Indian Trust Funds: Improvements Made in Acquisition of New Asset and
Accounting System But Significant Risks Remain (Letter Report,
09/15/2000, GAO/AIMD-00-259).

Pursuant to a congressional request, GAO evaluated the Department of the
Interior's effort to acquire and develop its new Trust Asset and
Accounting Management System (TAAMS), focusing on: (1) actions Interior
has taken in response to GAO's previous reviews and recommendations; and
(2) the business and technical risks and challenges still confronting
Interior as it begins to deploy TAAMS.

GAO noted that: (1) Interior is taking positive actions to address
several of the concerns raised in GAO's 1999 reports as well as
additional concerns raised in GAO's subsequent briefings on TAAMS
testing efforts; (2) Interior has taken actions to strengthen its
testing processes; (3) Interior is beginning to develop an information
systems architecture for Indian trust management which is needed to
ensure that trust fund and related systems are interoperable, function
together efficiently, and are cost-effective over their life cycles; (4)
these actions are good steps toward enhancing management of the TAAMS
effort; (5) however, there are a number of major business and technical
challenges that still put the TAAMS effort at considerable risk; (6) as
to the business challenges, Interior has not yet completed actions
designed to enhance overall trust fund management, including its efforts
to revamp policies and procedures for the entire trust management cycle
and to address long-standing internal control weaknesses; (7) both
efforts could have a significant impact on TAAMS because they may
require new features to be built into the system or current capabilities
to be modified and, in turn, increase the risk of introducing new system
defects; (8) fixing such defects late in the development effort can be
very costly; (9) in addition, Interior has not reengineered the business
processes that TAAMS is to support even though these processes were
designed in a very different system environment; (10) without taking
time now to examine and revise its business processes, Interior will not
be able to maximize the benefits that can be gained from TAAMS and it
may perpetuate outmoded ways of doing business; (11) on the technical
side, there are actions underway to strengthen Interior's management of
the TAAMS project, such as hiring a contractor to evaluate data
integrity and the adoption of more disciplined testing processes; and
(12) there are challenging undertakings still ahead.

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

 REPORTNUM:  AIMD-00-259
     TITLE:  Indian Trust Funds: Improvements Made in Acquisition of
	     New Asset and Accounting System But Significant Risks
	     Remain
      DATE:  09/15/2000
   SUBJECT:  Trust funds
	     Systems design
	     Native Americans
	     Information resources management
	     Indian lands
	     Internal controls
	     Funds management
	     ADP procurement
	     Federal agency accounting systems
	     Reengineering (management)
IDENTIFIER:  Dept. of the Interior Trust Asset and Accounting
	     Management System
	     BIA Land Records Information System
	     BIA Integrated Records Management System

******************************************************************
** 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/AIMD-00-259

Accounting and Information
Management Division

B-284142

September 15, 2000

The Honorable Ben Nighthorse Campbell
Chairman
Committee on Indian Affairs
United States Senate

The Honorable Slade Gorton
Chairman, Subcommittee on Interior and Related Agencies
Committee on Appropriations
United States Senate

This report responds to your request that we evaluate the Department of the
Interior's effort to acquire and develop its new Trust Asset and Accounting
Management System (TAAMS). TAAMS is part of a broader Interior effort to
address problems that have beset Indian trust management for many decades,
including inadequate accounting and management information systems; backlogs
in asset appraisals, ownership determination, and recordkeeping; and poor
internal controls. TAAMS itself is being developed to address deficiencies
in the recording, collecting, and accounting for revenues related to the
department's management of about 54 million acres of Indian trust land.

In April and July of 1999, we reported1 that Interior did not follow sound
acquisition management practices in the early stages of the TAAMS effort
and, as a result of its poor planning, it could not ensure that TAAMS would
cost effectively meet trust management needs or adequately mitigate
development risks. In more recent briefings to your offices, we reported
that Interior was not following sound practices in conducting system and
user acceptance tests for TAAMS. As agreed with your offices, we assessed
(1) actions Interior has taken in response to our previous reviews and
recommendations and (2) the business and technical risks and challenges
still confronting Interior as it begins to deploy TAAMS.

Interior is taking some positive actions to address several of the concerns
raised in our 1999 reports as well as additional concerns raised in our
subsequent briefings on TAAMS testing efforts. For example, Interior has
adopted a generally accepted methodology for developing a project such as
TAAMS and developed plans for mitigating risks, transitioning to the new
system, and validating and maintaining the integrity of the system
throughout its useful life. In addition, Interior has taken actions to
strengthen its testing processes; for example, by expanding tests to assess
how the system responds to unexpected conditions. Lastly, Interior is
beginning to develop an information systems architecture for Indian trust
management which is needed to ensure that trust fund and related systems are
interoperable, function together efficiently, and are cost-effective over
their life cycles.

These actions are good steps toward enhancing management of the TAAMS
effort. However, there are a number of major business and technical
challenges that still put the TAAMS effort at considerable risk. As to the
business challenges, Interior has not yet completed actions designed to
enhance overall trust fund management, including its efforts to revamp
policies and procedures for the entire trust management cycle and to address
long-standing internal control weaknesses. Both efforts could have a
significant impact on TAAMS because they may require new features to be
built into the system or current capabilities to be modified and, in turn,
increase the risk of introducing new system defects. Fixing such defects
late in the development effort can be very costly. The internal controls
effort is also critical to ensuring the accuracy and completeness of TAAMS
data on an ongoing basis. In addition, Interior has not reengineered the
business processes that TAAMS is to support even though these processes were
designed in a very different system environment. Without taking time now to
examine and revise its business processes, Interior will not be able to
maximize the benefits that can be gained from TAAMS and it may perpetuate
outmoded ways of doing business.

