Courthouse Construction: Sufficient Data and Analysis Would Help Resolve
the Courtroom-Sharing Issue (Letter Report, 12/14/2000, GAO/GAO-01-70).

The judiciary is in the midst of a multibillion-dollar courthouse
construction program. New courthouses are being constructed to
accommodate new judgeships created because of increasing caseloads and
to replace obsolete courthouses occupied by existing judges. For many
years there has been a debate about whether district judges could share
courtrooms to save taxpayer dollars without compromising effective
judicial administration. In 1997 GAO issued a report calling for better
courtroom use data and analysis to enhance facility planning.In
response, the Administrative Office of the U.S. Courts (AOUSC),
contracted with Ernst & Young to conduct a study of the judiciary's
facilities program in 1999. As part of the study AOUSC asked Ernst &
Young to conduct a thorough analysis of courtroom utilization,
assignment, and sharing by judges. GAO was then asked to review the
report to determine whether the study provided sufficient data and
analysis to show if courtroom sharing would be feasible. GAO found that
the Ernst & Young study did not provide the type of data and analysis
that GAO and other research organizations such as the Rand Institute for
Civil Justice and the Federal Judicial Center have determined would be
needed to help resolve the courtroom-sharing issue.

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  GAO-01-70
     TITLE:  Courthouse Construction: Sufficient Data and Analysis
	     Would Help Resolve the Courtroom-Sharing Issue
      DATE:  12/14/2000
   SUBJECT:  Government facility construction
	     Federal courts
	     Data collection
	     Judges
	     Trials
IDENTIFIER:  Judiciary Space and Facilities program

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Testimony.                                               **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
******************************************************************

GAO-01-70

GAO United States General Accounting Office

Report to Congressional Committees

December 2000 COURTHOUSE CONSTRUCTION

Sufficient Data and Analysis Would Help Resolve the Courtroom- Sharing Issue

GAO- 01- 70

Page 1 GAO- 01- 70 Courtroom Sharing

Contents Letter 3

Results in Brief 7 Background 9 Limited Data and Analysis Leave the
Courtroom- Sharing Issue

Unresolved 11 EY'S Analysis of Opportunities for Sharing was Problematic 16
Conclusions 20 Recommendations for Executive Action 21 Matter for
Congressional Consideration 21 EY and Judiciary Comments and Our Evaluation
22 EY Comments 22 AOC Comments 25 Comments by the Chair, Judicial Conference
Committee on

Security and Facilities 34 Appendixes Appendix I: Objective, Scope, and
Methodology 38

Appendix II: Comments From Ernst & Young 40 Appendix III: Comments From the
Administrative Office of the

U. S. Courts 43 Appendix IV: Comments From the Chair, Judicial Conference,

Committee on Security and Facilities 57

Page 2 GAO- 01- 70 Courtroom Sharing

Page 3 GAO- 01- 70 Courtroom Sharing United States General Accounting Office

Washington, D. C. 20548 Page 3 GAO- 01- 70 Courtroom Sharing

Lett er

December 14, 2000 The Honorable Ben Nighthorse Campbell Chairman The
Honorable Byron L. Dorgan Ranking Minority Member Subcommittee on Treasury
and General Government Committee on Appropriations United States Senate

The Honorable C. W. Bill Young Chairman The Honorable David Obey Ranking
Minority Member Committee on Appropriations House of Representatives

The Honorable Bud Shuster Chairman The Honorable James L. Oberstar Ranking
Democratic Member Committee on Transportation and Infrastructure House of
Representatives

The Honorable Bob Franks Chairman The Honorable Bob Wise Ranking Democratic
Member Subcommittee on Economic Development, Public Buildings, Hazardous
Materials, and Pipeline Safety Committee on Transportation and
Infrastructure House of Representatives

This report responds to Senate and House reports 1 that direct us to review
and comment on part of the May 2000 Ernst & Young (EY) study on the 1 S.
Rep. No. 106- 500, at 66 (2000); H. R. Rep. No. 106- 756, at 65 (2000).

Page 4 GAO- 01- 70 Courtroom Sharing

judiciary's space and facilities program. 2 Specifically, we were asked to
review the study as it relates to courtroom use and sharing. The House
report directed us to report our results to the House Committee on
Appropriations and the House Committee on Transportation and Infrastructure.
The Senate report directed us to provide our results to Congress. We are
providing our results to the Chairman and Ranking

Member of the Senate Appropriations Subcommittee on Treasury and General
Government to fulfill the Senate report requirement. The Chairman and
Ranking Democratic Member of the House Subcommittee on Economic Development,
Public Buildings, Hazardous Materials, and Pipeline Safety also asked us to
review the study as it pertains to courtroom use and sharing.

The judiciary is in the midst of a multibillion- dollar courthouse
construction initiative. New courthouses are being built to accommodate new
judgeships created because of increasing caseloads and to replace

obsolete courthouses occupied by existing judges. For several years, there
has been much debate about whether district judges could share courtrooms-
operate in a courthouse with fewer courtrooms than judges- to save taxpayer
dollars without compromising effective judicial administration. There has
been a belief among certain key stakeholders outside the judiciary- various
subcommittees and Members of Congress as

well as the Office of Management and Budget (OMB)- that courtroom sharing
may be possible and could lead to cost savings. The judiciary has instituted
a policy for sharing courtrooms among visiting judges- judges

from other locations who temporarily use courtrooms- and senior judges who
have reduced caseloads. 3 However, the judiciary and other key stakeholders
believe that the judiciary should retain its one- judge, onecourtroom policy
for active district judges because of the negative effects courtroom sharing
may have on effective judicial administration. Nonetheless, many
stakeholders and some organizations that have done research in the area have
recognized that existing data and analysis on courtroom use were limited and
could not resolve the courtroom- sharing debate, and more data and analysis
were needed.

2 Independent Assessment of the Judiciary's Space and Facilities Program,
Ernst & Young, May 2000. 3 The judiciary has two categories of district
judges who hear cases and use courtrooms. Active district judges carry full
caseloads. District judges with senior status, who we refer to in this
report as senior judges, have resigned from their active judgeships, but
continue to

carry a full or partial caseload.

Page 5 GAO- 01- 70 Courtroom Sharing

In 1997, we issued a report that called for better courtroom use data and
analysis to enhance facility planning and decisionmaking. 4 In that report,
we recognized that the process for administering justice is dynamic and
complex and that the availability of a trial courtroom is an integral part
of the judicial process because judges need the flexibility to resolve cases

efficiently. However, we also noted that trial courtrooms, because of their
size and configuration, are expensive to construct- EY estimated in its
study that a typical courtroom costs about $1. 5 million. In doing the work
for the 1997 report, we found that the judiciary had not compiled data on
how often and for what purposes courtrooms were actually used and did not
have analytically based criteria for determining how many and what types of
courtrooms were needed to effectively administer justice. Given this, we
analyzed various detailed records and documents at seven geographically
dispersed court locations- Dallas, TX; Miami, FL;

Albuquerque, Santa Fe, and Las Cruces, NM; San Diego, CA; and Washington, D.
C.- to develop data on actual courtroom use. Our 1997 analysis of actual
courtroom use for trials and nontrial activities 5 at these locations
suggested there may be opportunities to reduce costs by

building fewer full- sized trial courtrooms. Although our results were not
generalizable to all district courtrooms, our analysis showed that during
1995, courtrooms on average were used for trials less than one- third of 250
federal workdays. Active district judges used their courtrooms, on average,
about 65 percent of the days for trial and nontrial purposes; senior judges
used their courtrooms only about 38 percent of the days. We also noted that
for a significant number of days, the courtrooms were used for only 2 hours
or less, and at least one courtroom in each of two courthouses was unused

every workday of the year. However, in reporting these results, we
recognized that our data collection and analysis effort was limited in scope
in that we visited only seven locations and collected data for only a 1-
year period. Because of the lack of data, we were also unable to reflect
other factors in our analysis, such as latent use- using an available
courtroom and the scheduling of that courtroom as leverage to encourage case
settlements- and scheduling uncertainties that the judiciary believes are
important in determining the number of courtrooms needed. These 4 Courthouse
Construction: Better Courtroom Use Data Could Enhance Facility Planning and
Decisionmaking (GAO/ GGD- 97- 39, May 19, 1997).

5 Nontrial activities include events such as arraignments and motion
hearings.

Page 6 GAO- 01- 70 Courtroom Sharing

limitations prevented us from conclusively determining whether and to what
extent courtroom sharing may be feasible.

Consequently, we recommended that the judiciary design and implement cost-
effective research to fully examine the courtroom use issue. Specifically,
we recommended that as part of this research, the judiciary

should (1) establish criteria for determining effective courtroom
utilization and a mechanism for collecting and analyzing data at a
representative number of locations so that trends can be identified over
time and better

insights obtained on court activity and courtroom usage; and (2) design and
implement a methodology for capturing and analyzing data on latent use,
courtroom scheduling, and other factors that may substantially affect the

relationship between the availability of courtrooms and judges' ability to
effectively administer justice. The development of actual courtroom use data
may be time- consuming and difficult to pursue because of the lack of
readily available court records or other mechanisms for capturing
information on how courtrooms were used. However, given the results of our
1997 courtroom use report and the significant taxpayer dollars being

invested to replace obsolete facilities and accommodate new judgeships, this
research would be worth pursuing because significant savings could result if
fewer million- dollar trial courtrooms need to be constructed. Further,
without actual courtroom use data that could be used to analyze

the potential for sharing, the debate over sharing will continue to be based
on judgment that is not, in our view, supported by a methodologically sound
analysis of empirical evidence.

In commenting on our 1997 report, the Administrative Office of the U. S.
Courts (AOC), which is the judiciary's administrative arm, said the
judiciary should aggressively monitor both the effects of its ongoing
efforts to reduce space usage and its newly adopted initiatives related to
courtroom sharing among senior and visiting judges before it embarks on an
extensive and time consuming study of courtroom use. AOC went on to say that
the

continuing study of the need for courtrooms should be conducted by those who
are most knowledgeable in this area, that is, federal judges. 6 In 1999, to
address the courtroom- sharing issue and identify ways to improve its space
and facilities efforts, AOC contracted with EY to conduct a study of the
judiciary's facilities program. As part of the study, AOC asked EY to
conduct a thorough analysis of courtroom utilization, assignment, and

sharing by judges. As agreed with your offices, our objective was to 6 AOC's
comments and our evaluation of them are contained in our 1997 report.

