2000 Census
Complete Costs of Coverage Evaluation Programs Are Not Available Gao ID: GAO-03-41 October 31, 2002To assess the quality of the population data collected in the 2000 Census, the U.S. Census Bureau conducted the Accuracy and Coverage Evaluation (A.C.E.) program, which focused on a survey of housing units designed to estimate the number of people missed, counted more than once, or otherwise improperly counted in the census. GAO reviewed the life cycle costs of the A.C.E. program and its predecessor, the Integrated Coverage Measurement (ICM) program. GAO found that the original estimated cycle costs of conducting the ICM/A.C.E. programs were $400 million. The first evidence for the original $400 million estimate is in the original budget justifications for fiscal year 2000. The bureau based its estimates of ICM/A.C.E. costs on assumptions about the needs for personnel and benefits, contractual services, travel, office space, equipment, and other costs necessary to conduct and support operations of the programs. The budgeted amounts that GAO identified from bureau records for conducting the ICM/A.C.E. programs are $277 million through fiscal year 2003. The obligated costs that GAO identified from bureau records for conducting the ICM/A.C.E. programs are $207 million through fiscal year 2001. $58 million of budgeted funds for the ICM/A.C.E. programs that GAO identified from bureau records were not obligated through fiscal year 2001. The ICM/A.C.E. program-related costs that GAO identified from bureau records for the 1998 dress rehearsal were $11 million budgeted and $9 million obligated.
GAO-03-41, 2000 Census: Complete Costs of Coverage Evaluation Programs Are Not Available
This is the accessible text file for GAO report number GAO-03-41
entitled '2000 Census: Complete Costs of Coverage Evaluation Programs
Are Not Available' which was released on October 31, 2002.
This text file was formatted by the U.S. Government Accountability
Office (GAO) to be accessible to users with visual impairments, as part
of a longer term project to improve GAO products' accessibility. Every
attempt has been made to maintain the structural and data integrity of
the original printed product. Accessibility features, such as text
descriptions of tables, consecutively numbered footnotes placed at the
end of the file, and the text of agency comment letters, are provided
but may not exactly duplicate the presentation or format of the printed
version. The portable document format (PDF) file is an exact electronic
replica of the printed version. We welcome your feedback. Please E-mail
your comments regarding the contents or accessibility features of this
document to Webmaster@gao.gov.
This is a work of the U.S. government and is not subject to copyright
protection in the United States. It may be reproduced and distributed
in its entirety without further permission from GAO. Because this work
may contain copyrighted images or other material, permission from the
copyright holder may be necessary if you wish to reproduce this
material separately.
Report to Congressional Requesters:
United States Government Accountability Office:
GAO:
October 2002:
2000 Census:
Complete Costs of Coverage Evaluation Programs Are Not Available:
GAO-03-41:
Contents:
Letter:
Results in Brief:
Background:
Scope and Methodology:
Life Cycle Costs:
Seven Issues Related To ICM/A.C.E. Programs:
Agency Comments and Our Evaluation:
Appendixes:
Appendix I: Budgeted and Obligated Life Cycle Costs by Framework and
Project for Fiscal Years 1996 through 2003:
Appendix II: Comments From the Department of Commerce:
Tables:
Table 1: Census 2000 ICM/A.C.E. Program Budgeted Life Cycle Costs for
Fiscal Years 1996 through 2003:
Table 2: Census 2000 ICM/A.C.E. Program Obligated Life Cycle Costs for
Fiscal Years 1996 through 2001:
Figures:
Figure 1: Budgeted Costs of ICM/A.C.E. Programs 11 Figure 2: Obligated
Costs of ICM/A.C.E. Programs 12 Figure 3: Obligated Costs of ICM/A.C.E.
Programs by Category 13
Abbreviations:
A.C.E.: Accuracy and Coverage Evaluation:
CAMS: Commerce Accountability Management System:
FTE: full-time equivalent:
ICM: Integrated Coverage Measurement:
OMB: Office of Management and Budget:
PES: Post Enumeration Survey:
United States General Accounting Office:
Washington, D.C.: 20548:
October 31, 2002:
The Honorable William Lacy Clay, Jr.:
House of Representatives:
The Honorable Carolyn B. Maloney:
House of Representatives:
To assess the quality of the population data collected in the 2000
Census, the U.S. Census Bureau conducted the Accuracy and Coverage
Evaluation (A.C.E.) program, which focused on a survey of housing units
designed to estimate the number of people missed, counted more than
once, or otherwise improperly counted in the census. As agreed with
your offices, this report responds to your joint request to review
A.C.E. life cycle costs. Total life cycle costs for the 2000 Census
cover a 13-year period from fiscal year 1991 through fiscal year
2003[Footnote 1] at an estimated cost of $6.5 billion, adjusted to 2000-
year dollars. The A.C.E. program was first included in bureau program
documents in November 1998 and funded for fiscal years 2000 through
2003. A predecessor program, Integrated Coverage Measurement (ICM),
began in May 1995 and was funded by the bureau for fiscal years 1996
through 1999. We considered both programs to be within the scope of
your request.
As agreed with your offices, we focused our review on the following
seven questions concerning ICM/A.C.E. program life cycle costs.
1. What were the original estimated life cycle costs for the ICM/A.C.E.
programs?
2. What was the source and support for $400 million in life cycle costs
reported by the bureau for the ICM/A.C.E. programs?
3. How were ICM/A.C.E. program costs estimated?
4. How much did the bureau budget[Footnote 2] for the ICM/A.C.E.
programs?
