Federal-Aid Highways: FHWA Needs a Comprehensive Approach to	 
Improving Project Oversight (31-JAN-05, GAO-05-173).		 
                                                                 
The federal-aid highway program provides over $25 billion a year 
to states for highway and bridge projects, often paying 80	 
percent of these projects' costs. The federal government provides
funding for and oversees this program, while states largely	 
choose and manage the projects. Ensuring that states effectively 
control the cost and schedule performance of these projects is	 
essential to ensuring that federal funds are used efficiently. We
reviewed the Federal Highway Administration's (FHWA) approach to 
improving its federal-aid highway project oversight efforts since
we last reported on it in 2002, including (1) FHWA's		 
oversight-related goals and performance measures, (2) FHWA's	 
oversight improvement activities, (3) challenges FHWA faces in	 
improving project oversight, and (4) best practices for project  
oversight.							 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-05-173 					        
    ACCNO:   A16497						        
  TITLE:     Federal-Aid Highways: FHWA Needs a Comprehensive Approach
to Improving Project Oversight					 
     DATE:   01/31/2005 
  SUBJECT:   Best practices					 
	     Best practices methodology 			 
	     Cost analysis					 
	     Federal aid for highways				 
	     Federal funds					 
	     Federal/state relations				 
	     Lessons learned					 
	     Performance measures				 
	     Program evaluation 				 
	     State-administered programs			 
	     Program goals or objectives			 
	     FHWA Federal-Aid Highway Program			 

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Product.                                                 **
**                                                              **
** 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-05-173

United States Government Accountability Office

  GAO	Report to the Chairman, Committee on Transportation and Infrastructure,
                            House of Representatives

January 2005

FEDERAL-AID HIGHWAYS

       FHWA Needs a Comprehensive Approach to Improving Project Oversight

                                       a

GAO-05-173

[IMG]

January 2005

FEDERAL-AID HIGHWAYS

FHWA Needs a Comprehensive Approach to Improving Project Oversight

  What GAO Found

FHWA has made progress in improving its oversight efforts since 2002, but
it lacks a comprehensive approach, including goals and measures that guide
its activities; workforce plans that support these goals and measures; and
data collection and analysis efforts that help identify problems and
transfer lessons learned. FHWA's 2004 performance plan established, for
the first time, performance goals and outcome measures to limit cost
growth and schedule slippage on projects, but these goals and measures
have not been effectively implemented because FHWA has not linked its
day-to-day activities or the expectations set for its staff to them, nor
is FHWA fully using them to identify problems and target its oversight.

FHWA undertook activities in response to concerns raised about the
adequacy of its oversight efforts that have both promising elements and
limitations. For example, while FHWA now assigns a project oversight
manager to each major project (generally projects costing $1 billion or
more) and identified skills these managers should possess, it has not yet
defined the role of these managers or established agencywide performance
expectations for them. While FHWA issued guidance to improve cost
estimating and began collecting information on cost increases, it still
does not have the capability to track and measure cost growth on projects.
Finally, although FHWA received direction to develop a more
multidisciplinary workforce to conduct oversight, it has not fully
incorporated this direction into its recruiting and training efforts.

FHWA faces challenges to improving its oversight that are in large part
rooted in the structure of the federal-aid highway program and in FHWA's
organization and culture. As such, they may be difficult to surmount. For
example, because the program does not link funding to states with the
accomplishment of performance goals and outcome measures, it may be
difficult for FHWA to define the role and purpose of its oversight. Also,
FHWA's decentralized organization makes it difficult to achieve a
consistent organizational vision. Human capital challenges affecting much
of the federal government have affected FHWA, particularly in its need to
transform its workforce to meet its evolving oversight mission. FHWA faces
an increased oversight workload in the years ahead as the number of major
projects grows and if provisions Congress is considering to increase
FHWA's responsibilities become law. Questions exist about FHWA's ability
to effectively absorb these new responsibilities, overcome underlying
challenges, and improve its oversight.

We identified selected best practices that could help FHWA develop a
framework for a comprehensive approach to project oversight. These include
establishing measurable goals to objectively and quantifiably assess
progress, making oversight managers accountable for the effective
implementation of these goals, providing professional training, and
collecting and transferring lessons learned.

United States Government Accountability Office

Contents

  Letter

Results in Brief
Background
FHWA Established Some OversightGoalsand Measures butHas Not

Effectively Implemented Them FHWA's Oversight Activities Have Promising
Elements and

Limitations FHWA Faces Challenges to Improving Oversight Best Practices
Can Help Improve Progress and Address Challenges

to Improving Project Oversight Conclusions Recommendations Agency Comments

1 4 7

15

21 32

38 42 43 43

  Appendixes

Appendix I: Scope and Methodology 45

Appendix II: FHWA FY 2004 Major Project Team Work Plan Summary 48

Appendix III:	GAO Contacts and Staff Acknowledgments 50 GAO Contacts 50
Staff Acknowledgements 50

Table Table 1:	Types of Projects Receiving FHWA Oversight versus State
Oversight

Figures Figure 1: Stages of a Highway or Bridge Project and State and FHWA
Roles and Approval Actions 10 Figure 2: Goals and Associated Measures
Articulated In FHWA's 2004 Performance Plan 17

Contents

Abbreviations

AASHTO American Association of State Highway and Transportation

Officials DOT Department of Transportation FHWA Federal Highway
Administration FMIS Financial Management Information System FTA Federal
Transit Association GPRA Government Performance and Results Act of 1993
ISTEA Intermodal Surface Transportation Efficiency Act of 1991 OIG Office
of Inspector General OMB Office of Management and Budget PDP professional
development program TEA-21 Transportation Equity Act for the 21st Century

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. However, 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.

A

United States Government Accountability Office Washington, D.C. 20548

January 31, 2005

The Honorable Don Young Chairman, Committee on Transportation and
Infrastructure House of Representatives

Dear Mr. Chairman:

The federal-aid highway program provides over $25 billion a year to states
for highway and bridge projects, often paying 80 percent of these
projects' costs. The program is federally financed and state administered;
that is, the federal government provides funding and oversees the program,
while the states largely choose and manage the projects. These projects
can take years of planning and environmental review, as well as the time
spent in the design and construction phases. With highway congestion
projected to worsen over the next 20 years and freight traffic expected to
double, widespread consensus exists on the need to maintain and improve
the nation's surface transportation infrastructure. In the longer term,
broader fiscal challenges face the nation, including federal and state
budget deficits and a fiscal crisis looming as the baby boom generation
retires, causing mandatory commitments to Social Security and Medicare to
consume a greater share of the nation's resources, squeezing funding
available for domestic discretionary programs. Thus, ensuring that states
effectively manage and control the cost and schedule performance of
federally-aided projects, and that federal funds are used efficiently and
effectively, is critical in light of these challenges.

The Federal Highway Administration (FHWA), under the Department of
Transportation (DOT), is responsible for overseeing the federal-aid
highway program through its headquarters in Washington, D.C., and division
offices located in every state, the District of Columbia, and Puerto Rico.
To carry out this role, FHWA reviews and approves the transportation plans
and environmental impact assessments that states periodically prepare,
reviews and approves states' property acquisition activities; and enforces
a variety of requirements, such as civil rights laws, that states accept
as a condition of federal aid. FHWA also oversees the design and
construction of federally-aided projects, but this oversight has evolved
over the years and currently focuses on two broad areas: (1) for selected
projects, direct review and approval of state design plans, contract
awards, and construction progress; and (2) reviews of state management
processes, also known as process reviews, to ensure that the states have
adequate controls to effectively manage federally-assisted projects. FHWA
also

enforces requirements for "major projects"-generally those projects
estimated to cost over $1 billion-such as reviewing and approving annual
finance plans required by law for these projects.

We and others have raised concerns regarding FHWA's oversight of the
federal-aid highway program. For example, from 1997 to 2002, we
highlighted several problems, including the lack of a link between FHWA's
oversight activities and its business goals; a lack of emphasis on FHWA's
part to ensuring that containing costs is an integral part of states'
project management; a lack of useful cost estimates, which typically have
not been reliable predictors of the total costs or financing needs of
projects; and a lack of information on the amount of and reasons for cost
increases. In particular, we have highlighted the challenges that major,
multibilliondollar projects pose. These projects are very costly and
complex-they require large commitments of public resources that may take
several years to garner; can involve complicated financing arrangements;
can be technically challenging to construct; and involve a wide range of
social, environmental, land-use, and economic challenges before and during
construction. To keep the projects on schedule and within budget, federal
and state officials must carefully oversee their development, planning,
and construction.

We also recognized actions Congress, DOT, and FHWA have taken to improve
FHWA's oversight in response to concerns raised about its efforts. For
example, in 1998, Congress required that states submit to the Secretary of
Transportation an annual finance plan for major projects, which was
intended to help provide an advance warning system for increased costs for
large transportation projects receiving federal funds. In 2001, FHWA took
steps to improve its oversight of these major projects by, among other
things, establishing a major projects team to coordinate and implement
oversight policies for major projects and designating project oversight
managers to oversee these projects. In addition, for its reviews of state
management processes, FHWA also began to conduct more risk assessments to
identify and prioritize risk areas and allocate resources.

FHWA will face an increased oversight workload in the years ahead if
provisions Congress is considering become law and as the number of
federal-aid projects grows. In 2003, DOT proposed legislation for the
reauthorization of the Transportation Equity Act for the 21st Century (TEA

21) that would increase FHWA's oversight responsibilities by requiring
FHWA to annually review states' financial management processes and to
periodically review how states plan and manage projects. The legislation

would also extend certain oversight requirements that currently pertain
only to major projects, to other large-dollar projects as well. This
proposal was largely adopted in bills that were separately approved by the
House and Senate in 2004, but which were not enacted before the
adjournment of the 108th Congress. In addition, during the course of our
review, the number of major projects FHWA is responsible for overseeing
increased from 15 to 21, and that number is expected to nearly double to
40 in the next several years.

At your request, we reviewed FHWA's approach to improving its federal-aid
highway project oversight efforts since 2002, including (1) FHWA's
oversight-related performance goals and measures, (2) FHWA's oversight
improvement activities, (3) challenges FHWA faces in improving project
oversight, and (4) best practices for project oversight.

To determine FHWA's approach to improving its oversight efforts since
2002, the last time we reported on this issue, we evaluated the agency's
strategic and performance plans, and supporting documents. Furthermore, we
documented and analyzed the status of FHWA's various project oversight
efforts. We interviewed officials at FHWA Headquarters and selected FHWA
division offices, state departments of transportation, and others to
document oversight implementation efforts and the challenges FHWA faces in
improving its project oversight. We also reviewed FHWA's use of financial
information from its Financial Management Information System (FMIS) to
track and analyze trends in cost growth on projects. We did not
independently assess the reliability of FMIS data as the Department's
Inspector General has reported on weaknesses in FHWA's financial
management and reporting processes, most recently in November 2004 as part
of the annual audit of DOT's consolidated financial statements. In
addition, our work focused primarily on FHWA's use of FMIS data for
oversight purposes, rather than relying on FMIS data to support our
findings and conclusions.

To provide information on project management best practices we conducted a
literature search to identify best practices related to project oversight
and evaluated their applicability to FHWA's oversight efforts. We
conducted our work from August 2003 through December 2004 in accordance
with generally accepted government audit standards. Appendix I provides
the details of our scope and methodology.