On the technical side, there are actions underway to strengthen Interior's
management of the TAAMS project, such as hiring a contractor to evaluate
data integrity and the adoption of more disciplined testing processes.
Nevertheless, there are challenging undertakings still ahead. For example,
Interior needs to ensure that its contractor is following disciplined
processes for developing and testing TAAMS. And, it needs to carefully
manage the addition of new TAAMS requirements as new versions of the system
are continually rolled out and tested. Moreover, without a detailed
information systems architecture to help guide the TAAMS effort, Interior
will need to carefully ensure that TAAMS and other trust fund systems are
compatible and in line with business needs.

We are making recommendations to Interior to address the business and
technical risks we identified before undertaking the second phase of the
TAAMS effort. This phase involves rolling out key trust asset and land
management functions such as lease, contract, and permit management; billing
and collection; accounts receivable; lease and contract payment
distribution; and land resource management.

In its comments on a draft of this report, the Department of the Interior
generally agreed with our assessment of the status of its efforts and of the
continuing challenges during the remainder of the TAAMS initiative.

The Secretary of the Interior is primarily responsible for administering the
government's trust responsibilities to tribes and Indians, including
managing about $3 billion in Indian trust funds and administering about
54 million acres of Indian trust lands. Management of the Indian trust funds
and assets has long been characterized by inadequate accounting and
information systems; untrained and inexperienced staff; backlogs in
appraisals, ownership determinations, and recordkeeping; the lack of a
master lease file and an accounts receivable system; inadequate written
policies and procedures; and poor internal controls.

In 1998, Interior began undertaking a major effort to resolve these
problems. This initiative now includes 11 projects. They are directed at
such things as improving systems, enhancing the accuracy and completeness of
Interior data regarding the ownership and lease of Indian lands, and
correcting deficiencies with respect to records management, training, policy
and procedures, and internal controls.

TAAMS is one of the 11 projects. TAAMS is expected to replace two legacy
systems that are currently being used to manage Indian trust assets: the
Land Records Information System (LRIS) and the Integrated Records Management
System (IRMS). LRIS supports the land title function by providing land
title-related information, such as ownership and encumbrances. It calculates
ownership interests, in fractional and decimal forms, used by Interior for
distribution of land revenue. IRMS supports the land resource management
function and is primarily used at the agency level for generating lease
bills and for income and revenue distribution to Indian owners. It contains
information on Indians, leases (for example, pasture, range, timber,
mineral, and mining leases), land ownership, oil and gas royalties, and
trust fund accounts.

According to Interior, the two mainframe-based systems are not integrated,
have no electronic interfaces, and duplicate much of the same information
(that is, information on ownership, land, and lease/encumbrances). Neither
fully or adequately supports all of the activities of land title and
resource management functions. And, in some regions, Interior has found that
they are not even being used to support the leasing and distribution
process.

The TAAMS strategy has evolved considerably since 1998. For example,
Interior originally envisioned that LRIS functions would not be part of
TAAMS. However, subsequent to its original improvement plan, Interior
decided that they should be integrated into TAAMS. Also, Interior is making
many more modifications to the commercial-off-the-shelf (COTS) software
package supporting TAAMS than originally anticipated.

Rather than acquiring, modifying, owning, and operating the COTS system
itself, Interior has hired a contractor to modify the system and to manage
it. Under this approach, Interior will provide the contractor with the land
and trust account data to be entered into the system and will perform its
trust management functions by remotely accessing contractor-provided
applications. At present, Interior expects to spend about $40 million on
TAAMS through fiscal year 2001.

Figure 1 shows the functions now planned for TAAMS, along with TAAMS planned
interaction with other Interior systems, and deployment dates. As noted in
the figure, the land title functionality of TAAMS has already been deployed
in one region.

Although TAAMS has been deployed, as of August 31, 2000, Interior did not
consider it to be the system of record.

We began our assessment of TAAMS in July 1998 while Interior was in the
process of specifying the system's functional requirements. At that time, we
found that Interior was not following accepted practices that would
(1) help ensure that TAAMS cost effectively met trust management needs and
(2) reduce development risks. For example:

ï¿½ Although Interior was acquiring information services and systems such as
TAAMS and the Trust Funds Accounting System, at a cost of about $60 million,
it had not defined an integrated architecture for Indian trust operations.
Architectures are comprehensive plans that systematically and completely
describe an organization's target business environment, both in logical
(e.g., missions, business functions, and information flows) terms and
technical (e.g., software, hardware, and communications) terms. The
Clinger-Cohen Act requires the Chief Information Officer of each major
federal agency to develop and maintain an information systems architecture.
Previous GAO reviews have shown that, without a defined architecture,
agencies are at risk of building and buying systems that are duplicative,
incompatible, and unnecessarily costly to maintain and interface.

ï¿½ Interior did not thoroughly analyze technical alternatives before choosing
a vendor to provide the asset and land records management service.
Specifically, it did not assess the desirability of satisfying its
requirements by (1) modifying legacy systems, (2) acquiring a COTS product
and using existing Interior infrastructure resources, (3) building a system
that would provide the necessary capability, and (4) acquiring a service.
The Clinger-Cohen Act of 1996 requires agencies to establish a process to
assess the value and risks of information technology investments and set
priorities for alternative projects.

ï¿½ Interior did not perform a gap analysis in surveying the availability of
COTS products. This analysis, which is an accepted practice, would
systematically and quantitatively compare and contrast COTS products against
Interior's requirements based on functional, technical, and cost
differences.

ï¿½ Interior did not require the contractor to demonstrate that the COTS
system could work with Interior-provided data or that the system could
interface with other Interior systems, which are also accepted practices.

ï¿½ Interior did not develop a risk management plan to address the possibility
that the new service would not meet performance or business requirements, be
able to work with Interior systems, and/or be delivered on schedule and
within budget.

ï¿½ Interior did not prepare a realistic project schedule for TAAMS.
Organizations following sound software acquisition practices would typically
(1) identify the specific activities that must be performed to produce the
various project deliverables, (2) identify and document dependencies, (3)
estimate the amount of time needed to complete activities, and (4) analyze
the activity sequences, durations, and resources requirements. By contrast,
Interior used the Interior Secretary's stated expectation that all Indian
trust fund-related improvements should occur within a 3-year period
beginning in 1998 as the starting point for developing the TAAMS project
schedule.

As Interior continued developing and testing TAAMS, we identified additional
problems and briefed your staffs on our observations in January and May
2000. Namely, we found that there were serious flaws in the way

Interior was planning and conducting its system tests (which verify that a
system satisfies functional requirements)2 and its first set of user
acceptance tests (which verify that the system operates correctly with
operational hardware and meets user needs).3 Without following disciplined
testing processes, Interior could not ensure the successful implementation
of TAAMS.

In particular, test plans were flawed because they were designed with the
assumption that no errors would be found. They also did not include tests of
invalid and unexpected conditions--known as boundary testing. This would
include assessing whether TAAMS would accept and process data that are
obviously erroneous, for example, future birth or death dates, invalid
fractions, incorrectly formatted social security or taxpayer identification
numbers, or tracts of land that contain "0" acres.

Furthermore, some obvious problems/defects that occurred as the tests were
conducted were ignored because testers assumed that the unanticipated
results were attributable to eccentricities or malfunctions of the computing
platform rather than to defects in the system being tested. Also, even
though the TAAMS contract requires TAAMS to comply with the Core Financial
Management System Requirements issued by the Joint Financial Management
Improvement Program (JFMIP), Interior did not use test processes already
developed by JFMIP to certify compliance with these requirements. Lastly,
test results were not thoroughly inspected and test defects were not
effectively tracked.

Interior's contract and its oversight of the contractor contributed to some
of these problems. For example, Interior did not require, and the contractor
did not provide, adequate documentation showing what boundary condition
tests had been conducted. In addition, the contract did not require the
contractor to follow disciplined development and testing processes.

An independent verification and validation review of the system and user
testing conducted for Interior by another contractor raised some of these
same concerns and additional ones, such as noting that Interior had not yet
tested whether the system would be able to handle the maximum user load or
whether it could interface with Interior's trust fund accounting and
minerals management systems.

In our April 1999 report and during our subsequent briefings, we made a
number of recommendations to the Secretary of the Interior to address the
weaknesses we identified. In particular, we recommended that Interior
develop an information systems architecture for Indian trust operations;
clearly define and validate functional requirements, security requirements,
and data requirements for TAAMS; develop and implement an effective risk
management plan; and ensure that all TAAMS project decisions are based on
objective data and demonstrated project accomplishments, and are not
schedule driven. To strengthen testing, our suggestions to Interior included
tracking defects to determine whether TAAMS is becoming stable; performing
additional boundary condition tests; thoroughly inspecting test results;
ensuring that tests are planned with the assumption that errors will be
found; and ensuring that tests plans clearly define expected results.

In conducting our review, we reviewed TAAMS project documents, including the
project implementation plan, requirements specifications, risk management
plan, and test plan and procedures. We also observed selected development
and testing activities, including user acceptance tests conducted in
February and April 2000 and system acceptance tests conducted in July,
September, and November 1999. We analyzed these plans and testing activities
using generally accepted practices and criteria, such as the Software
Engineering Institute's4 Software Acquisition Capability Maturity Model and
standards promulgated by the Institute of Electrical and Electronics
Engineers. We also discussed Interior's efforts to develop, test, and
implement TAAMS and deviations from accepted software acquisition and
development practices with the TAAMS project manager, Interior's Chief
Information Officer, and the TAAMS contractor.

In addition, we reviewed documents related to Interior's efforts to
establish policies and procedures and adequate internal controls related to
trust management, including Interior's original and revised improvement
plan,5 as well as documentation on current trust policies and procedures and
internal controls. We also interviewed the Policies and Procedures Project
Manager, the Internal Controls Project Manager, and other responsible
Interior officials to ascertain their perspective of the status of
Interior's trust management improvement efforts. We compared Interior's
planned improvement actions for internal controls with GAO's Standards for
Internal Control in the Federal Government and the Office of Management and
Budget's Circular A-123, revised, Management Accountability and Control. We
also compared Interior's work on internal system controls for TAAMS to the
methodology included in GAO's Federal Information Systems Control Audit
Manual and guidance issued by the National Institute of Standards and
Technology.

We conducted our work at the Department of the Interior, its Office of the
Special Trustee, and its Bureau of Indian Affairs (BIA), in both Washington,
D.C., and in Billings, Montana, and at the TAAMS contractor's facilities in
Dallas, Texas, from June 1999 through May 2000 in accordance with generally
accepted government auditing standards.

We provided a draft of this report to Department of the Interior officials
for their review and comment. We also received several technical comments
from BIA's Chief Information Officer. Interior's agency comments are
reprinted in appendix I to this report and we have incorporated changes as
appropriate.

TAAMS Effort

Interior has taken actions since our 1999 reports to address some of our
previous concerns and has begun taking critical steps necessary to instill
processes, practices, and discipline needed to successfully guide the TAAMS
effort. Overall, officials responsible for TAAMS have begun to recognize the
importance of following disciplined system acquisition, development, and
testing processes, which include developing thorough and realistic plans and
schedules, following structured approaches to determining requirements and
conducting tests, and independently assessing work being performed on the
system. This recognition is important. Without disciplined approaches to
software development and acquisition, agencies can easily fall prey to ad
hoc and chaotic processes that subject projects to continuing risks of cost
overruns, poor quality software, and schedule delays. However, taking needed
improvement steps can be difficult in organizations that are inexperienced
in development and acquisition and where few processes are defined.

Over the past year, Interior has adopted a life cycle model for TAAMS. Life
cycle programs define expectations for managing information technology
investments from conception, development, and deployment through maintenance
and support. Interior has adopted an "evolutionary prototyping" life cycle
model for its TAAMS effort, under which the system is designed, modified,
and tested in increments to identify and address user needs. Prototypes are
refined as many times as necessary to achieve the desired functionality
needed by users. This particular life cycle model has been adopted and
accepted, but effective adoption of this model requires organizations to
carefully manage associated risks such as unrealistic schedule and budget
expectations and poor user feedback--tasks which Interior has not carried
out adequately in the past for TAAMS.

Also, as noted in table 1, Interior has developed several plans and policies
key to enhancing TAAMS management, including a project management plan, risk
management plan, general operating policies, transition management plan,
quality assurance plan, configuration management plan and data management
plan. These plans generally conform to industry practice. While these plans
should have been developed before TAAMS was initiated to reduce the risks
associated with major acquisitions, they can still help Interior to better
manage the effort, measure the progress, and validate and maintain system
and data integrity. In addition, they can provide an example of the
discipline and structure needed to guide future information technology
acquisitions.

       Policy/plan               Definition          Importance to system
                                                           success
                                                   Documented project plans
                          Defines work to be       help organizations to
                          performed, establishes   define realistic time
                          necessary commitments,   frames and identify
 Project management plan  assigns                  responsibilities for key
                          responsibilities,        tasks, deliverables, and
                          establishes schedules,   resources. They provide
                          and performance          the yardstick by which
                          measures.                to measure the progress
                                                   of an effort.
                                                   Provides a disciplined
                          Identifies, assesses,    means to predict and
                          and documents risks      mitigate risks, such as
                          associated with cost,    the risk that the system
                          resource, schedule, and  will not (1) meet
 Risk management plan     technical aspects of the performance and business
                          project and determines   requirements, (2) work
                          the procedures that will with other systems
                          be used to manage those  belonging to the
                          risks.                   organization and/or (3)
                                                   be delivered on schedule
                                                   and within budget.
                          Provides general
                          background information
                          on the system including  Provides a framework for
 General operating        a summary of issues that the project and helps to
 policies                 the system intends to    identify which
                          correct. It also         management processes
                          provides the operational will be used.
                          policies governing the
                          management of system.
                                                   Provides the foundation
                                                   that is used to develop
                                                   detailed transition and
                                                   management planning,
                          Provides a blueprint for including planning for
 Transition/deployment    the transition to and    the implementation of
 management plan          deployment of a new      individual system
                          system.                  components. Also
                                                   provides a guide for
                                                   carrying out the
                                                   activities required for
                                                   installation and
                                                   operation of the system.
                          Defines processes for    Ensures that the system
                          reviewing and auditing   complies with applicable
                          the software products    processes, standards,
                          and activities to ensure and procedures. Reduces
                          that they comply with    the risk of software
 Quality assurance plan   the applicable           process and product
                          processes, standards,    standards not being met
                          and procedures and       and, in turn, the risk
                          providing staff and      that the system will
                          managers with results of cost more and take
                          their reviews and        longer to develop than
                          audits.                  necessary.
                                                   Enables organizations to
                                                   establish and maintain
                          Defines the process to   the integrity of the
                          be used for establishing system throughout its
 Configuration management product baselines and    life cycle. Prevents
 plan                     systematically           organizations from
                          controlling changes made producing and using
                          to those baselines.      inconsistent product
                                                   versions and creating
                                                   operational problems.
                          Defines the              Conveys the data
                          implementation effort    management strategy to
                          and actions required for locations that will be
 Data management plan     long-term management of  affected by the new
                          data. Describes how      system and facilitates
                          Interior will monitor    the planning of data
                          contractor operations    management activities
                          related to data.         such as data conversion.