5. What were the obligated life cycle costs for the ICM/A.C.E.
programs?
6. Were any budgeted funds for the ICM/A.C.E. programs not used as of
the end of fiscal year 2001, and if so, how much?
7. What were the ICM/A.C.E. program-related costs for the bureau dress
rehearsal in fiscal year 1998?
This report is part of a series of GAO studies on the lessons learned
from the 2000 Census that can help in the planning and development
effort now occurring for the 2010 Census.
Results in Brief:
Although the U.S. Census Bureau tracked some costs of conducting the
ICM/A.C.E. programs, it did not identify the complete life cycle costs
of the programs due to three factors. First, program costs prior to
fiscal year 1996 were not identified because the bureau considered
costs from earlier years to be part of its general research and
development efforts and the bureau did not assign unique project codes
to identify ICM/A.C.E. program and related costs in its financial
management system. Second, although $3.6 million of fiscal year 1996
obligated costs were identified in the bureau‘s financial management
system as an ICM special test, the bureau did not consider these costs
as part of the ICM/A.C.E. programs and it also classified these costs
as general research. We disagreed with the bureau on this point and
have included this amount in our report as part of the ICM/A.C.E. life
cycle costs we could identify. Finally, we were unable to identify
ICM/A.C.E. portions of costs, such as evaluations and data processing,
which the bureau included in other 2000 Census programs. Based on
available information for the seven questions, we found the following:
* The original estimated cycle costs of conducting the ICM/A.C.E.
programs were about $400 million.
* The first evidence for the original $400 million estimate is in the
original budget justifications for fiscal year 2000.
* The bureau based its estimates of ICM/A.C.E. costs on assumptions
about the needs for personnel and benefits, contractual services,
travel, office space, equipment, and other costs necessary to conduct
and support operations of the programs.
* The budgeted amounts that we identified from bureau records for
conducting the ICM/A.C.E. programs are about $277 million through
fiscal year 2003.
* The obligated costs that we identified from bureau records for
conducting the ICM/A.C.E. programs are about $207 million through
fiscal year 2001.
* About $58 million of budgeted funds for the ICM/A.C.E. programs that
we identified from bureau records were not obligated through fiscal
year 2001.
* The ICM/A.C.E. program-related costs that we identified from bureau
records for the 1998 dress rehearsal were about $11 million budgeted
and about $9 million obligated.
Due to limitations in the bureau‘s available financial data, our
responses to the seven specific questions identified in your request do
not include all ICM/A.C.E. program life cycle costs and are limited to
cost information covering fiscal years 1996 through 2003, except where
indicated.
The Department of Commerce expressed disagreement with how we had
presented answers to the seven questions, but did not comment on the
substance of our answers. Regarding its concern that our report implied
managerial or reporting weaknesses, our answers were not prepared with
the intent of drawing conclusions beyond the information presented. We
have addressed the comments in the ’Agency Comments and Our Evaluation“
section of this report and the full text of the comments and our
detailed evaluation is presented in appendix II.
Background:
Early in the 2000 Census cycle, the U.S. Census Bureau was researching
coverage measurement options for the 2000 Census, including the Post
Enumeration Survey (PES) methods used in past decennial censuses. The
bureau explored a number of design options aimed at improving data
accuracy while controlling costs. In 1993, the bureau was also
evaluating the feasibility of conducting a one-number census, which
combines the features of both the traditional head count and
statistical methods to produce a single count before the mandated
deadlines.In May 1995, the bureau announced that it would conduct a
sample survey of 750,000 housing units, called Integrated Coverage
Measurement (ICM), to estimate how many housing units and people it
would miss or count more than once in the 2000 Census.[Footnote 3] In
this initial design for the 2000 Census, the bureau planned to use
statistical methods to integrate the results of this survey with the
traditional census enumeration to provide a one-number census by
December 31, 2000. The U.S. Supreme Court ruled in January 1999 that 13
U.S.C. 195 prohibited the use of statistical sampling to generate
population data for reapportioning the U.S. House of
Representatives.[Footnote 4] However, the court‘s ruling did not
prohibit the use of statistical sampling for other purposes, such as
adjusting formulas to distribute billions of dollars of federal funding
to state and local governments.
Following the Supreme Court ruling, the bureau abandoned certain
statistical aspects of the ICM program, and announced the A.C.E.
program to assess the quality of the population data collected in the
2000 Census, using a smaller sample of 300,000 housing units.[Footnote
5] The bureau conducted A.C.E., which corresponded to the PES in past
censuses and the ICM in the original 2000 Census Plan, to measure and
correct the overall and differential coverage of the U.S. resident
population in the 2000 Census. Although A.C.E. was generally
implemented as planned, the bureau found that A.C.E. overstated the
census net undercount. This was due, in part, to errors introduced
during matching operations and from other remaining uncertainties. The
bureau has reported that additional review and analysis would be
necessary before any potential uses of A.C.E. data could be
considered.[Footnote 6]
Due to uncertainties or errors in the A.C.E. survey results, the acting
director of the bureau decided in separate decisions in March 2001 and
October 2001 that the 2000 Census tabulations would not be adjusted for
any purpose, including distribution of billions of dollars in federal
funding. These decisions were consistent with those for the 1990
Census, which was not adjusted due to other problems. According to
senior bureau officials, the bureau is continuing to evaluate issues
related to A.C.E. and the census, and the results of its evaluation are
expected to influence the bureau‘s planning for the 2010 Census.