Page 7 GAO- 01- 70 Courtroom Sharing

determine whether the EY study provided sufficient data and analysis to show
if, and to what extent, courtroom sharing may be feasible. To meet this
objective, we obtained the EY study and analyzed those areas that pertained
to courtroom use and sharing and discussed our analysis with AOC officials
and the EY staff who worked on the study. We also discussed the study with
representatives from OMB and the Congressional Budget Office (CBO). In
addition, we reviewed previous studies on or related to courtroom use as
well as AOC's contract file for the study. We did our work between June and
September 2000 in accordance with generally accepted government auditing
standards. A more detailed description of our objective, scope and
methodology is in appendix I. We received written

comments on a draft of this report from EY, AOC, and the Chair of the
Judicial Conference's Committee on Security and Facilities. These comments
are discussed near the end of this letter.

Results in Brief The EY study, as it pertained to the need for courtrooms,
was informative in that it provided current information on issues related to
courtroom use and the potential for sharing courtrooms. However, the EY
study was not designed to provide the type of data and analysis we and other
research organizations such as the Rand Institute for Civil Justice (Rand)
and the Federal Judicial Center (FJC), the judiciary's research arm, have
determined would be needed to help resolve the courtroom- sharing issue.

AOC did not specifically require EY to develop these data and analysis; and,
according to EY, to do so would have involved a level of resource commitment
and time frames that were beyond the scope of its work. According to EY,
such a data collection effort would have been extensive because court
records do not adequately track courtroom usage; the frequency of
proceedings; or uncertainties, such as delays and cancellations. In short,
there continues to be a lack of actual courtroom usage data and analysis
that show (1) how often and for what purposes

courtrooms are being used and (2) the impact that other factors- such as
courtroom scheduling uncertainties and latent use- may have in determining
the need for courtrooms.

According to our work and assessments by these other research organizations,
a cost- effective, empirical assessment that would generate actual courtroom
use data is key to informed decisionmaking about the feasibility of
courtroom sharing. Without these data, it is not possible to determine more
conclusively whether courtroom- sharing opportunities exist. It is important
that expert judgment and experience be used in interpreting empirical data
on courtroom use. However, it is also important

Page 8 GAO- 01- 70 Courtroom Sharing

that such judgment supplement, not serve as a substitute for, empirical data
and analysis on courtroom use. Given the significant taxpayer dollars being
invested to replace obsolete courthouse facilities and accommodate new
judgeships that are projected, it appears that the judiciary, Congress, and
the administration would be well served by a methodologically sound,
empirical study of the courtroom- sharing issue. Although the EY study did
not provide any new courtroom use data or perform the type of quantitative
analysis we and others have

recommended, it contained conclusions and recommendations that were based on
an analysis of the potential for courtroom sharing. Our review showed that
this analysis was problematic. The analysis, which suggested that five
active judges cannot share four courtrooms, was based on a mathematical
formula that (1) used a questionable flexibility factor to account for
various uncertainties that may affect a courtroom scheduling process, such
as whether a trial will actually take place and how long a trial

will last, and (2) incorrectly used courtroom usage data from our 1997
report. Although the study identified several uncertainties that may affect
courtroom scheduling, it provided no data, rationale, or analytical basis to
support the 20 to 25 percent flexibility factor it used in its formula other

than to say it was appropriate. According to EY staff, the 20 to 25 percent
flexibility factor was based on the professional judgment of the staff doing
the study, because there were limited empirical data available on how
various uncertainties actually affect the need for a courtroom.

Furthermore, the flexibility factor could be viewed as excessive,
considering that it was used in conjunction with a courtroom use measure we
developed for our 1997 report that already contained a degree of flexibility
that may have accommodated various uncertainties. Moreover,

the formula incorrectly used our courtroom usage measure as a lights- on
measure- the number of hours judges spent in the courtroom- when it was
actually a measure of workdays when there was any use at all, even if the
events lasted for less than an hour. Using our data as a lights- on

measure overstated the number of hours judges actually spent in the
courtrooms. It also did not recognize that a significant number of the
workdays had events that lasted only 2 hours or less, leaving sufficient
time to possibly accommodate unanticipated events. Both of these limitations

raised questions about the merits of the analysis and the study's
conclusions and recommendations regarding courtroom sharing. This report
contains recommendations to AOC that address the problems we identified. In
commenting on a draft of this report, AOC had several

Page 9 GAO- 01- 70 Courtroom Sharing

serious concerns with the draft and did not agree with our recommendations.
However, AOC did not provide any data, analysis, or rationale that would
give us an adequate basis for changing or dropping our recommendations.
Accordingly, we continue to believe that our recommendations are
appropriate. Furthermore, given AOC's reluctance to

implement our recommendations and the potential for savings if fewer trial
courtrooms were built, we believe Congress should consider requiring AOC to
provide persuasive courtroom use data and analysis, along with its views, to
justify the number of courtrooms being requested in future courthouse
construction projects before funding is approved.

Background The judiciary is in the midst of a multibillion- dollar
courthouse construction initiative. The judiciary's most recent 5- year
construction plan

proposed 50 new courthouse projects that are estimated to cost about $3
billion. The General Services Administration (GSA) is responsible for
building these facilities with input from various stakeholders, including
Congress, OMB, and the judiciary. The Judicial Conference of the United
States is the policymaking body for the judiciary. Currently, the judiciary
follows a practice that each active district judge is assigned a dedicated
courtroom, although the judiciary has taken steps to implement courtroom

sharing among visiting judges and senior judges who have reduced caseloads.
Over the years, there has been considerable debate about whether district
judges should share courtrooms in new courthouse facilities to save taxpayer
dollars- according to EY, each courtroom costs about $1. 5 million.

The debate over courtroom sharing has revolved around whether it would
negatively affect the judiciary's efficiency and effectiveness. As
previously mentioned, in 1997, we reported that our analysis of actual
courtroom use

for trial and nontrial activities at seven locations suggested there may be
opportunities to reduce costs by building fewer full- sized trial courtrooms
in the judiciary's multibillion- dollar courthouse construction initiative.
7 Our

report recommended that the judiciary fully examine the courtroom use issue
to form a better basis for determining the number and types of 7 At the
request of a House Subcommittee, we also issued a companion report-
Courthouse Construction: Information on the Use of District Courtrooms At
Selected Locations (GAO/ GGD- 97- 59R, May 19, 1997)- that provided data on
courtroom use at four additional locations. Our findings on courtroom use at
these additional locations were similar to our findings associated with the
initial seven locations previously discussed.

Page 10 GAO- 01- 70 Courtroom Sharing

courtrooms needed and whether each district judge needs a dedicated
courtroom. Before our report was issued, Rand and FJC had made similar
recommendations. In commenting on our 1997 report, AOC requested that we
recast the recommendation. Instead of recommending what it thought would be
a time- consuming and expensive study of the courtroom usage

issue, AOC requested that we recommend that the judiciary monitor the
implementation of its policy initiatives on courtroom sharing among senior
and visiting judges and facilities planning with regard to the impact on
case management and effectiveness in contributing to its efforts to reduce
space costs. We chose not to recast our recommendation as AOC requested; it
was our view that just monitoring these initiatives would be an incomplete
basis for courtroom construction decisions because it would not include

information and analysis on actual courtroom usage. Without actual courtroom
use data, there would always be questions about how many fullsized trial
courtrooms are really needed. There is a belief among certain key
stakeholders outside the judiciary- various subcommittees and Members of
Congress as well as OMB- that courtroom sharing may be possible and could
lead to cost savings. According to AOC officials, the judiciary and other
stakeholders, including certain Members of Congress and several U. S.
Attorneys, believe that the

complexities associated with courtroom availability and effective judicial
administration reduce the likelihood that sharing is feasible. The
judiciary's position is that some sharing may be possible among visiting
judges and senior district judges with reduced caseloads, but active
district judges and senior judges with full caseloads need dedicated
courtrooms. According to AOC officials, the judiciary and others believe
that in addition to trial and

nontrial activities, such as motion hearings and arraignments, other factors
of uncertainty affect the need for courtrooms and make courtroom sharing
more of a challenge. These factors include whether a case will go to trial,
how long a trial will take, and whether an emergency proceeding will require
immediate courtroom use. In addition, latent use would also have to be
considered in determining the need for courtrooms. However, there currently
is limited information to determine how often these factors may affect the
need for a courtroom and the degree to which they may actually impede
courtroom sharing. As we have reported in the past, the judiciary

lacks data and analysis on courtroom use and, is therefore not in a good
position to support its practice of providing a trial courtroom for every
judge.

Page 11 GAO- 01- 70 Courtroom Sharing

In 1999, in an effort to address the courtroom- sharing issue and improve
its space and facilities efforts, AOC contracted with EY to conduct a
comprehensive review of its facilities program. With regard to courtroom
sharing, AOC directed EY to conduct a thorough analysis of courtroom
utilization, assignment, and sharing by judges. The EY study, among other
things, concluded that courtroom sharing (1) would not be feasible in
courthouses with fewer than 5 district courtrooms and (2) would result in
significant scheduling problems in courthouses with 6 to 10 district
courtrooms. The study also recommended that the judiciary retain its policy
of providing a courtroom for every active district judge; however, it said
that sharing at large courthouses- those with more than 10 district judges-
may be possible and that senior judges can share courtrooms after

the first 2 years of senior status. Limited Data and Analysis Leave the
Courtroom- Sharing Issue Unresolved The EY study was informative in that it
provided current information on

issues related to courtroom use and the potential for sharing courtrooms. In
doing their work, EY staff, among other things, visited federal courthouses,
interviewed key stakeholders and judiciary personnel, and

conducted focus groups with judges. A top AOC official who was involved with
contracting for the study pointed out that as part of its work, EY visited
14 court locations, observed courtroom use and spoke with users, examined
calendars, analyzed statistical data and courthouse planning documents,
conducted interviews with dozens of individuals who had different
experiences and views, and held focus groups.

Nevertheless, the EY study did not provide the type of research we and other
research organizations, such as Rand and FJC, have said would be needed to
help resolve the courtroom- sharing issue. The EY study clearly stated that
EY did not attempt to collect new data on the use of courtrooms because, to
be nationally representative, such a data collection effort would have
involved a research commitment and a time frame that were beyond the scope
of the study. The study also said that Rand concluded that fully researching
the impact of courtroom sharing would require a 2- to 3- year study period.
Through discussion of the study with AOC and EY staff and a review of the
contract file for the study, we found that AOC did not specifically require
EY to develop the type of data and analysis we and

other organizations have recommended. According to a top AOC official
involved with contracting for the study, AOC was under pressure from OMB and
Congress to have the study completed within a year. According to this
official, this time frame and budget restraints prevented AOC from

Page 12 GAO- 01- 70 Courtroom Sharing

having EY develop the type of data and analysis we and other organizations
have recommended.