Interior also acted to strengthen processes for user acceptance tests that
were carried out in April 2000 in response to observations we made during
previous system and user acceptance tests. For example, Interior involved
land title officers from offices outside its Billings, Montana, office--a
TAAMS prototype site--to review TAAMS functionality and develop the test
plan. Involving these officers was important since their needs can differ
from the Billings land title officer's needs. Interior also performed
additional boundary condition testing on the land title functions of TAAMS,
such as assessing whether the system would accept future dates and entries
on tracts with "0" acres. In some cases, the application responded in the
expected manner while in others, it did not. Interior officials indicated to
us that they planned to make additional adjustments to the system to address
the problems we identified during boundary condition tests.

Lastly, Interior is now undertaking an effort to develop an information
systems architecture for Indian trust fund management. As of April 2000,
Interior had developed a detailed work plan, outlining the tasks and
milestones for completing nine phases of its architecture; appointed an
architecture project manager; and defined resource requirements, including
all the bureau/office staffing, contracting and other costs for the duration
of the project. Interior expects to complete the architecture by August
2001.

If effectively implemented, Interior's planned actions over the past year
will substantially enhance management over the TAAMS effort. However, there
are a number of major business and technical challenges that still put TAAMS
at considerable risk. Some of these relate to actions not yet taken by
Interior to enhance overall trust fund management. On the business side, for
example, Interior has not reengineered business processes which TAAMS is
being designed to support even though these processes use an older and a
very different system environment. Until it does so, Interior will not be
able to maximize the benefits that can be gained from TAAMS, and it may
perpetuate outmoded ways of doing business.

Also, Interior needs to clean up thousands of inaccurate, incomplete, and/or
outdated trust fund records before converting the data for TAAMS. Yet it has
only a few months left to do so before the planned implementation of phase I
of TAAMS. Further, Interior has not established needed policies or
procedures or adequate internal controls related to trust fund management.
Policies and procedures would cover activities ranging from the way leasing
and grazing agreements are awarded to the way Indian estates are probated.
Internal controls would help strengthen management and accountability over a
wide range of trust fund operations, such as cash management, records
management, disbursements, and data processing. Both the policy and
procedures and internal controls efforts could significantly affect TAAMS
since they may require new, costly features to be built into the system or
current capabilities to be modified. Also, the internal controls are
critical to prevent future trust fund data problems.

On the technical side, there are actions underway to strengthen Interior's
management of the TAAMS project, but challenging undertakings still lie
ahead. For example, Interior needs to ensure that its contractor is
following disciplined processes for developing and testing TAAMS. And, it
needs to carefully manage the addition of new TAAMS requirements as
prototypes are rolled out and tested. Moreover, until Interior has a
complete information systems architecture, it will continue to face the risk
of spending more money and time than necessary to ensure that TAAMS and
other trust fund systems are compatible with each other and in line with
business needs.

Each of these challenges is discussed in more detail below.

1. Reengineering business processes. To maximize the success of a new system
acquisition, organizations should consider the redesign of long-standing and
ineffective business processes. As we noted in our executive guide on
financial management,6 leading finance organizations have found that
productivity gains typically result from more efficient processes, not from
simply automating old processes. Moreover, the Clinger-Cohen Act of 1996
requires agencies to analyze the missions of the agency and, based on the
analysis, revise mission-related and administrative processes, as
appropriate, before making significant investments in information technology
used to support those missions.

The need to reengineer is especially critical with trust asset management
operations: Interior is moving from a mainframe, batch-oriented,
non-integrated system environment to a client-server, real-time, integrated
environment. Such changes demand rethinking old ways of doing business. For
example, (1) Interior may now need to redefine the duties of some staff
currently responsible for keying data into trust systems since TAAMS
combines the functionality of IRMS and LRIS, (2) processes requiring data to
be transmitted from field offices to area offices may be eliminated since
data can now be input into TAAMS by field offices directly, and (3) approval
processes may be revamped since transaction approval can be immediately
reflected in TAAMS. Without taking time now to reexamine and revise its
business processes, Interior will not be able to maximize the benefits that
can be gained from TAAMS and may perpetuate outmoded ways of doing business.

2. Ensuring data integrity. The accuracy, availability, and completeness of
trust fund records has been a long-standing problem and one which several of
Interior's projects are focused on. Tens of thousands of records on trust
fund accounts, for example, contain incorrect addresses for the account
holders or lack social security or taxpayer identification numbers. Also,
data on land title and resource management is inconsistent in terms of
completeness and availability. Not only does Interior need to ensure that
records are accurate, up-to-date, and complete before converting to TAAMS,
it needs to maintain data integrity over the useful life of the system.

Interior has begun addressing both challenges. First, a project management
decision has been made to delay the implementation of TAAMS at any site
other than Billings, Montana, until that site has developed an acceptable
data conversion plan. Second, it has hired a contractor to analyze TAAMS'
data once the system is deployed to ensure accuracy and completeness.
Interior expects that, as a result of both actions, data in TAAMS will
eventually be at least 95 percent accurate.7 To achieve this rate of
accuracy, Interior will need to ensure that it can clean up the tens of
thousands of trust fund records in time for TAAMS deployment this fall.
Also, as discussed below, adequate internal controls are essential to the
ongoing success of TAAMS.

3. Establishing internal controls. A strong system of internal controls is
key to maintaining data and system integrity on an ongoing basis. Internal
controls comprise the plans, methods, and procedures used to meet an
organization's missions, goals, and objectives. They serve as the first line
of defense in safeguarding assets and preventing and detecting errors and
fraud. However, reviews conducted over the past 15 years by us, Interior's
Inspector General, and independent accounting firms have found that there
are serious financial management and internal control problems permeating
every aspect of the trust management spectrum. As a result, Interior lacks
reasonable assurance that its trust operations are effective, efficient, and
in compliance with applicable laws; that its financial reporting to trust
account holders and the Congress is reliable; and that trust assets are
adequately safeguarded.