Bureau Appropriations and Accounting:
The bureau receives two appropriations from the Congress: (1) salaries
and expenses and (2) periodic censuses and programs. The salaries and
expenses appropriation provides 1-year funding for a broad range of
economic, demographic, and social statistical programs. The periodic
censuses and programs appropriation includes primarily no-year
funding[Footnote 7] to plan, conduct, and analyze the decennial
censuses every decade and for other authorized periodic activities.
Since fiscal year 1996, the bureau has prepared its annual budget
request for the 2000 Census in eight broad frameworks of effort that
were submitted to the Office of Management and Budget (OMB) and the
Congress. For management, program, financial, staffing, and performance
purposes, frameworks are further divided by the bureau into activities
and then projects within these activities.[Footnote 8]
The bureau accounts for the costs of conducting the ICM/A.C.E. programs
in its Commerce Administrative Management System (CAMS), which became
operational in fiscal year 1997. Bureau financial management reports
generated by CAMS have provided appropriated amounts, expended and
obligated amounts, and variances to a project level from fiscal year
1997 to the current period. The ICM/A.C.E. programs are an activity
comprised of eight projects contained within three frameworks. Fiscal
year 1996 was the first year the bureau set up a specific project code
to identify ICM program costs through fiscal year 1999. However, it was
difficult to identify the change to the A.C.E. program beginning in
fiscal year 2000 because the bureau did not change many of the project
descriptions in CAMS from the ICM program. As discussed in our December
2001 report, we identified specific control weaknesses for fiscal year
2000 related to the lack of controls over financial reporting and
financial management systems.[Footnote 9]
Scope and Methodology:
To meet the objective of responding to seven questions concerning
ICM/A.C.E. program life cycle costs, we reviewed and analyzed budget
and program data for all coverage measurement programs that existed
during the 2000 Census (for fiscal years 1991 to 2003), which included
the ICM and A.C.E. programs. We did not audit budget and other
financial data provided by the bureau. We also reviewed planning and
methodology documents and other available information in order to
determine the history of the programs. Also, we identified ICM and
A.C.E. project accounts and analyzed amounts by fiscal year using the
financial management reports generated by CAMS. We discussed the
results of our analysis with senior bureau officials and interviewed
bureau officials to obtain their views and observations regarding the
ICM and A.C.E. programs. It was not our objective to assess the
efficiency of expenditures and obligations against planned budget
appropriations.
We encountered several limitations in the scope of our work on this
assignment as follows.
* We were unable to determine the complete life cycle costs of the
ICM/A.C.E. programs because the bureau considered any ICM/A.C.E.related
costs from fiscal years 1991 through 1995 as part of its general
research and development programs and thus did not separately track
these costs. Although some costs were tracked in fiscal year 1996, the
bureau still considered these costs as research and development and did
not include these costs as ICM/A.C.E. program costs.
* We were further unable to identify ICM/A.C.E. portions of costs, such
as evaluations and data processing, which the bureau included with
other 2000 Census programs.
Our work was performed in Washington, D.C. and at U.S. Census Bureau
headquarters in Suitland, Maryland, from February 2002 through July
2002. Our work was done in accordance with U.S. generally accepted
government auditing standards. On November 17, 2002, the Department of
Commerce provided written comments on a draft of this report and we
have reprinted the comments in appendix II. Technical comments were
also provided by the department and incorporated into the report where
appropriate, but have not been reprinted.
Life Cycle Costs:
Although the bureau tracked some costs of conducting the ICM/A.C.E.
programs, we found that the bureau did not identify the complete life
cycle costs of the programs due to the following three factors.
First, the bureau only tracked the costs of conducting the ICM/A.C.E.
programs, which covers the period from fiscal year 1997 through 2003.
Although life cycle costs for the 2000 Census cover a 13-year period
from fiscal years 1991 through 2003, senior bureau officials said that
the ICM/A.C.E. program was not viable for implementation until fiscal
year 1997. Therefore, the bureau considered costs from earlier years as
part of its general research and development programs and the bureau
did not assign unique project codes to identify ICM/A.C.E. programs and
related costs in its financial management system.
Second, although $3.6 million of fiscal year 1996 obligated costs were
identifiable in the bureau‘s financial management system as an ICM
special test, the bureau did not consider these costs as part of the
ICM/A.C.E. programs. Instead, these costs were considered general
research and development. However, because the bureau separately
identified these costs as ICM program costs, we have included the $3.6
million as part of the ICM/A.C.E. program costs we could identify in
this report.
Finally, we were unable to identify the ICM/A.C.E. portions of costs,
such as evaluations and data processing, which the bureau included with
other 2000 Census programs. For example, in late fiscal year 2000 and
after, the bureau did not separate A.C.E. evaluations from its other
2000 Census evaluations in its financial management system. Bureau
officials stated that the contracts for evaluations included overall
2000 Census and A.C.E. evaluations, and did not have a separate code
identifying A.C.E. costs. Similarly, the bureau did not capture all
costs for items such as data processing by programs like ICM/A.C.E.
These type of operations were conducted for the 2000 Census overall,
were budgeted by framework, were not separated by program in the
bureau‘s financial management system, and were not allocated back to
individual projects. Therefore, we were unable to identify these types
of costs for the ICM/A.C.E. programs.