Results in Brief 	FHWA has made progress in improving its oversight
efforts since 2002, but it lacks a comprehensive approach to project
oversight, including oversight-focused goals and outcome measures that
guide its activities; an overall plan for its oversight activities that
responds to past concerns and is linked to these goals and measures;
workforce planning efforts that support these goals and measures; and data
collection and analysis efforts that identify problems, help target
resources, and transfer lessons learned. FHWA's 2004 performance plan
established, for the first time, performance goals and outcome measures to
limit the magnitude of cost growth and schedule slippage for major
projects, but these goals and measures have not been effectively
implemented because FHWA has not linked the day-today oversight activities
of its division offices or the expectations set for its staff to them. For
example, none of the three division offices we visited that are currently
overseeing major projects had established any link in its unit performance
plan between its activities and FHWA's goal and measure to limit cost
increases and schedule slippage on major projects. FHWA also established
performance goals and measures to contain costs on projects other than
major projects, but it is not yet using these goals to identify problems
on a state-by-state or project-by-project basis in order to target its
oversight activities. FHWA also uses cost and schedule estimates developed
relatively late in a project's development-at the point at which the
project is ready to go to construction-as a baseline for measuring its
progress in limiting cost growth. However, by the time a project goes to
construction, a public investment decision has effectively been made, as
substantial funds will have been spent on designing the project and
acquiring property, and much of the increases in a project's costs may
have already occurred.

In response to concerns raised about the adequacy of its oversight
efforts, FHWA undertook a series of oversight-related activities that have
both promising elements and limitations. For example, FHWA now routinely
assigns a project oversight manager to each major project. It established
core competencies identifying the skills that project oversight managers
should possess, which are to serve as a guide for selecting these
managers. However, FHWA has not yet defined the role of project oversight
managers or established agency-wide performance expectations for them.
Instead, expectations for this position were developed locally at division
offices, and none of the three division offices' expectations for
oversight managers that we reviewed specifically tasked the manager with
achieving FHWA's goals and measures to limit cost increases and schedule
slippage. Without this link between performance expectations and goals, it
is unclear how the

project oversight managers will improve oversight of major projects and
what training and development are needed to achieve the desired
performance. In another positive step, in June 2004, FHWA issued guidance
to state transportation agencies to assist them in applying sound cost
estimating practices, including guidance on developing more realistic
early cost estimates. However, this guidance is voluntary and applies only
to major projects, and it is too soon to tell whether it will be effective
in improving the quality of estimates. DOT proposed legislation empowering
FHWA to develop national standards for estimating project costs, which was
included in the surface transportation reauthorization bills separately
passed by the House and the Senate in 2004, but which were not enacted
before the adjournment of the 108th Congress. FHWA also started tracking
information on the cost growth of major projects, and it recently started
developing some cost information on all federal-aid highway projects over
$10 million. However, FHWA still does not have the capability to track and
measure cost growth on projects it oversees, which limits its ability to
evaluate the reasons why cost growth occurs, identify problems and
solutions, and transfer lessons learned. FHWA has also incorporated more
risk assessments into its process reviews of state transportation
agencies, but we found that some division offices are not using their risk
assessments to guide their reviews. The DOT Inspector General recently
reported that FHWA's risk assessments were voluntary and did not provide a
systematic approach for assessing program risks throughout the agency.
Finally, although FHWA received congressional-committee direction to
restructure its workforce to develop a multidisciplinary approach to
oversight, it has made limited progress in doing so, as it has not fully
incorporated such an approach into its human capital planning, recruiting,
or training efforts.

FHWA faces several challenges to improving its oversight-challenges rooted
in the structure of the federal-aid highway program as well as FHWA's
organization and culture-that may be difficult to surmount. The
federal-aid highway program provides states with broad flexibility in the
use of federal funds. Although DOT has articulated goals for the program
such as improving safety and reducing the growth of traffic congestion,
the program does not have the mechanisms to link funding levels with the
accomplishment of specific performance-related goals and outcomes. As a
result, it may be difficult for FHWA to define its role and the purpose of
its oversight. FHWA's long-standing culture of partnership with the states
also poses challenges; FHWA and state officials believe this partnership
has helped to build trust and respect between the state agencies and FHWA
and has resulted in projects being planned and built more efficiently and
effectively. However, there is a potential downside-when a project

overseer becomes an active partner in a project, the arms-length,
independent perspective that can be important to effective project
oversight can be lost. Another challenge is FHWA's decentralized
organization, which can make implementing a consistent oversight vision
and strategies throughout FHWA difficult. Human capital challenges that
affect much of the federal government have also affected FHWA's ability to
improve its oversight, particularly in its need to transform its workforce
and culture to support its evolving oversight mission. Finally, FHWA
officials said that a challenge to improving oversight is that legislation
passed in 1991 and 1998 has, in their view, sent mixed messages regarding
FHWA's and states' roles, leading to confusion as to the extent of FHWA's
authority and responsibilities over state-managed highway projects.
Language in reauthorization legislation proposed by DOT and separately
passed by the House and the Senate in 2004 but not enacted may help
clarify FHWA's oversight role if it is reintroduced and enacted by the
109th Congress.

We identified selected best practices that are relevant to FHWA's efforts
to improve project oversight through analyzing our past work on
effectively managing and overseeing projects, as well as reports by the
National Research Council, the Office of Management and Budget (OMB), and
others. While some of these best practices are beginning to be reflected
in FHWA's activities, as a whole, they could provide a framework for
moving to a comprehensive approach to project oversight. The best
practices we identified include (1) establishing measurable project
oversight goals and communicating these goals down through all levels of
the agency; (2) establishing project oversight manager roles,
responsibilities, and accountability measures based on oversight goals;
(3) providing professional training and a career path in oversight
management; and (4) identifying and transferring lessons learned.
Establishing measurable project oversight goals and communicating these
goals down through all levels of an agency illustrates how an agency will
execute its oversight mission and establishes what its oversight is
designed to accomplish. Establishing oversight manager roles,
responsibilities, and accountability measures based on oversight goals
links efforts to goals and makes managers accountable for proper project
oversight. Providing professional training in oversight management could
ensure that managers develop the skills necessary for conducting their
oversight activities. Providing a career path would allow project managers
to develop their abilities as they progress through increasingly
challenging assignments. Identifying lessons learned from the successes
and setbacks that occur on projects-and

transferring those lessons to other projects-can prevent mistakes from
being repeated and promote improved oversight.

FHWA will face an increased oversight workload in the years ahead if
provisions Congress is considering become law and as the number of
federal-aid projects grows. Given the limitations present in FHWA's
oversight approach today, questions exist about its ability to effectively
absorb new responsibilities and to improve its oversight efforts over
federal-aid highway projects in the years ahead. Moreover, absent a
comprehensive approach, FHWA is unlikely to overcome the structural,
organizational, and cultural challenges it faces and to fully address the
concerns raised about the adequacy of its oversight efforts. To address
these limitations, we are recommending that FHWA link its day-to-day
activities and the expectations set for its staff to its goals and outcome
measures; develop an overall plan for its oversight initiatives that is
tied to its oversight-related goals and measures; improve the use and
performance of project oversight managers by centrally defining their role
and responsibilities; and develop the capability to track and measure
costs over the lives of projects to identify problems, help target
resources, and transfer lessons learned.

We provided a draft of this report to DOT and met with FHWA officials,
including the Deputy Administrator, to obtain their comments on the draft.
FHWA generally agreed with the facts and conclusions in the report and our
characterization of the challenges FHWA faces in improving its project
oversight. FHWA officials emphasized that although we highlighted
potential drawbacks associated with both its culture of partnership with
the states and its decentralized organization, this partnership and
organization are also major strengths of the federal-aid highway program
that will allow the agency to absorb potential new responsibilities, help
overcome challenges, and improve program oversight in the future through a
more comprehensive approach. FHWA officials did not take a position on our
recommendations, but they stated that they would be taking them under
advisement. They also suggested some technical and clarifying comments
that we incorporated into the report as appropriate.

Background	Federal funding for highways is provided to the states mostly
through a series of formula grant programs collectively known as the
federal-aid highway program. Periodically, Congress enacts multiyear
legislation that authorizes the nation's surface transportation programs,
including highways, transit, highway safety, research, and motor carrier
programs. In

1998 Congress enacted TEA-21, which authorized $172.4 billion for the
federal-aid highway program from fiscal years 1998 through 2003. The
program expired on September 30, 2003, and it has been extended by six
short-term extensions, the most recent extending the program until May 31,
2005. During the 108th Congress, both the House and Senate approved
separate legislation to reauthorize the federal-aid highway program;
however, the reauthorization legislation was not been enacted before the
adjournment of the 108th Congress. The bill approved by the House
authorized $226.3 billion for the federal-aid highway program for fiscal
years 2004 through 2009, an increase of about 31 percent over TEA-21,
while the bill approved by the Senate authorized $256.4 billion, an
increase of about 49 percent.1 Because both bills contained funding
increases, it is likely that the number of federal-aid highway projects
will rise in the next several years.

FHWA administers the federal-aid highway program and distributes most
highway funds to the states through annual apportionments established by
statutory formulas contained in law. Once FHWA apportions these funds,
they are available to be obligated for construction, reconstruction, and
improvement of highways and bridges on eligible federal-aid highway routes
and for other purposes authorized in law. About 1 million of the nation's
4 million miles of roads are eligible for federal aid; including the
161,000 mile National Highway System, of which the 47,000 mile Interstate
Highway System is a part.2 While FHWA administers the program, the
responsibility for choosing projects generally rests with state
departments of transportation and local planning organizations. The states
have considerable discretion in selecting specific highway projects and in
determining how to allocate available federal funds among the various
projects they have selected. For example, section 145 of title 23 of the
United States Code describes the federal-aid highway program as a
federally-assisted state program and provides that the federal
authorization of funds, as well as the availability of federal funds for
expenditure, shall not infringe on the states' sovereign right to
determine the projects to be federally financed.

1S. 1072, 108th Cong.S:1101 (2004); and H.R. 3550, 108th Cong.,S:1101
(2004).

2The 1 million miles of roads eligible for federal aid accounted for about
85 percent of the vehicle miles traveled on the nation's roadways in 2002.
The 3 million miles of roads that are generally ineligible are
functionally classified as local roads or rural minor collectors.

A highway or bridge construction or repair project usually has four
stages: (1) planning, (2) environmental review, (3) design and property
acquisition, and (4) construction. FHWA reviews and approves long-term and
shortterm state transportation plans and programs, environmental
documents,

and the acquisition of property for all highway projects.3 However, its
role in overseeing the design and construction of projects varies. On
selected projects, FHWA exercises what is often considered "full"
oversight, meaning that FHWA (1) prescribes design and construction
standards, (2) approves design plans and estimates, (3) approves the
selection of the contract award, (4) periodically inspects the progress of
construction, and (5) renders final acceptance on projects when they are
completed. However, relatively few projects are subject to this full FHWA
oversight. The last two authorizations, the Intermodal Surface
Transportation Efficiency Act of 1991 (ISTEA) and TEA-21, devolved an
increasing amount of responsibility to the states. Under current law FHWA
exercises full oversight of certain high-cost interstate system projects,
while states oversee design and construction on other federal-aid
projects.

The stages of a highway or bridge project and the corresponding state role
and FHWA approval actions are shown in figure 1.

3Specifically, FHWA approves state short-term transportation plans and
reviews state and metropolitan planning processes.

Figure 1: Stages of a Highway or Bridge Project and State and FHWA Roles
and Approval Actions

Source: GAO.

The types of projects for which FHWA exercises full oversight as compared
with state oversight are shown in table 1.

