260.92
1716.73
709.63
1821.67
810.89
1593.37
825.62
1814.31
1090.74
871.65
718.83
1508.68
937.93
1893.48
998.68
788.80
2536.03
2218.57
1961.34
1821.67

143.00
404.94
515.74
166.00
383.03
435.79
175.00
187.00
577.52
277.96
498.55
486.66
199.00
289.18
493.49
541.64
190.25
314.17
150.78
719.31
476.51
220.00
325.20
216.00
208.00
259.50
323.80
226.00
778.22
396.73
544.14
226.75
587.96
269.50
531.18
276.00
614.08
372.09
298.00
247.50
521.81
330.00
668.18
352.69
280.00
900.48
793.02
729.44
678.46

25.02%
25.15%
25.23%
25.44%
26.13%
26.23%
26.59%
26.70%
26.73%
26.90%
27.01%
27.02%
27.26%
27.78%
27.79%
27.82%
28.20%
28.28%
28.59%
28.75%
29.36%
30.14%
30.38%
30.44%
30.50%
30.75%
31.33%
31.44%
31.46%
31.46%
31.70%
31.95%
32.28%
33.24%
33.34%
33.43%
33.85%
34.11%
34.19%
34.43%
34.59%
35.18%
35.29%
35.32%
35.50%
35.51%
35.74%
37.19%
37.24%

74.98%
74.85%
74.77%
74.56%
73.87%
73.77%
73.41%
73.30%
73.27%
73.10%
72.99%
72.98%
72.74%
72.22%
72.21%
72.18%
71.80%
71.72%
71.41%
71.25%
70.64%
69.86%
69.62%
69.56%
69.50%
69.25%
68.67%
68.56%
68.54%
68.54%
68.30%
68.05%
67.72%
66.76%
66.66%
66.57%
66.15%
65.89%
65.81%
65.57%
65.41%
64.82%
64.71%
64.68%
64.50%
64.49%
64.26%
62.81%
62.76%

MCCDA Sick Leave Usage
October 11, 2006
Page 11 of 15
10/28/1996
1/11/1999
6/5/2000
5/17/1999
4/24/2000
8/19/1985
10/6/1997
10/31/1983
8/8/1983
10/11/1999
3/3/1998
3/18/1985
4/2/2001
7/12/1994
9/19/1988
12/12/2005
6/1/1998
3/23/1987
10/6/1997
3/14/1983
4/14/1997
10/26/1998
5/4/1998
10/31/2005
10/10/1995
4/11/1994
4/14/1986
7/10/1995
5/15/1989
6/13/2005
8/10/1987
10/18/2004
4/2/1984
5/24/1994
6/15/1987
2/23/1998
6/3/1985
1/22/2003
11/27/1995
8/23/1993
4/24/1990
8/14/1989
5/16/2005
6/14/2001
10/2/2002
9/19/1988
10/26/1998
3/14/1983
3/23/1998

945.29
733.56
599.16
700.42
610.21
2020.52
855.08
2193.58
2215.68
661.76
816.15
2061.02
519.99
1165.97
1724.09
68.91
792.48
1867.70
855.08
2254.34
901.10
753.81
799.84
79.96
1046.29
1190.16
1957.92
1070.49
1661.49
116.78
1830.88
179.38
2153.08
1178.85
1845.61
818.25
2040.77
346.40
1033.67
1250.92
1571.02
1637.56
124.15
500.79
375.85
1724.09
753.81
2254.34
810.89

356.50
277.25
233.50
276.00
242.00
823.56
366.50
940.74
958.96
288.00
368.49
934.16
236.74
533.64
792.83
32.00
378.00
900.95
423.25
1185.76
479.40
405.00
438.50
44.00
580.04
686.47
1143.13
649.70
1012.29
71.26
1127.45
111.48
1351.75
744.69
1202.05
540.16
1388.30
242.00
730.30
904.01
1146.04
1204.46
92.00
373.50
280.84
1319.52
593.50
1807.96
671.50

37.71%
37.80%
38.97%
39.40%
39.66%
40.76%
42.86%
42.89%
43.28%
43.52%
45.15%
45.33%
45.53%
45.77%
45.99%
46.44%
47.70%
48.24%
49.50%
52.60%
53.20%
53.73%
54.82%
55.03%
55.44%
57.68%
58.39%
60.69%
60.93%
61.02%
61.58%
62.15%
62.78%
63.17%
65.13%
66.01%
68.03%
69.86%
70.65%
72.27%
72.95%
73.55%
74.11%
74.58%
74.72%
76.53%
78.73%
80.20%
82.81%