Seven Issues Related To ICM/A.C.E. Programs:
Due to the limitations in the bureau‘s data, our responses to the seven
specific questions identified in your request do not include all
ICM/A.C.E. life cycle costs and are limited to available cost
information covering fiscal years 1996 through 2003, except where
indicated, and exclude such costs as A.C.E. evaluations and some data
processing. The following sections include our responses to the seven
questions on ICM/A.C.E. program life cycle costs.
1. What were the original estimated life cycle costs for the ICM/A.C.E.
programs?
The bureau originally estimated the costs of conducting the ICM program
to be about $400 million when it planned to use statistical methods to
integrate the results of a survey based on 750,000 housing units with
the traditional census enumeration to provide a one-number census. This
original estimate included fiscal years 1997 through 2003. However,
this estimate was incomplete, as the bureau did not include program
costs prior to fiscal year 1997 because it considered them as general
research and development costs. The bureau also combined costs for
A.C.E. evaluation and data processing with other program costs in
different frameworks.
The U.S. Supreme Court ruled in January 1999 that statistical sampling
could not be used to generate population data for reapportioning the
House of Representatives. As a result of the ruling, in June 1999, as
part of its amended fiscal year 2000 budget request, the bureau
decreased the ICM/A.C.E. program by about $214 million, due to a
reduction in the sample size from 750,000 to 300,000 housing units. We
could not identify from bureau records an original estimate for only
A.C.E. life cycle costs for completing the scaled-back survey.
2. What was the source and support for $400 million in life cycle costs
reported by the bureau for the ICM/A.C.E. programs?
In 1995, the bureau estimated life cycle costs for the 2000 Census in
13 frameworks; however, bureau documents did not break out the
frameworks into activities and projects.[Footnote 10] The first
evidence for the $400 million cost estimate for conducting the
ICM/A.C.E. program for the 2000 Census was submitted as part of the
original fiscal year 2000 budget justification for overall census
operations to the Congress in February 1999. This original budget was
prepared based on the initial design for ICM, which planned to
incorporate statistical methods to integrate the results of a survey
based on 750,000 housing units with the traditional census enumeration
to provide a one-number census.
3. How were ICM/A.C.E. program costs estimated?
According to bureau officials, estimates of ICM/A.C.E. costs are based
on assumptions about the needs for headquarters and support staff and
related benefits, contractual services, travel, office space, and
equipment costs necessary to conduct and support operations of the
program. The bureau used an electronic cost model to calculate many of
the estimates for the ICM/A.C.E. programs.
For personnel costs, the A.C.E. program costs were divided into costs
for data collection and costs for headquarters full-time equivalent
(FTE) staff and support staff as follows.
* The A.C.E. field staff needed to conduct each A.C.E. data collection
operation included enumerators, crew leaders, field operations
supervisors, and assistants. The cost model was designed to estimate
the number of field staff positions, hours, FTEs, salary costs, and
mileage costs. In the cost model, each operation had its own distinct
production assumptions based on the data collection needs for that
operation. Based on operational needs, the bureau determined the
assumptions for production rates, mileage rates, production and
training days, and hours worked per day.
* The magnitude of the A.C.E field production labor was determined by
the A.C.E. field operation workload. Based on the workload, the number
of A.C.E. enumerators was calculated for each operation. Then, based on
the number of enumerators, the bureau determined the number of crew
leaders, field operations supervisors, and assistants needed. The
number of production positions became the bureau‘s basis for the number
of staff to be trained. The number of positions, both production and
trainee, was then used to estimate the salary cost as a function of the
total production and trainee hours and applicable labor rates.
* Once labor rates were determined, a percentage was used to calculate
benefit costs.
For nonpersonnel costs, the bureau estimated the costs based on the
following.
* Contract costs were estimated based upon procurement needs for goods
and services, including contractors hired to assess the feasibility of
A.C.E. operations and to evaluate the results of the program.
* Travel costs were estimated using the numbers of production and
trainee positions to calculate the average miles per case and the
mileage reimbursement rate.
* Office space estimates were based on the number of people who needed
space, the number of square feet per person, and the cost per square
foot.
* Equipment and supply costs were based on the needs of each employee
and the specific needs of each A.C.E. operation. This included laptop
computers that were provided to field data collection staff to conduct
interviews and to monitor the operational progress of the program.
4. How much did Census budget for the ICM/A.C.E. programs?
As shown in figure 1, we identified from bureau records budgeted
amounts of $276.5 million for conducting the ICM/A.C.E. programs. Of
this amount, $64.2 million was for the ICM program from fiscal year
1996 through 1999, and $212.3 million was for the A.C.E. program from
fiscal year 2000 through 2003. Also, see table 1 in appendix I for
additional details of ICM/A.C.E. budgeted costs by framework and
project.
Figure 1: Budgeted Costs of ICM/A.C.E. Programs (Dollars in Millions):
This figure is a pie chart showing budgeted costs of ICM/A.C.E.
programs (dollars in millions).
A.C.E. (fiscal years 2000-2003): $212.3: 77%;
ICM (fiscal years 1996-1999): $64.2: 23%;
Total budgeted costs $276.5.
[See PDF for image]
Source: GAO analysis of unaudited bureau data.
[End of figure]
5. What were the obligated life cycle costs for the ICM/A.C.E.
programs?
As shown in figure 2, we identified from bureau records obligated
amounts of $206.9 million, of which $58.4 million was for the ICM
program from fiscal year 1996 through 1999, and $148.5 million was for
the A.C.E. program for fiscal years 2000 and 2001. We did not include
obligated costs for fiscal year 2002 as they are not yet final and
fiscal year 2003 obligations have yet to be incurred. Also, see table 2
in appendix I for additional details of obligated costs for the
ICM/A.C.E. programs.