Table 1: Types of Projects Receiving FHWA Oversight versus State Oversight

Percent of federal Design and highway funds construction Type of project
Mileage obligated in 2002 oversight Exceptions

               Interstate   47,000      12         FHWA      Certain types of 
                   System                     oversight     projects, or      
                                                             projects below a 
                                                            dollar threshold, 
                                                               where FHWA and 
                                                           state determine    
                                                           state oversight is 
                                                             appropriate      

                    National    115,000    45    State may      State or FHWA 
                     Highway                        assume   determines state 
                     System,                     oversight   oversight is not 
              non-Interstate                                  appropriate     
                      routes                               

               Federal-aid    798,000   42   State shall     State determines 
              highways off                        assume   state oversight is 
              the National                     oversight   not appropriate    
                   Highway                               
                    System                               

Source: GAO analysis.

According to FHWA, the agency retains the responsibility to oversee all
federally-aided highway and bridge projects, including projects for which
FHWA does not exercise oversight over the design and construction phases.
FHWA conducts oversight of state transportation programs through a variety
of means, including process reviews-reviews of state management processes
to ensure that states have adequate controls to effectively manage
federally-assisted projects. States and FHWA execute stewardship and
oversight agreements to define their respective oversight
responsibilities.

TEA-21 contains an additional oversight requirement for so-called "major
projects"-generally those estimated to cost at least $1 billion. Since
TEA21 was enacted in 1998, states must submit finance plans to DOT
annually for such projects, based on detailed estimates of the costs to
complete the project and on reasonable assumptions about future increases
in such costs. FHWA developed guidance that requires states to include in
these finance plans a total cost estimate for the project, adjusted for
inflation and annually updated; estimates about future cost increases; a
schedule for completing the project; a description of construction
financing sources and revenues; a cash flow analysis; and a discussion of
other factors, such as

how the project will affect the rest of the state's highway program. FHWA
approves these plans as a condition of federal aid. As of November 2004,
11 of the 21 current major projects had finance plans. Approved finance
plans will be required for the other projects prior to FHWA authorizing
federal funds for construction.4 FHWA forecasts that another 19 major
projects, estimated to cost from $34 billion to $60 billion, will be
starting over the next several years and will also require finance plans.

Over the past several years, we and others have identified problems with
FHWA's oversight of major projects and other large highway and bridge
projects. For example, in 1997, we reported that the overall amount of and
reasons for cost increases on highway and bridge projects could not be
determined because data were not readily available from FHWA or the
states.5 We found, however, on many of the projects for which we could
obtain information, that costs had increased, sometimes significantly, and
that several factors accounted for the increases. In addition, initial
cost estimates were not reliable predictors of a project's total cost or
financing needs because they were developed at the environmental review
stage, and their purpose was to compare project alternatives, not to
develop reliable cost estimates. We further reported that cost containment
was not an explicit statutory or regulatory goal of FHWA's oversight;
therefore, the agency had done little to ensure that cost containment was
an integral part of the states' project management. In our May 2002
testimony before the Highways, Transit, and Pipelines Subcommittee of your
Committee, we reported that FHWA had begun to improve its oversight by
implementing Congress' finance plan requirements for major projects and
introducing risk-based decision making into its oversight of states'
processes on other projects.6 However, we also reported that FHWA had not
yet developed performance goals or measurable outcomes linking its
oversight activities to its business goals, and that goals and strategies
for containing costs could improve accountability and make cost
containment an integral part of how states manage projects over time.
Furthermore, we stated that opportunities existed for improving the
quality of cost estimating and

4A finance plan has not been prepared for one major project that is
nearing completion, because it predates the requirement for these plans.

5GAO, Transportation Infrastructure: Managing the Costs of Large-Dollar
Highway Projects, GAO/RCED-97-47 (Washington, D.C.: Feb. 27, 1997).

6GAO, Transportation Infrastructure: Cost and Oversight Issues on Major
Highway and Bridge Projects, GAO-02-702T (Washington, D.C.: May 1, 2002).

developing reliable and accurate information on the extent and nature of
projects' cost performance to help direct federal oversight efforts. Our
work identified several options for enhancing the oversight of major
projects. Reports by DOT's Office of Inspector General, as well as reviews
by state audit and evaluation agencies, have also shown that the
escalating costs and management of major projects continue to be a
problem. For example, the Inspector General has issued several reports on
FHWA's oversight and stewardship of major projects, such as the Central
Artery/Tunnel project in Massachusetts and the Woodrow Wilson Bridge in
Virginia and Maryland. More recently, the Inspector General reported signs
of improvement in FHWA's stewardship over major projects but identified
improvements needed in eight areas, including developing more reliable
cost estimates, managing project schedules better, strengthening efforts
to prevent and detect fraud, and refocusing FHWA's efforts on project
management and financial oversight.7

Partly in response to concerns that we and others have raised, in addition
to the provisions Congress enacted in TEA-21, DOT also took further
action. In 2000 the Secretary of Transportation established a task force
to review oversight mechanisms and processes for major transportation
projects across DOT. Among other things, the task force recommended that
DOT improve the skills and qualifications of staff overseeing major
projects and conduct more rigorous financial reviews of such projects.
Although DOT did not formally implement the task force's recommendations,
FHWA responded to the task force report by establishing a major projects
team in Washington, D.C., to assist FHWA's division offices in reviewing
financial plans and overseeing major projects and by assigning project
oversight managers to each of the major projects. In addition, in 2003,
DOT proposed new legislation as part of its TEA-21 reauthorization
proposal requiring that (1) states submit a project management plan as
well as an annual financial plan for any project with an estimated total
cost of $1 billion or more or any other project at the discretion of the
Secretary; (2) states develop financial plans for any project receiving
over $100 million in federal funds; (3) FHWA perform annual reviews of
state transportation programs' financial management and periodic reviews
of state project delivery systems for planning and managing projects; and
(4) DOT develop minimum standards for estimating project costs and perform
periodic reviews of state practices

7Management of Cost Drivers on Federal-aid Highway Projects, Statement of
the Honorable Kenneth H. Mead, Inspector General, U.S. Department of
Transportation, May 8, 2003.

for estimating costs and awarding contracts. This proposal was largely
adopted in bills that were separately approved by the House and the Senate
in 2004 but that were not enacted before the adjournment of the 108th
Congress.

To meet the requirements of the Government Performance and Results Act of
1993 (GPRA), DOT establishes goals and outcome measures for the programs
under its jurisdiction, including the federal-aid highway program, through
its strategic and performance plans. GPRA requires agencies to complete
strategic plans in which they define their missions, establish
outcome-oriented goals, and identify the strategies that will be needed to
achieve those goals. GPRA also requires agencies to prepare annual
performance plans to articulate goals for the upcoming fiscal year that
are aligned with their long-term strategic goals. The establishment of
goals and measures is a valuable tool for guiding an agency's strategies
and resource allocations and for establishing accountability for the
outcomes of its dayto-day activities. As our prior work has shown,
measuring performance allows organizations to track the progress they are
making toward their goals and gives managers crucial information on which
to base their organizational and management decisions. When an agency's
day-to-day activities are linked to outcome measures, these measures can
create powerful incentives to influence organizational and individual
behavior. In prior work, we found that leading agencies that successfully
link their activities and resources also seek to establish clear
hierarchies of performance goals and measures. Under these hierarchies, an
agency links the goals and outcome measures for each organizational level
to successive levels and ultimately to the agency's strategic goals.
Without this link, managers and staff throughout the organization will
lack straightforward roadmaps showing how their daily activities can
contribute to attaining organization wide strategic goals.8

8GAO. Executive Guide: Effectively Implementing the Government Performance
and Results Act, GAO/GGD-96-118 (Washington, D.C.: June 1, 1996).

  FHWA Established Some Oversight Goals and Measures but Has Not Effectively
  Implemented Them

FHWA established measurable, outcome-oriented goals and measures related
to cost and schedule performance for the first time in its 2004
performance plan, but FHWA has not effectively implemented these goals and
measures in order to improve oversight. Specifically, FHWA has not linked
its day-to-day oversight activities to its goals for major projects, and
it has not yet used its goals and measures for nonmajor projects to
examine the performance of states or particular projects. FHWA also uses
estimates developed relatively late in a project's development as its
baseline for measuring its performance on achieving cost and schedule
goals; thus, it does not task itself with controlling cost and schedule
slippage during the early stages of a project's development.

    FHWA Recently Established Goals and Measures

In December 2000, DOT issued a task force report concluding that a
significant effort was needed to improve the oversight of major projects
and recommending that DOT incorporate goals for its oversight efforts into
its performance plans as well as into the plans of FHWA.9 In 2002, we
reported that FHWA had not yet developed performance goals or measurable
outcomes linking its oversight activities to its business goals and that
goals and strategies for containing costs could improve accountability and
make cost containment an integral part of how states manage projects over
time.10

FHWA has made some improvements over the past several years in developing
goals and performance measures related to cost and schedule performance of
federal-aid highway projects. In its fiscal year 2002 performance plan,
FHWA included a strategic goal of organizational excellence that had among
its many strategic objectives the aim to improve organizational
performance. Since that time, from fiscal year 2003 to fiscal year 2005,
FHWA's performance plans have specifically identified under the
organizational excellence heading a general oversight goal to improve
project oversight and stewardship so as to realize more cost efficient
federal-aid funds administration and project management and more effective
use of funds in terms of return on investment. In its fiscal year 2004
performance plan, DOT for the first time established goals and

9Report of the ONE DOT Task Force on Oversight of Large Transportation
Infrastructure Projects; December 2000.

10GAO-02-702T.

outcome measures specifically related to achieving cost and schedule
targets for its transportation projects.11 FHWA incorporated these goals
and measures into its performance plan for highway projects, establishing,
for the first time, goals and measures for major projects that are outcome
oriented and measurable and clearly define containing project costs and
schedules as an integral part of FHWA's oversight mission. Figure 2 shows
the goals and associated measures articulated in FHWA's fiscal year 2004
performance plan.12

11Between fiscal years 2001 and 2003 FHWA did have an outcome measure
under the organizational excellence goal to reduce unexpended balances.
However, this measure did not specifically relate to cost and schedule
targets on transportation projects. Unexpended balances are the annual
amount of federal-aid funds obligated but not expended on projects of $1
million or more and with no billing activity for more than a year.

12For FY 2005, FHWA made its measure to prevent high growth in project
costs more specific by adding that the total percentage of cost growth for
construction projects will be less than 10 percent above the original
estimate on construction projects over $10 million.

Figure 2: Goals and Associated Measures Articulated In FHWA's 2004
Performance Plan

While linking day-to-day activities to goals and measures is an important
element of implementing goals and measures by ensuring that they are being
used as a framework to guide the activities, we found no evidence that
FHWA has linked the day-to-day activities of its division offices to its
goal and measure for major projects. In our visits to the three division
offices that were overseeing a major project, we found a lack of
documented goals, strategies, or measures showing how the division
offices' activities supported and furthered the goals and measures
articulated in FHWA's 2004 performance plan. While each division office
had developed its own individual unit fiscal year 2004 performance plan,
there was no link in these plans between the division offices' activities
and FHWA's goal and measure for major projects: that is, to meet 95
percent of schedule milestones and cost estimates for major projects or to
miss them

                     Source: FHWA's 2004 Performance Plan.

    FHWA Has Not Effectively Implemented its Goals and Measures

by less than 10 percent.13 Furthermore, in these three division offices,
the project oversight managers were not specifically tasked, as part of
their duties and responsibilities, with implementing or furthering the
articulated cost and schedule performance goals for major projects.