Interior has made some progress in resolving its internal control
weaknesses. For example, it has identified and catalogued audit findings and
recommendations related to weaknesses and mapped them to current improvement
efforts. And it has selected an approach for developing the methodology that
will be used by its new Risk Management Program Office to monitor internal
controls. However, its overall progress has been slow. For example, Interior
originally expected to map identified weaknesses to current improvement
efforts by October 31, 1998, but it did not complete this until April 30,
2000. It originally expected to be able to complete the project and
establish continuing quality assurance by June 30, 1999. Now, however, it
does not expect to do this until December 2000.

Many of the internal controls now being reviewed by Interior--such as
segregation of duties, supervisory review, system security, and project
payment management--relate to requirements that should have been defined
early in the TAAMS effort. Because they were not defined early by Interior,
TAAMS was developed based on the current control environment, long known to
be inadequate. As a result, like the policies and procedures effort,
Interior may have to modify TAAMS after deployment to accommodate new
controls, thereby increasing development risks and costs. Also, until
adequate internal controls are in place to ensure the accuracy,
availability, and completeness of trust fund data, Interior will not be able
to fully ensure the integrity of TAAMS on an ongoing basis.

4. Establishing policies and procedures for trust fund operations. According
to Interior, proper management of Indian trust assets has been hampered by a
lack of comprehensive, consistent, up-to-date regulations, policies, and
procedures covering the entire trust business cycle, from management of
trust assets to distribution of trust income. Interior also maintains that
contemporary federal environmental protection statutes have placed agencies
with little direct previous experience in managing Indian trust resources in
the position of significantly affecting the use and disposition of Indian
trust resources. According to Interior, this has resulted in program gaps
and divergent practices, and a corresponding inability to ensure that
consistent, sound policies and procedures are applied across the department.

Some progress has recently been made on this issue. For example, in August
1999, Interior assigned responsibility for a policies and procedures
improvement project to an official with experience in developing and
implementing policies and procedures in the federal government. In addition,
Interior has reviewed four sets of BIA regulations--on leasing, grazing,
probate, and funds held in trust--and drafted proposed revisions. However,
like the internal controls effort, Interior's overall progress in addressing
this issue has been fairly limited. In 1998, Interior envisioned publishing
by September 30, 2000, new and revised policies and procedures in all trust
management areas where they are needed. Now, however, it does not expect to
accomplish this until June 30, 2004. The Policy and Procedures project
office was not fully staffed until July 2000, almost 2 years after Interior
began its policies and procedures initiative. We found that Interior's slow
progress in establishing internal controls and policies and procedures was
partly attributable to the effort not getting priority attention from senior
Interior managers. Such attention is needed to help build consensus among
disparate stakeholders across the department and to ensure that the projects
are adequately staffed and supported.

Interior officials have explained that initially the scope of this project
was greatly underestimated and that their revised plans provide for a more
thorough review of the conduct of trust functions. Nevertheless, these
delays pose a significant risk for TAAMS. Certain business rules cannot be
defined until the policies and procedures effort has resolved questions such
as whether to record known defective documents and whether to distribute
income to owners who don't have a trust relationship with the federal
government. The business rules that result from such policy and procedures
decisions may require modification to TAAMS. Any rework of this nature bears
the risk of introducing new defects into the system.

5. Developing TAAMS without a complete architecture. Not having a complete
information systems architecture to guide TAAMS and other projects under its
improvement effort will continue to be a major challenge for Interior.
Architectures enable organizations to know their portfolio of desired
systems and to develop a clear understanding of how these systems will
collectively support and carry out business objectives. Moreover, they help
ensure that systems are interoperable, function together efficiently, and
are cost-effective over their life cycles. As noted previously, major
federal agencies are required under the Clinger-Cohen Act to develop and
maintain information systems architectures.

While the absence of an architecture does not guarantee the failure of TAAMS
or other system modernization efforts, it does greatly increase the risk
that Interior will spend more money and time than necessary to ensure that
its systems are compatible with each other and in line with business needs.
As a result, Interior will need to take extra precautions to ensure that
TAAMS can work effectively with other trust fund systems and with Interior's
existing information technology infrastructure. In February 2000, Interior
recognized that this would be a challenging task since TAAMS and the two
systems with which it is supposed to interface--TFAS and MMS--had different
owners, different software vendors, and different program objectives. It
also recognized that its plan to purchase two off-the-shelf systems
independently (TAAMS and TFAS) and interface them with an existing system
(MMS) had inherent difficulties from its inception.

6. Defining and managing TAAMS requirements. To help ensure successful
acquisition of software-intensive service, information technology experts
recommend that organizations establish and maintain a common and unambiguous
definition of requirements (e.g., performance, function, help desk
operations, data characteristics, security, etc.) among the acquisition
team, the service users, and the contractor. The requirements must be
consistent with one another, verifiable, and traceable to higher-level
business or functional requirements. As noted in our previous report, poorly
defined, vague, or conflicting requirements can result in a service which
does not meet business needs or which cannot be delivered on schedule and
within budget.