Figure 2: Obligated Costs of ICM/A.C.E. Programs (Dollars in millions):
This figure is a pie chart showing obligated costs of ICM/A.C.E.
programs.
A.C.E. (fiscal years 2000-2001): $148.5: 72%;
ICM (fiscal years 1996-1999): $58.4: 28%;
Total obligated cots: $206.9.
[See PDF for image]
Source: GAO analysis of unaudited bureau data.
[End of figure]
As shown in figure 3, of the $206.9 million of obligated ICM/A.C.E.
program costs through fiscal year 2001, 65 percent or about $135
million were for salaries and benefits. The next largest category was
for contractual services, which constituted about $22.3 million or 11
percent of ICM/A.C.E. program costs. The third largest category was for
equipment, which constituted about $22 million, or 11 percent of
ICM/A.C.E. program costs. Other costs - including office space, travel,
and supplies - made up about $27.6 million or 13 percent of program
costs.
Figure 3: Obligated Costs of ICM/A.C.E. Programs by Category (Dollars
in Millions):
This figure is a pie chart showing obligated costs of ICM/A.C.E.
programs by category (dollars in millions).
Salaries and benefits: $135.0;
Other: $27.60;
Contractual services: $22.30;
Equipment: $22.0;
Total obligated costs: 206.9.
[See PDF for image]
Source: GAO analysis of unaudited bureau data.
[End of figure]
6. Were there any budgeted funds for the ICM/A.C.E. programs not used
as of the end of fiscal year 2001, and if so, how much?
About $57.7 million of budgeted funds that we identified from bureau
records for the ICM/A.C.E. programs were not obligated through fiscal
year 2001.
* For fiscal years 1996 and 1997, there were no unused funds for the
ICM program.
* For fiscal year 1998, about $2.7 million remained unobligated for the
ICM program because of the following reasons.
- $1.5 million was due to the dress rehearsal housing unit follow-up
workload being smaller than anticipated; a bureau bonus
program[Footnote 11] not being implemented although budgeted; and less
mileage reimbursement than budgeted under project code 6205 (ICM Dress
Rehearsal).$400,000 was due to unused budgeted funds for salaries
related to project code 6352 (ICM Coverage Measurement).
- $400,000 was due to unused budgeted funds for salaries and a delay in
awarding contract services under project code 6444 (ICM Procedures and
Training).
- $400,000 was due to unused budgeted funds for regional office manager
and assistant manager salaries and travel costs due to delays in hiring
under project code 6480 (ICM Collection).[Footnote 12]
* For fiscal year 1999, about $3.6 million budgeted for the ICM program
remained unobligated due to the following reasons.
- About $2.3 million related to project code 6480 (ICM Collection) was
not used, including $1.6 million due to unspent salaries related to
hiring delays, hiring fewer staff than authorized for selected
positions, and hiring qualified candidates at less than budgeted
levels. Another $0.7 million was due to less mileage reimbursement than
budgeted.
- About $1.2 million was due to equipment costs and hardware for 2000
being less than budgeted under project code 6608 (ICM Processing).
* For fiscal year 2000, about $42.5 million remained unobligated for
the A.C.E. program, consisting of almost $40 million for project code
6480 (A.C.E. Collection), which was budgeted for the program but was
not used primarily because of the following reasons.
- About $32 million was due to unspent salaries and benefits for office
staff in field offices from hiring fewer positions and hiring at lower
grades than budgeted and from lower data collection costs due to a
reduction in cases requiring personal visits.
- About $4 million was due to the fact that contract obligations for
laptop computers and support services were less than budgeted.
- About $2 million resulted from lower GSA rents than budgeted.
- For fiscal year 2001, about $8.9 million of unobligated funds
remained for the A.C.E. program, consisting mostly of $6.4 million for
project code 6480 (A.C.E., Collection), which was budgeted for the
program but was not used.
7. What were the ICM/A.C.E. program-related costs for the bureau dress
rehearsal in fiscal year 1998?
As shown in appendix I, the ICM program-related costs for the 1998
dress rehearsal were captured under project code 6205 (ICM Dress
Rehearsal). Of the total $10.8 million budgeted, we were able to
identify obligations of $9.4 million from bureau records. Most of these
obligations were incurred in fiscal year 1998, with some follow-up
amounts in the first quarter of fiscal year 1999. According to bureau
officials, the dress rehearsal project activities included data
collection and case management, data processing, and implementation of
estimation operations. This project also covered the implementation of
ICM and some elements of A.C.E. at the Sacramento, California, and
Menominee County, Wisconsin dress rehearsal sites.
Agency Comments and Our Evaluation:
The Department of Commerce comments expressed disagreement with how we
presented answers to the seven questions in the report, but did not
comment on the substance of our answers. It said that our report‘s
conclusions imply financial management or reporting failures and
suggest specific control weaknesses in the bureau‘s financial
management systems. It also said we inferred an inability to properly
manage from large unexplained discrepancies between budgeted and
obligated amounts for the ICM/A.C.E. programs. Our answers were not
prepared with the intent of drawing conclusions beyond the information
presented and we did not make interpretive conclusions or qualitative
judgments about the ICM/A.C.E. programs. Although not within the scope
of this report, our December 2001 report identified internal control
weaknesses for fiscal year 2000 related to the bureau‘s lack of
controls over financial reporting and financial management
systems.[Footnote 13]
The department‘s written comments and our more detailed evaluation of
its concerns are presented in appendix II.