In 1997, we recommended that the judiciary design and implement
costeffective research to fully examine the courtroom use issue to form a
better basis for determining the number and type of courtrooms needed as
well as whether each district judge needs a dedicated courtroom. We reported
that this effort should include

establishing criteria for determining effective courtroom utilization and a
mechanism for collecting and analyzing data at a representative number of
locations so that trends can be identified over time and better insights
obtained on court activity and courtroom use;

designing and implementing a methodology for capturing and analyzing data on
latent use, courtroom scheduling, and other factors that may substantially
affect the relationship between the availability of courtrooms and judges'
ability to effectively administer justice;

using these data and criteria to explore whether the one- judge,
onecourtroom practice is needed to promote efficient courtroom management or
whether other courtroom assignment alternatives exist; and establishing an
action plan with time frames for implementing and

overseeing these efforts. The courtroom use data we developed for our 1997
report, although not generalizable to all federal district courtrooms,
suggested that there may be opportunities for the judiciary to reduce costs
by building fewer trial courtrooms. We reported that opportunities to reduce
costs would depend on the potential impact or benefits and costs of options,
such as instituting

courtroom- sharing practices; changing the configuration of courtrooms by
building a mix of full- size and smaller, less expensive courtrooms or
hearing rooms; or holding meetings or proceedings in facilities other than

trial courtrooms, possibly by using audio or video technology. Research
prior to the issuance of our report also called for better data and analysis
in this area. In March 1996, a study commissioned by AOC was released that
used quantitative methods to recommend that the judiciary

and GSA continue to build one courtroom for every active district judge. 8
The Judicial Conference's Committee on Court Administration and Case
Management asked FJC, the judiciary's research arm, to critique this study.
In August 1996, FJC praised the report for pointing out some of the limits
of current data and the complexities of dealing with matters such as

Page 13 GAO- 01- 70 Courtroom Sharing

courtroom scheduling. 9 However, FJC concluded that the “limitations
of the analysis, some of which are acknowledged in the report, substantially
limit its value as a basis for any policy decisions.” FJC concluded
that the findings and recommendations went beyond the data presented and
that other more useful techniques might have been developed.

Rand also expressed concern about the March 1996 AOC- commissioned report.
In September 1996, Rand issued a project memorandum, prepared under contract
with AOC, that reviewed available research on courtroom

sharing. 10 Rand found that previous research had been limited and did not
resolve the courtroom- sharing issue. Similar to FJC, Rand questioned
whether the March 1996 report explored various analytical techniques that
could be applied to the courtroom- sharing concept. These techniques
included advanced computer simulations using detailed, actual data on

how courtrooms are used. Rand also questioned the assumption in the report
that an additional scheduler would have to be employed by the judiciary in
courthouses operating under a sharing scenario. Most

importantly, Rand emphasized the need for further study on courtroom use
issues, stressing the need for the judiciary to understand the effects of
courtroom sharing on the judicial system when making facility decisions.
Rand concluded the following:

“Making decisions without such an understanding presents two kinds of
risks. On the one hand, reducing the courtroom- per- judge ratio may
unacceptably impair the ability of the federal court system to meet its
judicial obligations and may have other potentially negative effects. On the
other hand, not reducing the ratio may forego an opportunity to save
taxpayer dollars.”

Rand suggested that the judiciary, Congress, AOC, and GSA would be well
served by a methodologically sound, empirical study that would require
investigating the effects of varying the courtroom- to- judge ratio. A more
detailed description of the March 1996 AOC- commissioned study, the FJC

8 Edward H. Leekley and William T. Rule II, The Impact of Providing Fewer
Than One Courtroom Per Judgeship: Report to the Committee on Security,
Space, and Facilities of the Judicial Conference of the United States,
Washington, D. C., March 1996. 9 Federal Judicial Center Research Note on
The Impact of Providing Fewer Than One Courtroom Per Judgeship, Federal
Judicial Center, August 28, 1996. 10 Terence Dunworth and James S. Kakalik,
Research on Courtroom Sharing, Project Memorandum, Rand Institute for Civil
Justice, PM- 598- 1- ICJ, September 1996.

Page 14 GAO- 01- 70 Courtroom Sharing

critique, and the Rand report is contained in our May 1997 report on
courtroom use. 11

As previously mentioned, AOC did not require EY to carry out the research we
and others recommended. The EY staff we interviewed said that such an effort
would have involved much more time and resources than they had allotted to
the subject, which was part of a broader study they were tasked

with completing. Given this, there continues to be a lack of actual
courtroom usage data that show (1) how often and for what purposes
courtrooms are being used and (2) the impact that other factors- such as
courtroom scheduling uncertainties and latent use- may have in determining
the need for courtrooms. Although expert judgment should be considered in
determining the merits of courtroom sharing, it is clear to us that the
courtroom- sharing issue is not going to be resolved without better data and
analysis on actual courtroom use.

Terence Dunworth, a co- author of the 1996 Rand study and consultant to AOC
on courtroom use issues, agreed that the existing research did not resolve
the issue of whether courtroom sharing can take place without adverse
consequences. Dunworth said that the judiciary still needs to invest in an
empirical assessment of the courtroom- sharing issue;

otherwise, decisions will continue to be made on the basis of opinion and
judgment. Given the results of our 1997 report on courtroom use and the
significant taxpayer dollars that will continue to be invested to replace
obsolete facilities and accommodate growth, it is our view that the

judiciary, Congress, and the administration would be well served by a
methodologically sound analysis of the courtroom- sharing issue that is
based on empirical evidence.

This type of study may be time consuming and involve an extensive effort to
develop actual courtroom use data because of the lack of readily available
court records or other mechanisms for capturing information on how
courtrooms were used. In addition, once these data are collected, the study
would have to carefully consider the type of analytical techniques to be
used. For example, in EY's opinion, it would not be possible to use the

regression analysis technique because of the difficulties of predicting
trial length. However, as Rand indicated, advanced forms of computer
simulations show the best potential for exploring the feasibility of overall
courtroom sharing. Despite these challenges, this research would be worth

11 GAO/ GGD- 97- 39.

Page 15 GAO- 01- 70 Courtroom Sharing

pursuing because significant savings could result if fewer million- dollar
trial courtrooms need to be constructed now and in the future. Further,
without actual courtroom use data that could be used to analyze the
potential for sharing, the debate over sharing will continue to be based on
speculation rather than on a methodologically sound analysis of empirical
evidence.

It is important to recognize that AOC has taken some actions that relate to
courtroom sharing, such as the EY study, policy changes made by the
judiciary to consider sharing among senior judges with reduced caseloads,
and changes made to the judiciary's courthouse construction design guide to
consider sharing opportunities. Regarding sharing by senior judges, each
federal circuit now has some type of policy to encourage courtroom sharing
among senior judges on the basis of criteria established by each

circuit. These criteria include workload, the number of years judges are
expected to continue working, and an evaluation of how courtrooms are used
in existing facilities. According to the EY study, 38 of the proposed

projects in the judiciary's 5- year construction plan that EY analyzed
anticipate courtroom sharing, with 274 courtrooms planned for 347 judges.
With the exception of one of these projects, this sharing involves only
senior and visiting judges and not active district judges. Nonetheless, we
view these initiatives as steps in the right direction. However, despite
these

steps and AOC's and EY's efforts to contact various stakeholders during the
study on the judiciary's space and facilities program, the judiciary has not
reached up- front agreement with key stakeholders on what type of
costeffective research could be pursued- including development of study
objectives, potential methodologies, and reasonable approaches- that would
help resolve the debate over the courtroom- sharing issue and identify the
full potential of courtroom sharing. We made recommendations to the
judiciary in this report to do the needed research to help resolve the
courtroom- sharing issue. In commenting on a

draft of this report, AOC disagreed with the recommendations because it
believes that implementing them would be costly and unproductive. However,
AOC provided no support for its assertions and offered no other options,
other that to say that the judiciary has no interest in constructing
courtrooms that are not needed. Given AOC's reluctance to do the needed
research and the potential savings that could be derived from building fewer
expensive trial courtrooms, Congress should consider requiring AOC to
provide persuasive courtroom use data and analysis, along with its views, to
justify the number of courtrooms being requested for each proposed
courthouse construction project before funding is approved.

Page 16 GAO- 01- 70 Courtroom Sharing

The actual courtroom use data and analysis could supplement the standardized
courtroom utilization studies that are currently required by the annual
appropriations acts that fund courthouse construction projects. Since fiscal
year 1997, appropriations acts providing funding for courthouse construction
have contained this requirement. 12 However, the

utilization studies provided were general in nature and were primarily
limited to identifying courtroom assignments for judges. They did not
contain actual data and analysis on how often and for what purposes
courtrooms were being used or the impact that other factors- such as
courtroom scheduling uncertainties or latent use- may have on determining
the need for courtrooms.

EY'S Analysis of Opportunities for Sharing was Problematic

Although the EY study did not provide new courtroom use data or perform the
type of quantitative analysis we and others have recommended, it contained
conclusions and recommendations that were based on an

analysis of the potential for courtroom sharing. Our review of this analysis
showed that it was problematic. The analysis was based on a mathematical
formula- which EY acknowledged was simplistic in the study- that used a
questionable flexibility factor to account for various uncertainties that
may affect courtroom scheduling and incorrectly used data from our 1997
report on courtroom use. Both limitations raised questions about the merits
of the analysis and the study's (1) conclusions that courtroom sharing would
not be feasible in courthouses with fewer than 5 district courtrooms and
would result in significant scheduling problems in

courthouses with 6 to 10 district courtrooms; and (2) recommendations that
the judiciary should retain the 1- judge, 1- courtroom policy for active
district judges and provide 1 courtroom for every 2 senior judges.

More specifically, in using its formula to conclude that five judges could
not share four courtrooms, EY used a flexibility factor of 20 to 25 percent
to account for various uncertainties that may affect courtroom scheduling,
such as whether a trial will actually take place, how long a trial will
last, and whether an emergency proceeding will require immediate courtroom
use. However, the study was silent on how the flexibility factor was
determined. The study provided no data, rationale, or analytical basis to
support the 20 to 25 percent flexibility factor, saying only that EY decided
that the estimates were appropriate. Thus, there is no link between the

12 See Treasury and General Government Appropriations Act for Fiscal Year
2000, Public Law No. 106- 58, sect.404, 113 stat. 430, 453 (1999) for this
provision.

Page 17 GAO- 01- 70 Courtroom Sharing

basis for EY's judgment- stakeholder input and observations- and the
specific percentages that EY decided to use. EY staff said that they
developed the estimate using their professional judgment based on the

work they did on the study, including discussions with judges and other
judiciary officials, because there were limited empirical data available on
how various uncertainties actually affect the need for a courtroom. Also, it
is interesting to note that if a 15 percent flexibility factor were used in
the formula, the results suggest that five judges could share four
courtrooms. 13 Furthermore, as discussed below, the 20 to 25 percent
flexibility factor

could be viewed as excessive considering that it was used in the formula in
conjunction with a courtroom use measure that already contained a degree of
flexibility that may have accommodated various uncertainties. Without

better data on the impact that these uncertainties may have on the need for
a courtroom, it is difficult to gauge whether they are obstacles to the
potential for sharing.