When it first undertook the TAAMS effort, Interior did not follow a sound
process for defining requirements. For example, in April 1999, we reported
that Interior did not define high-level functional requirements for projects
contained in its improvement plan to help guide requirements development
processes for each of the individual projects. It also did not define some
specific requirements relating to security, data management, and conflicts
of interest for TAAMS. Since then, Interior has enhanced and strengthened
processes for defining requirements for TAAMS, principally by following the
new life cycle methodology. However, a risk associated with this approach is
that requirements are (1) defined iteratively rather than before development
begins and (2) defined by users alone rather than through a more formal
process involving users as well as business managers and key documents such
as the information systems architecture and a concept of operations, which
describes overall quantitative and qualitative system characteristics. This
introduces the possibility that important requirements will not be defined
until late in development and/or the possibility of "feature creep," that
is, users add more requirements than are necessary as they react to
protoypes.

Also, as noted earlier, Interior has not yet dealt with some important
policy and procedures questions that should have been addressed early in
development. For example, it has not yet determined how to distribute income
received on tracts of land which have "mixed ownership"8 even though its
field offices apply different rules to this situation. It has also not
decided how the system should handle official records that contain data
known to be erroneous.9 As we stressed in our April 1999 report, poorly
defined, vague, or conflicting requirements can result in a service which
does not meet business needs or which cannot be delivered on schedule and
within budget. As a result, it will be essential for Interior to promptly
resolve the policy and procedures issues and then define the corresponding
business rules surrounding TAAMS and carefully manage requirements as new
versions are rolled out.

7. Ensuring that plans and processes needed to successfully develop TAAMS
are followed. While Interior has started establishing plans and processes
needed to successfully develop and implement TAAMS, it is unclear whether
these tools can be effectively implemented. Moreover, even if Interior does
successfully implement disciplined processes, it will still need assurance
that its contractor is following disciplined processes. To date, an
independent assessment of whether the TAAMS contractor is doing so has not
been made.

Interior has requested the TAAMS contractor to follow "ISO 9000" standards10
relating to system testing to the greatest extent possible when performing
its internal tests. These standards are recognized worldwide as a measure of
an organization's ability to produce quality products and can provide a
useful benchmark to measure a contractor's performance.

However, because Interior has only asked the contractor to use the ISO
standards related to testing, this does not provide assurance that the
contractor will follow other processes that are critical to successful
development such as configuration management, risk management, and
requirements definition. Moreover, Interior has not yet taken steps to
independently assess its own software development and acquisition
capabilities. Organizations following best practices conduct such
assessments periodically, using accepted methodologies, such as the
Capability Maturity Models developed by Carnegie Mellon's Software
Engineering Institute. Such methodologies enable organizations to assess
whether they have the necessary process discipline in place to repeat
earlier successes on similar projects and to form specific improvement
plans.

In April 1999, after reviewing the Bureau of Land Management's (BLM) efforts
to develop and deploy its Automated Land and Mineral Record System, we
recommended that Interior obtain an independent assessment of BLM's systems
acquisition capabilities, and ensure that it uses sound system acquisition
processes.11 Interior agreed that it needed to do this and to implement our
recommendation.

Interior is taking significant actions to strengthen TAAMS management and it
has begun to recognize the value of following disciplined processes in
developing and testing the system. But much remains to be done before
Interior moves on to the second phase of TAAMS (which involves developing
and deploying a range of trust asset functions beyond land title functions,
including lease/contract/permit management, billing and collection, accounts
receivable, lease/contract payment distribution, and land resource
management functions) in order to help ensure that TAAMS will operate
efficiently and effectively.

Interior needs to ensure that the new processes and policies it has
established for developing and testing TAAMS are carefully adhered to.
Otherwise, TAAMS may well fail to meet current cost, schedule, and
performance goals. In addition, Interior needs to address larger business
challenges--developing effective policies and procedures for trust fund
management and addressing long-standing internal controls weaknesses--that
directly impact TAAMS and more importantly, the business processes to be
administered through TAAMS. Until these issues are addressed, data
inaccuracies will continue to plague the trust management process--whether
or not TAAMS is deployed. Further, the longer these efforts take to
complete, the costlier and riskier modifications to TAAMS are likely to be.
And finally, Interior needs to take time now to reexamine and reengineer
business processes. Such efforts could facilitate other initiatives integral
to strengthening trust fund management, such as defining policies and
procedures and controls. Moreover, without rethinking old ways of doing
business, Interior may well find that TAAMS will be limited in optimizing
trust fund operations.

To ensure that TAAMS is fully aligned with trust management needs and that
Interior's use of the system is optimized, we recommend that the Secretary
of the Interior direct the Assistant Secretary for Indian Affairs to work
with the Special Trustee for American Indians to do the following before
phase II of TAAMS.

ï¿½ Examine and revise business processes supported by TAAMS.

ï¿½ Properly develop and implement data conversion plans.

ï¿½ Evaluate and revise policies, procedures, and internal controls relating
to TAAMS; ensure that top trust fund managers across the department
participate in this effort; and ensure that any needed modifications to
TAAMS are made and tested.

To facilitate Interior's adoption of disciplined software and acquisition
processes, we recommend that the Secretary of the Interior direct the Chief
Information Officer to do the following before phase II of TAAMS.

ï¿½ Evaluate existing software development and acquisition processes against
the Capability Maturity Models developed for these activities by the
Software Engineering Institute; implement disciplined processes where they
are lacking; and regularly assess progress in this regard.