62.29%
62.20%
61.03%
60.60%
60.34%
59.24%
57.14%
57.11%
56.72%
56.48%
54.85%
54.67%
54.47%
54.23%
54.01%
53.56%
52.30%
51.76%
50.50%
47.40%
46.80%
46.27%
45.18%
44.97%
44.56%
42.32%
41.61%
39.31%
39.07%
38.98%
38.42%
37.85%
37.22%
36.83%
34.87%
33.99%
31.97%
30.14%
29.35%
27.73%
27.05%
26.45%
25.89%
25.42%
25.28%
23.47%
21.27%
19.80%
17.19%

MCCDA Sick Leave Usage
October 11, 2006
Page 12 of 15
5/4/1998
1/29/1990
3/23/1998
1/6/1997
5/4/1998
5/4/1998
11/14/2005
9/19/2005
11/28/2005
3/6/2006
11/28/2005
10/31/2005
Totals
Average

799.84
1593.37
810.89
926.88
799.84
799.84
76.28
91.00
72.59
46.82
72.59
79.96

665.00
1335.49
716.00
832.00
720.00
744.00
75.50
92.00
73.50
48.00
76.00
84.00

83.14%
83.82%
88.30%
89.76%
90.02%
93.02%
98.98%
101.09%
101.25%
102.53%
104.69%
105.06%

16.86%
16.18%
11.70%
10.24%
9.98%
6.98%
1.02%
-1.09%
-1.25%
-2.53%
-4.69%
-5.06%

549,235.14
1,225.97

85,174.97
190.12

15.51%

84.49%

MCCDA Sick Leave Usage
October 11, 2006
Page 13 of 15
ARTICLE 10
SICK LEAVE
(2004-2010 Agreement)
1. Accrual. Employees shall accrue sick leave at the rate of four (4) hours for each semimonthly pay period worked, to be used in the event of his or her illness or illness of a member
of his or her immediate household. Sick leave may be accrued on an unlimited basis. Absence
due to sickness in excess of three (3) days must be verified by a physician's certificate at the
request of the County.
2. Bereavement Leave. In addition to regular sick leave, an employee shall be granted not
more than three (3) days' leave of absence with payment at the regular rate of pay for working
time missed during such three (3) day period in the event of death in the immediate family of
the employee. If such funeral is beyond 350 miles from the City of Portland, Oregon, the
employee may be granted up to three additional days of paid leave for travel. Such leave with
pay shall be for the purpose of making household adjustments or to attend funeral services.
3. Immediate Family. For purposes of the immediately preceding paragraph only, an
employee's immediate family shall be defined as spouse, domestic partner, parents, children,
grandchildren, brother, sister, grandparents, father-in-law, mother-in-law, sister-inlaw, or brother-in-law. For purposes of this section, a domestic partner’s children shall be
treated as children of the employee if, before their death, the employee legally adopted them
or they regularly lived with the employee and domestic partner for at least six (6) months
immediately prior to the death as part of a joint familial unit to which the employee regularly
contributed financial support and parental guidance. Further, the legally recognized parents
and siblings of the domestic partner shall be treated as in-law equivalents of the employee. In
the event of death involving relationships other than those set forth above, under exceptional
circumstances, a leave of absence may be granted by the Sheriff upon request.
4. Reporting of Sick Leave. Any employee who must be absent due to illness shall notify the
supervisor on duty as early as possible but no later than one (1) hour before the beginning of
his or her shift. Failure to so report may result in loss of pay for the day involved. For the
purposes of this paragraph, the beginning of shift is defined as the start of the preliminary
security briefing, if the employee is scheduled to attend such a briefing.
5. Maternity Sick Leave. The use of sick leave during pregnancy shall be subject to the same
standard as any other illness or injury under the terms of this Article except as provided by
Section 8 of this Article (Parental Sick Leave).
6. Other Sick Leave Provisions.
A. Used sick leave shall be charged on the basis of forty (40) hours per week,
or: (1) ten (10) hours per day for (4) day work week employees, or (2) eight (8) hours
per day for five (5) day work week employees.

MCCDA Sick Leave Usage
October 11, 2006
Page 14 of 15
B. Sick leave charges in excess of accrued sick leave credits may be charged against
earned and available annual leave or leave without pay at the employee's option.
Leaves without pay shall be subject to the approval of management.
C. Sick leave shall be charged to the nearest full hour.
7. Sick Leave In Application to Final Average Salary. In accordance with the terms of ORS
237.153, accumulated unused sick leave will be applied to final average salary.
8. Parental Sick Leave. During the term of a parental leave mandated by Oregon Law, the
employee on such leave may use accumulated sick leave up to twelve (12) weeks following
birth or adoption of a child. The leave may extend for the full twelve (12) weeks regardless of
parental leave taken by the other parent.
9. Sick Leave Records. The medical or psychological condition that is the employee’s reason
for his or her use of sick leave shall be considered confidential information to the extent
required by the Americans With Disabilities Act or other applicable law.
10. Saved Holiday Bonus for limited use of Sick Leave Effective July 1, 2001, employees who
have worked full time for the entire preceding fiscal year are eligible to receive saved holiday
time as a bonus incentive for low sick leave usage, as specified below:
1. Eligible employees who use no more than eight (8) hours of sick leave in a fiscal year
will receive sixteen (16) hours of personal holiday time for use after July 15 of the
following fiscal year; those who use more than eight (8) hours, but no more than sixteen
(16) hours of sick leave will receive eight hours of personal holiday time for use after
July 15 of the fiscal year.
2. Use of saved holiday bonus time will be governed by the provisions of
Article 8, Section 4, specifically to include the provision requiring use in the
same fiscal year in which it was accrued.
Effective July 1, 2003, employees who worked four (4), ten (10) hour shifts and who have
worked full time for the entire preceding fiscal year are eligible to receive saved holiday time as
a bonus incentive for low sick leave usage, as specified below:
1. Eligible employees who use no more than ten (10) hours of sick leave in a fiscal year
will receive twenty (20) hours of personal holiday time for use after July 15 of the
following fiscal year; those who use more than ten (10) hours, but no more than twenty
(20) hours of sick leave will receive eight hours of personal holiday time for use after
July 15 of the fiscal year.
2. Use of saved holiday bonus time will be governed by the provisions of
Article 8, Section 4, specifically to include the provision requiring use in the
same fiscal year in which it was accrued.

MCCDA Sick Leave Usage
October 11, 2006
Page 15 of 15
Note Regarding Sick Leave and Overtime
Until 2006, time spent on sick leave was considered “hours worked” for the computation of
overtime. In other words, if an employee worked four days (32 hours) and called in sick on
their fifth day (8 hours), they were considered to have “worked” 40 hours that week, and any
additional hours worked would be paid at the overtime rate.
Following recent negotiations between the county and MCCDA, the parties agree that effective
July 1, 2006, “Authorized work” hours for overtime eligibility “shall exclude paid leave charged
to sick use, but shall include all other paid leave” [e.g., vacation, personal holiday, comp time].

Appendix F

Overtime by Month
20,000
18,000
16,000
12,000
10,000
8,000
6,000
4,000
2,000

FY03

FY04

FY05

ne
Ju

ay
M

ril
Ap

ar
M

ua

Fe

br

ch

ry

y
ar
nu

Ja

em
be

r

r
ec
D

N

ov

em
be

er
ob

O

ct

be
r
em

pt
Se

Au

gu

st

ly

Ju

Hours

14,000

Appendix G

Department of Community Justice

MULTNOMAH COUNTY OREGON
Administrative Services
501 SE Hawthorne Boulevard, Suite 250
Portland, Oregon 97214
(503) 988-3701 phone
(503) 988-3990 fax
MEMORANDUM
TO:

Joanne Fuller, Director
Department of Community Justice
Michael D. Schrunk, District Attorney
Multnomah County
Steve Liday, Assistant Director
Department of Community Justice
Dave Koch, Assistant Director
Department of Community Justice

FROM:

Shaun Coldwell, Business Services Manager
Department of Community Justice

DATE:

September 15, 2006

RE:

CORRECTIONS GRAND JURY AND INDEPENDENT INVESTIGATION
Department of Corrections 1145 Funding
Juvenile Detention Beds

This memo is in response to the September 5, 2006 memorandum to Mike Schrunk on the above
subject. Specifically, I am providing further information to the grand jury on Department of
Corrections 1145 Funding and the daily rate for our juvenile detention beds.
Department of Corrections 1145 Funding
The allocation of State DOC funds for the 2005-2007 biennium replicated the amount budgeted for
the prior biennium, at $190 million statewide. Multnomah County’s share of that total amount was
reduced from 2003-2005, due to a shift in the proportion of offenders compared to the other counties
in the State (from 25.62% to 22.61%). The State does not provide a formula for determining the
portion of this funding that should be allocated to local control and other felony supervision offenders
within the counties. For several years, Multnomah County has based the split of DOC funding
between community supervision, local control offenders in jails, and the Local Public Safety
Coordinating Council on historical funding, which has driven a shared allocation of 65.05% to DCJ,
34.12% to MCSO and 0.83% to LPSCC.

Corrections Grand Jury and independent investigation
September 15, 2006
For the 2005-2007 biennium, the formula results in the following:
Biennium
2005-2007
State DOC Community Corrections Allocation
Multnomah County Share (Biennial)

190,221,843
22.6049%
42,982,409

Department of Community Justice
Sheriff's Office
LPSCC

27,959,842
14,664,137
358,387

DCJ
MCSO
LPSCC

2005-2006
13,700,323
7,185,427
175,610

DCJ
MCSO
LPSCC

2006-2007
14,259,519
7,478,710
182,778

DOC Actual Cost Study
The Department of Corrections has convened a work group of Oregon Association of Community
Corrections Directors (OACCD), which for the past year has been working to determine the “actual
cost” of providing supervision, sanctions, treatment and services to felony offenders on community
supervision in the State. At the same time, State DOC has been working on a parallel project with the
Oregon State Sheriff’s Association (OSSA) to determine the “actual cost” of jail beds. This
memorandum addresses the former study.
The State is projecting a proposed increase in DOC funding due to increased population and cost of
living. This would result in a higher allocation using the current formula.
However, the current formula fails to capture the costs of sanctions, services and treatment for
offenders; these costs have been included in the analysis for the Actual Cost Study. The study has
not at this time been finalized. The Governor will decide whether to include the results of the study in
his budget, and at what level. All indications are that revisions to the formula based on the Actual
Cost Study would increase funding for community corrections.
Multnomah County is actively involved with statewide committees working on this question, i.e. State
Department of Corrections, Association of Oregon Counties and the Community Corrections Funding
Committee.
Juvenile Detention Beds
The September 5 memorandum, referenced above, addressed the cost of Multnomah County juvenile
detention beds, and cited a figure of $450 per bed per day. The memo used that figure to make an
assumption of revenue loss due to renting beds to Washington and Clackamas Counties at a lesser
amount. I believe that this figure was driven by dividing the number of juvenile detention beds
currently occupied by the total cost of the program offer. Further detail work needs to be done to

2

Corrections Grand Jury and independent investigation
September 15, 2006
isolate the cost of the detention beds, since that program offer includes costs for community
alternatives to detention and facility costs for other residential programs, as well as other costs.
The chart below breaks detention costs by operations to support occupied beds, spread across 80
beds, and debt service costs to pay for construction, which are spread across all beds, both occupied
and unoccupied.
Operations of the detention facility include intake and admissions, detention staff, support staff, food,
equipment and supplies, and facilities maintenance and utilities.
The Juvenile Detention bed day rate, including administration and support costs, is $308.60 per bed
per day. That figure includes the cost of debt service on the certificates of participation, issued to pay
for facility construction1.
The Auditor’s FY 05 Service Efforts and Accomplishments Report provides data for the average cost
per bed day in detention (excludes treatment programs, adjusted for inflation):
$290.59 for FY 02
$311.23 for FY 03
$287.37 for FY 04
$268.37 for FY 05
At the time of construction of the facility, Multnomah County formed an agreement with Washington
and Clackamas Counties to provide for their use of the regional juvenile detention facility.
Clackamas County paid up front a share of capital construction for 10 beds ($750,000), so their
remaining obligations cover operations and maintenance on the facility, and detention program
operations.
Washington County is currently paying an annual amount of $256,325 for debt service payments on
the construction costs for 26 beds (10 beds at $102,716, by agreement dated November 1993 and 16
beds at $153,609, by agreement dated July 1997).
Clackamas and Washington Counties currently contract for beds in the detention facility at 14 beds
each. The current lease agreement for both Washington and Clackamas counties is for $132.19 for
the first 10 beds and $191.00 for the next 4 beds.
When debt service costs are taken out of the formula, the per bed per day operating rate is $291.70.

1

The debt service cost of $996,923 is spread over the total 191 available beds in the facility. Current
occupied bed level is 80 beds, not including residential treatment.

3

Corrections Grand Jury and independent investigation
September 15, 2006
Juvenile Detention Beds Analysis
Total Cost of Detention Operations (excl debt)
Admin cost on above 4.78%
Support cost on above 13.46%

7,203,599
344,332
969,604

Cost including Admin and Support

8,517,535

Per Bed Cost @ 80 Beds (annual)
Per Bed Cost @ 80 Beds (daily)

$

Cost of Debt Service (spread over 191 beds)
Admin cost on above 4.78%
Support cost on above 13.46%

106,469
291.70
996,923
47,653
134,186

Cost including Admin and Support

1,178,762

Per Bed Cost @ 191 Beds (annual)
Per Bed Cost @ 191 Beds (daily)

$

6,172
16.91

Total Bed Cost per day

$

308.60

*General Population, not including RAD or SRTP, not including community shelter beds

4

 

 

Prisoner Education Guide side
Advertise here
The Habeas Citebook: Prosecutorial Misconduct Side