In addition to our concerns with the flexibility factor, we also noted that
EY incorrectly used the courtroom use data from our 1997 report. EY's
mathematical formula used the 65 percent courtroom use measure for active
district judges as an actual, lights- on courtroom use measure- the number
of hours judges spent in the courtrooms- when in fact it was a percentage of
workdays that the courtroom was used for any activity. As pointed out in our
report, if courtroom events took less than an hour and the courtroom was
unused for the rest of the day, we credited the courtroom with a full day of
use for that day. We did not pursue a lights- on measure because, at the
time of our review, court records did not allow us to determine the exact
number of hours courtrooms were actually used for trials. However, we were
able to determine the number of hours and the

specific days that courtrooms were used for nontrial activities, and in our
report we presented a separate analysis showing how many of these days had
events lasting a total of 2 hours or less. Because trial and nontrial times
were recorded differently, we chose to present the actual courtroom use data
in terms of the percentage of workdays used. Consequently, by using our
courtroom measure as a lights- on measure, EY overstated the number of hours
judges spent in the courtrooms and did not recognize that a significant
number of workdays had events that lasted 2 hours or less.

13 It is important to note that it was not our intent to suggest that five
judges can actually share four courtrooms, but rather to demonstrate the
limitations associated with EY's analysis.

Page 18 GAO- 01- 70 Courtroom Sharing

When EY used our courtroom use measure as a lights- on measure in its
formula, the results suggested that five judges could not share four
courtrooms. For example, the EY study contained the following calculation:
.65 (utilization rate) + .25 (flexibility factor) = .9 (courtrooms per
judge)

.9 x 5 = 4. 5 (equivalent courtrooms) This analysis showed that using the 25
percent flexibility factor and rounding the 4.5 figure upward result in the
conclusion that five judges would need five courtrooms.

Because EY's analysis is predicated on using a lights- on measure of
courtroom use, we tested EY's formula to determine if EY's incorrect use of
our data affected its results. We converted our 65- percent measure to more
of a lights- on measure by removing some of the hours for which we were
certain that no activity occurred. Specifically, when the courtrooms were

used for 2 hours or less for nontrial activities on a given day, we credited
the courtrooms with 2 hours of use instead of a full day, as was done for
our 1997 report. This yielded a courtroom use percentage of about 50
percent, which, although not a pure lights- on measure, is closer to the
type

of measure upon which EY's formula was predicated. 14 Using a 50- percent
use measure and incorporating EY's 25- percent flexibility factor, EY's
analysis suggested that five judges could in fact share four courtrooms, as
shown in the following formula: .50 (adjusted utilization rate) + .25
(flexibility factor) = .75 (courtrooms per judge) .75 x 5 = 3.75 (equivalent
courtrooms) EY staff acknowledged that there was a discrepancy in their
portrayal of our data as a lights- on measure. However, the EY staff said
that they understood our data and, for the purpose of their analysis, the
discrepancy was insignificant. They said they used the 65- percent figure
because it was the best available measure of actual courtroom use. We do not
share EY's view that this discrepancy was insignificant. As mentioned
before, we 14 The objective, scope, and methodology section of this report,
which is contained in appendix I, fully explains how we made this
adjustment. Also, as mentioned before, it was not our intent to suggest that
five judges can actually share four courtrooms, but rather to

demonstrate the limitations associated with EY's analysis.

Page 19 GAO- 01- 70 Courtroom Sharing

recorded a day of use even if courtroom events lasted less than an hour. For
example, about one- half of the days reflected in the 65 percent had only
nontrial activity, and most of these days involved events that lasted 2
hours or less. Given this, there were a significant number of days when the
courtrooms had time that may have been sufficient to accommodate various
uncertainties or other events that may have been scheduled.

Readers of the EY study could easily make the mistake of assuming that
judicial business was actually being conducted 65 percent of the available
hours. As shown above, if we adjusted our measure to more of a lights- on
measure, the analysis would show a different result. The imprecise portrayal
and use of our data, combined with the judgmental flexibility factor, raised
questions about the merits of the analysis and EY's conclusions and
recommendations regarding courtroom sharing.

We also noted that EY reported that 91 percent of courthouses have five or
fewer active district judges, and EY used this fact to provide additional
support for its recommendation that the judiciary retain the one- judge,
onecourtroom practice. In EY's opinion, the potential for courtroom sharing
throughout the country is small because most courthouses have five or fewer
judges. We agree that, intuitively, sharing in smaller facilities may

present more of a challenge for the judiciary than sharing in larger ones.
However, we also believe that better data and analysis are needed to
definitively determine the circumstances under which sharing is actually

feasible. It is also interesting to note that even though 91 percent of the
courthouses may have five or fewer active district judges, as EY reported,
about 40 percent of all current, active district judges are located in the
remaining 9 percent of the courthouses. From this perspective, the data show
that a considerable percentage of active district judges are in large
courthouses where, according to EY, sharing is likely to be more feasible.

We also were concerned about EY's use of the courthouse in Brooklyn, NY, to
illustrate that scheduling courtrooms in an actual sharing scenario would be
problematic. AOC officials told us that they urged EY to use Brooklyn as a
case study because it was the only facility that was operating with fewer
courtrooms than active district judges. Appendix B of EY's study points out
that this facility has been partially demolished in

preparation for a new courthouse and that it was difficult for EY to
determine the extent to which operational difficulties were caused by
courtroom sharing or by inadequate facilities. However, in the body of the
study, EY said that operational difficulties in the Brooklyn courthouse
demonstrated that courtroom sharing has been costly to administer in terms
of staff time, disruptions, and ineffective administration of justice.

Page 20 GAO- 01- 70 Courtroom Sharing

Given the presence of the other variables related to the inadequacy of the
facility, it is our view that the Brooklyn courthouse may not be a good
example for demonstrating courtroom- sharing feasibility.

We also noted that the study provided data and analysis directly related to
courtroom sharing for senior judges and in large courthouses with more than
10 district judges. 15 EY suggested that sharing is possible in these areas
and, in particular, recommended that the judiciary provide one courtroom for
every two senior judges. However, as previously discussed, the analysis used
to determine if active district judges can share courtrooms- which was
similarly applied in analyzing larger courthouses and senior judges' use-
was problematic. As a result, although data from

our 1997 report indicated low levels of use among senior judges with reduced
caseloads, we cannot comment on the merits of EY's recommended 2- to- 1
ratio. In our view, a reasonable ratio of courtrooms to

senior judges and the specific circumstances under which senior judges
should share courtrooms are still unknown. Regarding large courthouses, it
seems from an intuitive standpoint that the potential for sharing does
increase with the number of judges in a facility. However, better data and

research are still needed to demonstrate empirically that this is the case.
Conclusions Courtroom sharing in the federal judiciary has been a highly
visible and much- debated issue in recent years across all three branches of
government. Proponents of courtroom sharing argue that sharing is feasible
because of low use levels and that taxpayer dollars could be saved by
reducing the number of courtrooms needed. Opponents of courtroom sharing
argue that the complexities associated with courtroom availability and
effective judicial administration reduce the likelihood that sharing is
feasible. We and certain other organizations have reported that more
research is needed to develop data on the major factors that influence
courtroom use and availability. Without these data and analyses, it will be
difficult to reach consensus on if and to what extent courtroom sharing
could be instituted. Furthermore, the judiciary, key stakeholders, and the

organizations that have done research in this area have not reached any
upfront agreement on what type of cost- effective research should be done.

15 EY's study defines a large courthouse in two slightly different ways: The
executive summary defines a large courthouse as having “more than ten
district judges,” but chapter IV refers to them as having “ten
or more active judges.” We used “more than ten district
judges” for the purpose of our review.

Page 21 GAO- 01- 70 Courtroom Sharing

Although the EY study provided useful information on the various factors
that may affect courtroom sharing, it was not the type of research needed to
help resolve the courtroom- sharing issue. Also, the study's limitations
raised questions about the sufficiency of the data and analyses presented to
support its conclusions and recommendations regarding courtroom sharing. We
recognize that expert judgment should be considered in determining the
merits of courtroom sharing. However, it is clear to us that the courtroom-
sharing issue is not going to be resolved without better data and analysis
on actual courtroom use, including factors that may affect the

need for a courtroom, such as trial length and whether a trial will actually
take place. Also, the credibility of any further research could be enhanced
if it recognized the views of all key stakeholders in developing the study's
objectives, methodologies, and approaches for doing the work. Costeffective
research along the lines of what we and others have recommended has not been
performed and is needed if the government is to make informed, sound
decisions on courtroom use and sharing issues.

Although AOC is reluctant to do this type of cost- effective research, we
continue to believe it should be done. Given the judiciary's position, we
believe that Congress, as a minimum, should have access to actual courtroom
use data and analysis for locations where new courthouse projects are
proposed so that the judiciary's justification for the number of courtrooms
in these projects can be assessed before funding is approved.

Recommendations for Executive Action

We recommend that the Director, AOC, in conjunction with the Judicial
Conference's Committee on Court Administration and Case Management and
Committee on Security and Facilities, design and implement costeffective

research more in line with the recommendations in our 1997 report. These
recommendations, which were similar to those made by Rand and FJC, were
aimed at developing the type of data that would convincingly illustrate
whether, and under what circumstances, opportunities for courtroom sharing
exist. We also recommend that AOC establish an advisory group made up of
interested stakeholders and experts to assist in identifying study
objectives, potential methodologies, and reasonable approaches for doing
this work.

Matter for Congressional Consideration

Given the controversy surrounding the courtroom- sharing issue, the
potential savings that could be derived if fewer expensive trial courtrooms
are built, and AOC's reluctance to design and implement cost- effective
research to help resolve the issue, Congress should consider requiring AOC

Page 22 GAO- 01- 70 Courtroom Sharing

to provide persuasive courtroom use data and analysis, along with its views,
to justify the number of courtrooms being requested in future courthouse
construction projects before funding is approved.

EY and Judiciary Comments and Our Evaluation

We received written comments on a draft of this report from EY on November
9, 2000. (See app. II.) We received written comments from AOC's Associate
Director on November 3, 2000. (See app. III.) We also received

written comments on October 31, 2000, from the Chair of the Judicial
Conference's Committee on Security and Facilities. (See app. IV.) OMB did
not provide comments on the draft. An overall description of the comments
and our evaluation are discussed below. In addition to our evaluation of

AOC's major comments below, appendix III contains comments on other specific
points AOC made. In October 2000, EY and AOC also provided us with oral
technical comments on a draft of this report, which we incorporated where
appropriate.

EY Comments In its letter, EY provided general comments to add context to
the work it did and clarify its overall objectives and scope. EY explained
the design and objectives of its study and pointed out that the chapter on
courtroom utilization was part of a broader, comprehensive review of
courthouse planning. EY said that the chapter on courtroom utilization
should not be read and/ or used separately from the entire report. EY went
on to say that empirical data and research on courtroom utilization were
sparse and that the judiciary reported that data on courtroom utilization
were generally

unavailable. EY said that it considered various approaches that others had
developed to attempt to study this issue empirically; but it became clear
that such approaches would entail a data collection effort that was beyond
the scope, time frame, and resource commitment of its engagement. Given
this, EY said that it used the empirical assessment of courtroom utilization
in federal district courts that was available in our 1997 report, which
provided a measure of courtroom utilization. EY said that it recognized that
the assessment had limitations, but it attempted to build on

it by examining what factors would influence the ability of district judges
to share courtrooms, given the levels of utilization measured by GAO. EY
said that the value of this type of analysis was that it showed that even in
high levels of utilization, some courtroom sharing would be possible,
although complex to implement. EY said that its view was that it would be
possible for larger courthouses and senior judges to share courtrooms.
However, it

Page 23 GAO- 01- 70 Courtroom Sharing

did not feel that its analysis was sufficient to recommend formally to the
Judicial Conference that it change its national policy of providing one
courtroom per active district judge.

In addition to general comments, EY had specific comments related to issues
discussed in our report. For example, EY recognized that the 20 to 25
percent flexibility factor used in its mathematical formula to determine if
five judges could share four courtrooms is not based on empirical data and
analyses. EY said that on the basis of discussions with stakeholders and
visits to courthouses, it identified factors that would require some degree
of flexibility in its mathematical formula. EY said that the flexibility
factor is a judgmental decision that cannot, at this time, be supported by
empirical

data. We share EY's view that some degree of flexibility may be needed to
account for uncertainties that may affect the need for a courtroom, such as
the trial length and whether trials will occur. However, a more reliable
flexibility factor that would better reflect these uncertainties has not yet
been developed. Therefore, we had concerns with EY's use of a subjective

flexibility factor in its mathematical formula, given that the results of
EY's analysis using the formula formed the basis for its conclusions and
recommendations on the merits of courtroom sharing. As we stated in our
report, had EY used a flexibility factor of 15 percent instead, the results
would have suggested that five judges could have shared four courtrooms. We
continue to believe that a persuasive assessment of the potential for

sharing should be rooted in actual data on how often and for what purposes
courtrooms are used. This could lead to a more analytically based assessment
of the interrelationship of various events and a more defensible estimate of
the flexibility needed to account for uncertainties.

In commenting on our assessment that the study provided no data, rationale,
or analytical basis to support the flexibility factor, EY said that it
listed multiple factors in the study that supported the need to take
flexibility into account when considering courtroom sharing. EY explained

that because these factors were derived from interviews with stakeholders,
including judges, they constitute a rationale, particularly in the absence
of empirical data. We agree that the factors EY listed generally constitute
a rationale for why EY believed a flexibility factor was needed. However,
our concern was that no rationale was provided for the specific flexibility
factor- 20 to 25 percent- that EY used in its formula. On the basis of EY's
comment, we clarified this point in our report. We continue to believe that
the EY report does not provide an adequate rationale for how EY

determined that a flexibility factor of 20 to 25 percent should be used to
account for the factors EY identified. Without such a rationale, there is no

Page 24 GAO- 01- 70 Courtroom Sharing

link between the basis for EY's judgment- stakeholder input and
observations- and the specific percentages that EY decided to use.

In commenting on our conclusion that it incorrectly used data from our 1997
report on courtroom usage, EY acknowledged that it did not adequately
describe our data in its assessment. However, EY continues to believe that
its assessment of the potential for courtroom sharing is appropriate, in
part, because the GAO data could not distinguish between factors that affect
partial days and multiple days in the scheduling process. We do not see how
distinguishing between factors that affect partial days and multiple days
has any bearing on the interpretation of our data. As stated in this report,
EY's courtroom utilization model used our 65 percent courtroom use measure
as an actual lights- on courtroom use measure- the number of hours judges
spent in the courtrooms- when in fact it was a percentage of workdays that
the courtrooms were used for any activity, even if the activities lasted
less then an hour. Using our courtroom use

measure as a lights- on measure overstated the number of hours that judges
spent in the courtrooms. Also, it does not recognize that on a significant
number of the days, the courtrooms were used for only 2 hours or less.

Regarding our point that the Brooklyn courthouse may not be a good example
to demonstrate courtroom- sharing feasibility, EY said that its assessment
noted the limitations of using Brooklyn as a case study and that the
limitations were disclosed. We recognized in our report that EY

identified the limitations of using the Brooklyn case study in appendix B of
its study. However, as indicated in our report, the body of EY's study does
not mention these limitations and is definitive in saying that the Brooklyn
courthouse demonstrated that sharing has been costly to administer in terms
of staff time, disruptions, and ineffective administration of justice. As a
result, readers of the chapter on courtroom sharing may mistakenly conclude
that the problems in Brooklyn were due to courtroom sharing

alone and not the presence of operational difficulties caused by inadequate
facilities. In our October meeting with EY to discuss the draft report, EY
staff said that on the basis of our concern, they would have considered
modifying the body of their report to reflect the limitations with using
Brooklyn to draw conclusions about sharing.

Finally, EY said that the chapter on courtroom utilization should not be
read and/ or used separately from its entire report because its analysis on
courtroom utilization was part and parcel of a comprehensive review of
courthouse planning. Although we were specifically requested to examine

Page 25 GAO- 01- 70 Courtroom Sharing

the courtroom utilization chapter, we did review it in the context of the
other chapters in the report.

AOC Comments In commenting on a draft of this report, AOC disagreed with our
conclusion that the EY study did not provide the type of data or analyses
needed to resolve the courtroom- sharing issue. AOC also disagreed with our
recommendation to design and implement cost effective research to develop
actual courtroom use data that we believe could more convincingly illustrate
whether, and under what circumstances,

opportunities for courtroom sharing exist. AOC said that the EY study was
comprehensive; it believes that no further statistical study would be
productive, expert judgment needs to be applied, and that it is highly

unlikely that we would accept any result short of our previous conclusions.
AOC also said that decisions on courtrooms should not be based on purely
economic reasons, sharing beyond what the judiciary has agreed to would

impair the effective and fair administration of justice, and those who think
that the judiciary is unfit to determine its requirements might ask
themselves why.

We do not believe that AOC's comments are persuasive and continue to believe
that our conclusions and recommendations are appropriate and are fully
supported by the evidence and analysis in our report. As discussed in this
report, EY's assessment of the courtroom- sharing issue was not the

type of research we and others- including Rand and FJC- have concluded would
be needed to resolve the courtroom- sharing issue. The EY report itself
essentially supports this assessment. EY's analysis of the courtroomsharing
issue was problematic, as we discussed in our report, and neither EY nor AOC
provided any additional data or analysis that resolve the problems we
identified.

We do not believe, as AOC indicated, that decisions on courtroom sharing
should be made solely on the basis of empirical data, nor do we believe that
(1) cost savings should override considerations related to the effective
administration of justice, (2) courtroom- sharing decisions should be made
without the application of expert judicial judgment, or (3) existing data
support the need for active district judges to share courtrooms. Our
position is that given the amount of time expensive trial courtrooms appear
to go unused, more should be done to determine if sharing is feasible
without impairing the judicial process. Also, it is our view that decisions
on courtroom sharing should entail the application of expert judicial
judgment to complement more empirical data on courtroom use than currently
exists

Page 26 GAO- 01- 70 Courtroom Sharing

and should reflect more sophisticated methodological approaches than were
included in EY's study. Contrary to AOC's view, we are not locked into a
preconceived conclusion or specific methodological approach. We do believe,
however, that the courtroom- sharing issue needs to be addressed through a
more empirically based study than EY's, and, we note that EY itself
characterized its study as “ï¿½exploring the context of courtroom
sharingï¿½.”

Furthermore, although we agree with AOC that courtroom sharing should not be
implemented if it impedes the effective administration of justice or
infringes on constitutional rights, AOC provided no evidence or analyses to
show that this would be the case. In addition, EY itself raised questions

about drawing conclusions from the Brooklyn courthouse experience. Finally,
in neither this nor our prior reports have we questioned the judiciary's
fitness to determine its requirements. At the same time, it is a normal part
of the congressional appropriations and oversight process to review and
question the basis for the stated resource requirements of any federal
agency or department, including the judiciary. Currently, the

judiciary has presented virtually no empirical data to support its position
that it is essential for each active district judge to have a dedicated
courtroom. We are not, as AOC suggests, proponents of courtroom sharing, but
we are proponents of having persuasive data and analysis for making informed
decisions about the use of public resources. It would be both

appropriate and desirable for the judiciary to be directly involved with any
independent research that we and others are recommending; decisions on
courtroom sharing should not be made without full consideration of the
judiciary's views. However, AOC remains reluctant to undertake the type of

research we and others have said is necessary to make informed decisions
about the number of courtrooms needed. Consequently, we believe Congress
should consider requiring AOC to provide persuasive courtroom use data and
analysis, along with its views, to justify the number of courtrooms
requested for future courthouse construction projects before funding is
approved.

In its November 3 letter, AOC raised seven specific concerns with our draft
report. These concerns, and our evaluation, are discussed below as well as
in the GAO Comments section that follows AOC's letter in appendix III.
Stakeholder Involvement in the EY Study

AOC's first major concern was that our report did not credit the judiciary's
actions to involve stakeholders in the EY study. AOC states that the

Page 27 GAO- 01- 70 Courtroom Sharing

judiciary has taken the concerns raised by a few stakeholders about
courtroom needs seriously and, among other things, undertook the study by EY
in 1999. AOC said the EY study was comprehensive and inclusive and that the
study had a requirement for the consultant to seek out and consider the
views of all stakeholders. AOC said that the judiciary invited 39 officials
in Congress and the executive branch to participate in the assessment and
that EY interviewed many to consider their suggestions in framing its
approach. AOC went on to say that GAO wrongly faults the judiciary with
failing to obtain stakeholder agreement on the study

objectives and approaches. AOC is correct that our draft did not recognize
AOC's and EY's efforts to contact various stakeholders in doing its overall
study on the judiciary's space and facilities program, and we modified our
report to recognize these

efforts. However, our point that agreement has not been reached with
stakeholders on the type of study that should be done to help resolve the
courtroom- sharing issue remains valid, as evidenced by our discussions

with AOC, OMB, congressional staff, and one of the authors of the Rand
study. Further, when EY staff met with us at the beginning of the study, we
gave them an overview of our prior work and the recommendations we made
related to the type of research that we believe is still worth pursuing.

However, the EY staff who interviewed us did not ask us to comment on a
specific, proposed approach or methodology for its study. Thus, we were not
given the opportunity to comment on EY's planned approach prior to its
implementation.

Portrayal of Stakeholder Views AOC's second major point was that our draft
report did not accurately

portray the views of stakeholders. AOC stated that the draft suggested that
there is a courtroom debate with two distinct sides- with the judiciary on
one side and most other stakeholders outside the judiciary on the other
side. AOC went on to say that those directly involved in the justice system
have voiced concerns similar to those of the judiciary. AOC provided
examples of others, such as a number of U. S. Attorneys and the President of
the Federal Bar Association, who are concerned about the merits of courtroom
sharing. AOC said that to suggest that the judiciary stands

nearly alone in its views about courtrooms is simply not accurate. AOC's
point is valid. It was not our intent to portray the judiciary as the only
opponent of courtroom sharing, and we modified the text to reflect the fact
that other stakeholders besides the judiciary have concerns about sharing.
On the other hand, AOC also said that a few individuals who are involved
with funding have expressed a view on sharing different from the

Page 28 GAO- 01- 70 Courtroom Sharing

judiciary's view. We believe that AOC's characterization understates the
concerns that have been expressed. As AOC later acknowledges, OMB proposed
courtroom sharing in projects included in the President's budget request.
The Senate report that mandated our work dedicates a section to the
courtroom- sharing issue that states that AOC “fails to pursue a
policy of fiscal restraint” and indicates that the Appropriations
Committee “will continue to pursue all avenues with respect to cost
containment.” The House Subcommittee on Economic Development, Public
Buildings, Hazardous Materials, and Pipeline Safety has also raised
questions on the sharing issue. In addition, CBO has explored the courtroom-
sharing issue. Finally, AOC cited comments made by those opposed to
courtroom sharing that discussed the negative effects they believe sharing
would have on the administration of justice. These concerns are important
and need to be explored, but AOC provided no additional evidence on the
validity of these concerns or that would address the courtroom use results
discussed in our 1997 reports on courtroom utilization. Recognition of
Judiciary

Actions to Share Courtrooms and Potential for Additional Savings

AOC's third area of major concern was that our report gives insufficient
recognition to the fact that the judiciary's actions have already reduced
the number of courtrooms planned in new facilities and that it
oversimplifies the potential for additional savings. AOC pointed out that
the EY study reported that some sharing already occurs in the judiciary. It
also stated that for future projects, 274 courtrooms are planned for 347
judges. AOC

said that this important information is not mentioned until near the end of
our report. AOC also stated that the idea that further reductions in
courtrooms will result in savings is not as simple as we suggested in our

report. AOC said that according to the EY study, if courthouses are not
built to accommodate a court's operational needs and future growth
requirements, substantial future expenditures could be incurred sooner. AOC
said that if even one annex or new courthouse is required a few years hence
because an insufficient number of courtrooms were provided in a courthouse
project, millions of additional dollars would have to be spent. Therefore,
AOC said that to avoid added costs, it is as important not to underbuild the
number of courtrooms in a new or expanded courthouse as it is not to
overbuild. According to AOC, courtrooms are built to be used for at least 50
years, but the number of courtrooms built in a new facility is based on only
a 10- year projection of need. According to AOC, with the

growing workloads in the courts, short- term savings can readily be
overshadowed by additional costs to expand facilities later.

Page 29 GAO- 01- 70 Courtroom Sharing

In raising these concerns, it is important to note that the sharing AOC
refers to, including the sharing reflected in almost all future projects, is
for visiting judges and senior judges who carry reduced caseloads, not
active

district judges. Only one location among the future projects- Seattle- plans
sharing among active district judges. According to the AOC official who
oversaw the EY study, it is not yet clear whether sharing will ever actually
occur in Seattle. To reflect AOC's concern, we modified our report to
recognize the judiciary's actions on sharing earlier in our report. That is,

the judiciary has decided that some sharing is feasible. However, the issue
of whether active district judges can share courtrooms was the central issue
addressed by EY's analysis and its resulting conclusion and recommendation
that the judiciary retain the one- judge, one- courtroom policy. AOC and EY
officials told us during our review that Brooklyn was the only location
where active district judges were currently sharing. As mentioned earlier,
EY acknowledged that various limitations made it difficult to determine if
the problems in Brooklyn were due to sharing. Regarding potential savings,
AOC did not provide evidence to support its assertion that building fewer
courtrooms in a project will accelerate the

need for future expansion. The EY quote that AOC uses to make this point
states that additional expenditures could be incurred sooner if courthouses
are not built to accommodate future growth requirements. We did not say in
this report that courthouses should not be built to accommodate future
growth. Instead, our work has shown that the number of courtrooms that

may be needed to accommodate growth efficiently while maintaining effective
judicial administration may be less than the number that are currently being
constructed. Our assessment was based on the low levels of courtroom use we
observed and documented in our prior work.

Furthermore, in our 1997 report, we noted that of the six locations with
more than one trial courtroom, all courtrooms at any location were seldom
used for trials and nontrial activities the same day. For example, of the
250 workdays in 1995, Miami and Washington each had at least one unused
courtroom on each of the workdays. We believe that judgments about how many
courtrooms are needed to efficiently accommodate the judiciary's needs
without compromising effective judicial administration would be better
supported by sound methodological analyses of empirical data on

how often and for what purposes courtrooms are actually used. It is also
important to point out that the EY economic analysis AOC mentioned was
inconclusive. The analysis suggested that any potential savings from
constructing fewer courtrooms could be offset by the costs of hiring
schedulers to enable courtroom sharing to take place. To be

Page 30 GAO- 01- 70 Courtroom Sharing

conclusive, such an analysis would need to consider exactly how many fewer
courtrooms could be built, and how many- if any- new personnel would have to
be hired as schedulers. These are empirical questions that could be
addressed in the type of cost- effective research we and others have
recommended needs to be done to help resolve the courtroom- sharing issue.
Focus on GAO's 1997 Report AOC's fourth major concern was that the report's
organization and presentation pay attention to promoting GAO's earlier
conclusions rather than to presenting a thorough consideration of EY's
approach to the

subject. AOC states that from the start, our report focuses on restating our
conclusions and recommendations from our 1997 report instead of on the
current assignment to review the EY study. AOC said that nearly half of the

draft report's text has little or nothing to do with the EY study. AOC said
that it had previously documented serious reservations about the accuracy of
our earlier assessment and the validity of conclusions that are repeated in
this report.

We believe that AOC's concern about the organization of our report is not
valid. Given the complexity of this issue, we would be remiss if we did not
include a discussion of what we and other researchers- including Rand and
FJC- have reported in this area. In meeting our objective to determine
whether the EY study provided sufficient data and analysis to show if, and
to what extent, courtroom sharing may be feasible, we used the results of
our prior work and the research of others as criteria for evaluating the EY

study. This is a common and widely accepted practice in social science
research. In addition, because EY relied heavily on data from our 1997
report for its analysis, we believed that readers of this report needed
information on our earlier study to fully understand EY's use of our data
and our assessment of EY's study. Nonetheless, on the basis of AOC's
concern, we added an explanation of why we used our past work and that of
others in the detailed discussion of our objective, scope, and

methodology contained in appendix I. It is also important to point out that
in its comments on our 1997 report, although AOC disagreed with our
conclusions and recommendations, it did not take issue with the accuracy of
our data on courtroom use, as was implied in its comments related to this
report.

Page 31 GAO- 01- 70 Courtroom Sharing

Empirical Data Versus Expert Judgment

AOC's fifth major concern was that our report overstates the likelihood that
more data will resolve the courtroom- sharing issue. AOC said that our
report reflects scant respect for expert judgment, dismisses the
professional judgment applied by EY, and shows a high degree of skepticism
about the value of the opinions of judges and other primary users and
schedulers of courtrooms. AOC went on to say that there are many questions
for which data will not provide answers and that the EY study describes many
factors relevant to the courtroom issue that are highly variable,
unpredictable, or cannot be measured empirically.

AOC's explanation of this concern inaccurately portrays our view of the role
that methodologically sound analysis of empirical data can play in helping
to resolve the courtroom- sharing issue and contains several assertions that
it has not provided sufficient evidence to support. Two important facts in
this debate are that available data show that courtroom use levels appear
low, and trial courtrooms are expensive to build. To date, there has been
much debate by experts on both sides of the issue. However, this debate-
which reflects the expert judgment that AOC mentions- has resulted in an
impasse regarding whether sharing is feasible. Our view is that more
empirical research could be done to further the debate and

provide empirical support for either side's argument. We do not, as AOC
asserted, believe that this research will provide all the answers or that
there is no place for expert judgment. We modified our report to make these
points clearer. We have a number of other points that raise questions about
the validity of

AOC's fifth major concern. AOC mentions factors that are difficult to
measure and are highly variable and implies that these issues cannot be
studied empirically. It is our view that this has not been conclusively
determined. In fact, our interviews with EY staff and review of AOC
documents show that EY proposed a study of courtroom sharing similar to what
Rand proposed. A Rand- type study would attempt to study these factors
empirically. This proposal was rejected because the additional cost and time
frame involved went beyond the scope of EY's broader study of

the judiciary's space and facilities program, not because it was
definitively determined that these factors could not be studied empirically.
We also disagree with AOC's statement that most of the analysis we have
recommended would be theoretical, not empirical. Although a methodologically
sound analysis of this issue- such as modeling or the computer simulations
that Rand suggests- would involve making assumptions, it would be based on
data on how courtrooms are actually

Page 32 GAO- 01- 70 Courtroom Sharing

used and thus would be considered an empirical assessment according to
standard practice in social science research. Further, although AOC believes
it is impossible to find a scientific way to measure what does not happen,
we believe that social science research has been developed to do just that.
In fact, the approach we discuss in our

report is not dissimilar to the one that AOC uses in its long- range
planning model for predicting its space needs. AOC forecasts its future
workload using the available data on past workload and statistical models
that are based on standard theories and assumptions. Essentially, AOC uses
scientific methods to measure what has not happened. In another example, the
Bureau of Prisons (BOP) routinely uses computer simulations to estimate the
effects of potential changes in sentencing legislation on the

size of the federal prison population. The results of these simulations are
used to project the financial impact of potential legislative changes.
According to BOP, the results of these simulations are also used in
deliberations by Congress and other bodies, such as the U. S. Sentencing