As agreed with your offices, unless you announce its contents earlier,
we plan no further distribution of this report until 7 days after its
issuance date. At that time, we will send copies to the Chairman and
Ranking Minority Member of the Senate Committee on Governmental
Affairs, the House Committee on Government Reform, and the House
Subcommittee on Civil Service, Census, and Agency Organization. We will
also send copies to the Director of the U.S. Census Bureau, the
Secretary of Commerce, the Director of the Office of Management and
Budget, the Secretary of the Treasury, and other interested parties.
This report will also be available on GAO‘s home page at [hyperlink,
http://www.gao.gov].
If you or your staffs have any questions concerning this report, please
contact me at (202) 512-9095 or by e-mail at kutzg@gao.gov or Roger R.
Stoltz, Assistant Director, at (202) 512-9408 or by e-mail at
stoltzr@gao.gov. A key contributor to this report was Cindy Brown-
Barnes.
Signed by:
Gregory D. Kutz:
Director:
Financial Management and Assurance:
Appendix I: Budgeted and Obligated Life Cycle Costs by Framework and
Project for Fiscal Years 1996 through 2003:
Table 1: Census 2000 ICM/A.C.E. Program Budgeted Life Cycle Costs for
Fiscal Years 1996 through 2003:
[See PDF for image]
Note: ICM projects became A.C.E. projects in fiscal year 2000 and
thereafter.
[A] This full year original cost estimate of the A.C.E. program was
developed around May 2001. On June 12, 2002, a senior official told us
that the bureau was revising this amount to about $2 million to cover
the period October 1 through December 31, 2002, when A.C.E. program
will end.
Source: GAO compiled from U.S. Census Bureau financial management
records.
[End of table]
Table 2: Census 2000 ICM/A.C.E. Program Obligated Life Cycle Costs for
Fiscal Years 1996 through 2001:
[See PDF for image]
Source: GAO compiled from U.S. Census Bureau financial management
records.
Note: ICM projects became A.C.E. projects in fiscal year 2000 and
thereafter.
[End of table]
Appendix II: Comments from the Department of Commerce:
The Secretary of Commerce:
Washington, D.C. 20230:
October 17, 2002:
Mr. Gregory D. Kutz:
Director, Financial Management and Assurance:
U.S. General Accounting Office:
Washington, D.C. 20548:
Dear Mr. Kutz:
The Department of Commerce appreciates the opportunity to comment on
the General Accounting Office draft report entitled "2000 Census:
Complete Costs of Coverage Evaluation Programs Are Not Available." The
Department's commerce on this report are enclosed.
Sincerely,
Signed by:
Donald L. Evans:
Enclosure:
Comments from the U.S. Department of Commerce:
U.S. Census Bureau:
U.S. General Accounting Office draft report entitled 2000 Census:
Complete Costs of Coverage Evaluation Programs Are Not Available:
General Comments on the Board:
The U.S. Census Bureau takes exception with the central conclusions of
this report, as they imply that large sums of money associated with the
Integrated Coverage Measurement (ICM) and Accuracy and Coverage
Evaluation (A.C.E.) Programs for Census 2000 were not properly reported
as being spent on those programs. Moreover, the General Accounting
Office (GAO) suggest that the Census Bureau was unable to properly
estimate full-cycle costs for these programs that this inability,
coupled with the fact that there was almost $5 million in budgeted but
unobligated funds for ICM/A.C.E. programs during FY 1998-2001,
indicates that there were specific control weaknesses in the Census
Bureau's financial management systems. The GAO's apparent remedy for
this would be for the Census Bureau to allocate ICM/A.C.E. program
expenditures to codes established for this program at an earlier point
in the census cycle.
(See Comment 1):
The Census Bureau respectfully disagrees with this position. We believe
that we have accurately captured the ICM/A.C.E. costs attributable to
Census 2000 in our financial planning. Prior to 1996, the Census Bureau
had not sufficiently defined the specific coverage measurement program
that ultimately became the ICM. Consequently, the Census Bureau did not
allocate any expenditures to ICM project codes. This is in keeping with
Census Bureau procedures regarding a wide range of decennial census
operations. It underscores an important point: namely, that the GAO's
inability to identify portions of the costs attributable to ICM/A.C.E.
program is not demonstrative of a financial or reporting failure; it
simply indicates that the Census Bureau's system and categorization of
Census 2000 expenditures do not permit the parsing out of the specific
breakdowns requested.
(See Comment 2):
To be sure, the dual system estimation (DSE) methodology cental to the
ICM/A.C.E. program was under development. In fact, the Census Bureau
had been developing DSE methodology and other sampling methodologies
for many years. But it was logical for us to allocate expenditures
associated with this and other decennial programs to a general research
and development activity code early in the decade when we were
developing and testing decennial census operations. Costs for the ICM
Special Test in 1996, which the GAO specifically addresses, also should
be understood in the context of the Census Bureau's ongoing development
of DSE procedures. It was only after Census 2000 operations were
finalized in 1997 that a specific code was established for the ICM
program as a component of Census 2000. It is, therefore, misleading for
the GAO to include the costs for the development of DSE methodologies
early in the decade in reporting the full-cycle costs for the
ICM/A.C.E. programs.