This absence of a link between activities and goals and measures was in
noticeable contrast to the link that the division offices had established
between their activities and the three areas of work that FHWA has
designated as its "vital few" priorities. FHWA's vital few priorities,
which consist of safety, congestion mitigation, and environmental
stewardship and streamlining, are areas that FHWA has determined are key
priorities and that it accordingly highlights in its performance plans as
areas where the agency has identified performance gaps that must be
addressed if FHWA is to be successful. Perhaps in line with this emphasis,
FHWA has developed a better link between its division offices' activities
related to these vital few priorities and its goals related to these vital
few priorities. For example, all seven of the division offices we visited
had unit plans that linked their activities to all three of FHWA's vital
few priorities. This link was established through listing specific
unit-level activities and measures that were designed to meet unit goals
that mirrored the national performance plan's goals for its vital few
priorities. For example, for the vital few priority of safety, FHWA's
fiscal year 2004 performance plan set a performance goal of reducing
highway fatalities to no more than 1.38 per 100 million vehicle miles
traveled. The fiscal year 2004 performance plan for one division office
tasked itself with five performance objectives to address this national
goal, including such objectives as improving accident rates involving
roadway departures, increasing the capability of FHWA and state engineers
in highway safety design, and reducing pedestrian fatalities.14 One or
more division-level performance measures and several specific activities
were identified for each of these five division objectives, and
performance expectations set for key division staff identified which of
these activities they were responsible for performing.

In addition to not linking its activities to its goal for major projects,
FHWA has also not yet used its goals and outcome measures to help it
identify and

13Four of the division offices we visited were not at the time overseeing
a major project; therefore, the major project goal and measure did not
apply to their responsibilities at that time.

14The other two objectives were to increase the number of high-accident
intersection improvement projects and to support the state's safety plan.

correct problems on the vast majority of projects that are not considered
major projects. In 2004, FHWA did not develop numerical goals or outcome
measures related to nonmajor projects, nor did it assess the cost and
schedule performance of projects on a state-by-state or project-by-project
basis in order to gain a clear picture of whether certain states or
projects have more cost or schedule overruns than others in order to
target its oversight activities. Instead, FHWA officials told us that
while FHWA's major projects team recently started developing this
state-by-state information, FHWA relies on the division offices to monitor
costs of individual contracts and take action as appropriate. However,
these officials could not say with certainty whether their division
offices were carrying out this monitoring function, or what kinds of
corrective measures were being applied. FHWA officials also said that the
agency relies on FHWA's division offices to execute formal oversight
agreements with the states to ensure that they are working to control
costs. However, none of the oversight agreements of the seven division
offices we visited reflected an agreement between FHWA and the states to
do this. As we concluded our review, FHWA officials stated that in
response to issues we raised, FHWA would begin sharing information with
its division offices and begin discussing appropriate solutions or actions
the divisions can take to address incidences of cost growth.

For fiscal year 2005, FHWA made its cost-related goal for nonmajor
projects more specific by adding the outcome measure that the total
percentage of cost growth for all construction projects over $10 million
will be less than 10 percent above the estimated cost when the project
went to construction. FHWA's preliminary information indicates that the
agency is, in the aggregate, meeting its goal; however, sharing
information with its division offices about variations in state contract
costs could help FHWA target its oversight efforts. For example, FHWA's
information also shows that about 1 in 5 of the 492 contracts approved for
construction in fiscal year 2003 exceeded the 10 percent threshold in
fiscal year 2004. One contract exceeded the threshold by 160 percent. Our
analysis of FHWA's information also shows that some states may be more
effectively controlling the costs of federal-aid highway contracts than
others. For example, in one state, 6 of 9 contacts over $10 million had
exceeded the threshold, while in another state, all of the contracts were
under the threshold. While opportunities exist for FHWA to use this
information to better target its oversight efforts, it faces challenges in
doing so in light of

weaknesses recently reported by the DOT Inspector General's Office in its
financial management and reporting processes.15

    FHWA Does Not Measure its Performance over the Full Life of Projects

FHWA uses cost and schedule estimates developed relatively late in a
project's development-at the point at which the project is ready to go to
construction-as a baseline for measuring its performance. We have
discussed our concerns with FHWA's use of later estimates as its baseline
measure in earlier work.16 We have recognized that developing early
estimates is difficult; however, we have pointed out that using this late
estimate as a baseline for measuring cost growth provides a misleading
picture of actual cost growth. This is because cost estimates developed
much earlier in the project-for example, at the environmental review
stage-are used to make the public investment decision regarding the
project. By the time the project goes to construction, a public investment
decision effectively has been made, as substantial funds will have been
spent on designing the project and acquiring property, commitments will
have already been made to the public, and much of the increases in a
project's costs may have already occurred.

Moreover, by measuring its performance only after construction begins,
FHWA is not tasking itself with or establishing any accountability for
controlling cost growth during the part of the process where it exercises
direct oversight responsibility. Rather, it has focused its goals on the
phases of the project where it exercises less oversight. This is because
while FHWA is responsible for reviewing and approving certain state
transportation plans, environmental impact assessments, and the
acquisition of property for all projects, its role in approving the design
and construction of projects varies.

15We did not independently assess the reliability of FMIS data as the
Department's Inspector General has recently reported on weaknesses in
FHWA's financial management and reporting processes as part of the annual
audit of DOT's consolidated financial statements. In addition, our work
focused primarily on FHWA's use of FMIS data for oversight purposes,
rather than relying on FMIS data to support our findings and conclusions.

16GAO/RCED-97-47 and GAO-02-702T.

  FHWA's Oversight Activities Have Promising Elements and Limitations

FHWA and its major projects team undertook a number of activities to
improve its oversight efforts, which the major projects team documented in
its workplan summary (see app. II). Activities undertaken in response to
prior concerns included increasing the use of project oversight managers,
issuing guidance to states for improving cost estimates throughout the
life of projects, developing some information on cost growth of major and
other large projects, incorporating more risk assessments into its reviews
of state management processes, and attempting to address
congressionalcommittee direction to develop a multidisciplinary approach
to its oversight. FHWA's activities in these areas have promising elements
and limitations.

    FHWA Established Competencies for Project Oversight Managers but Did Not
    Establish Roles or Consistent Performance Expectations

FHWA has taken some positive steps in its use of project oversight
managers for major projects, but it has not yet defined the role of
project oversight managers or established agency wide performance
expectations for them. Currently, FHWA has assigned project oversight
managers to 14 of the 21 active major projects, compared with 7 project
oversight managers and 14 major projects in 2002. An FHWA official said
that 6 project oversight manager positions would be advertised soon for
the other projects and would be filled within 6 months.17 In August 2002,
it issued a core competency framework to identify the technical,
professional, and business skills that project oversight managers should
possess and to serve as a guide for selecting and developing these
managers. This core competency framework defines the skills and supporting
behaviors of project oversight managers in areas such as project and
financial management, contract administration, and program laws, and it
specifies the desired proficiency level for each competency at each grade
level.

FHWA has also taken steps to provide guidance and tools for project
oversight managers, including an online resource manual and other guidance
on reviewing project management plans and finance plans. It also made
major projects team staff available to assist the project oversight
managers in completing their reviews of such plans, and it sponsored
annual meetings for project oversight managers to share experiences.
Additionally, FHWA identified external training opportunities to help
managers reach or improve their core competency skills. FHWA sent a

17One major project, which is nearing completion, predates the major
project requirements, and a project oversight manager was not assigned to
it.

listing of these opportunities to project oversight managers via email and
invited these staff to enlist in courses that interested them. For the
future, FHWA's 2004 major projects team work plan summary envisions a
variety of additional activities to improve the effectiveness of project
oversight managers, including working with universities and training
vendors to establish a skill set development and certification program to
ensure that all project oversight managers acquire the same critical
skills and to establish a career path for them. According to FHWA, having
a career path would make the position of project oversight manager a more
attractive career option because it would provide opportunities to work
with more challenging projects and provide promotion opportunities so that
managers could advance within FHWA while staying in the project management
track.

However, there are limitations with FHWA's efforts so far. While the core
competencies define the skills that project oversight managers are
expected to possess, they do not define what the managers should do to
oversee a major project. FHWA has not yet articulated the role of project
oversight managers or established agency wide performance expectations for
them. In prior work, we established that setting performance expectations
that are linked to goals is important, as a specific alignment between
performance expectations and organizational goals helps individuals see
the connection between their daily activities and organizational goals.18

According to FHWA officials, project oversight managers are assigned to
the division offices, and each division office defines what its project
oversight manager does. At the three division offices we visited that had
major projects and project oversight managers, none had set performance
expectations for the project oversight manager that specifically tasked
the project oversight manager with achieving the goals and outcome
measures for the major projects. Project oversight managers and division
officials stressed the project oversight managers' close, hands-on
involvement with the state transportation agencies in the project, on an
almost daily basis. For example, project oversight managers and other
division office staff help state transportation agencies prepare finance
and project management plans, get involved in design, participate in
community outreach, and brief local political leaders on major projects.
However, the

18GAO, Results-Oriented Culture: Creating a Clear Linkage between
Individual Performance and Organizational Success, GAO-03-488 (Washington,
D.C.: Mar. 14, 2003).

extent to which the activity of the project oversight managers supported
DOT's cost and schedule goals was not clear.

Finally, without clear roles, responsibilities, and performance
expectations for project oversight managers that are clearly linked to
FHWA's goals, it is unclear what training is most needed to enable project
oversight managers to improve their performance and meet the agency's
goals. Our guidance for assessing training efforts cites the need for
training efforts to be an integral part of the strategic and performance
planning process and to focus on reaching the agency's goals, rather than
being implemented ad hoc.19 Currently, the training opportunities FHWA
offers to project oversight managers are identified by the major projects
team and are voluntary. There is no program of required courses-staff can
choose which courses they would like to take, or take no courses at all.
In March 2004, the head of FHWA's major projects team sent an e-mail to
the oversight managers advising them of available training. To date, three
project oversight managers and one other division office engineer have
each volunteered to take one or two courses.

FHWA officials told us they eventually plan to establish a certification
program for project oversight managers and to introduce a project
oversight manager skills-set or career path to make project management a
more attractive career option by setting out opportunities for more
challenging projects, and providing promotion opportunities. However, as
of December 2004, FHWA does not have a time frame for implementing its
plans, and officials told us these activities would not be implemented
without additional resources.

    FHWA Provided Guidance to States on Developing Cost Estimates but this
    Guidance is Voluntary and Covers Only Major Projects

In another positive step since 2002, FHWA has provided guidance to state
transportation agencies to assist them in applying sound cost estimating
practices, including guidance in developing more realistic early cost
estimates. However, this guidance is voluntary and covers only major
projects, and we found evidence that there is some resistance by FHWA
officials to focusing on developing earlier cost estimates. In past work,
we have identified problems related to FHWA's lack of accurate cost
estimates for projects. For example, in 1997, we found that cost increases
occurred on projects, in part, because the initial cost estimates were not
reliable

19GAO, Human Capital: A Guide for Assessing Strategic Training and
Development Efforts in the Federal Government, GAO-04-546G (Washington,
D.C.: Mar. 24, 2004).

predictors of the total costs or financing needs. Rather, these estimates
were developed for the environmental review-the purpose of which is to
compare project alternatives, not to develop reliable cost estimates. In
addition, each state used its own methods and included different types of
costs in developing its estimates, since FHWA had no standard requirements
for preparing cost estimates. Since that time, in 2003, FHWA surveyed its
division offices on cost estimating practices in their states and found a
variety of approaches to developing cost estimates, including manually
compiling estimates from historical data, using estimated quantity or cost
per mile calculations, or utilizing various externally or internally
developed software; one state reportedly lacked any formal process.
Similarly, the American Association of State Highway and Transportation
Officials (AASHTO) reported widely varying practices among the states in
developing cost estimates.