Commission and the Department of Justice. Value Added by Additional Study

AOC's sixth major concern was that our report uses confusing figures to
support the idea that another extensive study is worth pursuing. AOC said
that there is no compelling justification provided in our report to explain
why another study of the subject would be cost effective. AOC said that we
acknowledged the study would be lengthy and costly and referred to data that
it believes show that the potential for sharing is small given the size of
most courthouses. We disagree with AOC's contention that some of our

figures are confusing. Our point was to show that although 91 percent of
courthouses have five or fewer active district judges, the remaining 9
percent of these courthouses contain 40 percent of all active district
judges. Therefore, close to one- half of active district judges are located
in courthouses where EY concluded that sharing is likely to be more
feasible. Furthermore, although we agree that sharing in smaller courthouses
would pose more of a challenge, our work showed that the mathematical
analysis EY used to examine the potential for courtroom sharing in smaller
courthouses was problematic. As a result, we disagree with AOC's premise
that it has been definitively determined that sharing at smaller courthouses

is not possible. Regarding the cost- effectiveness of future research, we
disagree with AOC's view that our report does not contain a compelling
justification. As stated in our report, available data suggest that
courtrooms are not used

Page 33 GAO- 01- 70 Courtroom Sharing

very often, the need for new judgeships will continue to grow, and
significant amounts of taxpayer dollars are being spent to replace obsolete
facilities and accommodate growth. Cumulatively, these factors form a basis
to suggest that further research would be worth pursuing because

significant savings could result if fewer million- dollar trial courtrooms
need to be constructed. In our 1997 report on courtroom use, we said that
AOC workload projections indicated that the number of district judgeships
could double or perhaps increase more significantly by the year 2020- from
647 judgeships in 1996 to between 1,280 and 2,410 judgeships over this
period. Although AOC officials said that these previous projections were no

longer valid, they acknowledged that workloads would continue to increase.
AOC told us that for budgeting purposes, it was planning for an 8percent
increase in judgeships by 2006. Although the timing and rate of increase in
new judgeships remain unclear, it is apparent that the judiciary is planning
to accommodate growth and replace obsolete facilities in the

future, as evidenced by its multibillion- dollar courthouse construction
initiative.

Related to the cost- effectiveness of further research, AOC also incorrectly
stated that we acknowledged in our report that such a study would be costly.
We recognized that data collection for the effort would be timeconsuming
because of the lack of data on courtroom use, but we did not conclude that
such an effort would be too costly to consider. In fact, the cost of such a
study is unknown because the objectives, scope, and methodology have not
been determined. And, whether such a study would

be considered costly would have to be determined within the context of the
potential savings that may be achieved by building fewer expensive trial
courtrooms. It is our view that AOC's reasoning that such research is too

costly is not supported by facts and analysis and that at least some of the
data we believe are important could possibly be gathered without incurring
any or significant additional costs. For example, the judiciary could ask
court personnel to track courtroom use on a real- time basis for a period of
time. Such an approach would not require a special, separate study in which
researchers would have to examine records or interview personnel to obtain
historical data.

Criteria for Effective Courtroom Use AOC's seventh major concern had to do
with part of a recommendation

made in our 1997 report to establish criteria for determining effective
courtroom use. AOC said that our proposal to establish courtroom use
criteria is alarming. AOC said that if efficiency was valued more in our
society than fairness, fewer courtrooms could be built. AOC went on to say

Page 34 GAO- 01- 70 Courtroom Sharing

that focusing on use and cost of courtrooms may overlook more fundamental
values and indicated that it is neither possible nor feasible to establish a
standard for courtroom use.

AOC's basis for this concern appears to be its assumption that delays and
other problems would occur if courtroom usage criteria were adopted.
However, AOC did not provide any convincing evidence to support its view. We
are not suggesting, as AOC pointed out, that the judicial system should
operate under factory- like standards. In fact, we stated in this report
that efficiency and cost savings should not override considerations related
to the administration of justice. However, we do not find it unreasonable
for

the judiciary to develop and analyze actual courtroom use data so that
informed decisions could be made on whether courtroom sharing is feasible.
As part of collecting and analyzing courtroom use data, it is important that
criteria be developed to determine what constitutes efficient and effective
courtroom use so that judiciary officials can consistently interpret usage
levels and apply an agreed- upon standard to identify the

number of courtrooms needed to conduct efficient operations without
compromising effective judicial administration. Establishment of criteria
would not preclude the use of expert judgment which could be applied to
empirical data and sound analyses to draw conclusions and determine
appropriate actions.

Comments by the Chair, Judicial Conference Committee on Security and

Facilities The Honorable Jane R. Roth, Chair of the Committee on Security
and

Facilities of the Judicial Conference of the United States, also provided
comments on a draft of the report. Ms. Roth said that in her capacity as the
Chair of the Committee on Security and Facilities, she was familiar with the
methodologies EY used to analyze the subject of courtroom sharing and

was satisfied that the consultant's approach was sound. She went on to say
that in her view, the recommendations made by EY regarding the courtroom
needs of active district judges were reasonable. She said that given the
complexities of the subject of courtroom sharing, she does not agree that a
matter of such importance to the administration of justice can

be answered by the collection and manipulation of data. She said that there
are too many unmeasurable ingredients to achieve the clarity of results or
the uniform answer that is sought by GAO. She said that she concurred with
the comments provided by AOC on our draft report. We agree that the issue of
courtroom sharing cannot be resolved only

through the collection and manipulation of data. However, we strongly
believe that the development of actual courtroom data is a critical starting

Page 35 GAO- 01- 70 Courtroom Sharing

point to fully understanding the issue. That is, knowing how often and for
what purposes courtrooms are actually used and how the various events
interrelate would form the basis for further research on the potential for
courtroom sharing and the application of expert judgment. In our view, it
seems reasonable for decisionmakers to want information on actual courtroom
use, especially given the high cost to construct trial courtrooms.

In fact, in our 1997 report, we pointed out that FJC believed that “it
seems likely the judicial branch can expect the current pressure for
economy, efficiency, and effectiveness to continue and quite probably to
intensify” and that “expenditures for features beyond the most
Spartan will have to be defended with hard data.” Furthermore, FJC
proposed major changes to the judiciary's regular data collection “so
that the elements, dynamics, and

effects of court operations can be substantively reported without assembling
an ad hoc study each time a specific aspect of the system is questioned and
singled out for scrutiny.” Without better data, it is difficult to
make informed decisions about the feasibility of courtroom sharing. Given
this, the debate over courtroom sharing will continue to be based on
speculation and qualitative judgment that is not based on the
methodologically sound analysis of empirical evidence.

We are sending copies of this report to the Chairmen and Ranking Minority
Members of congressional committees with jurisdiction over the judiciary and
GSA. We are also sending copies of this report to the Honorable Jane R.

Roth, Chair of the Judicial Conference's Committee on Security and
Facilities; the Honorable John W. Lungstrum, Chair of the Judicial
Conference's Committee on Court Administration and Case Management; the
Honorable L. Ralph Mecham, Director, AOC; the Honorable Jacob J.

Lew, Director, OMB; and the Honorable David J. Barram, Administrator, GSA.
We will also send copies to interested congressional committees and make
copies available to others on request.

Page 36 GAO- 01- 70 Courtroom Sharing

Major contributors to this report were James G. Cooksey, Martin H. de
Alteriis, David E. Sausville, and Gerald Stankosky. If you or your staff
have any questions, please contact me on (202) 512- 8387 or at ungarb. ggd@
gao. gov. Bernard L. Ungar Director, Physical Infrastructure

Page 37 GAO- 01- 70 Courtroom Sharing

Page 38 GAO- 01- 70 Courtroom Sharing

Appendix I

Appendi xes Objective, Scope, and Methodology Appendi x I

Our objective was to determine whether the Ernst & Young (EY) study provided
sufficient data and analyses to show if, and to what extent, courtroom
sharing may be feasible. To meet this objective, we obtained and analyzed
the EY study, with a focus on the parts of the study that pertained to
courtroom use and sharing, including the executive summary; chapter IV,
“Assessing the Need for Courtrooms”; appendix A, “Study
Approach and

Methodology”; and appendix B, “Brooklyn Courthouse Case
Study.” We assessed EY's interpretation and use of data from our 1997
report and examined and tested the mathematical formula EY developed and
applied to determine whether courtroom sharing was feasible. We used our
1997 report and research done by other organizations as criteria to evaluate
the EY study. In focusing on the parts of the study that pertained to
courtroom use and sharing, we reviewed those elements that were directly
related to supporting the study's conclusions and recommendations on
courtroom

use and sharing. These included the mathematical formula EY used to examine
the potential for sharing, the Brooklyn courthouse example, and the
qualitative arguments that EY developed to support its positions. In order
to test the EY formula used to illustrate courtroom sharing opportunities,
we adjusted data from our 1997 report to reflect a measure that more closely
resembled lights- on use- that is, the number of hours judges spent in the
courtrooms. We used this measure in EY's formula to determine how it would
have affected EY's results. In our 1997 report, we

had credited courtrooms with a full day of use when there was any trial or
nontrial activity, even if it lasted less than an hour. To adjust our data
to more of a lights- on measure, we were able to remove some of the hours
for which we were certain that no activity had occurred. Specifically, when
there were 2 hours or less of nontrial activity, we credited the courtroom
with 2 hours of use instead of a full day's use as was done for our 1997

report. We made this adjustment relying completely on the data available in
our 1997 report and on the basis of an 8- hour day. This adjustment creates
more of a lights- on measure, but it still is conservative because it
continues to contain time with no use because we credited the courtrooms
with a full 2 hours of use even though less than 2 hours of use could have
occurred.

Also, we made no adjustment to the days when there was nontrial time that
lasted more than two hours but may not have lasted the whole day. In
addition, we made no adjustment for days when trials took place because our
data source did not always allow us to determine the actual number of hours
trials were held each day. We reviewed the qualitative information presented
in the study, including

information on issues related to the potential for courtroom sharing and

Appendix I Objective, Scope, and Methodology

Page 39 GAO- 01- 70 Courtroom Sharing

the Brooklyn, NY, courthouse case study. We examined EY's conclusions and
recommendations in relation to the data and analysis presented. We
interviewed officials from the Administrative Office of the U. S. Courts
(AOC), Office of Management and Budget, and Congressional Budget Office to
obtain their views on the EY study and courtroom- sharing issues.

We also interviewed the EY staff who worked on the study and one of the
authors of a 1996 Rand Institute for Civil Justice (Rand) study on courtroom
use. 1 We reviewed previous studies on or related to courtroom use by us,
AOC, Rand, and the Federal Judicial Center (FJC) to determine if the EY
study represented the type of research and analysis that we and

others have recommended. We also examined AOC's contract file for this
study. We did our work between June 2000 and September 2000 in accordance
with generally accepted government auditing standards. We received written
comments on a draft of this report from EY, AOC, and the Chair of the
Judicial Conference's Committee on Security and Facilities. An overall
description of these comments and our evaluation are discussed near the end
of the letter. Appendix III also contains comments on other specific points
made by AOC.

1 Terence Dunworth and James S. Kakalik, Research on Courtroom Sharing,
Project Memorandum, Rand Institute for Civil Justice, PM- 598- 1- ICJ,
September 1996.

Page 40 GAO- 01- 70 Courtroom Sharing

Appendix II Comments From Ernst & Young Appendi x II

See pp. 22- 23. Now on pp. 6- 7.

Appendix II Comments From Ernst & Young

Page 41 GAO- 01- 70 Courtroom Sharing

See pp. 23- 24. Now on p. 8. Now on p. 8.

Appendix II Comments From Ernst & Young

Page 42 GAO- 01- 70 Courtroom Sharing

See p. 24. See pp. 23- 24. See p. 24.

Now on p. 8. Now on p. 8.

Now on pp. 19- 20. Now on pp. 19- 20.

Page 43 GAO- 01- 70 Courtroom Sharing

Appendix III Comments From the Administrative Office of the U. S. Courts
Appendi x I II

Note: GAO's comments supplementing those in the report text appear at the
end of this appendix.

See pp. 26- 27. Now on p. 15.

Appendix III Comments From the Administrative Office of the U. S. Courts

Page 44 GAO- 01- 70 Courtroom Sharing

See pp. 27- 28.

Appendix III Comments From the Administrative Office of the U. S. Courts

Page 45 GAO- 01- 70 Courtroom Sharing

Appendix III Comments From the Administrative Office of the U. S. Courts

Page 46 GAO- 01- 70 Courtroom Sharing

See pp. 28- 30.

Appendix III Comments From the Administrative Office of the U. S. Courts

Page 47 GAO- 01- 70 Courtroom Sharing

See p. 30.

Appendix III Comments From the Administrative Office of the U. S. Courts

Page 48 GAO- 01- 70 Courtroom Sharing

See GAO comment 1. See GAO comment 2.

See GAO comment 3. See GAO comment 4.

Appendix III Comments From the Administrative Office of the U. S. Courts

Page 49 GAO- 01- 70 Courtroom Sharing

See GAO comment 5. See pp. 31- 32.

Appendix III Comments From the Administrative Office of the U. S. Courts

Page 50 GAO- 01- 70 Courtroom Sharing

See pp. 32- 33.

Appendix III Comments From the Administrative Office of the U. S. Courts

Page 51 GAO- 01- 70 Courtroom Sharing

See pp. 33- 34.

Appendix III Comments From the Administrative Office of the U. S. Courts

Page 52 GAO- 01- 70 Courtroom Sharing

See GAO comment 6.

Appendix III Comments From the Administrative Office of the U. S. Courts

Page 53 GAO- 01- 70 Courtroom Sharing

Appendix III Comments From the Administrative Office of the U. S. Courts

Page 54 GAO- 01- 70 Courtroom Sharing

The following are GAO's comments on AOC's letter dated November 3, 2000.

GAO Comments 1. AOC said that our report had the effect of suggesting that
GAO began this current task with the premise that there was only one valid
approach to studying the courtroom issue. Our objective was to determine
whether the EY study provided sufficient data and analysis to show if, and
to what extent, courtroom sharing may be feasible. As stated in the report,
certain

stakeholders and organizations we identified that have done research in this
area recognize that existing data and analysis on courtroom use were limited
and could not resolve the courtroom- sharing debate and that more data and
analysis were needed. Given this, our focus was to examine the study from
the perspective of whether it provided this type of data and

analysis. 2. AOC said that our report ignored or dismissed without much
attention significant parts of EY's analysis. AOC went on to mention EY's
use of a study completed in 1998 by the National Center for State Courts
that demonstrated the unpredictability of trial length. AOC also mentioned
the difficulty of predicting accurately whether a scheduled case will settle
before trial. Although we did not specifically mention the National Center
for State Courts study, we discuss various uncertainties throughout the

report- such as whether a trial will take place, how long a trial will last,
and whether an emergency proceeding will require the immediate use of a
courtroom- that may affect courtroom scheduling. The important issue here is
that the impact these uncertainties actually have on the need for a
courtroom is unknown because there is very little data and analysis. For
example, the judiciary cites predicting trial length as a major uncertainty
in scheduling courtrooms. However, it has not provided any convincing data
or analysis showing how this uncertainty would affect a scheduling process,
especially considering that available data suggest that courtroom

use is low. As previously mentioned, our data showed that district
courtrooms we examined were used for trials, on average, less than onethird
of the 250 federal work days in 1995 and that the use of courtroom for
trials varied by location. Also, our data showed that in 1995, at least one
courtroom in two courthouses was unused every workday of the year.

3. AOC said that EY concluded that “data analysis alone cannot
adequately assess the effect of courtroom availability on settlement rates,
trial delays and delivering justice.” AOC said that EY decided to
examine “the factors

that determine how easily judges can share courtrooms, and the potential

Appendix III Comments From the Administrative Office of the U. S. Courts

Page 55 GAO- 01- 70 Courtroom Sharing

cost and operational impact of courtroom sharingï¿½ to establish whether
courtroom sharing would significantly reduce courtroom construction cost and
the implications for administration of justice.” We agree that data
analysis alone cannot resolve the courtroom- sharing issue. However, as
pointed out in the report, a cost- effective empirical assessment that would
generate actual courtroom use data and analysis is key to informed
decisionmaking about the feasibility of courtroom sharing. Without these
data, it will be difficult to determine more conclusively whether
courtroomsharing opportunities exist. 4. AOC said that our report does not
mention Rand's recognition that expert input on the implications for the
administration of justice and

consideration of cost and potential benefits would be major components of
research on courtroom sharing. AOC said that EY addressed these aspects and
we did not. AOC also said that EY obtained expert input through

interviews and focus groups and through observing the impact of courtroom
sharing in a district and discussing the implication with judges, staff,
attorneys, and others affected. We disagree that we did not recognize many
of EY's efforts in doing its study. We pointed out in our report that the EY
study was informative and that EY staff, among other things, visited

federal courthouses, interviewed key stakeholders and judicial personnel,
and conducted focus groups with judges. Our report also said that EY visited
14 locations, observed courtroom use and spoke with users, examined
calendars, analyzed statistical data and courthouse planning documents,
conducted interviews with dozens of individuals who had different
experiences and views, and held focus groups. To address AOC's concern,
however, we clarified our report to reflect our point that expert judgement
is important but that it needs to supplement, not be a substitute

for, empirically based data and analysis on courtroom use. 5. AOC said that
we denigrated EY's use of the empirical information it collected related to
the Brooklyn case study. We disagree- our intent was to show that EY had
reservations about using Brooklyn to assert that the problems there were due
entirely to sharing. As pointed out in the report, we had concerns about
EY's use of the Brooklyn courthouse to illustrate that scheduling courtrooms
in an actual courtroom- sharing scenario would

be problematic. We explained that in appendix B of EY's study, EY points out
that the Brooklyn facility has been partially demolished in preparation for
the new courthouse and that it was difficult for EY to determine the extent
to which operational difficulties were caused by courtroom sharing or by
inadequate facilities. However, in the body of the study, EY said that
operational difficulties in the Brooklyn courthouse demonstrated that

Appendix III Comments From the Administrative Office of the U. S. Courts

Page 56 GAO- 01- 70 Courtroom Sharing

courtroom sharing has been costly to administer in terms of staff time,
disruptions, and ineffective administration of justice. Given the presence
of the other variables related to the inadequacy of the facility, it was our
view

that the Brooklyn courthouse may not be a good example for demonstrating
courtroom sharing feasibility. EY staff we interviewed agreed with our
assessment, and we stand by our observation.

6. AOC's conclusion states that for those who may be largely unfamiliar with
the judicial process, it seems unquestionably wasteful for judges to have
dedicated courtrooms. AOC then goes on to present analogies to demonstrate
why each judge needs a courtroom, comparing the reasons judges cannot share
courtrooms to the reasons neighbors do not share a

car and workers do not share phones. AOC's point is that cars and phones are
not used full- time, yet it is accepted that they will not be shared because
of the unpredictable nature of their use. AOC's reasoning is problematic
because cars and phones cost much less than courtrooms, which EY reported
cost $1.5 million each. Furthermore, to use AOC's

analogy, if the cars in question were used by public officials and data on
the use of the cars were comparably as low as available data on courtroom
use, we believe having one car for two or more officials would be worth

considering to save taxpayer dollars. This would certainly be a logical
reason for an organization to maintain a “pool” of vehicles for
staff to use when needed rather than providing a dedicated vehicle to each
person. Also, AOC's observation that it seems unquestionably wasteful to
those outside the judiciary for judges to have dedicated courtrooms is
exactly why we believe the judiciary needs better justification for this
practice.

Page 57 GAO- 01- 70 Courtroom Sharing

Appendix IV Comments From the Chair, Judicial Conference, Committee on
Security and Facilities Appendi x I V

(240411) Lett er

See pp. 34- 35.

Ordering Information The first copy of each GAO report is free. Additional
copies of reports are $2 each. A check or money order should be made out to

the Superintendent of Documents. VISA and MasterCard credit cards are
accepted, also. Orders for 100 or more copies to be mailed to a single
address are discounted 25 percent.

Orders by mail: U. S. General Accounting Office P. O. Box 37050 Washington,
DC 20013

Orders by visiting: Room 1100 700 4th St. NW (corner of 4th and G Sts. NW)
U. S. General Accounting Office Washington, DC

Orders by phone: (202) 512- 6000 fax: (202) 512- 6061 TDD (202) 512- 2537

Each day, GAO issues a list of newly available reports and testimony. To
receive facsimile copies of the daily list or any list from the past 30
days, please call (202) 512- 6000 using a touchtone phone. A recorded menu
will provide information on how to obtain these lists.

Orders by Internet: For information on how to access GAO reports on the
Internet, send an e- mail message with “info” in the body to:
info@ www. gao. gov or visit GAO's World Wide Web home page at: http:// www.
gao. gov

To Report Fraud, Waste, or Abuse in Federal Programs

Contact one: Web site: http:// www. gao. gov/ fraudnet/ fraudnet. htm e-
mail: fraudnet@ gao. gov 1- 800- 424- 5454 (automated answering system)

United States General Accounting Office Washington, D. C. 20548- 0001

Official Business Penalty for Private Use $300

Address Correction Requested Bulk Rate

Postage & Fees Paid GAO Permit No. GI00
*** End of document. ***