The GAO also cites discrepancies between the $400 million cost estimate
provided in early 1999 and the budgeted amount totaling roughly $277
million, both of which were calculated through FY 2003. Following the
January 1999 Supreme Court riling prohibiting the use of sampling to
produce the apportionment counts, the Census Bureau revised its plans
for its coverage measurement program. However, the Census Bureau had
already prepared its FY 2000 budget request for Census 2000 activities,
which included a cost estimate of $400 million for the ICM program for
FY 1997-2003. The Census Bureau's revised plans called for a coverage-
measurement survey-the A.C.E.-of less than half the size of the ICM.
This major revision to the Census Bureau's coverage measurement program
explains the large discrepancy between the $400 million cost estimate
provided in early 1999 and the budgeted accounts for totaling roughly
$277 million.[Footnote 14] The Census Bureau subsequently revised its
budget request for Census 2000, based on the requirements for the
A.C.E. program.
(See Comment 3):
In addition, the GAO reports both budgeted and obligated full-cycle
costs for the ICM/A.C.E. programs. The obligated full-cycle cost total
of approximately $207 million covers the period FY 1996-2001, given
that obligation for FY 2002 and FY 2003 are not yet known. The budgeted
full-cycle cost total of approximately $277 million covers the period
FY 1996-2003, although the FY 2003 amount has been revised recently.
The GAO suggested that the difference between these two totals
demonstrates the Census Bureau's inability to properly manage and
record expenditures relating to the ICM/A.C.E. programs. Again, we
disagree with the GAO's interference. First of all, the totals cover
two different time periods, with budgeted costs of these programs for
FY 2002 and FY 2003 totaling $12.5 million. Much of the remainder of
unobligated funds (approximately $58 million) can be attributed to the
$42.5 million in unobligated funds from FY 200 alone. While the Census
Bureau had made reductions to its budgeted amounts for the A.C.E.
Collection based on the smaller sample size of the A.C.E., the conducts
of the A.C.E. data collection operations in FY 2000. This result is
consistent with what was achieved for similar Census 2000 field
operations conducted during FY 2000.
(See Comment 4):
Finally, the GAO notes that ICM/A.C.E. portions of costs relating to
data processing and evaluations are not captured in the full-cycle cost
calculations. We are not aware of any post-FY 1995 data processing
costs related to ICM/A.C.E. programs that are not included in the
project codes that the GAO used to calculate the full-cycle costs. With
regard to evaluations, the GAO has correctly stated that there is not a
separate project code for evaluations relating to the ICM/A.C.E.
programs. It is not possible to parse out the total costs of the
evaluations relating to these programs, because evaluations of all
Census 2000 programs and operations were assigned the same project
code. Once again, the inability to identify the portions of these costs
attributable to ICM/A.C.E. programs is not demonstrative of a financial
management or reporting failure, it simply indicates that the Census
Bureau's system and categorization of Census 2000 expenditures do not
permit the parsing out of the specific breakdowns requested. If it were
possible to discern thee costs, based on our experience, we believe
they would prove to be a relatively small percentage of the full-cycle
costs of the ICM/A.C.E. program.
(See Comment 5):
The following are GAO‘s comments on the letter dated October 17, 2002,
from the Department of Commerce.
GAO Comments:
1. Our report does not make interpretive conclusions or qualitative
judgments about the ICM/A.C.E. programs. With bureau assistance, we
compiled unaudited budgeted and obligated amounts for projects that the
bureau reported in its financial management system as being ICM/A.C.E.
related. Our review of these reported costs indicated that life cycle
costs of the ICM/A.C.E. programs were not complete due to three factors
as discussed in the body of our report. One of the factors we cited
that contributed to incomplete life cycle costs was $3.6 million of
fiscal year 1996 obligated costs for an ICM special test. In its
comments, the bureau pointed out that prior to fiscal year 1996 it had
not defined the coverage measurement program, did not allocate any
expenditures to the ICM project codes, and could not identify any costs
prior to fiscal year 1996. Thus, it was the bureau‘s decision to not
track specific costs during this time period and to consider them as
general research.
We also stated that $57.7 million of budgeted funds were not obligated
or spent through fiscal year 2001, and, with input from bureau
officials, we obtained reasons why these funds were not spent. The
bureau did not take exception to these facts in its response and we
noted no improprieties in this report.
Regarding a reference to specific control weaknesses in its financial
management systems, the scope of this report did not include an
assessment of internal control weaknesses in the bureau‘s financial
management systems. However, in a December 2001 report, we identified
specific internal control weaknesses for fiscal year 2000 related to
the bureau‘s lack of controls over financial reporting and financial
management systems.14
2. We still disagree with the bureau on this point, as we stated in the
draft report. Because these costs were separately tracked by a specific
ICM project code in the bureau‘s financial management system, we
included them in the costs of the ICM/A.C.E. programs that we could
identify.
3. We did not cite discrepancies between the $400 million original cost
estimate of the ICM/A.C.E. programs provided in early 1999 and the $277
million budgeted amount we identified for fiscal years 1996 through
2003. An objective of our report was to determine what were the
original estimated life cycle costs for the ICM/A.C.E. programs. The
earliest amount that we could identify from bureau records was $400
million and in our report we explained that this amount was estimated
by the bureau before the January 1999 Supreme Court decision. As a
result of this decision and as disclosed in our report, the bureau
decreased the ICM/A.C.E. program by about $214 million due to a
reduction in the sample size from 750,000 to 300,000 housing units.