In June 2004, FHWA issued guidance that articulated the importance of
developing realistic early cost estimates that would be more stable as a
project progresses. Specifically, FHWA's guidance stated that it is
important that care be taken to present an achievable estimate even in the
early stages of project development, because logical and reasonable cost
estimates are necessary to maintain public confidence and trust throughout
the life of a major project. Moreover, the guidance recognized that cost
increases over and above the early planning and environmental estimates
for major transportation projects have become of increasing concern to
congressional and political leaders, federal and state top managers, and
auditing agencies. In addition to recognizing the difficulty of developing
more accurate cost estimates early in the project, this guidance includes
such components as what should be included in an estimate, how it should
be approved, factors to include in contingencies, and other information.
This guidance may help states move towards more consistent and reliable
cost estimates during the earlier planning phases when decisions are being
made about whether or not to go forward with the project, as well as the
project's potential design and construction.

FHWA also established help teams that travel to states that ask for
assistance in creating better estimates. For example, in March 2003 FHWA
was asked by the Kentucky and Indiana transportation departments for help
in reviewing the accuracy and reasonableness of the initial cost estimate
to complete the Ohio River Bridges project. This project includes two new
bridges over the Ohio River that would link eastern Louisville, Kentucky,
and Clark County, Indiana, with additional interchange improvements. FHWA
staff helped state officials identify the need for

revised cost estimates and more realistic completion dates based on such
factors as more realistic right-of-way costs, needed environmental
mitigation, revised contingencies, and updated inflation rates. A team of
federal and state staff working with consultants recommended that the
total cost estimate of the project be revised from $1.6 billion to $2.5
billion and that its expected completion date be revised from 2017 to
2020. State officials accepted these recommendations.

While these cost estimating guidance and assistance efforts represent a
positive step, it is too early to tell whether they will actually improve
cost estimating efforts in most states. Furthermore, there are indications
that there is some resistance among FHWA officials and states to
emphasizing the importance of more accurate early estimates in practice.
For example, some FHWA officials with whom we spoke said that costs cannot
be accurately estimated early because issues such as public opposition to
a project or unforeseen environmental mitigation procedures that are
determined necessary are likely to drive up the cost of a project. They
said early estimates should not be used as a basis for monitoring project
costs. Other FHWA officials believed that the estimate developed at the
conclusion of the design phase, as the project is ready for construction,
is the only realistic estimate to be used as a baseline. Some FHWA
officials told us that resolving concerns about cost estimates is more a
matter of managing public expectations, so that the public understands
that early estimates are not reliable and cannot be counted on, and that
the actual cost will exceed early estimates. AASHTO also believes that
accurately estimating costs at the early stages of a project can be a
challenge. According to a May 2004 AASHTO report, property acquisition
needs and environmental and regulatory requirements may not be fully known
early on, becoming clear only as the project progresses. Public input can
contribute to additional features being added to projects, known as "scope
creep," and litigation can delay a project, adding to costs because of
inflation.

We recognize that many challenges exist to developing more realistic early
estimates that more accurately reflect the expected cost of a project.
However, as we have also reported, relying on estimates prepared as a
project is ready to move to construction is too late in the process, as
substantial funds may have already been spent on designing the project and
acquiring property, and a public investment decision may, in effect,
already have been made. FHWA's guidance recognizes that steps can be taken
to take uncertainties into account when developing early cost estimates
through such means as developing contingencies. Some states have begun

taking action to improve the reliability of early cost estimates. For
example, Washington State's Cost Estimate Validation Process uses project
teams to identify risk factors, along with costs and mitigation strategies
for each factor. These results are then entered into a computer-based
modeling program that produces a range and a project cost estimate at the
90 percent confidence level, rather than a single dollar cost estimate.

DOT's proposed legislation for the reauthorization of TEA-21 in 2003
included provisions empowering the Secretary to develop minimum standards
for estimating project costs and to perform periodic reviews of state
practices for estimating project costs. These provisions were adopted in
bills that were separately approved by the House and the Senate in 2004
but that were not enacted before the adjournment of the 108th Congress.
According to FHWA officials, if these provisions are adopted, the
provisions may require them to move beyond voluntary guidance and issue
regulations covering states' practices for estimating costs.

    FHWA Has Started to Collect Some Cost Information, but it still Lacks the
    Capability to Determine the Extent of and Reasons for Cost Growth on
    Projects

FHWA has started to collect some cost information on some projects, but it
still lacks the capability to determine the extent of and reasons for cost
growth on projects so that it can better focus its oversight efforts. In
1997 we reported that cost growth occurred on projects, but the extent
could not be determined because FHWA's information system for highway
projects could not track total costs over the life of a project.20 In
2002, we testified that this information was still not available and noted
that recent congressional attempts to gather complete and accurate
information about the extent of and the reasons for cost growth had met
with limited success.21 In response to these concerns and requests from
Congress for data, FHWA has begun to collect project cost data, but it has
not substantially improved its ability to monitor total costs on projects.

FHWA has undertaken two efforts to collect information on the cost
performance of federally financed projects. First, it has started tracking
information on cost growth of major projects. The small number of these
projects allows the tracking to be done manually on a table containing
cost and schedule information for key aspects of each major project.
Second, FHWA has developed aggregated cost information on construction

20GAO/RCED-97-47. 21GAO-02-702T.

contracts over $10 million on a state-by-state basis. FHWA has done this
by comparing the current estimated costs of all contracts over $10 million
in each state with the engineering estimate developed before the contract
was awarded. However, as mentioned earlier, the state-by-state information
FHWA has developed has not yet been used to measure performance or target
its oversight efforts.

In spite of this progress, FHWA still does not have the capability to
measure the extent of and reasons for cost growth on projects. FHWA's
principal vehicle for tracking project costs is its financial management
system. This system is an accounting system, not a project information
system, and it tracks federal reimbursements by contract rather than by
project. Because one project can include many contracts over many years,
and the system does not automatically link contracts to projects, FHWA has
little easily accessible information to help it determine the total
overall costs of each project, other than the major projects it tracks
individually outside of its financial management system. In one case, FHWA
division staff told us that because FHWA's financial management system
does not track costs by project, the division developed its own
spreadsheet to track project costs.

Our recent work confirmed FHWA's continued difficulty with tracking cost
growth on projects. We randomly selected 14 contracts from 7 division
offices and asked FHWA's division offices to identify the project related
to each contract. We then requested consolidated cost information on the
14 projects. FHWA took an average of more than 3 months-and up to 6
months-to provide us this information for 12 of the 14 projects, and it
was unable to provide us complete cost information on the other 2
projects. (See app. I for more details.) The primary reason for FHWA's
difficulty in providing us with this information was that FHWA and state
staff could not easily or electronically compile information on a
project-by-project basis. For example, one division office said it had to
develop and run special transaction reports and manually extract the
information we wanted because the support files for the information were
at different locations, including a state district office, state
transportation agency offices, and comptroller offices. Another told us it
had to take the extra step of either combining or separating contracts in
order to compile information by project, which resulted in more "hand
work." Another said that files on contracts for one project were kept in
different locations depending on the stage of the project that the
contract was related to. As a result, quite a bit of staff time was tied
up as they attempted to get information from multiple departments of the
state transportation agency. FHWA's continued difficulties in maintaining
accurate and complete data to determine the

extent of cost growth on projects limit its ability to evaluate why cost
growth occurs, identify problems and solutions, target its oversight
efforts, and transfer lessons learned.

    FHWA Has Established Risk Assessments, but These Risk Assessments Are Not
    Always Used

FHWA expects its division offices to use some form of risk assessment to
help guide its reviews of state management processes, also known as
process reviews. However, risk assessments are not always being used
consistently or effectively. As we reported in 2002, FHWA issued a policy
in June 2001 encouraging its division offices to prioritize the risks in
the transportation programs in their states and to direct their oversight
efforts based on these results. The policy did not require a specific risk
assessment approach but allowed division offices flexibility in developing
an approach with their state agencies. FHWA considered its establishment
of risk assessment practices at the division offices to be the first of a
two-phased approach that would lead to an overall risk management program
for FHWA, which was still under consideration within FHWA's leadership as
of November 2004.

Each of the seven division offices we visited had developed a risk
assessment approach, and five out of seven of the offices were using these
risk assessments to guide their process reviews. However, at two division
offices, the results had not been used to direct their process reviews.
Staff at one division office we visited reported that although they had
been doing risk assessments for a few years, they did not use the results
to target state activities for review. Instead, they targeted state
activities for review by meeting with state officials to draw up an
intuitive list of state operations for process reviews. Similarly, another
division office had drafted a risk assessment approach, but it had not yet
tried to use it. Division office staff were skeptical that it would yield
better results than their own more intuitive approach to identifying which
state program operations warranted a process review.

In addition, in November 2004 the DOT Inspector General reported that
FHWA's risk assessments were voluntary and did not provide a systematic
approach for assessing program risks throughout the agency. The Office of
Inspector General (OIG) reported that risks assessments varied
significantly in the scope and methodology used and how the assessment
results were rated and classified. As a result, some major programs were
not reviewed, and risk assessment results were not reliable or comparable
across states. To improve FHWA's process for managing risk, the OIG
recommended that FHWA require all division offices to conduct risk

assessments and that it issue guidance to division offices to ensure risk
assessments are conducted more strategically and with a disciplined
methodology. The OIG further recommended that FHWA analyze trends within
individual risk assessments to identify agency wide issues and problems
and establish a systematic follow-up process to ensure that oversight
attention is given to high-risk areas.22 FHWA was in the process of
reviewing and responding to the OIG's recommendations when we concluded
our review.

    FHWA Has Begun to Take Steps to Develop a Multidisciplinary Approach to
    Oversight, but its Efforts so Far Have Been Limited

In February 2003, in the Conference Committee Report for the DOT fiscal
year 2003 continuing appropriations, the conferees expressed continuing
concern about FHWA's management of major projects, and in particular, a
concern that FHWA's traditional engineering focus had inhibited oversight
in such areas as financing, cost control, and schedule performance.23
Accordingly, FHWA was directed to evaluate the range of disciplines and
skills within its staff and to develop a strategy for achieving a more
multidisciplinary approach towards its oversight activities, including
identifying staff with such skills as financing and cost estimation.

However, FHWA's human capital plan does not incorporate strategies for
developing a workforce to support a more multidisciplinary oversight
approach. In prior work, we noted that the process of strategic workforce
planning addresses two critical needs: (1) aligning an organization's
human capital program with its current mission and programmatic goals; and
(2) developing long-term strategies for acquiring, developing, and
retaining staff to achieve programmatic goals.24 To some extent, FHWA's
human capital plan does this for the agency's current vital few priorities
of safety, congestion mitigation, and environmental stewardship. But the
agency's oversight mission is not truly incorporated into the plan. FHWA's
human capital plan acknowledges the congressional-committee direction FHWA
received to develop a more multidisciplinary approach to oversight. The
plan states that this approach will require the development or acquisition
of

22DOT Office of Inspector General, Managing Risk in the Federal-Aid
Highway Program, MH-2005-012 (Washington, D.C.: Nov. 19, 2004).

23U.S. House of Representatives Conference Report 108-10, Making Further
Continuing Appropriations for the Fiscal Year 2003, and Other Purposes
(Feb. 12, 2003). p. 1263.

24GAO, Human Capital: Key Principles for Effective Strategic Workforce
Planning, GAO04-39 (Washington, D.C.: Dec. 11, 2003).

new skills, specifically in the areas of financing, funds accountability,
project-level cost control, schedule performance, process management, and
transportation planning. However, FHWA's human capital plan does not
relate these needed skills to the skills possessed by its present
workforce, nor does it address how these skills will be developed or
acquired. Instead, FHWA's human capital plan is essentially a plan for
replacing individuals in its current key occupations whom it expects to
lose through attrition over a 5-year period.25

Additionally, strategies for developing a multidisciplinary approach were
not reflected in FHWA's guidance to its division offices for developing
their workforce plans. This year, FHWA required its division offices and
other units to prepare a workforce plan for the upcoming 3-to-5 year
period identifying anticipated skill gaps in their workforce. However, the
guidance FHWA provided did not mention the multidisciplinary skills that
FHWA had identified in its human capital plan. As we have pointed out in
prior work, when planning for the future, leading organizations go beyond
simply replacing individuals and engage in broad, integrated planning and
management efforts that focus on strengthening both current and future
organizational capacity.26 This is particularly important for FHWA, as its
traditional engineering focus has drawn congressional committee concern
that has led to direction to develop a multidisciplinary approach towards
its oversight activities.

Similarly, FHWA's recruiting efforts do not incorporate strategies for
developing a more multidisciplinary approach to project oversight. Like
its human capital plan, FHWA's recruitment plan for 2003 through 2005 is
primarily a plan for hiring to fill the agency's traditional occupations.
The recruitment plan does not set any specific goals or objectives for
acquiring needed multidisciplinary skills that FHWA articulated in its
human capital plan, such as project level cost control, schedule
performance, process management, and transportation planning. Under the
recruiting plan, the development of a multidisciplinary approach is
addressed through FHWA's professional development program (PDP). FHWA's
PDP, which historically focused on engineers, is a 2-year program that
provides developmental

25The plan focuses on what FHWA has identified as its mission critical
occupations: civil engineers, planners, environmental specialists,
financial management, engineering technicians, realty specialists, and
transportation specialists.

26GAO, Human Capital: Succession Planning and Management is Critical
Driver of Organizational Transformation, GAO-04-127T (Washington, D.C.:
Nov. 1, 2003).

assignments and on-the-job and classroom training for entry-level staff.
Officials told us that PDP staff are now being given assignments allowing
them to develop a broader range of skills at the start of their careers,
including assignments to division offices with major projects. They also
note that over recent years FHWA has been hiring fewer engineers for its
PDP programs and more staff from other backgrounds.

The other principal component of FHWA's response to congressional
committee direction to develop a multidisciplinary approach to project
oversight is training, but the agency has made limited progress in
developing new courses to bring new skills to its workforce. Only two new
training courses were being developed specifically to address needed
skills-a course on project cost estimation and a course on project
management for managers in division offices. As of November 2004, both
courses were being pilot tested.27 As we noted earlier, it is important
for training to be an integral part of an agency's performance planning
process to ensure that it contributes to reaching agency goals. However,
in its fiscal year 2005 performance plan, FHWA allows divisions the
discretion to decide whether or not to participate in multidisciplinary
training for its project oversight managers and professional development
program staff. In addition, as noted earlier, FHWA has identified and
offered external training courses to project oversight managers, but to
date only a few managers and other key division staff have expressed an
interest. Even so, FHWA human resources officials we spoke to told us they
believed that the congressional committee's direction to develop a
multidisciplinary approach to project oversight has been largely met
through their already existing training efforts. These efforts include
making courses available on risk assessment techniques, conducting process
reviews, and implementing financial management improvements.

In addition to FHWA's limited progress in developing strategies for
meeting this congressional-committee direction, FHWA has not fully
embraced the need to develop a more multidisciplinary approach to
oversight. FHWA human resources officials we spoke to believed the concern
that FHWA's workforce is centered on engineering at the expense of other
project oversight skills is based on a misperception-that is, not
recognizing that FHWA engineers take on many other tasks not strictly
related to

27FHWA officials told us the project management course is targeted to
about 100 FHWA officials, and project cost estimating will be provided to
about 450 staff, both over the next 2 years.

engineering. Furthermore, two division office officials we spoke to in the
course of our work questioned the need for FHWA to focus on
multidisciplinary skills. One division administrator commented that
"multidisciplinary" means that a person can do many things, and therefore
that division's staff was already multidisciplinary. The deputy
administrator in another division questioned what was meant by
multidisciplinary skills, believing there was no guidance from
headquarters on this.

  FHWA Faces Challenges to Improving Oversight

FHWA's efforts to improve oversight face several challenges. These
challenges stem from the structure of the federal-aid highway program and
the culture of partnership that has resulted between FHWA and the states.
These challenges also stem from FHWA's decentralized organization, human
capital challenges that mirror those faced throughout government, and
FHWA's perception that it has received conflicting signals on its
oversight role over the years. Because these challenges are in large part
rooted in FHWA's organization and culture, and in the structure of the
program it administers, they may be difficult to surmount.

    Structure of the Federal-Aid Highway Program

Because the federal-aid highway program is a state-administered, federally
assisted program, it provides states broad flexibility in deciding how to
use their funds, which projects to pick, and how to implement them.
Furthermore, states are exempt from FHWA oversight on design and
construction of many projects. Although DOT has articulated goals and
outcome measures for the federal-aid highway program, such as improving
safety and reducing the growth of traffic congestion, FHWA must implement
and achieve these goals through a program over which it exercises limited
control. Our past work across government programs has shown that in
programs that have limited federal control, agencies face challenges to
ensure that federal funds are efficiently and effectively used.28 We have
also found that these challenges can be successfully overcome, in some
cases, by ensuring that the program has clear goals and strong analytical
data to measure program results. However, as stated earlier, FHWA's
efforts both to implement its goals and to collect and analyze data on
project costs have fallen short.

28GAO, Managing for Results: Measuring Program Results That Are Under
Limited Federal Control, GAO/GGD-99-16 (Washington, D.C.: Dec. 11, 1998).

Exacerbating this challenge is the fact that, as our August 2004 report
highlighted, the federal-aid highway program does not have the mechanisms
to link funding levels with the accomplishment of goals and outcome
measures that DOT has articulated.29 We have also reported that although a
variety of tools are available to help measure the potential performance
outcomes, such as those that measure the costs and benefits of
transportation projects, such potential outcomes often do not drive
investment decisions, as many political and other factors influence
project selections.30 For example, the law in one state requires that most
highway funds, including federal funds, be distributed equally across all
of the state's congressional districts. Consequently, the structure of the
federalaid highway program provides no way to measure how funding provided
to the states is being used to accomplish particular outcomes, such as
reducing congestion or improving safety, and little assurance that
projects most likely to accomplish goals and outcome measures articulated
by DOT will be funded. The absence of such a link may make it more
difficult for FHWA to define its role, the purpose of its oversight, and
what its oversight is designed to accomplish.

In August 2004, we reported that policy makers may wish to consider
realigning the federal-aid highway program's design, structure, and
funding formulas to take into account the program's goals and to include
greater performance and outcome oriented features. We also said that such
consideration could include the appropriate roles of the federal and state
governments, including what type of administrative structure for
overseeing the federal-aid highway program would best ensure that the
performance goals are measured and accomplished. Our report provided
Congress with a matter for congressional consideration and said that the
proposed National Commission to assess future revenue sources to

29GAO-04-802.

30GAO, Surface Transportation: Many Factors Affect Investment Decisions,
GAO-04-744 (Washington, D.C.: June 30, 2004).

support the Highway Trust Fund might be an appropriate vehicle through
which to examine these options.31

Culture of Partnership 	Consistent with the structure of a state
administered, federally assisted program, FHWA has developed a culture of
partnership with the states. This culture of partnership dates back to the
Federal-Aid Road Act of 1916, when the program was funded through a 50
percent federal and 50 percent state matching share. This partnership
approach recognizes that states select, plan, and build projects, while
FHWA ensures that federal laws and other requirements are followed by
maintaining a close, hands-on involvement with state transportation
agencies in delivering projects. FHWA and state officials believe that
over the years this partnership has helped to build trust and respect
between state transportation agencies and FHWA and ensure that priorities
such as safety and the environment are addressed, and has resulted in
projects being built more economically and efficiently.

However, there is a potential down side to this partnership approach. When
a project overseer becomes an active partner in a project, an arms-length,
independent perspective can be lost. In fact, FHWA's partnership approach
to project oversight has failed in the past. FHWA had an oversight manager
on the Central Artery/Tunnel Project in Boston, Massachusetts, a project
that experienced widely-reported cost increases, growing from around $2.3
billion in the mid-1980s to almost $15 billion by 2004. In March 2000, an
FHWA task force charged with reviewing FHWA's oversight of the project
found that FHWA had been caught unaware earlier that year when the state
revealed an estimated $1.4 billion cost increase. The task force
attributed this to FHWA's over reliance on trust between itself and the
state, reporting that FHWA's partnership approach failed to achieve
independent and critical oversight of the project.

FHWA officials acknowledged that independence is critical to effective
oversight and also acknowledged the need to closely monitor the
performance and independence of their project oversight managers on an

31This commission, proposed by the administration and included in both the
Senate and House reauthorization bills approved during 2004 but not
enacted before the adjournment of the 108th Congress, is to consider how
the program is financed and the role of other stakeholders in financing
it. As we reported, the appropriate program structure and mechanisms for
delivering that funding are important components of making these
decisions.

ongoing basis. However balancing the role of overseer and partner can be
difficult. In one state we visited, the division's oversight manager for a
major project had business cards that identified him as a member of the
state's project team-with the project's logo, Website, and e-mail address
printed on the card-rather than as a federal employee. Only his position
title on the card, "FHWA Project Administrator," identified him as an FHWA
employee, rather than as a state employee. Ensuring that FHWA oversight
personnel maintain an independent perspective is especially critical given
the current lack of linkage between FHWA's performance goals and the roles
and expectations of its project managers.

Another potential challenge presented by FHWA's culture of partnership
with the states is that it may have prevented FHWA from considering other
models for project oversight-including some models in use within DOT. For
example, the Federal Transit Administration (FTA) uses competitively
selected engineering firms as oversight contractors to monitor major mass
transit projects costing over $100 million. During the project's design,
the contractor reviews the grantee's plan for managing the project and
determines whether the grantee has the technical capability to complete
the project. Once FTA approves the plan, the contractor monitors the
project to determine whether it is progressing on time, within budget, and
according to plan. In prior work, we noted that FTA's project management
oversight program benefited both the agency and the grantees carrying out
the projects.32 As another example, DOT established a Joint Program Office
to help carry out the Transportation Infrastructure Finance and Innovation
Act Program, which provides credit assistance to states and other project
sponsors for surface transportation projects. This office reviews and
evaluates proposed projects for participation in the program, reviews
financial plans and progress reports during project construction, monitors
the project sponsor's credit, and coordinates site visits and other
oversight activities with DOT field offices.

FHWA's Organization 	FHWA administers the federal-aid highway program
through a decentralized division office structure and delegates much of
FHWA's decisionmaking and program implementation to those offices.
Therefore, FHWA's division administrators enjoy wide latitude to implement
their programs. FHWA has had a field office in every state since 1944,
and,

32GAO, Mass Transit: Project Management Oversight Benefits and Future
Funding Requirements, GAO/RCED-00-221 (Washington, D.C.: Sept. 15, 2000).

according to FHWA and state officials, this arrangement gives maximum
flexibility to the people closest to the customer and to the issues to
make decisions best suited to particular needs and situations. According
to FHWA officials, this decentralization of decisionmaking and program
implementation to the division offices increased after 1998 and the
passage of TEA-21, which eliminated FHWA's nine regional offices.

While this flexibility may have benefits, decentralization presents
challenges for the implementation of a consistent national leadership
vision and strategies. These long-standing organizational arrangements may
have contributed to such conditions as the lack of uniform performance
expectations for project oversight managers, widely varying methods used
to develop cost estimates for projects, and different approaches to doing
risk assessments. Some limitations are by design. For example, while
FHWA's fiscal year 2005 performance plan discusses multidisciplinary skill
training for its oversight managers and professional development program
staff, it also specifically grants division administrators the discretion
about whether to participate. FHWA officials acknowledged the challenges
of consistently implementing national level goals and programs among the
many division offices.

Human Capital	Our 2003 update to our High-Risk Series of reports
recognizes that strategic management of human capital continues to be a
high-risk area government wide.33 Although considerable progress has been
made since we first designated human capital a government wide high-risk
area in 2001, federal human capital strategies are not yet appropriately
constituted to drive the transformation that is needed across the federal
government. Among the challenges agencies face are the need to improve
their ability to acquire, develop, and retain talent, and the need to
better and more fully integrate these and other human capital efforts with
agencies' missions and program goals.

For FHWA, this government wide challenge manifests itself in a number of
ways, including the need to transform its workforce and culture to meet
its evolving mission. FHWA's workforce partnered with the states to build
the Interstate Highway System from 1956 into the 1990s. FHWA needed
engineering skills to perform tasks, such as detailed reviews of design

33GAO, High-Risk Series: Strategic Human Capital Management, GAO-03-120,
(Washington, D.C.: Jan. 1, 2003).

plans and inspections of construction progress to ensure that national
uniformity in terms of design and safety was established throughout the
interstate system. These skills were especially important because,
according to FHWA and state officials, state transportation agencies did
not have the equivalent capability to do the job at that time. In recent
years Congress has recognized the increased capacity of state
transportation agencies and increasingly delegated approval authorities to
the states, including the authority over design and construction decisions
for most projects. As a consequence, FHWA's oversight role and mission
have evolved to include, for example, greater reliance on broad reviews of
state management processes.

As FHWA's oversight role and mission evolves, FHWA faces the challenge of
transforming its workforce and culture to evolve with this role and
mission. In our discussions with FHWA field staff, we noted reluctance
among some FHWA staff to focus on these broader reviews that FHWA
increasingly relies on because they see these as less important than the
traditional tasks of reviewing design plans and inspecting the progress of
construction. Division office officials in two states we visited told us
that change has been an issue for its more tenured staff. For example, the
Administrator at one office had begun to hire staff with a variety of
skills, while officials at the other office saw a need for more
specialists, including staff with financial expertise. Officials also said
some staff have resisted doing process reviews because they see it as
functioning as auditors rather than as partners with the state in
delivering projects, which is how they prefer to be seen. Overcoming these
challenges will become even more important in the years ahead should
proposed legislation increasing FHWA's oversight responsibilities be
enacted.

FHWA's Perception of In 2001, a FHWA task force concluded that changes in
the agency' s Conflicting Legislative oversight role mandated by highway
program authorizations enacted in Direction on Oversight Role 1991 and
1998 had resulted in internal confusion and wide variation in

interpretations by FHWA personnel covering the agency's roles andand
Responsibilities responsibilities in overseeing projects.34 In 2002, we
reported that FHWA could not say whether it had resolved the internal
confusion and variations

34FHWA Stewardship/Oversight Task Force Final Report, March 20, 2001.

in interpretations of the agency's oversight role identified by the task
force.35

During our review we found that some confusion continues, as some of the
FHWA personnel we spoke to expressed the view that Congress has sent mixed
messages about the extent to which it would like to see FHWA oversee
projects. According to some division and headquarters FHWA officials,
federal laws over the years have required FHWA to withdraw from direct
oversight of most projects, while at the same time, legislation has
increased the oversight requirements for major projects, resulting in
mixed signals. Changes that were proposed by DOT and passed by the House
and the Senate in 2004 but not enacted before the adjournment of the 108th
Congress could, if reintroduced and enacted by the 109th Congress, help
clarify FHWA staff's perception of their oversight role by, for example,
mandating reviews of state financial system, developing cost estimating
standards, and cascading requirements for major projects to other
projects. Enactment of these provisions would also provide Congress the
opportunity to provide a more detailed explanation of and purposes for
these provisions regarding FHWA's role versus the states' role in
overseeing cost and schedule performance of federal-aid highway projects
in the legislative history accompanying the reauthorization bill. As we
stated in our 2002 testimony, such clarification would be helpful.

  Best Practices Can Help Improve Progress and Address Challenges to Improving
  Project Oversight

Reports and analyses published by us, OMB, and the National Research
Council suggest a set of best practices that agencies can benefit from in
conducting effective oversight of large infrastructure projects such as
those in the federal-aid highway program overseen by FHWA. While these
reports and analyses tend to focus more on overall project management,
there are elements in each of them that relate specifically to improving
project oversight. From our review of these reports and analyses, we
identified four best practices that are particularly applicable to FHWA's
oversight efforts and that FHWA officials and decision makers can consider
to help effectively oversee large infrastructure projects and states'
financial and management processes. While some of these best practices are
beginning to be reflected in FHWA's activities, as a whole, they could
provide a framework for moving to a comprehensive approach to project
oversight. These best practices are 1) establishing measurable project

35GAO-02-702T.

oversight goals and communicating these goals down through all levels of
the agency, 2) establishing project oversight manager role and
accountability based on oversight goals, 3) providing professional
training and a career path, and 4) learning lessons and transferring them.

    Establishing Measurable Project Oversight Goals and Communicating These
    Goals through All Agency Levels

As we discussed earlier, agencies seeking to make oversight a priority
should establish measurable project oversight goals that help it carry out
its mission and define what its oversight is designed to accomplish-and
should communicate these goals down through all levels of the agency.
Having measurable goals gives managers the means to objectively and
quantifiably assess progress toward achieving certain outcomes. If an
agency relies only on general goals to guide its efforts, the agency will
not have any way of determining whether it achieves those goals since it
has not first identified a way to quantify or measure the outcome. Once
these goals are established, agencies should communicate these goals down
to all levels of the agency. One way to ensure that the goals are
communicated effectively is to link the agency's day-to-day activities to
these goals. Our 1998 report on leading practices in capital
decision-making added that clear communication of an organization's vision
and goals is a prerequisite for success. Top-level officials develop the
organization's priorities and communicate them downward to subunits within
the organization. Based on these goals, managers at all levels work to
produce plans and activities that outline their individual strategies for
achieving top-level goals.36

    Establishing Project Oversight Manager Role and Accountability Based on
    Oversight Goals

Once an agency establishes its oversight goals, it should incorporate
those goals into its strategies and activities by making oversight
managers accountable for the effective implementation of the goals. We
recently recommended that Amtrak adopt policies and procedures for
managing infrastructure projects that, among other things, include
mechanisms to ensure accountability for a project's success. We stated
that such mechanisms should clearly indicate the individuals responsible
for implementing the project, the expectations for their performance, the
ways their performance will be measured, and the potential consequences
for

36GAO. Executive Guide: Leading Practices in Capital Decision-Making,
GAO-AIMD-99

32. (Washington D.C.: Dec. 14, 1999).

failing to meet expectations.37 In this report, we noted that some of the
railroads we had contacted tied pay and personnel decisions to
performance, holding project managers directly responsible for the
project's success and failure. In other previous work, we have also noted
that how such pay for performance efforts are done, when they are done,
and the basis on which they are done can make all the difference in
whether such efforts are successful.38 In addition, in other prior work,
in 2000, we found a number of emerging benefits from the use of
resultsoriented performance agreements for executives, including, among
other things, providing results-oriented performance information to serve
as the basis for executive performance evaluations. 39

    Providing Professional Training and a Career path

Professional training enables oversight staff to understand their expected
roles in achieving the agency's oversight goals. Having a view of a future
career is also desirable for the development of oversight staff. In 1999
the National Research Council reported that the Department of Energy could
improve its project performance by developing skills, training
opportunities, and a career path in project management. The report added
that the agency needed to establish criteria and standards for selecting
and assigning project managers, including documentation of training, and
should require that all project managers be trained and certified.40 In
prior work, we have found that an agency's training program should be
linked to achieving the agency's strategic goals, while specific training
for each individual should be based on his or her developmental needs.41

37GAO. Intercity Passenger Rail: Amtrak's Management of Northeast Corridor
Improvements Demonstrates Need for Applying Best Practices, GAO-04-94
(Washington, D.C.: Feb. 27, 2004).

38GAO, Human Capital: Implementing Pay for Performance at Selected
Personnel Demonstration Projects, GAO-04-83. (Washington, D.C.: Jan. 23,
2004).

39GAO, Managing for Results: Emerging Benefits From Selected Agencies' Use
of Performance Agreements, GAO-01-115 (Washington D.C.: Oct. 30, 2000).

40National Research Council, "Improving Project Management in the
Department of Energy," National Academy Press, 1999.

41GAO-04-546G.

    Lessons Learned and Transferred

Effective oversight also requires a proactive approach to establishing
evaluation mechanisms, collecting information, and transfering lessons
learned on an ongoing basis. Learning from past successes and mistakes and
sharing that information with decision makers, agency officials, and
project managers is a critical element for effective oversight. Our 1996
executive guide to help agencies implement GPRA reported that agencies
analyzing the gap between where they are and where they need to be to
achieve desired outcomes can target those processes that are in most need
of improvement, set realistic improvement goals, and select an appropriate
process improvement technique such as benchmarking. Benchmarking compares
an internal agency process with those of private and public organizations
that are thought to be the best in their fields.42 In addition, our 1998
report on leading practices in capital decision making also found that
agencies could evaluate and compare results with goals by using financial
and non-financial criteria that link its overall goals and objectives. In
2000, we reported that agencies conducting program evaluations improved
their measurement of program performance or understanding of performance
and how it might be improved.43 In addition, our Executive Guide on
Capital Decision-Making identified practices federal agencies can
implement to enhance their evaluation processes.44 In 1997, OMB stated in
its Capital Programming Guide45 that agencies should be able to document
and support the accomplishment of the respective agency goals. Agencies
can also evaluate the planning and procurement process to determine
whether a project accurately predicted the desired benefits 3 to 12 months
after it has become operational. The Guide added that conducting a project
post-implementation review that evaluates the success or failure of
projects serves as an assessment. The review compares actual results
against planned cost, returns, and risk. The results are used to calculate
a final return on investment, determine whether any additional project
modifications may be necessary, and provide lessons learned for changes

42GAO, Executive Guide: Effectively Implementing the Government
Performance and Results Act, GAO/GGD-96-118. (Washington D.C.: June 1996).

43GAO, Program Evaluation: Studies Helped Agencies Measure or Explain
Program Performance. GAO/GGD-00-204. (Washington D.C.: Sept. 29, 2000).

44GAO, Executive Guide: Leading Practices in Capital Decision-Making,
GAO-AIMD-99

32. (Washington D.C.: Dec. 14, 1998).

45Capital Programming Guide, Office of Management and Budget. July 1997.
The Guide is a supplement to Circular A-11, which the Office of Management
and Budget issued to help agencies integrate and implement GPRA
requirements.

to the agency's capital programming processes and strategy. Finally, the
National Research Council's 1999 report stated that agencies such as the
Department of Energy should transfer knowledge gained about cost
estimating techniques, project review processes, change control
mechanisms, and performance metrics from one project to another.

Conclusions	FHWA has made progress since 2002 in improving its oversight
efforts, including its direct oversight of major projects and its broader
reviews of state management processes that are used to oversee states'
management of most other projects. For example, FHWA's actions to enhance
the capabilities of project oversight managers overseeing major projects
and to incorporate risk assessments into its reviews of state management
processes are both positive steps towards improving oversight. Most
significantly, FHWA has established, for the first time, goals and
measures that clearly make containing project costs and schedules an
integral part of how FHWA conducts its oversight.

However, despite promising results, FHWA's efforts have also had
limitations. FHWA still lacks a comprehensive approach to ensuring that
its oversight of federal-aid highway projects supports the efficient and
effective use of federal funds. A comprehensive approach would avail
itself of best practices and would include (1) goals and outcome measures
with activities and performance expectations set for its staff that are
linked to these goals and measures; (2) an overall plan for FHWA's
oversight initiatives and activities that responds to past concerns raised
about its program and is tied to its goals and measures; (3) workforce
planning efforts that support the goals, measures, and overall plan; (4)
centrally defined roles and responsibilities for key staff, such as
oversight managers for major projects; and (5) the capability to track and
measure costs over the life of projects in order to identify problems,
help target resources, and transfer lessons learned. Without such a
comprehensive approach, FHWA cannot ensure that its varied activities are
resulting in tangible improvements in the quality of its oversight and in
the performance of federal-aid projects. Furthermore, without a
comprehensive approach, FHWA is not able to articulate what it wants its
oversight to accomplish, the composition of its workforce to accomplish
it, and how it will measure whether its efforts have or have not been
successful. Thus, it is limited in its ability to ensure that its
oversight efforts are meeting its organizational goals, that these efforts
address concerns that have been raised, and that they result in more
effective and efficient use of federal funds.

Although broader questions exist about the structure of the federal-aid
highway program and the role of FHWA, the agency will face considerable
increases in its oversight responsibilities in the years ahead,
particularly if the proposals made by DOT and considered by Congress
become law. Given the limitations present today, questions exist about the
ability of FHWA to effectively absorb these new responsibilities and to
improve its oversight of the federal-aid highway program in the years
ahead. Moreover, absent a comprehensive approach, FHWA is unlikely to be
able to overcome the structural, organizational, and cultural challenges
it faces in effectively overseeing the federal-aid highway program.

Recommendations	In order to establish a comprehensive approach to project
oversight, we recommend that the Secretary of Transportation direct the
Administrator, FHWA, to take the following four actions:

o 	link FHWA's day-to-day activities and the performance expectations set
for its staff to its goals and outcome measures;

o 	develop an overall plan for its oversight initiatives that is tied to
its goals and measures, along with priorities and time frames, and that
includes workforce planning efforts that support these goals and measures;

o 	improve the use and performance of project oversight managers by
centrally defining their role and responsibilities; and

o 	develop the capability to track and measure costs over the life of
projects to help identify the extent of and reasons for problems, target
resources, and transfer lessons learned.

Agency Comments 	We provided a draft of this report to DOT and met with
FHWA officials, including the Deputy Administrator, to obtain their
comments on the draft. FHWA officials generally agreed with the facts and
conclusions in the report and our characterization of the challenges FHWA
faces in improving its project oversight. FHWA officials emphasized that
although we highlighted potential drawbacks associated with both its
culture of partnership with the states and its decentralized organization,
this partnership and organization are also major strengths of the
federal-aid program that will allow the agency to absorb potential new
responsibilities, help overcome challenges, and improve program oversight
in the future

through a more comprehensive approach. FHWA officials did not take a
position on our recommendations, but they stated that they would be taking
them under advisement. They also suggested some technical and clarifying
comments that we incorporated into the report as appropriate.

We are sending copies of this report to the Honorable Norman Mineta,
Secretary of Transportation. We will also make copies available to others
upon request. In addition, the report will be available at no charge on
the GAO Web site at http://wwwgao.gov.

If you have any questions about this report, please contact me at
[email protected], or (202) 512-6570 or contact Steve Cohen at
[email protected] or (202) 512-4864. GAO contacts and acknowledgments are
listed in appendix III.

Sincerely yours,

Katherine Siggerud Director, Physical Infrastructure

Appendix I

Scope and Methodology

We reviewed the Federal Highway Administration's (FHWA) approach to
improving its federal-aid highway project oversight efforts since 2002,
including (1) FHWA's oversight-related performance goals and measures, (2)
FHWA's oversight improvement activities, (3) challenges FHWA faces in
improving project oversight, and (4) best practices for project oversight.

We reviewed FHWA's oversight-related goals and measures by evaluating
Department of Transportation (DOT) and FHWA strategic and performance
plans, and supporting documents, from 2001 through 2004. We also reviewed
FHWA's annual performance reports from 2002 and 2003 and current OMB
President's Management agenda documents. We also reviewed FHWA and DOT
fiscal year 2005 budgets. As criteria in reviewing this information we
used GAO published guidelines and prior reports, including GAO's 2001
Performance Guide and GAO's 2003 Results Oriented Culture and GAO's 2003
Human Capital reports.

To review FHWA's oversight improvement activities we documented and
analyzed the status of FHWA's various project oversight efforts since 2002
using FHWA's FY 2004 Work Plan Summary from the major projects team (see
app. II). We also reviewed FHWA's use of financial information from its
Financial Management Information System (FMIS) to track and analyze trends
in cost growth on projects. We did not independently assess the
reliability of FMIS data as the Department's Inspector General has
reported on weaknesses in FHWA's financial management and reporting
processes, most recently in November 2004 as part of the annual audit of
DOT's consolidated financial statements. In addition, our work focused
primarily on FHWA's use of FMIS data for oversight purposes, rather than
relying on FMIS data to support our findings and conclusions. In addition,
to document continued difficulty in tracking cost growth on projects, we
randomly selected 14 contracts from seven division offices, each of which
had an estimated total cost of between $25 million and $50 million. We
then asked FHWA's division offices to identify the project related to each
contract (each contract was part of a different project, so there were 14
projects), and requested consolidated cost information on the 14 projects.
FHWA took an average of more than 3 months-and up to 6 months-to provide
us this information for 12 of the 14 projects, and it was unable to
provide us complete cost information on the other 2 projects. Finally, we
also interviewed officials at FHWA Headquarters, selected FHWA division
offices, state departments of transportation, and other officials to
document oversight implementation efforts.

Appendix I Scope and Methodology

We performed work at seven FHWA division offices and states located in
Colorado, Georgia, Missouri, Nevada, Pennsylvania, Washington, and
Wisconsin. We selected these 7 FHWA division offices and corresponding
states by selecting states that had a current or planned major project and
some that did not; states with large as well as relatively small
federal-aid highway programs in terms of funding; large and small FHWA
division offices as measured by the number of staff; and division offices
and states that FHWA and the American Association of State Highway and
Transportation Officials (AASHTO) officials had recommended because of
ongoing initiatives related to project oversight and management.

To document and review the challenges FHWA faces in improving its project
oversight we used our past work and interviewed FHWA headquarters,
division office and state transportation program officials. We also
interviewed AASHTO officials and state audit and evaluation organizations
across the country.

To address the use of best practices as a framework for the oversight of
large highway infrastructure projects, we conducted a literature search in
2004 to identify best practices related to oversight management. The
literature included our previous reports and guidelines on best practices
related to project management. It also included publications from the
Office of Management and Budget (OMB) that provided detailed guidance to
federal agencies on planning, budgeting, acquisition, and management of
capital assets and from the National Research Council addressing methods
the Department of Energy could implement to improve its project
management, including oversight of environmental restoration, waste
management, and construction projects. From this literature search, we
compiled the list of best practices that can provide FHWA with a
comprehensive approach and basic framework for effectively overseeing
highway projects. For the first practice of establishing measurable
project oversight goals we used information from two of our reports
related to the Government Performance and Results Act and another report
related to leading practices in capital decision-making. For the second
practice of establishing project oversight manager role and accountability
based on oversight goals, we used our report related to improving project
management for Amtrak and another of our reports on performance
agreements. For the third practice of providing professional training and
a career path, we used a National Research Council report on improving
project management at the Department of Energy. For the fourth practice of
learning lessons and transferring them, we used information from the
National Research Council report mentioned above, the GAO report on

Appendix I Scope and Methodology

leading practices in capital decision-making, another GAO report on
program evaluations, and OMB guidance in Circular A-11 and its Capital
Programming Guide.

Appendix II

FHWA FY 2004 Major Project Team Work Plan Summary

Activity

Issue final project management plan guidance
Develop cost estimating guidance
Issue major project delivery template
Establish independent cost estimating review program
Establish major project cost and schedule measures
Monitor project cost growth
Coordinate with other transportation programs
Reevaluate major project finance plan guidance
Develop finance plan guidance for projects over $100m
Identify trends in finance plans
Host new FHWA hires
Manage project oversight manager positions
Develop clear statement of team objectives
Implement major project team marketing plan
Develop working relationships with other DOT agencies
Develop executive level major projects training
Develop risk-based conflict management model
Facilitate organizational career track for oversight managers
Monitor selected projects
Review project finance plans and annual updates
Conduct training and outreach
Establish risk management program
Promote oversight best practices
Restructure oversight group website
Establish and lead oversight working group
Establish cross-functional oversight website
Identify risk assessment training opportunities
Identify national trends in process reviews
Establish oversight agreement repository
Develop new generation of oversight performance measures
Develop performance pilot measures
Facilitate the development of cost growth countermeasures
Prepare major project team report to Congress
Conduct monthly project tracking activities
Develop major project continuum model

Status

Completed
Completed
Under Development
Under Development
No Work Started
Ongoing
Ongoing
Under Development
No Work Started
Under Development
Ongoing
Ongoing
Completed
Ongoing
Ongoing
Under Development
No Work Started
No Work Started
Ongoing
Ongoing
Ongoing
Under Development
Ongoing
Under Development
Under Development
Under Development
Under Development
No Work Started
Completed
No Work Started
No Work Started
No Work Started
Ongoing
Ongoing
Under Development

Appendix II FHWA FY 2004 Major Project Team Work Plan Summary

(Continued From Previous Page)

Activity

Establish skill sets development program for oversight managers
Monitor large projects in preconstruction stage
Identify preconstruction project milestones
Establish major project speakers' bureau
Establish major project briefing repository
Establish training and development opportunities for new staff
Organize annual project oversight managers meeting
Participate in National Engineer's Week
Serve as structural liaison
Conduct security vulnerability assessments
Establish national best practice and lessons learned program
Post best practices and lessons learned material on website
Produce best practices and lessons learned bulletins

Status

Under Development
No Work Started
No Work Started
No Work Started
No Work Started
No Work Started
Ongoing
Ongoing
Ongoing
Ongoing
Under Development
Under Development
No Work Started

                                  Source: GAO.

Appendix III

                     GAO Contacts and Staff Acknowledgments

GAO Contacts	Kate Siggerud (202) 512-2834 Steve Cohen (202) 512-4864

Staff 	In addition to those named above, Sam Abbas, Catherine Colwell, Pat
Dalton, Don Kittler, Alex Lawrence, Sara Ann Moessbauer, John Rose,

Acknowledgements	Stacey Thompson, and Alwynne Wilbur made key
contributions to this report.

GAO's Mission	The Government Accountability Office, the audit, evaluation
and 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 The fastest and easiest way to obtain copies of GAO
documents at no cost

is through GAO's Web site (www.gao.gov). Each weekday, GAO postsGAO
Reports and newly released reports, testimony, and correspondence on its
Web site. To Testimony have GAO e-mail you a list of newly posted products
every afternoon, go to

www.gao.gov and select "Subscribe to Updates."

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. Government Accountability 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, Contact:
  Waste, and Abuse in Web site: www.gao.gov/fraudnet/fraudnet.htm

E-mail: [email protected] Programs Automated answering system: (800)
424-5454 or (202) 512-7470

Congressional	Gloria Jarmon, Managing Director, [email protected] (202)
512-4400 U.S. Government Accountability Office, 441 G Street NW, Room 7125

Relations Washington, D.C. 20548

Public Affairs	Paul Anderson, Managing Director, [email protected] (202)
512-4800 U.S. Government Accountability Office, 441 G Street NW, Room 7149
Washington, D.C. 20548
*** End of document. ***