ï¿½ Ensure that contractors used by Interior to develop software systems have
implemented disciplined software development processes.

ï¿½ Define and manage the requirements that TAAMS should meet using accepted
processes. Once the requirements have been adequately defined, perform a gap
analysis to assess whether TAAMS is capable of providing the necessary
functionality and what modifications, if any, are necessary to address
Interior's needs. If modifications are needed, then Interior should develop
the cost, schedule, and performance impacts of making those modifications.

To ensure that TAAMS can properly interoperate with other trust fund
systems, we reiterate our recommendation to the Secretary to develop an
information systems technology architecture for trust fund operations. In
the interim, we recommend that the Secretary direct the Chief Information
Officer to (1) perform an analysis of the infrastructure necessary to
support the TAAMS application and ensure its adequacy and (2) ensure that
TAAMS can interface with TFAS and MMS systems.

We provided a draft of this report to Department of the Interior officials
for their review and comment. In their response, Interior officials did not
comment on our recommendations, but stated that they generally agreed with
our assessment of the status of Interior efforts on the TAAMS project and of
the continuing challenges they face during the remainder of this initiative.

We are sending copies of this report to Senator Daniel K. Inouye, Vice
Chairman, Senate Committee on Indian Affairs and to Senator Robert C. Byrd,
Senator Joseph I. Lieberman, Senator Ted Stevens, Senator
Fred Thompson, Representative Dan Burton, Representative George Miller,
Representative David Obey, Representative Henry A. Waxman, Representative
C.W. Bill Young, and Representative Don Young, in their capacities as
Chairmen or Ranking Minority Members of Senate and House Committees. We are
also sending copies of this report to the Honorable Jacob J. Lew, Director
of the Office of Management and Budget; the Honorable Bruce Babbitt,
Secretary of the Interior; and other interested congressional committees and
Members of Congress. Copies will be made available to others upon request.

If you have questions regarding this report, please contact me or
Mike Koury, Assistant Director, at (202) 512-9508. Other key contributors to
this report include Wendy Albert, Naba Barkakati, Cristina Chaplain, Charles
Norfleet, and Maria Zacharias.
Keith A. Rhodes
Chief Technologist

Comments From the Department of the Interior

(913863)
  

1. Indian Trust Funds: Interior Lacks Assurance That Trust Improvement Plan
Will Be Effective (GAO/AIMD-99-53 , April 28, 1999) and Indian Trust Funds:
Challenges Facing Interior's Implementation of New Trust Asset and
Accounting Management System
(GAO/T-AIMD-99-238 , July 14, 1999).

2. System testing may also stress the system in ways that may be unnoticed
by the user but are critical to the proper functioning of the system. It is
also designed to apply conditions, such as exchanging data with other
systems, which are hard to duplicate through a user interface.

3. This type of testing focuses on how the system handles typical scenarios
or transactions rather than extreme conditions and generally assumes that
system testing has been performed in an acceptable manner.

4. The Software Engineering Institute is recognized for its experience in
software development and acquisition processes. It has also developed
methods and models that can be used to define disciplined processes and
determine whether an organization has implemented them.

5. Trust Management Improvement Project: High Level Implementation Plan
(U.S. Department of the Interior, July 1998) and Trust Management
Improvement Project: High Level Implementation Plan (U.S. Department of
Interior, Revised and Updated February 29, 2000).

6. Executive Guide: Creating Value Through World-class Financial Management
(GAO/AIMD-99-45 , Exposure Draft, August 1999).

7. The TAAMS Data Management Plan establishes a preconversion goal of 95
percent "good data." This means that data quality reports will identify no
more than five errors per 100 records. This is a cumulative error condition
covering all critical data items and record relationships. The post
conversion data cleanup goal is to obtain a 98.5-percent good data rate on
data within TAAMS.

8. According to Interior officials, some interests in some trust lands are
not trust interests. These ownership interests are not considered "trust
interests" because the interest is classified as a "fee-simple" interest.
This situation can occur, for example, when non-Indians marry Indians and
inherit interests in trust land when their spouses die or when a trust owner
elects to convert a trust interest, with Interior approval, to a fee-simple
interest. BIA lacks a written policy on whether income generated through BIA
efforts on mixed-ownership land should be distributed to all owners (trust
and fee-simple) or only to trust owners.

9. For example, according to Interior officials, the probate process, which
determines what amounts of a deceased Indian's land ownership interests
should be conveyed to which heirs, sometimes produces an obviously erroneous
conveyance document. That is, the conveyance document may actually show an
amount of ownership interest being conveyed that is different from the
amount that had actually been owned. In such a case, the land title system
user will notice this discrepancy when the document is to be entered into
the TAAMS database, and must at that point either (1) enter the obviously
erroneous data into TAAMS because it is contained on an "official"
conveyance document and then request a corrected conveyance document from
the probate authority or (2) forgo entering the data from this official
document, return it to the probate authority, and await a corrected
document. BIA Land Title officials have told us that this situation occurs
with some frequency and in the absence of a business rule, may be handled
inconsistently at different BIA offices.

10. The International Organization for Standardization (ISO) is a worldwide
federation of national standards bodies from some 130 countries. ISO 9000
standards provide a framework for quality management and quality assurance.

11. Land Management Systems: Major Software Development Does Not Meet BLM's
Business Needs (GAO/AIMD-99-135 , April 30, 1999).
*** End of document. ***