4. We did not suggest that the difference between $277 million of
budgeted life cycle costs and $207 million of obligated life cycle
costs demonstrated the bureau‘s inability to properly manage and record
expenditures relating to the ICM/A.C.E. programs. As presented in our
report, the budgeted amount of $277 million included fiscal years 1996
through 2003 and the obligated amount of $207 million included amounts
for 2 fewer fiscal years (1996 through 2001). As the bureau pointed out
in its response, it is too soon to determine obligated amounts for
fiscal years 2002 and 2003 that were budgeted for $12.5 million.
Variances for the remaining $57.7 million of unspent funds are
discussed in comment 1.
5. The bureau agreed that it did not capture the life cycle costs of
evaluations for the ICM/A.C.E. programs because evaluations for all
2000 Census programs were charged to one project code. However, the
bureau believes that data processing costs were included in the life
cycle costs of the ICM/A.C.E. programs and stated that not being able
to identify portions of these costs is not demonstrative of a financial
management or reporting failure. We agree with the bureau that some
data processing costs were captured in the life cycle costs of the
ICM/A.C.E. programs as evidenced by project codes for ICM/A.C.E. data
processing for procedures, training, and processing as part of
Framework 5. However, we do not believe that all data processing costs
were included. Similar to evaluation costs, the bureau attributed much
of its computer hardware and support costs to overall 2000 Census
programs, and did not allocate costs to specific projects or programs.
[End of section]
Footnotes:
[1] Proposed fiscal year 2003 funding was included in the President‘s
Budget in February 2002, but has not yet been approved by the Congress.
[2] Since fiscal year 1996, the bureau has received its appropriation
based on eight broad frameworks of effort and then establishes budgets
for individual activities and projects within frameworks.
[3] U.S. Census Bureau, The Reengineered 2000 Census (Suitland, MD: May
19, 1995).
[4] Department of Commerce v. United States House of Representatives,
525 U.S. 316 (1999).
[5] U.S. Census Bureau, Updated Summary Census 2000 Operational Plan
(Suitland, MD: Feb. 24, 1999).
[6] U.S. General Accounting Office, 2000 Census: Coverage Evaluation
Matching Implemented as Planned, but Census Bureau Should Evaluate
Lessons Learned, GAO-02- 297 (Washington, D.C.: Mar. 14, 2002).
[7] No-year funds are available for their original purpose until they
are either expended, rescinded, transferred, or reprogrammed, or the
account is closed.
[8] For example, for fiscal year 2000, the bureau used 8 frameworks, 23
activities, and 119 projects.
[9] U.S. General Accounting Office, 2000 Census: Analysis of Fiscal
Year 2000 Budget and Internal Control Weaknesses at the U.S. Census
Bureau, GAO-02-30 (Washington, D.C.: Dec. 28, 2001).
[10] U.S. Census Bureau, The Reengineered 2000 Census (Suitland, MD:
May 19, 1995).
[11] This program was intended to provide certain temporary employees a
cash incentive to stay at the bureau until the end of 2000 Census
operations.
[12] This project covers the implementation of certain ICM/A.C.E. field
operations and support activities.
[13] GAP-02-30.
[14] This amount includes the $3.3 million for the ICm Special Test
that the Census Bureau does not include in its calculations.
[15] GAO-02-30.
GAO‘s Mission:
The General Accounting Office, the investigative arm of Congress,
exists to support Congress in meeting its constitutional
responsibilities and to help improve the performance and accountability
of the federal government for the American people. GAO examines the use
of public funds; evaluates federal programs and policies; and provides
analyses, recommendations, and other assistance to help Congress make
informed oversight, policy, and funding decisions. GAO‘s commitment to
good government is reflected in its core values of accountability,
integrity, and reliability.
Obtaining Copies of GAO Reports and Testimony:
The fastest and easiest way to obtain copies of GAO documents at no
cost is through the Internet. GAO‘s Web site [hyperklink,
http://www.gao.gov] contains abstracts and full text files of current
reports and testimony and an expanding archive of older products. The
Web site features a search engine to help you locate documents using
key words and phrases. You can print these documents in their entirety,
including charts and other graphics.
Each day, GAO issues a list of newly released reports, testimony, and
correspondence. GAO posts this list, known as ’Today‘s Reports,“ on its
Web site daily. The list contains links to the full-text document
files. To have GAO e-mail this list to you every afternoon, go to
[hyperlink, http://www.gao.gov] and select ’Subscribe to daily E-mail
alert for newly released products“ under the GAO Reports Order GAO
Products heading.
Order by Mail or Phone:
The first copy of each printed report is free. Additional copies are $2
each. A check or money order should be made out to the Superintendent
of Documents. GAO also accepts VISA and Mastercard. Orders for 100 or
more copies mailed to a single address are discounted 25 percent.
Orders should be sent to:
U.S. General Accounting Office:
441 G Street NW, Room LM:
Washington, D.C. 20548:
To order by Phone:
Voice: (202) 512-6000:
TDD: (202) 512-2537
Fax: (202) 512-6061
To Report Fraud, Waste, and Abuse in Federal Programs Contact:
Web site: [hyperlink, http://www.gao.gov/fraudnet/fraudnet.htm]:
E-mail: fraudnet@gao.gov:
Automated answering system: (800) 424-5454 or (202) 512-7470:
Public Affairs:
Jeff Nelligan, managing director, NelliganJ@gao.gov:
(202) 512-4800:
U.S. General Accounting Office:
441 G Street NW, Room 7149:
Washington, D.C. 20548:
