GAO-11-283, Intercity Passenger Rail: Recording Clearer Reasons for Awards Decisions Would Improve Otherwise Good Grantmaking Practices


This is the accessible text file for GAO report number GAO-11-283 
entitled 'Intercity Passenger Rail: Recording Clearer Reasons for 
Awards Decisions Would Improve Otherwise Good Grantmaking Practices' 
which was released on April 11, 2011. 

This text file was formatted by the U.S. Government Accountability 
Office (GAO) to be accessible to users with visual impairments, as 
part of a longer term project to improve GAO products' accessibility. 
Every attempt has been made to maintain the structural and data 
integrity of the original printed product. Accessibility features, 
such as text descriptions of tables, consecutively numbered footnotes 
placed at the end of the file, and the text of agency comment letters, 
are provided but may not exactly duplicate the presentation or format 
of the printed version. The portable document format (PDF) file is an 
exact electronic replica of the printed version. We welcome your 
feedback. Please E-mail your comments regarding the contents or 
accessibility features of this document to Webmaster@gao.gov. 

This is a work of the U.S. government and is not subject to copyright 
protection in the United States. It may be reproduced and distributed 
in its entirety without further permission from GAO. Because this work 
may contain copyrighted images or other material, permission from the 
copyright holder may be necessary if you wish to reproduce this 
material separately. 

United States Government Accountability Office: 
GAO: 

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

Intercity Passenger Rail: 

Recording Clearer Reasons for Awards Decisions Would Improve Otherwise
Good Grantmaking Practices: 

GAO-11-283: 

GAO Highlights: 

Highlights of GA0-11-283, a report to the Chairman, Committee on 
Transportation and Infrastructure, House of Representatives. 

Why GAO Did This Study: 

The American Recovery and Reinvestment Act of 2009 (Recovery Act) 
appropriated $8 billion for high and conventional speed passenger 
rail. The Federal Railroad Administration (FRA), within the Department 
of Transportation (the department), was responsible for soliciting 
applications, evaluating them to determine program eligibility and 
technical merits, and selecting awards, which were announced in 
January 2010. 

This report examines the extent to which FRA (1) applied its
established criteria to select projects, (2) followed recommended 
practices for awarding discretionary grants, and (3) communicated 
outcomes to the public, compared with selected other Recovery Act 
competitive grant programs. To address these topics GAO reviewed 
federal legislation, FRA documents, and guidance for other competitive 
grant programs using Recovery Act funds. GAO also analyzed data 
resulting from the evaluation and selection process and interviewed a 
cross-section of FRA officials and applicants. 

What GAO Found: 

FRA applied its established criteria during the eligibility and 
technical reviews, but GAO could not verify whether it applied its 
final selection criteria because the documented rationales for 
selecting projects were typically vague. Specifically, FRA used 
worksheets and guidebooks that included the criteria outlined in the 
funding announcement to aid in assessing the eligibility and technical 
merit of applications. FRA also recorded general reasons for selecting 
applications and publicly posted broad descriptions of the selected 
projects. However, the documented reasons for these selection 
decisions were typically vague or restated the criteria listed in the 
funding announcement. In addition, there were only general reasons 
given for the applications not selected or for adjusting applicants' 
requested funding amounts. FRA subsequently provided GAO with more 
detailed reasons for several of its selection decisions, but this 
information was not included in the department's record of its 
decisions. Documentation on the rationales for selection decisions is 
a key part of ensuring accountability and is recommended by the 
department as well as other federal agencies. Without a detailed 
record of selection decisions, FRA leaves itself vulnerable to 
criticism over the integrity of those decisions—an important 
consideration, given that passenger rail investments have a very 
public profile. 

FRA also substantially followed recommended practices when awarding 
grants, including communicating key information to applicants prior to 
the competition, planning for the competition, using a merit review 
panel with certain characteristics, assessing whether applicants were 
likely to be able to account for grant funds, notifying applicants of 
awards decisions, and documenting the rationale for awards decisions 
(albeit generally). For example, FRA issued a funding announcement 
that communicated key pieces of information, such as eligibility, 
technical review, and selection criteria. FRA officials also conducted 
extensive outreach to potential applicants, including participating in 
biweekly conference calls, providing several public presentations on 
the program, and conducting one-on-one site visits with potential 
applicants. According to FRA, officials used lessons from a number of 
other grant programs when developing its approach to reviewing and 
selecting projects. 

FRA publicly communicated outcome information similar to other 
Recovery Act competitive grant programs we examined, including 
projects selected, how much money they were to receive, and a general 
description of projects and their intended benefits. Only one of the 
programs GAO examined communicated more outcome information on 
technical scores and comments; however, this program used a much 
different approach to select awards than FRA used to select intercity 
passenger rail awards. According to officials, FRA did not disclose 
outcome information from the technical reviews because officials were 
concerned that releasing reviewers' names and associated scores could 
discourage them from participating in future grant application reviews. 

What GAO Recommends: 

GAO recommends that FRA create additional records to document the 
substantive reasons behind award decisions to better ensure
accountability for its use of federal funds. In commenting on a draft 
of this report, the department agreed to consider our recommendation. 
The department also provided technical comments, which were 
incorporated as appropriate. 

View [hyperlink, http://www.gao.gov/products/GA0-11-283] or key 
components. For more information, contact Susan Fleming at (202) 512-
2834 or flemings@gao.gov. 

[End of section] 

Contents: 

Letter: 

Background: 

FRA Applied Its Established Criteria to Determine Eligibility and 
Assess Technical Merit, but Selection Rationales Were Typically Too 
Vague to Assess: 

FRA Substantially Met Recommended Practices for Awarding Discretionary 
Grants: 

FRA Publicly Communicated at Least as Much Outcome Information as 
Other Competitively Awarded Recovery Act Grant Programs: 

Conclusions: 

Recommendation for Executive Action: 

Agency Comments: 

Appendix I: Extent to Which Recovery Act Projects Align with
Statutory and Other Goals: 

Appendix II: Scope and Methodology: 

Appendix III: Difference Between the Amounts Requested and
Estimated Awards by State: 

Appendix IV: Additional Results from Our Statistical Analysis of
Award Decisions: 

Tables: 

Table 1: Technical Review Criteria: 

Table 2: Selection Criteria: 

Table 3: Recommended Practices FRA Followed: 

Table 4: Recovery Act Applications Supporting High Speed Rail 
Categories by Future Corridor Speed 5 Years After Project Completion: 

Table 5: Amounts Awarded, Obligated, and Spent on First Round Track 1 
and 2 Recovery Act Projects, as of December 31, 2010: 

Table 6: FRA Plan for Obligating and Spending Recovery Act Funds and 
Amounts Obligated and Spent, as of December 31, 2010: 

Table 7: Guidance and Reports Used To Identify Recommended Government 
Practices: 

Table 8: Recovery Act Discretionary Grant Programs Reviewed: 

Table 9: Difference between the Amounts Requested and Estimated Awards 
by State, as of January 2010: 

Table 10: Difference between the Amounts Requested and Estimated 
Awards by State, as of December 2010: 

Table 11: Applications Selected by Technical Review Score, and Odds 
and Odds Ratios Derived from Them: 

Table 12: Applications Selected by Track, and Odds and Odds Ratios 
Derived from Them: 

Table 13: Applications Selected by Requested Amount, and Odds and Odds 
Ratios Derived from Them: 

Table 14: Applications Selected by State and State Group, and Odds and 
Odds Ratios Derived from Them: 

Table 15: Odds Ratios from Bivariate and Multivariate Logistic 
Regression Models by Technical Review Score, Track, Amount Requested, 
and State and State Group: 

Figures: 

Figure 1: Approach that FRA Used to Assess Applications and Select 
Recovery Act Recipients: 

Figure 2: Differences between Proposed and Requested Award Amounts, in 
Percents: 

Figure 3: Number of Selected Applications by Technical Review Score: 

Figure 4: HSIPR Reported Outcomes Compared to Other Competitively 
Awarded Recovery Act Programs: 

Abbreviations; 

department: Department of Transportation: 

FRA: Federal Railroad Administration: 

HSIPR: high speed intercity passenger rail: 

PRIIA: Passenger Rail Investment and Improvement Act of 2008: 

Recovery Act: American Recovery and Reinvestment Act of 2009: 

[End of section] 

United States Government Accountability Office: 
Washington, DC 20548: 

March 10, 2011: 

The Honorable John Mica: 
Chairman: 
Committee on Transportation and Infrastructure: 
House of Representatives: 

Dear Mr. Chairman: 

A recent influx of federal funds has breathed new life into the 
prospect of developing an expanded national passenger rail network in 
the United States. Specifically, the American Recovery and 
Reinvestment Act of 2009 (Recovery Act) appropriated $8 billion—-
significantly more than Congress provided in recent years-—to develop 
high speed and intercity passenger rail service.[Footnote 1] Interest 
in these funds was high, and in January 2010 the Federal Railroad 
Administration (FRA)-—an agency within the Department of 
Transportation (the department)-—selected 62 applications in 23 states 
and the District of Columbia to receive the money.[Footnote 2] The 
vast majority (almost 90 percent) of the $8 billion awarded went to 
develop new or substantially improved passenger rail corridor 
projects, which, in several cases, expect to deliver high speed rail 
service reaching speeds of more than 150 miles per hour. The remaining 
funding generally went to projects focusing on upgrades and 
improvements to existing rail service (typically up to 79 miles per 
hour). 

With the Recovery Act funding, FRA recognized that it needed to 
transform itself from essentially a rail safety organization to one 
that can make and oversee multibillion dollar investment choices. 
[Footnote 3] This report assesses how FRA made the first of those 
choices and ensured that national investment goals are being met. 
[Footnote 4] It focuses on the extent to which FRA (1) applied its 
established criteria to select projects; (2) followed recommended 
practices for awarding discretionary grants; and (3) communicated 
outcomes to the public, compared with selected other Recovery Act 
competitive grant programs. These topics are the main focus of the 
report. In addition, we are also reporting on the extent to which 
selected projects align with legislative and federal goals. (See 
appendix I.) 

Our overall approach to addressing these topics was to (1) review 
publicly available information, such as federal legislation, plans, 
and other guidance, about the high speed intercity passenger rail 
program's evaluation, selection and communication approach, and 
compare it to practices used by other competitive grant programs; (2) 
review documents that FRA used in reviewing applications and selecting 
awardees to determine the extent to which FRA applied its established 
criteria; (3) analyze FRA data on technical review scores to determine 
the statistical relationship between some of FRA's published criteria 
and the selection decisions; and (4) interview a cross-section of 
officials from 12 of the 40 states and the District of Columbia which 
submitted either a preapplication or an application for Recovery Act 
funding (selected to reflect a range of application outcomes, award 
amounts, number of applications, and geographic location), a random 
sample of 18 of the 44 department reviewers which included at least 
one person from each applicant review panel, and other FRA officials 
who oversaw the evaluation and selected awards.[Footnote 5] We focused 
our review on projects selected by FRA in January 2010 and funded 
through the Recovery Act, which included applications submitted for 
ready-to-go projects (called "track 1a"), the completion of 
environmental and preliminary engineering requirements necessary to 
prepare projects for future funding (called "track 1b"), and projects 
to develop new high speed rail or intercity passenger services or 
substantially upgrade existing corridor service (called "track 2"). 
[Footnote 6] We assessed the reliability of FRA's scoring data by 
conducting a series of data tests, reviewing documents and reports 
about FRA's data systems, and speaking with officials familiar with 
the data. We determined that these data are sufficiently reliable for 
our reporting purposes. (Additional information on our scope and 
methodology is contained in appendix II.) 

We conducted this performance audit from April 2010 to March 2011 in 
accordance with generally accepted government auditing standards. 
Those standards require that we plan and perform the audit to obtain 
sufficient, appropriate evidence to provide a reasonable basis for our 
findings and conclusions based on our audit objectives. We believe 
that the evidence obtained provides a reasonable basis for our 
findings and conclusions based on our audit objectives. 

Background: 

FRA is the primary federal agency responsible for overseeing safety in 
the railroad industry, as well as for distributing federal funds for 
intercity passenger rail service. FRA also administers federal 
operating and capital grants to the National Railroad Passenger 
Corporation (known as Amtrak), which have averaged between $1 billion 
and $1.3 billion per year since fiscal year 2003. FRA also approves 
Railroad Rehabilitation and Improvement Financing loans and Rail Line 
Relocation and Improvement Capital grants, and is the granting agency 
for the $120 million in fiscal year 2008 and fiscal year 2009 capital 
funds for intercity passenger rail projects. 

Recent legislation has vastly increased the federal role in and 
federal funds for developing intercity passenger rail service. The 
Passenger Rail Investment and Improvement Act of 2008 (PRIEA), enacted 
in October 2008, authorized more than $3.7 billion for three different 
federal programs for high speed rail, intercity passenger rail, and 
congestion reduction grants.[Footnote 7] PRIEA also called for FRA to 
create a preliminary national rail plan within 1 year after passage of 
the act as well as a long-range national rail plan that promotes an 
integrated and efficient national rail system. FRA released a 
preliminary national rail plan in October 2009 and a subsequent 
progress report in September 2010.[Footnote 8] 

The Recovery Act, enacted in February 2009, appropriated $8 billion 
for the three PRIIA-established intercity passenger rail programs. 
[Footnote 9] Unlike PRIIA, which authorized an 80 percent federal 
share for a project's capital costs, the Recovery Act provided up to 
100 percent federal funding available for obligation through fiscal 
year 2012 and expenditure through fiscal year 2017.[Footnote 10] The 
Recovery Act required that the department develop a strategic plan to 
use these funds. In April 2009, FRA released its strategic plan for 
developing high speed rail in America and distributing federal funds. 
[Footnote 11] 

PRIIA and the Recovery Act created new responsibilities for FRA to 
plan, award, and oversee the use of new federal funds for intercity 
passenger rail. In response, FRA launched the high speed intercity 
passenger rail (HSIPR) program in June 2009 by issuing a funding 
announcement and interim guidance, which outlined the requirements and 
procedures for obtaining federal funds.[Footnote 12] FRA further 
outlined the vision and goals of the program through a number of 
outreach events and meetings, including seven regional workshops and 
more than 25 one-on-one site visits and conference calls with 
potential state applicants.[Footnote 13] States expressed a great deal 
of enthusiasm for the new program, requesting $102 billion across 278 
preapplications, which FRA used to gauge initial interest and 
anticipate its staffing needs to manage the program. States, including 
the District of Columbia, ultimately submitted 229 applications for 
$57.8 billion in Recovery Act funds.[Footnote 14] 

FRA asked applicants to submit Recovery Act project applications under 
three tracks: 1a, 1b, and 2. Track 1 was intended to primarily address 
economic recovery goals, and could either focus on ready-to-go 
projects (track la) or the completion of environmental and preliminary 
engineering requirements necessary to prepare projects for future 
funding (track 1b). Track 2 focused on much larger, long-term projects 
to develop new high speed rail services or substantially upgrade 
existing corridor service. While track 1 and track 2 applications were 
submitted and reviewed at different times, FRA used a similar approach 
to assess them, and applied the same criteria during three independent 
steps: eligibility determination, technical review, and selection. 
[Footnote 15] (See figure 1.) 

Figure 1: Approach that FRA Used to Assess Applications and Select 
Recovery Act Recipients: 

[Refer to PDF for image: time line] 

June 2009: 
FRA launches the HSIPR program. 
FRA issues interim guidance and a funding announcement with the 
requirements and procedures for obtaining federal funds. 

July 2009: 
Preapplications deadline. 
FRA received 278 preapplications expressing initial interest from 40 
states and the District of Columbia totaling $102 billion. 

August 2009: 
Track 1: 
Applications due to FRA. 
FRA received 184 applications requesting $6.9 billion. 

August-September 2009: 
Track 1: 
Eligibility determination. 
FRA determined 156 applications requesting $3.5 billion were eligible. 

September 2009: 
Track 1: 
Technical review. 
Twelve panels, each comprised of three individuals, assessed eligible 
applications on six pre-established technical review criteria. 

Track 1: 
Preliminary selection. 
Senior department and FRA officials considered the results of the 
technical review as well as four additional pre-established selection 
criteria. 

October 2009: 
Track 2: 
Applications due to FRA. 
FRA received 45 applications requesting $50.9 billion. 

October-November 2009: 
Track 2: 
Eligibility determination. 
FRA determined 23 applications requesting $20.6 billion were eligible. 

November-December 2009: 
Track 2: 
Technical review. 
One panel comprised of eight individuals assessed eligible 
applications on six pre-established technical review criteria. 

December 2009: 
Track 2: 
Preliminary selection. 
Senior department and FRA officials considered the results of the 
technical review as well as four additional pre-established selection 
criteria. 

January 2010: 
Track 1 and 2: 
Final decisions. 
Senior department and FRA officials examined projects from all tracks 
and recommended the selection of specific projects. The Secretary of 
Transportation concurred with these recommendations. 

Track 1 and 2: 
Award announcement. 
FRA announced the selection of 62 applications from 23 states and the 
District of Columbia for the $8 billion in Recovery Act funds. 
* Track 1: 48 applications were selected for $0.9 billion. 
* Track 2: 14 applications were selected for $7.0 billion. 

Source: GAO presentation of FRA data. 

[End of figure] 

The eligibility determination was conducted by panels of officials 
experienced in environmental requirements or passenger and commuter 
rail. These officials used worksheets to aid in assessing application 
completeness and determining whether applicants and proposed projects 
were eligible to receive funds. Eligibility panels also made 
preliminary determinations as to whether applicants had substantially 
completed environmental requirements and whether the projects they 
submitted were ready to begin. For the most part, applications deemed 
not yet ready or ineligible were not forwarded for technical review; 
because the track 1 eligibility and technical review periods 
overlapped, there were two Recovery Act applications that received 
technical review scores and were later deemed not yet ready or 
ineligible and removed from award consideration. 

The technical review was conducted by panels of officials with 
experience in several fields, such as passenger and commuter rail, 
grants management, and environmental requirements. The technical 
review differed slightly for track 1 applications, which were reviewed 
by 12 panels composed of three reviewers, and track 2 applications, 
which were reviewed by a single panel of eight reviewers. For both 
tracks, reviewers used guidebooks to assess applications against six 
technical review criteria: (1) transportation benefits, (2) economic 
recovery benefits, (3) other public benefits (e.g., environmental 
quality and energy efficiency), (4) project management approach, (5) 
sustainability of benefits, and (6) timeliness of project completion. 
(See table 1.) The guidebooks provided detailed descriptions of what 
was included within each of these criterion, as well as step-by-step 
instructions on reviewing applications that included a suggested 
scoring method using a scale from one (lowest) to five (highest). For 
example, the track 2 guidebook suggested applications that included 
more than one major weakness, were nonresponsive, or failed to address 
a particular criterion be given a technical review score of a one for 
that criterion. Applications that technical panelists determined were 
responsive, and included major and minor strengths and no major or 
very few minor weaknesses in a particular criterion, were to be given 
a technical review score of a five for that criterion. After 
completing an individual evaluation of each application, reviewers 
convened within their panel to discuss their overall thoughts on the 
application and technical review scores for each criterion, which they 
could revise based on input from other panelists. To arrive at a final 
score for each application, FRA officials used a formula that averaged 
individual scores and weighted the scores based on established 
priorities identified in the funding announcement.[Footnote 16] In 
addition, program officials standardized track 1 application scores to 
correct for potential inconsistencies across review panels.[Footnote 
17] 

Table 1: Technical Review Criteria: 

Criteria: Transportation benefits; 
Types of factors to be considered by reviewers when assessing the 
application: 
* Supports the development of high speed passenger rail and generates 
improvements to intercity passenger rail.
* Reduces congestion across other modes of transportation.
* Encourages integration across other modes of transportation, such as 
connections at airports, bus terminals, and subway stations.
* Promotes equipment standardization, signaling, communication, and 
power.
* Provides for cost sharing across benefiting rail users, including 
freight and commuter railroads, host railroads, and state and local 
government.
* Improves the overall safety of the transportation system. 

Criteria: Economic recovery benefits; 
Types of factors to be considered by reviewers when assessing the 
application: 
* Promotes business opportunities, including the short- and long-term 
creation and preservation of jobs.
* Increases efficiency by promoting technological advances.
* Avoids reduction in the essential services provided by states. 

Criteria: Other public benefits; 
Types of factors to be considered by reviewers when assessing the 
application: 
* Contributes to environmental quality and energy efficiency and 
reduces dependence on foreign oil.
* Promotes livable communities, including integration with high-
density, livable developments (e.g., central business districts with 
access to public transportation, pedestrian, and bicycle networks). 

Criteria: Project management approach; 
Types of factors to be considered by reviewers when assessing the 
application: 
* Assesses that the applicant has the financial, legal, and technical 
capacity to implement the project.
* Considers the applicant's experience in administering similar grants.
* Examines the soundness of cost methodologies, assumptions, and 
estimates.
* Examines whether the application is complete and includes 
comprehensive supporting documentation, such as a schedule for project 
implementation, a project management plan, agreements with key 
partners, an explanation of progress towards completing environmental 
requirements, and any completed engineering work. 

Criteria: Sustainability of benefits; 
Types of factors to be considered by reviewers when assessing the 
application: 
* Considers the quality of the financial and operating service plans.
* Examines the reasonableness of revenue and operating maintenance 
cost forecasts, and estimates for user and nonuser benefits.
* Assesses that funds are available to support operating costs.
* Considers agreements with key partners, such as the proposed 
operator and railroads that own pieces of infrastructure necessary to 
achieve benefits. 

Criteria: Timeliness of project completion; 
Types of factors to be considered by reviewers when assessing the 
application: 
* Assesses whether the project is ready-to-go, will be completed on 
time, and will deliver the proposed benefits. 

Source: GAO summary of FRA's funding announcement. 

[End of table] 

After the technical review, senior department and FRA officials—Deputy 
Secretary, Under Secretary for Policy, FRA Administrator, and FRA 
Deputy Administrator, among others—selected projects to recommend to 
the Secretary of Transportation. They considered the technical review 
scores along with four additional pre-established selection criteria 
identified in the funding announcement: (1) region and location, (2) 
innovation, (3) partnerships, and (4) track type and funding round. 
(See table 2.) HSIPR program officials gave five briefings to senior 
officials on the results of the technical review and possible factors 
to consider in making award decisions, such as potential project cost, 
service speed, shared benefits, and readiness. Program officials also 
provided additional information, including funding scenarios, facts 
sheets on individual applications, and corridor maps upon request. 
According to FRA, senior officials considered this information when 
making their recommendations, but did not numerically score or rank 
applications. 

Table 2: Selection Criteria: 

Criteria: Region and location; 
Factors to be considered by senior department and FRA officials when 
assessing the application: 
* Ensures projects are distributed across the country, in both small 
and large population centers.
* Ensures integration and augmentation of projects across the 
nationwide transportation network.
* Provides assistance to economically distressed regions. 

Criteria: Innovation; 
Factors to be considered by senior department and FRA officials when 
assessing the application: 
* Pursues new technology where the public return on investment is 
favorable.
* Promotes domestic manufacturing, supply, and industrial development.
* Develops passenger rail engineering, operating, planning, and 
management capacity. 

Criteria: Partnerships; 
Factors to be considered by senior department and FRA officials when 
assessing the application: 
* Emphasizes organized partnerships with joint planning and 
prioritization of investments when projects span multiple states.
* Encourages creative approaches to workforce diversity and use of 
disadvantaged and minority businesses. 

Criteria: Track type and funding round; 
Factors to be considered by senior department and FRA officials when 
assessing the application: 
* Preserves funds for track 2 projects, as well as future funding 
rounds, if possible. 

Source: GAO summary of FRA's funding announcement. 

[End of table] 

On January 27, 2010, the FRA Administrator recommended 62 applications 
for funding and the Secretary of Transportation concurred with these 
recommendations. On January 28, 2010, DOT announced the selections. 
[Footnote 18] The selections were spread across several types of 
intercity passenger rail, including projects for emerging high speed 
rail (operating at speeds up to 90 miles per hour), regional corridors 
(operating at speeds between 90 and 124 miles per hour), and core 
express corridors (operating at speeds between 125 and 250 miles per 
hour or more). For example, the department selected one project to 
receive $35 million to rehabilitate track and provide service from 
Portland to Brunswick, Maine at speeds up to 70 miles per hour. 
Another project was to receive more than $50 million to construct 11 
miles of dedicated passenger rail track near Rochester, New York, 
which will allow for service speeds up to 110 miles per hour. A third 
project was selected to receive almost $2.3 billion to initiate the 
first part of the California's high speed rail system, which will 
allow for more than 200 miles per hour service between Los Angeles, 
San Francisco and the Central Valley, and eventually, San Diego. These 
selections were consistent with the criteria in PRIIA, the Recovery 
Act, and FRA's strategic plan, which included broad goals that gave 
FRA discretion in developing a national passenger rail system. 
(Additional information on the legislative and program goals, and how 
the selected projects fit into them, is contained in appendix I.) 

FRA Applied Its Established Criteria to Determine Eligibility and 
Assess Technical Merit, but Selection Rationales Were Typically Too 
Vague to Assess:	 

FRA applied its established criteria to determine eligibility and 
assess applications' technical merit. However, its rationales for 
selecting projects were typically too general to determine how it 
applied the additional selection criteria. When asked for more 
information on certain applications, FRA provided specific reasons for 
its selection decisions, but, in our opinion, creating a detailed, 
comprehensive record alongside the final selections is preferable. 
Officials reported that they used the technical review scores as a 
starting point from which to apply each of the four selection 
criteria, which is partially supported by our analysis of FRA data. 
For example, we found that applications receiving a higher technical 
review score were about seven to eight times more likely to be 
selected for an award compared to those receiving a lower technical 
review score. 

FRA Applied Its Established Criteria to Determine Eligibility and 
Assess Technical Merit: 

We found that FRA applied eligibility criteria established in its 
funding announcement when determining whether applications were 
eligible. Specifically, eligibility criteria listed in the funding 
announcement aligned with criteria outlined in the worksheets used by 
the panelists to verify that applications were eligible. Panelists 
were given separate worksheets to conduct the track 1 and track 2 
eligibility reviews, and each of these worksheets included eligibility 
criteria listed in the funding announcement. For instance, as outlined 
in the funding announcement, the track 1 worksheet required 
eligibility panelists to indicate if the application was submitted on-
time, by an eligible applicant, and with all of the required 
supporting documents. Similarly, the track 2 worksheet included 
questions regarding applicant eligibility, qualifications, and 
construction grant prerequisites which aligned with the eligibility 
criteria listed in the funding announcement. 

FRA also applied the established technical review criteria 
communicated in its funding announcement by including these criteria 
in the guidebooks provided to technical panelists to assess the 
technical merits of each application. Specifically, the guidebooks FRA 
provided to panelists for reviewing track 1a, 1b, and 2 applications 
were divided into six sections that aligned with each of the six 
technical review criteria listed in the funding announcement. 
Moreover, the criteria within these sections of the guidebook often 
matched the criteria in the funding announcement very closely and, in 
some cases, word-for-word. For example, the funding announcement 
stated that an applicant's experience administering similar projects 
would be considered under the technical review criteria of project 
management approach, which was included word-for-word in the project 
management approach section of the guidebooks.[Footnote 19] We spoke 
to at least one representative from each technical review panel; these 
representatives confirmed that panelists used the criteria listed in 
the guidebooks and did not use other criteria during their evaluation 
of individual applications. 

Officials Reported Using Technical Review Panel Results and Selection 
Criteria to Make Awards Recommendations; Decision Rationales Provided 
Little Insight into Selections: 

Senior department and FRA officials recommended to the Secretary of 
Transportation applications to receive awards and the proposed amounts 
of the awards. When deciding which applications to recommend for 
awards, senior FRA officials told us that they used the results of the 
technical review panels and the four selection criteria. These four 
criteria were described in FRA's June 2009 funding announcement as: 
(1) region and location (e.g., ensuring geographic balance, 
integration into the nationwide transportation network, and assistance 
to economically distressed regions), (2) innovation (e.g., pursuing 
new technology with a favorable public return, promoting domestic 
manufacturing, and developing human capital capacity for sustainable 
rail development), (3) partnerships (e.g., multi-state planning and 
investment and workforce diversity), and (4) tracks and round timing 
(e.g., longer-term track 2 corridor development balanced with ready-to-
go track 1 investments). For example, officials stated that they used 
the innovation criterion to select applications with higher proposed 
speeds of service. In particular, senior officials reported using this 
criterion to reinforce the selection of the California and Florida 
intercity passenger rail projects, which were the only eligible 
projects with the potential for service above 150 miles per hour. In 
another example, officials reported that they applied the partnership 
criterion by assessing applicants' track record with implementing 
large transportation projects as well as demonstrated relationships 
with key stakeholders, such as private railroads. 

Senior FRA officials stated they developed their funding amount 
recommendations based on their professional judgment and national high 
speed intercity passenger rail program goals. Officials told us they 
accounted for the risks related to the total cost of the project 
during selection discussions and weighed them against the overall 
policy goals of developing a national high speed passenger rail 
network. Officials also stated that they used their professional 
judgment about rail systems to recommend the award amounts for each 
application, paying particular attention to the amounts distributed to 
the large, track 2 projects, and that they are continuing to assess 
the effect of changes to the requested funding amounts during the 
scope of work negotiations with awardees. 

According to FRA, its rationales for selecting applications are 
recorded in a recommendation from the FRA Administrator to the 
Secretary of Transportation and in a memorandum from the Secretary to 
the Administrator concurring on the recommendations and specifying 
potential funding amounts. The rationales stated in these memorandums 
were typically vague, such as "aligns well with FRA's published 
evaluation criteria" and "will result in significant transportation 
benefits [and] preserve and create jobs." These rationales most often 
restated the criteria listed in the funding announcement generally 
(e.g., result in significant transportation benefits) rather than 
providing insight into why the department viewed projects as 
meritorious. 

In addition, the memorandums did not provide any information on why 
other applications were not recommended for selection, which prevents 
us from assessing how the department viewed the merits of successful 
applications over unsuccessful ones. For example, we found several 
instances in which, without documentation, it was difficult to 
determine the reasons why some projects were selected and others were 
not. Specifically, FRA decided not to select six track 1a applications 
from New York that received higher technical review panel scores, and 
selected a lower scoring track 1a application from the same applicant. 
FRA officials subsequently told us that the lower scoring application 
was selected for a number of reasons, including improving the 
reliability of the passenger trains on the rail line, ensuring that 
the project will become part of the infrastructure of any significant 
improvements to passenger rail service west of Albany, and improving 
the fluidity of both passenger and freight rail operations on this 
heavily used rail route. Similarly, FRA selected a lower scoring track 
1a application from Illinois, but not a relatively high scoring one. 
FRA officials subsequently told us that they selected this application 
because it is an essential part of a long-standing program of projects 
to improve the fluidity of rail traffic in the highly congested 
Chicago area. FRA officials also told us that the scope of the 
relatively high scoring track 1a application was included in Illinois' 
selected track 2 application. This level of information, which 
provides some insight into the merits of projects, was not included in 
the department's record of its decisions. 

In addition to the memorandums, FRA posted descriptions on its Web 
site of the selected projects, their expected benefits, and 
prospective award amounts. However, these descriptions are not 
particularly useful in understanding why these projects were selected 
because the cited benefits—-such as reducing travel times, increasing 
travel speed and ridership, providing attractive transportation 
alternatives, and creating jobs—-were supposed to be integral to all 
projects. For example, FRA's Web site describes one project as 
increasing on-time performance and ultimately allowing speeds of up to 
110 miles per hour on its segment, but does not give any indication 
why this project was meritorious. Other descriptions were similar. 

FRA also sent letters to individual applicants regarding its decision, 
and, if the application was not selected, a brief explanation as to 
why it was not selected. For example, a number of these letters 
explained that applications were not selected because they did not 
meet a prerequisite, had application materials that did not provide 
sufficient support for the proposed activities, or did not submit all 
application materials necessary to adequately evaluate the project. 
However, these letters did not provide further details on how the 
proposed projects did not meet the prerequisite, how the application 
materials were insufficient, or which application materials were not 
received. Other decision letters provided applicants with similarly 
broad explanations. FRA officials also told us that they called all 
applicants, as well as their state secretaries of transportation and 
state governors, to inform them of FRA's decisions. Several of the 
states that we contacted reported that the primary purpose of these 
calls was for FRA to provide feedback on their individual projects 
and, when requested, give explanations for why projects were not 
selected. While applicants stated that this information will be 
helpful during future application rounds, there is no required written 
record of these conversations and, therefore, they do not provide 
others with insight on why selection decisions were made. 

Documentation of agency activities is a key part of accountability for 
decisions.[Footnote 20] The department has a financial assistance 
guidance manual to assist agencies with administering awards 
competitions and which FRA officials told us that they used to develop 
the competition framework.[Footnote 21] The manual recommends that all 
discretionary project selections, such as the intercity passenger 
awards, include an explanation of how the projects were selected based 
on the established funding priorities, but does not lay out 
expectations for the level of explanation. In particular, the manual 
recommends that officials document decisions if projects with the 
highest priority are not funded. 

While the department documented its decisions, as required by its 
financial assistance manual guidance, the absence of an insightful 
internal record of the reasons behind award recommendations, and the 
final selections where they differ, can give rise to challenges to the 
integrity of the decisions made. While FRA was able to provide us with 
specific reasons on a case-by-case basis for why projects were 
selected, almost a year after these decisions were made, we believe 
creating a sufficiently detailed record has increased relevance in 
high-stakes, high-profile decisions, such as the intercity passenger 
rail awards competition in which there are vocal critics and ardent 
supporters of the program. 

Similar arguments apply for creating an internal record for amounts 
recommended for awards. FRA officials understood that the available 
Recovery Act funds were not sufficient to fully fund a number of the 
projects and sought to fund projects or portions of projects that 
could provide transportation benefits if no additional federal funds 
were available. For these decisions, FRA proposed awarding 10 states 
(including the District of Columbia) all (100 percent) of the funds 
they applied for, 8 states nearly all (91-99 percent) of the funds 
they applied for, 5 states some (47-86 percent) of the funds they 
applied for, and one state with slightly more (104 percent) than it 
applied for.[Footnote 22] (See figure 2. See also app. DI for dollar 
amounts associated with figure 2.) The applicant notification letters 
did not offer an explanation for why FRA proposed award amounts that 
differed from requests, and applicants we spoke with did not report 
that FRA had provided such information to them. Given that 
infrastructure projects have an inclination for cost growth, 
developing a record that explains why the recommended costs are 
appropriate for the proposed project provides integrity to the final 
decisions. 

Figure 2: Differences between Proposed and Requested Award Amounts, in 
Percents: 

Percent of requested funds proposed for award: 100 or more; 
Number of states: 9. 

Percent of requested funds proposed for award: 91-99; 
Number of states: 10. 

Percent of requested funds proposed for award: 47-86; 
Number of states: 5. 

Source: GAO analysis of FRA data. 

[End of figure] 

The current economic climate has also increased the importance of 
providing an internal rationale for large differences between 
requested funds and proposed award amounts. Many states have faced 
large budget deficits in 2010 that will require them to make difficult 
budget decisions about the future use of state funds, particularly 
where the Recovery Act awards will not provide all the funding 
expected to be needed to complete a project. For example, as of June 
2010, Florida had made $3 billion in budget cuts to close its budget 
deficit. For its high speed rail award, Florida is slated to receive 
less than half of what it said is needed to complete the proposed 
Tampa to Orlando High Speed Rail Express project. An official from the 
Florida Department of Transportation is hopeful that Florida will 
receive additional federal grants, but is unsure where the remaining 
funds will come from otherwise.[Footnote 23] Additionally, Washington 
state applied for 16 separate projects totaling $976 million and was 
selected to receive a composite award of $590 million. Washington 
state officials acknowledged that the award amount will not fund all 
16 of the projects, and have since reduced the scope of the 
application to the 11 projects that could be completed with the 
awarded amount while still providing the maximum benefit to the 
corridor.[Footnote 24] FRA officials stated they awarded amounts that 
differed from those requested in applications as a result of their 
recognition that many of the projects were based on preliminary work 
that was not well-refined, and that states differed in their ability 
to accurately estimate costs.[Footnote 25] In contrast, North Carolina 
received 4 percent more funding than originally requested. According 
to FRA, the additional funding was allocated to North Carolina for 
possibly adding additional train frequencies for a Recovery Act 
project. While we recognize that FRA may have developed these proposed 
award amounts for good reasons, without a written record of the 
department's rationale for these adjustments, after the fact 
reconstructions of funding amount decisions invite outside criticism 
of the decisions. 

Applications with Higher Technical Review Scores Were Typically Chosen 
Over Those with Lower Scores: 

One of your interests was in how the results of technical review 
panels aligned with final award decisions. As discussed earlier, while 
FRA considered the technical review panels to be an important part of 
its decision making, they were not the sole basis for selecting 
projects. This was detailed in FRA's funding announcement, which 
described how applications were first to be assessed against six 
technical review criteria and then final recommendations would be made 
using the technical review results and four selection criteria. 

While the technical review panel evaluations alone were not meant to 
designate final selections, we found that of 179 eligible Recovery Act 
applications, senior management recommended 92 percent (57 of 62) of 
higher scoring applications for funding; that is they received review 
panel scores of 3 or higher out of 5 possible points. (See figure 3.) 
Within these recommended applications, most received a technical 
review score of 3 or 4, and three of the five applications that 
received a technical review score of 5 were recommended for selection. 
One of the two applications that scored a 5 and was not selected for 
funding was included in a selected track 2 application. 

Figure 3: Number of Selected Applications by Technical Review Score: 

[Refer to PDF for image: stacked vertical bar graph] 

Score: 1; 
Applications Selected: 0; 
Applications Not Selected: 3. 

Score: 2; 
Applications Selected: 5; 
Applications Not Selected: 31. 

Score: 3; 
Applications Selected: 35; 
Applications Not Selected: 48. 

Score: 4; 
Applications Selected: 19; 
Applications Not Selected: 33. 

Score: 5; 
Applications Selected: 3; 
Applications Not Selected: 2. 

Source: GAO analysis of FRA data. 

[End of figure] 

In a few cases though, senior officials recommended applications that 
received a lower technical review score (i.e., a score of 2) because, 
according to FRA, they believed these projects included freight and 
commuter rail service partners that were willing to make cost 
contributions in line with their potential benefit share, were 
strategically important to other selected applications, or helped to 
achieve regional balance. These considerations were included in the 
four selection criteria senior department and FRA officials said they 
used to evaluate applications. For example, one of the two 
applications from Rhode Island requested $1.2 million to complete 
preliminary engineering and environmental reviews and received a lower 
overall technical review score, in part because technical reviewers 
did not believe the applicant had sufficiently quantified the 
transportation and economic recovery benefits. This application was 
later recommended for selection. According to FRA, senior officials 
recommended applications receiving lower technical review scores, such 
as this Rhode Island application, in part to achieve greater regional 
balance. Additionally, FRA indicated this particular application was 
one of the few applications proposed for the Northeast Corridor, which 
further supported the region/location selection criteria. In another 
instance, senior officials selected a track 2 application from 
California that requested $194 million for preliminary engineering and 
environmental requirements for a large corridor application that 
received a lower technical review panel score. According to FRA, 
senior officials recommended some applications receiving lower 
technical review scores due to the projects' strategic importance to 
other selected applications. Officials stated that they recommended 
the track 2 California application because the completion of 
preliminary engineering and environmental requirements were necessary 
to move forward on several other large California projects also 
recommended for an award. 

Officials also told us that some applications receiving a higher 
technical review score (i.e., 3, 4, or 5) were not selected in order 
to ensure regional balance, especially when an applicant had already 
been selected for other large awards. For example, a track 1a 
application from North Carolina received a higher technical review 
panel score due, in part, to the anticipated transportation benefits 
of increased ridership and on-time-performance, and the applicant's 
estimates that the project would create more than 400 new jobs. Most 
of the projects that North Carolina applied for under this application 
were also included as part of a larger, intercity passenger rail 
application that was later recommended for selection, and the state 
was awarded an estimated total of $545 million for high and 
conventional speed rail projects. Department and senior FRA officials 
reported that higher evaluated applications were not selected if the 
proposed project was already included in larger selected projects, to 
avoid duplicative selections. Another example was FRA's decision not 
to select a higher scoring track 1a application from Florida that 
requested $270 million to acquire 61 miles of right-of-way. This 
application was scored highly due in part to its immediate benefits 
and substantial contribution of state funds but, similar to North 
Carolina, Florida had already been awarded $1.25 billion for a 
separate large, track 2 corridor project. In addition, almost 90 
percent of the applications that scored a 4 and were not selected were 
submitted by applicants that had either already received a large award 
or had submitted a relatively high number of applications. 

To provide further insight into the attributes that were consistent 
with being selected for Recovery Act awards, we examined technical 
review score and application data using a statistical model and found 
that two out of four variables we included in our model, technical 
review scores and the number of applications submitted per state, were 
significantly related to the likelihood of an application being 
selected for an award.[Footnote 26] Applications with higher scores 
(i.e., scores of 3, 4, or 5) were about seven to eight times more 
likely to be selected than those with scores of 1 or 2. For example, 
an application receiving a technical review score of 5, the highest 
possible score, was more than nine times more likely to be selected 
for an award as an application receiving a technical review score of a 
1 or 2. This analysis supports statements from senior department and 
FRA officials indicating that the technical review scores were largely 
the basis for their selection deliberations. 

Additionally, we found that states submitting fewer applications 
(i.e., between one and three) were more than three times more likely 
to have their application selected than states submitting higher 
numbers of applications (i.e., between four and nine). This result 
suggests that selection officials attempted to spread the awards 
across different applicants, which is consistent with FRA's reported 
efforts to attain geographic distribution. However, the results 
differed somewhat for the four states that submitted 10 or more 
applications. In this case two of the states had a lower likelihood of 
being selected for an award than states submitting fewer than 10 
applications, while one state had a higher likelihood of being 
selected. One additional state had about the same likelihood of being 
selected as states submitting between four and nine applications. When 
asked about these differences across states, FRA officials said that 
the number of applications submitted did not affect their selection 
decisions. 

FRA Substantially Met Recommended Practices for	Awarding Discretionary 
Grants: 

We identified six recommended practices used across the federal 
government to ensure a fair and objective evaluation and selection of 
discretionary grant awards. These practices are based on policies and 
guidance used by the Office of Management and Budget and other federal 
agencies—including the department, and our work.[Footnote 27] FRA 
substantially followed these practices, including communicating key 
information to applicants, planning for the competition, using a 
technical merit review panel with desirable characteristics, assessing 
applicants' ability to account for funds, and notifying applicants of 
awards decisions. (See table 3.) In our opinion, FRA partially met one 
recommended practice: documenting the rationale for funding decisions. 
As discussed previously, we believe it would have been beneficial to 
provide more detail about the rationales for these decisions. 
According to FRA officials, the methods they used to evaluate and 
select applications were based on best practices collected from 
several other federal government agencies, which we believe likely 
helped them meet a number of the recommended practices we identified. 

Table 3: Recommended Practices FRA Followed: 

Practice: Communicate with potential applicants prior to the 
competition; 
Attributes of practice: Provide information prior to making award 
decisions on available funding, key dates, competition rules (i.e., 
eligibility, technical review, and selection criteria), funding 
priorities, types of projects to be funded, outreach efforts to new 
applicants and preapplication assistance; 
FRA followed? Yes. 

Practice: Plan for administering the technical review; 
Attributes of practice: Develop a plan for the technical review	that 
describes the number of panels and reviewers and includes methods for 
assigning applications to review panels, identifying reviewers, 
recording the results of the technical review, resolving scoring 
variances across panels, and overseeing the panel to ensure a 
consistent review; 
FRA followed? Yes. 

Practice: Develop a technical review panel with certain 
characteristics; 
Attributes of practice: Use a technical review panel consisting of 
reviewers who hold relevant expertise, do not have conflicts of 
interest, apply the appropriate criteria, and are trained; 
FRA followed? Yes. 

Practice: Assess applicants' capabilities to account for funds; 
Attributes of practice: Assess applicants' abilities to account for 
funds by determining if applicants meet eligibility requirements, 
checking previous
grant history, assessing financial management systems, and analyzing 
project budgets; 
FRA followed? Yes. 

Practice: Notify applicants of awards decisions; 
Attributes of practice: Notify unsuccessful and successful applicants 
of selection decisions in writing and provide feedback on applications; 
FRA followed? Yes. 

Practice: Document rationale for awards decisions; 
Attributes of practice: Document the rationale for awards decisions, 
including the reasons individual projects were selected or not 
selected and how changes made to requested funding amounts may affect 
applicants' ability to achieve project goals; 
FRA followed? Partially. 

Source: GAO analysis of federal agency guidance and the HSIPR 
evaluation and selection approach. 

[End of table] 

* Communicate with potential applicants prior to the competition. FRA 
issued a funding announcement that included information on the $8 
billion in available funding, key dates, the competition rules, the 
funding priorities and relative importance for each one, and the types 
of projects FRA would consider for federal grants. Applicants we spoke 
with praised FRA's communication and stated that FRA officials did a 
good job providing information and answering questions during the 
period leading up to the preapplication and application deadlines. For 
example, officials from several states indicated that FRA officials 
participated in biweekly conference calls, which were helpful in 
understanding the technical aspects of how to apply. Applicants also 
indicated that the outreach events, particularly the site visits, 
helped them refine their applications and ensure projects met program 
requirements. 

* Plan for administering the technical review. FRA developed two plans 
for determining technical merit: (1) the track 1 technical review used 
12 panels each comprised of three reviewers and (2) the track 2 
technical review used one panel of eight reviewers. Track 1 
applications were randomly assigned across the panels, while the track 
2 panel reviewed all of the eligible applications. FRA identified and 
asked for volunteers to participate in the technical reviews from 
within FRA and across several other agencies within the department. 
FRA officials also provided reviewers with guidebooks to document 
their application assessments and instructed them to input the 
results, including scores and comments, into a centralized database. 
FRA standardized final track 1 application scores to account for any 
unintentional differences in the way panels assessed and scored 
applications, but did not need to standardize track 2 scores because 
the review was conducted by a single panel. Finally, according to 
officials, FRA oversaw the review by examining technical review scores 
and comments, and conducting daily meetings with representatives from 
each panel to ensure panelists were consistently applying the criteria. 

* Develop a technical review panel with certain characteristics. FRA 
compiled technical review panels that included staff with background 
in several relevant fields, such as grants management, passenger and 
commuter rail, and environmental requirements, and made other 
knowledgeable staff available if panelists had questions. FRA 
officials stated that panelists were also required to sign or submit a 
previously completed conflict of interest form to attest to their 
independence. In addition, panelists were given guidebooks to assess 
applications that included the technical review criteria and were told 
by FRA program officials to apply only these criteria during their 
efforts. FRA also trained panelists during a 1-day orientation session. 

* Assess applicants' capabilities to account for funds. FRA required 
applicants to provide information on their ability to account for 
funds. Specifically, applicants were asked to describe their 
experience, if any, managing rail investment projects. If applicants 
reported that they did not have experience on projects similar to the 
one they were proposing, FRA instead asked applicants to include a 
plan for building the capacity to manage the project. The application 
also required applicants to provide information on their financial 
management capability, including previous audit results, and the 
applicants' ability to manage potential cost overruns and financial 
shortfalls. In addition, FRA required applicants to submit 
supplemental materials such as a detailed capital cost budget, which 
provided a breakdown of the activities included in each application 
and their anticipated cost. These pieces of information were assessed 
by FRA through an eligibility panel, to ensure the application was 
complete, and a technical review panel, to evaluate the applicants' 
overall ability to manage the project. 

* Notify applicants of awards decisions. FRA officials provided each 
applicant with a letter indicating which applications were selected 
and a general reason why individual applications were not selected. 
While FRA did not include estimated award amounts in these 
notification letters, this information was made publicly available on 
the department's Web site and distributed through a press release. In 
addition, most of the applicants we spoke with indicated that FRA 
provided informal feedback on applications via telephone calls shortly 
after the awards were announced. For example, an official from one 
applicant stated that FRA provided information on ways to improve 
applications that were not selected, which the applicant used when 
applying for funds in future rounds. 

* Document rationale for awards decisions. According to the guidance 
from the department, Department of Commerce, the Department of 
Education, and our work, agencies should document their rationale for 
award decisions. As stated previously, FRA documented how it applied 
the technical criteria for selected projects, and provided applicants 
with a general explanation for selecting or rejecting individual 
projects. However, as discussed in a previous section, in our view FRA 
typically did not clearly document specific reasons for selecting 
individual projects, reasons for not selecting other projects, or how 
changes made to requested funding amounts might affect applicants' 
ability to achieve project goals. 

According to FRA, officials used lessons from a number of other 
government programs when developing the method for evaluating and 
selecting projects. For example, one of the officials responsible for 
developing the funding announcement, technical review guidebooks, and 
the format of the technical review panels stated that he relied on his 
experience working with large transit grants to create a review that 
was both quantifiable and allowed for subjective professional 
judgment. In addition, this official noted that FRA examined the 
methods used by other agencies, such as the Department of Health and 
Human Services, the Department of Justice, and the Federal Transit 
Administration, to develop and implement a list of best practices for 
awarding discretionary grants. 

FRA Publicly Communicated at Least as Much Outcome Information as Other
Competitively Awarded Recovery Act Grant Programs: 

FRA publicly communicated outcome information, such as a list of 
awards and the award amounts, at a level similar to or greater than 
most other Recovery Act competitive grant programs that we examined. 
Specifically, FRA communicated information on award decisions to the 
public, but did not communicate the results of the technical review 
that had contributed to these decisions. Only one of the programs that 
we examined-—the Department of Education's State Innovation grants 
(known as Race to the Tope[Footnote 28])-—publicly communicated the 
results of its technical review, which include technical scores and 
comments; however, this program used a much different approach for 
selecting awardees than the HSIPR program. 

Members of Congress and the President have emphasized the need for 
accountability, efficiency, and transparency in the expenditure of 
Recovery Act funds and have made it a central principle of the act. 
However, the act did not define the attributes of transparency or how 
deeply an agency's actions should be transparent.[Footnote 29] We also 
did not find any non-Recovery Act requirement or guidance instructing 
federal programs to publicly disclose the reasons for their selection 
decisions. To assess the extent to which FRA publicly communicated 
outcome information, we compared the HSIPR program to 21 other 
Recovery Act competitive grant programs, including Race to the Top. 
(See figure 4.) We selected 20 of these programs randomly from a list 
of almost 200 competitively awarded grant programs that distributed 
Recovery Act funds.[Footnote 30] We included the 21st program, Race to 
the Top, because it was of interest to you. 

Figure 4: HSIPR Reported Outcomes Compared to Other Competitively 
Awarded Recovery Act Programs: 

[Refer to PDF for image: illustrated table] 

Outcomes communicated: Technical review (scores and comments); 
Degree communicated: HSIPR: Information was not communicated; 
Degree communicated: Other Recovery Act programs: Information was not 
communicated; 
Degree communicated: Race to the Top: Information was communicated. 

Outcomes communicated: Selection decisions (awards and award amounts); 
Degree communicated: HSIPR: Information was communicated; 
Degree communicated: Other Recovery Act programs: Information was 
Information was communicated for some programs, but not all; 
Degree communicated: Race to the Top: Information was communicated. 

Source: GAO analysis of publicly available data on discretionary 
Recovery Act program grants. 

[End of figure] 

FRA publicly communicated at least as much outcome information as all 
but one Recovery Act competitive grant programs we reviewed. 
Specifically, FRA publicly communicated through its Web site the 
selection decisions, including the amount of funds requested, general 
benefits from the project, and the potential award amounts for the 62 
Recovery Act applications that it selected. It did not communicate the 
results of the technical review. Out of the other 21 competitively 
awarded Recovery Act programs we examined, 13 communicated selection 
information similar to FRA, including awards and award amounts, but 
not the results of the technical review. For example, the Department 
of Health and Human Services' National Institutes of Health published 
a list of 21,581 award winners for nearly $9 billion, but, similar to 
the HSIPR program, did not report the results of the technical review. 
Eight other programs conveyed less information than FRA and did not 
publicly communicate the results of the technical review or the awards 
and award amounts. 

Race to the Top was the only program we examined that publicly 
provided the results of its technical review. These results, which 
were posted on the Department of Education's Web site, included scores 
and comments from reviewers for each applicant, but were not connected 
to individual reviewers by name. According to its Web site, the 
Department of Education decided to release this level of detailed 
information because the $4 billion Race to the Top program was larger 
than any other discretionary program the Department of Education had 
previously administered, and officials sought to ensure the highest 
level of integrity and transparency. Unlike the HSIPR program, 
however, Race to the Top used these scores as the sole basis for 
selecting awards and only chose applicants receiving the highest 
scores. As described in the previous section, the technical review 
scores were an important component for making HSIPR selection 
decisions, but did not include consideration of additional pre-
established selection criteria designed to ensure long-term success 
and sustainability of the program. As such, publishing them without 
additional decision making information on the specific reasons for 
selecting and not selecting individual applications could lead to 
erroneous conclusions about FRA's decisions. 

According to FRA officials, the results of the technical review were 
not communicated because department officials were concerned that 
associating technical review scores and comments with a specific 
reviewer could discourage reviewers from participating in future 
department competitive grant evaluations. Furthermore, in their view, 
this might also prevent reviewers in future funding rounds from 
providing candid evaluations. However, as the Race to the Top program
demonstrated, it would be possible for FRA to present overall 
technical panel review assessments or their individual comments 
without linking individuals' names to comments, if it chooses to do 
so. FRA officials stated that the anonymous disclosure of technical 
scores and comments would still prevent FRA and department leadership 
and staff from frankly expressing their individual judgments, as they 
might still be concerned over how theses opinions would reflect on the 
FRA and HSIPR program if they were made public. 

Conclusions: 

The $8 billion appropriated by the Recovery Act for the High Speed 
Intercity Passenger Rail program represents a large investment in the 
development of a national passenger rail network. FRA established a 
fair and objective approach for distributing these funds and 
substantially followed recommended discretionary grant award practices 
used throughout the government. The exception is what we view as 
incomplete documentation of why some applications were chosen and not 
others, and how FRA decided to distribute the funds at the time those 
decisions were made. This incomplete documentation is notable given 
the robust documentation of the other steps used to determine 
eligibility and assess technical merit. We believe that establishing a 
record that provides insight into why decisions were made, rather than 
merely restating general technical review and selection criteria, 
including amounts to be provided, would enhance the credibility of 
FRA's awards decisions to the extent that this record confirms that 
selected projects aligned with established criteria and goals. By not 
establishing this record, FRA invites skepticism about the overall 
fairness of its decisions, even if they are sound, and hinders 
meaningful disclosure of how it made its decisions, if it chooses to 
do so. 

Recommendation for Executive Action: 

To help ensure accountability over federal funds, we recommend that 
the Secretary of Transportation direct the Administrator of the 
Federal Railroad Administration to create additional records that 
document the rationales for award decisions in future HSIPR funding 
rounds, including substantive reasons (1) why individual projects are 
selected or not selected and (2) for changes made to requested funding 
amounts. 

Agency Comments: 

We provided a draft of this report to the Department of Transportation 
for its review and comment. The department told us that it carefully 
constructed the grant processes for the HSIPR program based on 
extensive review and consideration of best practices both within and 
outside the agency with the intent of providing a comprehensive and 
transparent process. The department indicated that its overall intent 
was to select the best projects that offered the greatest available 
and achievable benefit to the nation. The department told us that it 
would carefully consider our recommendation to determine if there are 
means to further enhance the transparency of its grant selection 
process with additional documentation, without creating a process that 
is unduly burdensome to administer. The department also offered 
technical comments which we incorporated as appropriate. 

As agreed with your office, unless you publicly announce the contents 
of this report earlier, we plan no further distribution until 30 days 
from the report date. We are sending copies of this report to 
congressional subcommittees with responsibilities for surface 
transportation issues; the Director, Office of Management and Budget; 
the Secretary of Transportation; and the Administrator of the Federal 
Railroad Administration. In addition, this report will be available at 
no charge on GAO's Web site at [hyperlink, http://www.gao.gov]. 

If you or your staff have any questions regarding this report, please 
contact me at (202) 512-2834 or flemings@gao.gov. Contact points for 
our Offices of Congressional Relations and Public Affairs may be found 
on the last page of this report. GAO staff who made key contributions 
to this report are Owen Bruce, Matthew Cook, Colin Fallon, Michele 
Fejfar, Maria Gaona, Grant Mallie, James Ratzenberger, Douglas Sloane, 
Matthew Voit, and Crystal Wesco. 

Sincerely yours, 

Signed by: 

Susan A. Fleming: 
Director, Physical Infrastructure Issues: 

[End of section] 

Appendix I: Extent to Which Recovery Act Projects Align with Statutory 
and Other Goals: 

We examined the extent to which American Recovery and Reinvestment Act 
of 2009 (Recovery Act) projects selected by the Federal Railroad 
Administration (FRA) align with legislative and the administration's 
goals to develop high speed and conventional rail networks.
Congress Provided Broad Goals with a Priority for High Speed 
Systems	Congress provided its most recent expectations for high and 
conventional speed rail in the Passenger Rail Investment and 
Improvement Act of 2008 (PRIIA) and the Recovery Act. In this regard, 
PRIIA speaks generally about supporting improvements to high and 
conventional speed rail and does not set out any expectations for 
relative attention to high and conventional speed passenger rail 
improvements. The Recovery Act appropriated $8 billion for both forms 
of rail service broadly. However, it required that FRA give priority 
to projects that support the development of intercity high speed 
service. Further, the act required that FRA develop a strategic plan 
that describes how FRA will use Recovery Act funding to improve and 
deploy high speed systems. FRA had wide latitude to achieve goals laid 
out in its strategic plan. 

FRA's Vision Describes Broad Goals for High Speed Rail, but Provides 
Limited Detail on How Goals Will Be Achieved: 

FRA has outlined its vision for developing intercity passenger rail 
service in its strategic plan, as required by the Recovery Act, and in 
both its preliminary national rail plan issued in 2009 and the plan's 
update nearly a year later. FRA's vision documents—the strategic plan 
and its updated national rail plan—described broad goals, such as for 
transportation, safety, and economic competitiveness, and established 
categories for the type of high speed rail projects it intends to 
support. For example, the strategic plan notes the high speed rail 
program aims to generate construction and operating jobs, while 
providing a steady market for various industries producing rail, 
control systems, locomotives, and passenger cars. In addition, the 
plan notes that investments in high speed rail can result in 
competitive trip times and rail transport can also result in higher-
density development as compared to other modes of transportation. 
Similarly, the updated national rail plan sets a goal of connecting 
communities through high speed rail while, among other things, 
reducing congestion, boosting economic growth, and promoting economic 
sustainability. 

However, these vision documents provide limited details on goals for 
the high speed rail program. For example, while the strategic plan 
emphasizes investments that will yield tangible benefits to rail 
performance and improve connections between different modes of 
transportation, it does not describe how and when FRA intends to 
realize these benefits. In addition, as we reported last June, the 
preliminary rail plan did not offer specific recommendations for 
future action and was designed to serve as a springboard for further 
discussion with states and freight railroads.[Footnote 31] While the 
update to this plan included improving rail performance as a goal and 
provided some measurements for high speed rail performance, such as 
competitive trip times, it did not provide any specific targets for 
these metrics, or any time line showing when FRA hopes to attain these 
improvements. 

FRAs Application Selection Is Consistent with the Recovery Act's 
Priority for High Speed Rail Service: 

Consistent with the Recovery Act's direction to give priority for high 
speed rail service, about half (45 percent) of the applications 
selected were for core express corridors (high speed service of 125-
250 miles per hour or more) or regional corridors (higher-speed 
service of 90-124 miles per hour) using categories of service similar 
to those FRA established in its vision documents.[Footnote 32] (See 
table 4.) FRA did not establish specific targets for the number of 
each type of project it intended to support.[Footnote 33] 

Table 4: Recovery Act Applications Supporting High Speed Rail 
Categories by Future Corridor Speed 5 Years After Project Completion: 

Category: Core express corridors; 
Top speed: 125-250 miles per hour or more; 
Number of projects meeting targeted corridor top speed: 5. 

Category: Regional corridors; 
Top speed: 90-124 miles per hour; 
Number of projects meeting targeted corridor top speed: 23[A]. 

Category: Emerging high speed rail; 
Top speed: Up to 90 miles per hour; 
Number of projects meeting targeted corridor top speed: 21. 

Source: GAO categorization of applications based on FRA information 
and applicant data. 

Note: We categorized these applications solely by future project speed 
5 years after completion using the speeds reported at the time the 
application was submitted. Thirteen applications did not provide 
information on anticipated top speed. These applications anticipate a 
variety of improvements, such as station rehabilitations, the 
reconfiguration of rolling stock, and track and grade crossing 
upgrades. FRA subsequently classified two of these projects as 
supporting core express corridors, six as supporting regional 
corridors, two of these projects as emerging routes, and the remaining 
three projects as contributing to more than one category. 

We relied on data submitted by applicants to FRA to assign 
applications to categories. We have reported that applicants for major 
infrastructure projects, such as high speed rail projects, often 
overstate benefits, such as speed of service. See [hyperlink, 
http://www.gao.gov/products/GA0-09-317]. 

Some projects may attain higher speeds than those reported in the 
applications following negotiations between FRA and the various 
states. In addition, because corridors may include multiple projects, 
the top speed of a corridor may exceed those for some of its component 
projects. 

[A] Five of these applications had an estimated future speed of 110 
miles per hour, but did not specify whether this speed would be 
achieved within the 5 years following project completion. 

[End of table] 

Selected Applications Reflect Short-Term Economic Recovery and Long-
Term Infrastructure Investment Goals: 

In addition to providing priority for high speed projects, the 
applications that FRA selected were consistent with near-term economic 
recovery goals and its long-term development goals. While most 
selected projects are short-term in nature and are intended to support 
economic recovery goals established by the Recovery Act, most funding 
was provided to several long-term, high speed corridor projects. 
Specifically, we found that 48 of the 62 applications selected were 
track 1 applications, which are smaller projects designed to be 
completed within 2 years. (See table 5.) These projects represent 
about 11 percent of the funding provided for high speed rail and 
intercity passenger rail through the Recovery Act. The remaining 89 
percent of funding was provided for 14 track 2 applications, which are 
primarily long-term, corridor projects. The funding allocation aligns 
with FRA's focus on long-term investments that will support 
development of a high speed passenger rail network as described in the 
funding announcement. 

Table 5: Amounts Awarded, Obligated, and Spent on First Round Track 1 
and 2 Recovery Act Projects, as of December 31, 2010: 
				
Project track: Track 1; 
Number of applications: 48; 
Amount awarded: $887 million; 
Amount obligated: $71 million; 
Amount spent: $0. 

Project track: Track 2; 
Number of applications: 14; 
Amount awarded: $7.025 billion; 
Amount obligated: $4.192 billion; 
Amount spent: $50 million. 

Project track: Total; 
Number of applications: 62; 
Amount awarded: $7.912 billion; 
Amount obligated: $4.263 billion; 
Amount spent: $50 million. 

Source: GAO analysis of FRA data. 

Note: This table does not include $15 million obligated or $10 million 
spent by FRA for contracts associated with the award and oversight of 
these grants. 

[End of table] 

While FRA announced in January 2010 awards of nearly $8 billion in 
grants for the program, many of these projects have only recently 
begun. As of December 31, 2010, FRA had obligated $4.2 billion, or 
about 54 percent of the funding awarded in January, and about $50 
million has been spent for projects selected under track 2.
In May 2009, FRA issued a plan for spending Recovery Act funds, which 
it updated in July 2010. FRA missed its May 2009 targets for 
obligations and spending through 2010 estimate because it had planned 
to announce awards—and begin obligating funds—in the autumn of 2009. 
However, FRA did not make those announcements until January 2010 and 
did not begin to obligate funds until May 2010. FRA then revised its 
estimates in July 2010. FRA surpassed the calendar year 2010 goals for 
obligating and spending funds in the July 2010 plan. (See table 6.) 
During calendar year 2010, FRA obligated about 11 times as much as 
anticipated in the July 2010 plan, while awardees have spent about 7 
times as much as planned over the same time period. The Recovery Act 
authorized obligation of funds through September 30, 2012, and FRA 
intends to obligate all funds by this date.[Footnote 34] 

Table 6: FRA Plan for Obligating and Spending Recovery Act Funds and 
Amounts Obligated and Spent, as of December 31, 2010: 

Amount obligated: 	
May 2009 plan: $6.002 billion; 
July 2010 plan: $400 million; 
Actual: $4.263 billion. 

Amount spent: 
May 2009 plan: $1.760 billion. 
July 2010 plan: $7 million; 
Actual: $51 million. 

Source: GAO analysis of FRA data. 

Note: This table does not include $15 million obligated or $10 million 
spent by FRA for contracts associated with the award and oversight of 
these grants. 

[End of table] 

Passenger rail investments are often long-term efforts that must be 
carried out in partnership between the state and others, notably 
private railroads. For example, in order to begin design and 
construction on many of these projects, grant recipients must 
negotiate and secure agreements with private freight railroads to use 
their tracks for passenger rail trains. However, officials from these 
railroads are concerned that sharing tracks would create safety risks 
and liability concerns, prevent freight expansion, and cause rail 
congestion. Some of the states have experienced delays finalizing 
these agreements with the railroads and, accordingly, have not 
completed agreements with FRA to obligate awarded funding. 

[End of section] 

Appendix II: Scope and Methodology Criteria Used to Select Projects: 

To determine the extent to which FRA applied its established criteria 
to select projects, we identified the criteria that it planned to use 
from its June 23, 2009, funding announcement outlining its evaluation 
and selection approach. We then compared these criteria to the 
worksheets and guidebooks that FRA used to determine eligibility and 
assess technical merit. Finally, we interviewed FRA officials who 
participated in evaluating and selecting projects to obtain 
information on whether and how they applied the established criteria. 
Specifically, we randomly selected 1 technical reviewer from each of 
the 12 track 1, 3, and 4 panels (12 out of 36 reviewers), and 6 of the 
8 reviewers from the track 2 panel. In addition, we interviewed senior 
FRA officials to further understand how senior Department of 
Transportation (the department) and FRA officials applied the 
selection criteria, selected projects, and determined the amount of 
funding provided for each project. We also asked FRA officials to 
provide reasons for why several lower scored applications were 
selected, while other higher scored applications were not. 

We conducted semi-structured interviews with officials from 10 of the 
40 states and the District of Columbia that submitted a preapplication 
or an application for track 1 and track 2 funding about how FRA 
communicated its approach to reviewing applications and award results. 
[Footnote 35] We selected these states on the basis of four 
characteristics: (1) the extent to which applicants progressed through 
the preapplication, application, evaluation, and selection stages; (2) 
geographic regions; (3) the number of applications submitted; and (4) 
the amount of funding. We also contacted officials in two additional 
states (Ohio and Washington) to understand the effect of FRA's funding 
decisions on the scope of these states' proposed rail program. 
[Footnote 36] Our efforts were limited to applications requesting 
funding under track 1 and track 2 of the High Speed Intercity 
Passenger Rail Program (HSIPR) in August 2009 and October 2009 and 
awarded Recovery Act funding for projects in January 2010. We did not 
review FRA's rationale for its decision in December 2010 to 
redistribute $1.195 billion from two projects in Ohio and Wisconsin to 
on-going high speed rail projects in 13 states. 

We also assessed whether FRA's approach to calculating reviewers' 
individual scores and compiling them for an overall panel score 
reflected the criteria and weights for each criterion as published in 
the funding announcement as well as the overall reliability of the 
data used to make these calculations. To do this, we reviewed 
documentation about the system used to collect the information and 
spoke with officials knowledgeable about the data. We found some 
inaccuracies in how FRA calculated the technical review scores. 
Specifically, we found that some standardized scores were incorrect 
due to the inclusion of three duplicate records and three applications 
deemed not yet ready or ineligible. In addition, we noted FRA 
incorrectly weighted some technical evaluation scores for applications 
submitted under track 1b. However, we determined that these errors 
would not materially affect our findings and for the purposes of 
examining the effect of the scores on application selection, we found 
the data to be sufficiently reliable. FRA officials said that they 
would correct their calculations for future rounds of rail funding.
Further we performed tests to determine the variables (e.g., technical 
review scores and number of applications submitted) that had a 
significant statistical relationship with being selected for an award. 
Our approach is described in appendix IV. 

Following Recommended Practices for Discretionary Grant Awards: 

To determine the extent to which FRA used recommended practices for 
awarding discretionary grants, we examined Office of Management and 
Budget guidance, guidance from several federal agencies, and our 
reports on this issue. (See table 7.) We identified key grant practices	
recommended across executive branch agencies and compared them to 
practices analyzed in our prior work.[Footnote 37] Specifically, we 
identified six recommended practices relating to (1) communicating 
with potential applicants prior to the competition, (2) planning for 
administering the review of applications, (3) developing a technical 
review panel with certain characteristics, (4) assessing applicants' 
abilities to manage grant funds, (5) notifying applicants of 
decisions, and (6) documenting reasons for award decisions. We 
compared these practices to information from the 2009 funding 
announcement, guidance to applicant reviewers, and to statements made 
by FRA officials regarding their implementation of their grants award 
program. For this effort, one analyst carried out the comparison and a 
second analyst verified the comparison results. Where differences 
existed, the two analysts discussed them and reached agreement. We 
also discussed the extent of FRA's use of several of these practices 
with the officials from our sample of 10 states. 

Table 7: Guidance and Reports Used To Identify Recommended Government 
Practices: 

Federal agency: 

Source: Department of Commerce; 
Guidance or report: Grants and Cooperative Agreements Manual (June 
2007). 

Source: Department of Energy; 
Guidance or report: Merit Review Guide for Financial Assistance 
(August 2007). 

Source: Department of Labor; 
Guidance or report: U.S. Department of Labor, Veterans' Employment and 
Training Service Guide to Competitive and Discretionary Grants (April 
2003). 

Source: Department of Transportation; 
Guidance or report: Financial Assistance Guidance Manual (March 2009). 

Source: Office of Management and Budget; 
Guidance or report: Office of Federal Financial Management Policy 
Directive on Financial Assistance Program Announcements, 68 FR 37370 
(June 23, 2003). 

GAO: 

Guidance or report: Runaway and Homeless Youth Grants: Improvements 
Needed in the Grant Award Process, [hyperlink, 
http://www.gao.gov/products/GA0-10-335] (Washington, D.C.: May 10, 
2010); 

Guidance or report: Discretionary Grants: Further Tightening of 
Education's Procedures for Making Awards Could Improve Transparency 
and Accountability, [hyperlink, 
http://www.gao.gov/products/GAO-06-268] (Washington, D.C.: Feb. 21, 
2006); 

Guidance or report: Grants Management Despite Efforts to Improve Weed 
and Seed Program Management, Challenges Remain, [hyperlink, 
http://www.gao.gov/products/GAO-04-245] (Washington, D.C.: Mar. 24, 
2004) 

Guidance or report: Education Discretionary Grants: Awards Process 
Could Benefit From Additional Improvements, [hyperlink, 
http://www.gao.gov/products/GAO/HEHS-00-55] (Washington, D.C.: Mar. 
30, 2000); 

Guidance or report: Standards for Internal Control in the Federal 
Government, [hyperlink, 
http://www.gao.gov/products/GAO/AIMD-00-21.3.1] (Washington, D.C.: 
November 1999). 

Governmentwide: 

Source: Grant Accountability Project[A]; 
Guidance or report: Grant Accountability Project, Guide to 
Opportunities for Improving Grant Accountability (October 2005). 

Source: GAO. 

[End of table] 

This project was initiated by the Domestic Working Group, which 
consists of 19 federal, state, and local audit organizations and is 
chaired by the Comptroller General of the United States. The purpose 
of the group is to identify current and emerging challenges of mutual 
interest and explore opportunities for greater collaboration within 
the intergovernmental audit community. 

Communication of Selection Results: 

To determine the extent FRA publicly communicated information about 
the results of its award competition, we compared the information it 
communicated to the public about its awards to the types of 
information communicated by a random sample of 20 other competitively 
awarded Recovery Act programs. (See table 8.) We selected the sample 
from 193 Recovery Act programs identified in the Catalog of Federal 
Domestic Assistance as competitive grant programs using Recovery Act 
funds.[Footnote 38] In addition, we compared the information 
communicated about FRA's awards to the information communicated by the 
Innovation Grants program (Race to the Top)—a discretionary grant 
program run by the Department of Education. We included the Race to 
the Top program because you expressed interest in it. 

Table 8: Recovery Act Discretionary Grant Programs Reviewed: 

Program: Broadband Technology Opportunities Program; 
Responsible federal agency: Department of Commerce. 

Program: Central Valley Project Improvement Act, Title XXXIV; 
Responsible federal agency: Department of the Interior. 
			
Program: Emergency Medical Services for Children (Recovery Act); 
Responsible federal agency: Department of Health and Human Services. 

Program: Emergency Watershed Protection Program; 
Responsible federal agency: Department of Agriculture. 

Program: Fish and Wildlife Coordination Act; 
Responsible federal agency: Department of the Interior. 

Program: Geologic Sequestration Training and Research Grant Program; 
Responsible federal agency: Department of Energy. 

Program: Grants to Health Center Programs (Recovery Act); 
Responsible federal agency: Department of Health and Human Services. 

Program: Head Start (Recovery Act); 
Responsible federal agency: Department of Health and Human Services. 

Program: National Geospatial Program: Building The National Map; 
Responsible federal agency: Department of the Interior. 

Program: National Railroad Passenger Corporation Grants; 
Responsible federal agency: Department of Transportation. 

Program: Office of Science Financial Assistance Program; 
Responsible federal agency: Department of Energy. 

Program: Pregnancy Assistance Fund Program; 
Responsible federal agency: Department of Health and Human Services. 

Program: Preventing Healthcare—-Associated Infections (Recovery Act); 
Responsible federal agency: Department of Health and Human Services. 

Program: Prevention and Wellness-—Leveraging National Organizations 
(Recovery Act); 
Responsible federal agency: Department of Health and Human Services. 

Program: Recovery Act Grants for Training in Primary Care Medicine and 
Dentistry Training and Enhancement; 
Responsible federal agency: Department of Health and Human Services. 

Program: Recovery Act Transitional Housing; 
Responsible federal agency: Department of Justice. 

Program: Science Grants for Basic Research, Educational Outreach, or 
Training Opportunities (Recovery Act); 
Responsible federal agency: National Aeronautics and Space 
Administration. 

Program: Senior Community Service Employment Program; 
Responsible federal agency: Department of Labor. 

Program: State Fiscal Stabilization Fund Race-to-the-Top Incentive 
Grants (Recovery Act); 
Responsible federal agency: Department of Education. 

Program: State Grants to Promote Health Information Technology 
(Recovery Act); 
Responsible federal agency: Department of Health and Human Services. 

Program: Trans-National Institutes of Health Research Support 
(Recovery Act); 
Responsible federal agency: Department of Health and Human Services. 

Source: GAO. 

[End of table] 

We first reviewed materials on FRA's Web site and other public 
releases, such as press releases and outreach presentations to 
determine what FRA publicly communicated. We then discussed these 
results with FRA officials to confirm our results. 

For each of the 21 other Recovery Act programs, we reviewed three 
public information sources: (1) the program's Catalog of Federal 
Domestic Assistance award announcement, (2) internet search results, 
and (3) Grants.gov, which provides information on more than 1,000 
grant programs.[Footnote 38] For each program, we searched these 
sources for information about final award results (project 
description, why the project was selected, and award amount) and for 
information that demonstrated how applications fared at different 
states of the process (eligibility determination and internal reviews, 
such as technical review panels). We defined the results of any 
technical review as either scores or comments, and when at least one 
of these elements was listed in at least one of the three sources of 
information, we concluded that technical review information was 
publicly communicated about the program. In carrying out this 
assessment, one analyst carried out the work and a second analyst 
independently performed the same tasks. The two analysts then compared 
their results and resolved any differences. The results of our 
comparison to a sample of other Recovery Act programs are not 
generalizable across all Recovery Act programs. 

Alignment with Statutory and Other Goals: 

To determine the extent to which HSIPR applications align with 
statutory and other goals, we reviewed federal laws, including the 
Recovery Act and PRIIA. We analyzed FRA's Federal Register notice 
describing its approach for selecting applications, its strategic 
vision for high speed rail, and its preliminary national rail plan and 
its subsequent update to gather information on any goals the agency 
has established for high speed rail networks and conventional service 
and the types of projects it seeks to support. We did not assess 
whether the applications selected by FRA will achieve the stated 
benefits or costs. 

We reviewed information submitted by applicants, namely the type of 
project proposed, the funding requested and awarded, and the estimated 
future speed of the projects. We used this data to sort projects into 
three categories developed by FRA: core express corridors, regional 
corridors, and emerging high speed rail routes. FRA's definition of 
top speeds within these categories overlap, which we modified slightly 
to provide discrete endpoints. 

Of the 62 applications selected by FRA, 13 did not provide data on 
anticipated top speed after project completion. These 13 applications 
include a variety of improvements, including station rehabilitations, 
the reconfiguration of rolling stock, and existing tracks and grade 
crossings upgrades for which one would not expect top speed 
information. We used these data as background on selected applications 
and did not assess them for reliability. We also reviewed FRA's 
Recovery Act plans and compared FRA goals for obligating and spending 
awarded funds to its actual rates of obligating and spending from 
January 2010 through December 2010. After reviewing a Department of 
Transportation Inspector General audit report on its financial 
management system and speaking with department officials familiar with 
the system, we determined that these data were sufficiently reliable. 

[End of section] 

Appendix III: Difference Between the Amounts Requested and Estimated 
Awards by State: 

In January 2010, FRA proposed to provide 18 of the 24 states, 
including the District of Columbia, selected for awards all or nearly 
all (91 percent or more) of the money that they requested. (See table 
9.) The agency proposed to provide one state, North Carolina, with 
slightly more than it requested and the remaining five states with 
amounts varying from 47 percent to 86 percent of the amounts requested. 

Table 9: Difference between the Amounts Requested and Estimated Awards 
by State, as of January 2010: 

State: North Carolina; 
Amount requested: $523.8 million; 
FRA proposed amount: $545.0 million; 
Difference (percent): $21.2 million (104%). 

State: District of Columbia; 
Amount requested: $2.9 million; 
FRA proposed amount: $2.9 million; 
Difference (percent): $0.0 (100%). 

State: Maryland; 
Amount requested: $69.4 million; 
FRA proposed amount: $69.4 million; 
Difference (percent): $0.0 (100%). 

State: Michigan; 
Amount requested: $40.3 million; 
FRA proposed amount: $40.3 million; 
Difference (percent): $0.0 (100%). 

State: New Jersey; 
Amount requested: $38.5 million; 
FRA proposed amount: $38.5 million; 
Difference (percent): $0.0 (100%). 

State: Pennsylvania; 
Amount requested: $25.7 million; 
FRA proposed amount: $25.7 million; 
Difference (percent): $0.0 (100%). 

State: Rhode Island; 
Amount requested: $1.2 million; 
FRA proposed amount: $1.2 million; 
Difference (percent): $0.0 (100%). 

State: Texas; 
Amount requested: $3.8 million; 
FRA proposed amount: $3.8 million; 
Difference (percent): $0.0 (100%). 

State: Virginia; 
Amount requested: $74.8 million; 
FRA proposed amount: $74.8 million; 
Difference (percent): $0.0 (100%). 

State: Wisconsin; 
Amount requested: $831.7 million; 
FRA proposed amount: $822.0 million; 
Difference (percent): -$9.7 million (99%). 

State: Indiana; 
Amount requested: $71.4 million; 
FRA proposed amount: $71 million; 
Difference (percent): -$0.4 million (99%). 

State: Iowa; 
Amount requested: $17.3 million; 
FRA proposed amount: $17 million; 
Difference (percent): -$0.3 million (98%). 

State: Illinois; 
Amount requested: $1.275.3 billion; 
FRA proposed amount: $1.233.0 billion; 
Difference (percent): -$42.3 million (97%). 

State: Connecticut; 
Amount requested: $41.1 million; 
FRA proposed amount: $40.0 million; 
Difference (percent): -$1.1 million (97%). 

State: Massachusetts; 
Amount requested: $72.9 million; 
FRA proposed amount: $70.0 million; 
Difference (percent): -$2.9 million (96%). 

State: New York; 
Amount requested: $157.4 million; 
FRA proposed amount: $150.0 million; 
Difference (percent): -$7.4 million (95%). 

State: Vermont; 
Amount requested: $52.7 million; 
FRA proposed amount: $50.0 million; 
Difference (percent): -$2.7 million (95%). 

State: Missouri; 
Amount requested: $33.3 million; 
FRA proposed amount: $31.0 million; 
Difference (percent): -$2.3 million (93%). 

State: Maine; 
Amount requested: $38.4 million; 
FRA proposed amount: $35.0 million; 
Difference (percent): -$3.4 million (91%). 

State: Oregon; 
Amount requested: $9.4 million; 
FRA proposed amount: $8.0 million; 
Difference (percent): -$1.4 million (86%). 

State: Ohio; 
Amount requested: $563.8 million; 
FRA proposed amount: $400.0 million; 
Difference (percent): -$163.8 million (71%). 

State: Washington; 
Amount requested: $976.4 million; 
FRA proposed amount: $590.0 million; 
Difference (percent): -$386.4 million (60%). 

State: California[A]; 
Amount requested: $4.766.0 billion; 
FRA proposed amount: $2.343.0 billion; 
Difference (percent): -$2.423.0 billion (49%). 

State: Florida; 
Amount requested: $2.654.0 billion; 
FRA proposed amount: $1.250.0 billion; 
Difference (percent): -$1.404.0 billion (47%). 

Source: GAO analysis of FRA data. 

[A] There were two different entities submitting applications for 
projects in California. The California Department of Transportation 
submitted track 1a and 1 b applications, and the California High Speed 
Rail Authority, a public agency established by California to develop 
high speed rail, submitted track 2 applications. 

[End of table] 

In December 2010, nearly a year after making these proposals, FRA 
announced that $1.195 billion in Recovery Act funds for high speed 
rail-representing most of the $810 million for Wisconsin's Milwaukee-
Madison corridor and $385 million for Ohio's Cincinnati-Columbus-
Cleveland "3C" route, originally designated for these states in 
January 2010-would be redirected to high speed rail projects already 
underway in 13 other states.[Footnote 39] In making these changes, FRA 
noted that Wisconsin has suspended work under its existing high speed 
rail agreement and the incoming governors in Wisconsin and Ohio have 
both indicated that they will not move forward to use high speed rail 
money received under Recovery Act. The adjusted amounts resulted in 
FRA proposing to provide all or nearly all of the original request 
amounts (91 percent or more) for one additional state (Oregon). (See 
table 10.) While most of the funding was redistributed to three states 
(California, Florida, and Washington), the total funding awarded to 
these three states was less than 80 percent of their original requests. 

Table 10: Difference between the Amounts Requested and Estimated 
Awards by State, as of December 2010: 
			
State: North Carolina; 
Applicant requested amount: $523.8 million; 
FRA adjusted amount: $546.5 million; 
Difference (percent): $22.7 million (104%). 

State: District of Columbia; 
Applicant requested amount: $2.9 million; 
FRA adjusted amount: $2.9 million; 
Difference (percent): $0.0 (100%). 

State: Illinois; 
Applicant requested amount: $1.275.3 billion; 
FRA adjusted amount: $1.275.3 billion; 
Difference (percent): $0.0 (100%). 

State: Indiana; 
Applicant requested amount: $71.4 million; 
FRA adjusted amount: $71.4 million; 
Difference (percent): $0.0 (100%). 

State: Iowa; 
Applicant requested amount: $17.3 million; 
FRA adjusted amount: $17.3 million; 
Difference (percent): $0.0 (100%). 

State: Maryland; 
Applicant requested amount: $69.4 million; 
FRA adjusted amount: $69.4 million; 
Difference (percent): $0.0 (100%). 

State: Michigan; 
Applicant requested amount: $40.3 million; 
FRA adjusted amount: $40.3 million; 
Difference (percent): $0.0 (100%). 

State: New Jersey; 
Applicant requested amount: $38.5 million; 
FRA adjusted amount: $38.5v; 
Difference (percent): $0.0 (100%). 

State: Oregon; 
Applicant requested amount: $9.4 million; 
FRA adjusted amount: $9.4 million; 
Difference (percent): $0.0 (100%). 

State: Pennsylvania; 
Applicant requested amount: $25.7 million; 
FRA adjusted amount: $25.7 million; 
Difference (percent): $0.0 (100%). 

State: Rhode Island; 
Applicant requested amount: $1.2 million; 
FRA adjusted amount: $1.2 million; 
Difference (percent): $0.0 (100%). 

State: Texas; 
Applicant requested amount: $3.8 million; 
FRA adjusted amount: $3.8 million; 
Difference (percent): $0.0 (100%). 

State: Vermont; 
Applicant requested amount: $52.7 million; 
FRA adjusted amount: $52.7 million; 
Difference (percent): $0.0 (100%). 

State: Virginia; 
Applicant requested amount: $74.8 million; 
FRA adjusted amount: $74.8 million; 
Difference (percent): $0.0 (100%). 

State: Maine; 
Applicant requested amount: $38.4 million; 
FRA adjusted amount: $38.3 million; 
Difference (percent): -$0.1 million (100%). 

State: Massachusetts; 
Applicant requested amount: $72.9 million; 
FRA adjusted amount: $72.8 million; 
Difference (percent): -$0.1 million (100%). 

State: Missouri; 
Applicant requested amount: $33.3 million; 
FRA adjusted amount: $33.2 million; 
Difference (percent): -$0.1 million (100%). 

State: New York; 
Applicant requested amount: $157.4 million; 
FRA adjusted amount: $157.3 million; 
Difference (percent): -$0.1 million (100%). 

State: Connecticut; 
Applicant requested amount: $41.1 million; 
FRA adjusted amount: $40.0 million; 
Difference (percent): -$1.1 million (97%). 

State: Washington; 
Applicant requested amount: $976.4 million; 
FRA adjusted amount: $751.5 million; 
Difference (percent): -$224.9v (77%). 

State: California[A]; 
Applicant requested amount: $4.766.0 billion; 
FRA adjusted amount: $2.967.0 billion; 
Difference (percent): -$1.799.0 billion (62%). 

State: Florida; 
Applicant requested amount: $2.654.0 billion; 
FRA adjusted amount: $1.592.3 billion; 
Difference (percent): -$1.061.7 billion (60%). 

State: Wisconsin; 
Applicant requested amount: $831.7 million; 
FRA adjusted amount: $44.0 million; 
Difference (percent): -$787.7 million (5%). 

State: Ohio; 
Applicant requested amount: $563.8 million; 
FRA adjusted amount: $15.0 million; 
Difference (percent): -$548.8 million (3%). 

Source: GAO analysis of FRA data. 

Note: Due to rounding the percentage difference between applicant 
requested and FRA adjusted amounts may equal 100, even when there is a 
small dollar difference. 

[A] There were two different entities submitting applications for 
projects in California. The California Department of Transportation 
submitted track la and 11a b applications, and the California High 
Speed Rail Authority, a public agency established by California to 
develop high speed rail, submitted track 2 applications. 

[End of table] 

[End of section] 

Appendix IV: Additional Results from Our Statistical Analysis of Award 
Decisions: 

This appendix contains information related to our statistical analyses 
of FRA and application data to examine possible relationships between 
several variables and FRA's selection decisions. 

Overview of the Data: 

We obtained the data for our analysis from the Application Review 
Module of GrantSolutions, the database FRA used to store application 
information and technical review scores. Our analysis examined all 206 
out of 259 submitted applications which FRA deemed eligible and ready 
to receive federal funds. Eligible applications included those 
requesting Recovery Act funds, tracks 1 and 2, as well as those 
requesting annual appropriations, tracks 3 and 4. We included track 3 
and 4 applications in our analysis because FRA reviewed, weighted, and 
calculated the results for tracks 1, 3, and 4 applications as a group 
rather than by distinct tracks.[Footnote 40] 

To assess the reliability of the data in Application Review Module, we 
reviewed database user manuals, spoke with officials knowledgeable 
about the data, and conducted a series of data tests. We found some 
inaccuracies in how FRA calculated the technical review scores. 
Specifically, we found that some final technical review scores were 
incorrect due to the inclusion of three duplicate technical review 
scores and three applications later determined to be not yet ready or 
ineligible.[Footnote 41] In addition, FRA had mistakenly applied 
incorrect weights to the track 1b application technical review scores, 
which resulted in 15 final scores that were one point higher than they 
should have been and another 5 final scores that were one point lower 
than they should have been. We determined that these errors would not 
materially affect our findings and for the purposes of examining the 
effect of scores on application selection, found these data are 
sufficiently reliable. 

Methodology: 

To determine the extent to which specific variables were related to 
the department's selection of applications, we considered a set of 
bivariate tables and conducted a series of bivariate and multivariate 
regression analyses.[Footnote 42] From the tables and regression 
analyses we were interested in determining how the department's 
decision to select an application for an award was affected by four 
variables: (1) the technical review scores, (2) application track, (3) 
the requested funding amount, and (4) the number of applications 
submitted by state or groups of states. 

Our analyses provide us with estimates, called odds ratios, which 
indicate the differences in the odds of applications being selected 
for an award across certain categories of the different variables we 
examined.[Footnote 43] An odds ratio of 1.0 would indicate that 
applications in different categories were equally likely to be 
selected for an award. An odds ratio of less than 1.0 implies that 
applications in the category to which the odds ratio applies were less 
likely to be selected relative to those they are being compared to 
(known as the "reference" category). For example, if applications 
receiving a technical review score of a 3 had an odds ratio of 0.5 it 
would indicate that they were half as likely to be selected for an 
award as applications that received a score of 1 or 2 (the reference 
category). Inversely, an odds ratio greater than 1.0 suggests that 
applications with that characteristic were more likely to be selected. 
For example, if applications receiving a technical review score of a 5 
had an odds ratio of 3.0, we would conclude that applications 
receiving that score were three times more likely to be selected 
relative to the reference category. The primary reason for preferring 
odds ratios to describe the relationships across variables is because 
the significance of the differences between specific odds ratios can 
be easily tested and the ratios can be re-estimated after considering 
other variables. 

Technical Review Score Affected the Selection of Applications: 

We first examined the effect of technical review scores on the 
likelihood of being selected for an award, and found that the odds of 
being selected for an award were in general greater for applications 
receiving higher technical review scores than for applications 
receiving lower ones. For the purposes of our analyses we combined 
scores of 1 and 2, as there were only five applications that had 
received a score of 1 and there was no evidence that they were 
significantly different from applications that had been assigned 
scores of 2, in terms of being selected for funding. A smaller 
percentage of the applications that received scores of 1 or 2 were 
selected for funding (16 percent selected) than applications that had 
received a score of 3 (46 percent selected) or 4 (42 percent 
selected), and applications that received a score of 5 had the highest 
percentage of being selected for funding (69 percent selected). In 
addition, the odds ratios of 4.30, 3.67, and 11.57 indicate that 
applications receiving a higher technical review score (3, 4, or 5, 
respectively) were at least three times more likely to be selected for 
an award than those receiving a lower technical review score (i.e., 1 
or 2). (See table 11.) 

Table 11: Applications Selected by Technical Review Score, and Odds 
and Odds Ratios Derived from Them: 
		
Technical review score: 1 or 2; 
Selected: No: 36; 
Selected: Yes: 7; 
Percent selected: 16%; 
Odds on	selected: 0.19; 
Odds ratios: reference. 

Technical review score: 3; 
Selected: No: 49; 
Selected: Yes: 41; 
Percent selected: 46%; 
Odds on	selected: 0.84; 
Odds ratios: 4.30. 

Technical review score: 4; 
Selected: No: 35; 
Selected: Yes: 25; 
Percent selected: 42%; 
Odds on	selected: 0.71; 
Odds ratios: 3.67. 

Technical review score: 5; 
Selected: No: 4; 
Selected: Yes: 9; 
Percent selected: 69%; 
Odds on	selected: 2.25; 
Odds ratios: 11.57. 

Source: GAO analysis of FRA data. 

Note: The differences across score categories are significant given 
the low probability associated with the likelihood ratio chi-square 
statistic calculated to test the independence of scores and selection 
(L2 = 17.14 with 3 df, P < 0.01). 

This table includes track 1 a, 1 b, 2, 3, and 4 applications and, 
therefore, does not match the data provided in the body of this report. 

[End of table] 

We followed several steps to calculate the odds ratios of 4.30, 3.67, 
and 11.57. First, we derived the odds that applications with certain 
technical review scores would be selected for an award. For example, 
to determine the selection odds for applications receiving a score of 
1 or 2, we divided the number of applications receiving a score of 1 
or 2 that were selected by the number applications receiving those 
scores that were not selected. Seven were selected for awards, whereas 
36 were not; the resulting odds (7/36) equal 0.19. This means that 19 
applications receiving a score of 1 or 2 would be selected for an 
award for every 100 that were not. By comparison, the odds on being 
selected for applications receiving a technical review score of 3 were 
41/49, or 0.84, which indicates that 84 applications receiving a score 
of 3 would be selected for an award for every 100 that were not. The 
odds ratio comparing these two odds is 0.84/0.19 equal to 4.30. This 
odds ratio suggests that the odds of being selected for an award are 
more than four times greater for applications receiving a technical 
review score of 3 than for applications receiving a score of 1 or 2. 

Application Track Affected the Likelihood of Applications Being 
Selected for Award: 

We also found that there were sizable differences in the likelihood of 
applications submitted under different tracks being selected for an 
award. While only between one-quarter and one-third of the 
applications in tracks 1a and 1b were selected, 61 percent of track 2 
applications were selected, as were nearly three-fourths of the 
applications in tracks 3 and 4. These differences are also apparent 
from looking at the odds and odds ratios in table 12. The odds on 
being selected for funding were slightly lower (by a factor of 0.77) 
for track 1b applications than for track la, but they were more than 
three times greater for track 2 applications than for track 1a 
applications, and more than five times greater for track 3 and track 4 
applications than for track 1a applications. (See table 12.) 

Table 12: Applications Selected by Track, and Odds and Odds Ratios 
Derived from Them: 

Track: 1a; 
Selected: No: 68; 
Selected: Yes: 33; 
Percent selected: 33%; 
Odds on	selected: 0.49; 
Odds ratios: reference. 

Track: 1b; 
Selected: No: 40; 
Selected: Yes: 15; 
Percent selected: 27%; 
Odds on	selected: 0.38; 
Odds ratios: 0.77. 

Track: 2; 
Selected: No: 9; 
Selected: Yes: 14; 
Percent selected: 61%; 
Odds on	selected: 1.56; 
Odds ratios: 3.21. 

Track: 3 or 4; 
Selected: No: 7; 
Selected: Yes: 20; 
Percent selected: 74%; 
Odds on	selected: 2.86; 
Odds ratios: 5.89. 

Source: GAO analysis of FRA data. 

Note: The differences across tracks are statistically significant 
given the likelihood ratio chi-square statistic calculated to test the 
independence of tracks and selection (L2 = 23.17 with 3 df, P < 0.01). 

[End of table] 

Amount Requested Had an Effect on Likelihood of Selection: 

There were also sizable differences in the likelihood of applications 
being selected based on the amount requested. Slightly more than half 
of the applications that requested less than $1 million were funded, 
as were exactly half of the applications that requested $50 million or 
more. At the same time, roughly 40 percent of applications requesting 
between $1 million and $10 million were funded, and less than one-
fourth of the applications requesting $10 to $50 million were funded. 
The odds and odds ratios indicate that applications requesting the 
lowest amounts were the most likely to be selected and that 
applications requesting the highest amounts were almost as likely as 
those requesting the lowest amounts to be selected. Applications 
requesting more than $1 million but less than $50 million were 
somewhat less likely to be selected than applications requesting less 
than $1 million or more than $50 million. (See table 13.) 

Table 13: Applications Selected by Requested Amount, and Odds and Odds 
Ratios Derived from Them: 

Amount requested: $100,000-999,999; 
Selected: No: 18; 
Selected: Yes: 21; 
Percent selected: 54%; 
Odds on	selected: 1.17; 
Odds ratios: reference. 

Amount requested: $1,000,000-9,999,999; 
Selected: No: 44; 
Selected: Yes: 29; 
Percent selected: 40%; 
Odds on	selected: 0.66; 
Odds ratios: 0.56. 

Amount requested: $10,000,000-49,999,999; 
Selected: No: 43; 
Selected: Yes: 13; 
Percent selected: 23%; 
Odds on	selected: 0.30; 
Odds ratios: 0.26. 

Amount requested: $50,000,000 or more; 
Selected: No: 19; 
Selected: Yes: 19; 
Percent selected: 50%; 
Odds on	selected: 1.00; 
Odds ratios: 0.86. 

Source: GAO analysis of FRA data. 

Note: The differences across requested amount categories are 
statistically significant given the likelihood ratio chi-square 
statistic calculated to test the independence of amount requested and 
selection (L2 = 11.66 with 3 df, P < 0.01). 

[End of table] 

Number of Applications per State Had Sizable Effect on the Likelihood 
of Application Selection: 

We examined the differences of applications being selected by each 
state and, ultimately, by the number of applications submitted in many 
of the states. We first considered the number of applications selected 
for each of the 34 states that submitted eligible applications. Of 
these 34, 4 states submitted more than 10 applications, 22 states 
submitted 3 or fewer applications, and 9 states submitted a single 
application. We found a statistically significant relationship 
indicating there was a much greater tendency for applications to be 
selected when they came from states in which a maximum of three 
applications were submitted.[Footnote 44] 

Given the low number of applications submitted by many of the states, 
however, we could not control for all of the differences between 
states in a multivariate analysis in which the effects of the other 
variables are estimated simultaneously. Therefore, we combined the 
states with smaller numbers of applications into two groups: one group 
contained states which submitted one to three applications and the 
other group contained states submitting four to nine applications. 
These groupings did not result in the loss of any significant 
information with respect to differences in the likelihood of 
applications being selected across states.[Footnote 45] The results of 
these state groupings indicate that the percentage of applications 
selected for funding from states with one to three applications (70 
percent selected) were considerably higher than the percentage of 
funded applications from states with four to nine applications (40 
percent selected). In addition, those states that submitted more than 
nine applications showed considerable differences in the percent of 
applications selected. California (37 percent selected) and Missouri 
(75 percent selected) had a relatively high percentage of applications 
selected, and New York (18 percent selected) and Washington state (5 
percent selected) had a relatively low percentages of applications 
selected. 

As in the previous tables, the odds and odds ratios give us the same 
sense of the association that the percentages reveal; the odds on 
being selected for funding were more than three times higher for 
applications from states submitting one to three applications than for 
applications from California or from states submitting four to nine 
applications (2.33/0.68, which equals 3.43). In addition, New York and 
Washington state were much less likely to likely to have an 
application selected than California and Missouri. (See table 14.) 

Table 14: Applications Selected by State and State Group, and Odds and 
Odds Ratios Derived from Them: 

State and state group: States with 1 to 3 applications; 
Selected for funding: No: 12; 
Selected for funding: Yes: 28; 
Percent selected: 70%; 
Odds on selected: 2.33; 
Odds ratios: 3.94. 

State and state group: States with 4 to 9 applications; 
Selected for funding: No: 31; 
Selected for funding: Yes: 21; 
Percent selected: 40%; 
Odds on selected: 0.68; 
Odds ratios: 1.14. 

State and state group: Missouri; 
Selected for funding: No: 3; 
Selected for funding: Yes: 9; 
Percent selected: 75%; 
Odds on selected: 3.00; 
Odds ratios: 5.06. 

State and state group: Washington; 
Selected for funding: No: 20; 
Selected for funding: Yes: 1; 
Percent selected: 5%; 
Odds on selected: 0.05; 
Odds ratios: 0.08. 

State and state group: New York; 
Selected for funding: No: 31; 
Selected for funding: Yes: 7; 
Percent selected: 18%; 
Odds on selected: 0.23; 
Odds ratios: 0.38. 

State and state group: California; 
Selected for funding: No: 27; 
Selected for funding: Yes: 16; 
Percent selected: 37%; 
Odds on selected: 0.59; 
Odds ratios: reference. 

Source: GAO analysis of FRA data. 

Note: The differences across states and groups of states are 
statistically significant given the likelihood ratio chi-square 
statistic calculated to test the independence of states and state 
groups and selection (L2 = 43.32 with 5 df, P < 0.01). 

[End of table] 

Regression Analyses Show that Technical Review Score and Number of 
Applications Submitted Significantly Affected the Likelihood of 
Selection: 

We also examined the data using bivariate and multivariate logistic 
regression models to estimate the effects of these different variables 
on the likelihood of applications being selected for funding. The 
bivariate models examine the effects of the technical review score, 
track, amount requested, and state or group of states from which the 
application arose, one at a time. The odds ratios from these models 
are the same as those produced from the observed frequencies in the 
different two-way tables described above. From these models, however, 
we obtain specific tests of the significance of the differences 
between each variable category to determine more generally whether 
there are differences between any of the categories. In the bivariate 
regression model we find that many, but not all, of the odds ratios 
describing these differences are significant. (See table 15.) 

Table 15: Odds Ratios from Bivariate and Multivariate Logistic 
Regression Models by Technical Review Score, Track, Amount Requested, 
and State and State Group: 

Variable category: Technical review score: 1 or 2; 
Bivariate logistic regressions: reference; 
Multivariate logistic regressions: reference. 

Variable category: Technical review score: 3; 
Bivariate logistic regressions: 4.30[A]; 
Multivariate logistic regressions: 6.88[A]. 

Variable category: Technical review score: 4; 
Bivariate logistic regressions: 3.67[A]; 
Multivariate logistic regressions: 7.53[A]. 

Variable category: Technical review score: 5; 
Bivariate logistic regressions: 11.57[A]; 
Multivariate logistic regressions: 9.36[A]. 

Variable category: Track: 1a; 
Bivariate logistic regressions: reference; 
Multivariate logistic regressions: reference. 

Variable category: Track: 1b; 
Bivariate logistic regressions: 0.77; 
Multivariate logistic regressions: 0.60. 

Variable category: Track: 2; 
Bivariate logistic regressions: 3.21[A]; 
Multivariate logistic regressions: 2.34. 

Variable category: Track: 3 or 4; 
Bivariate logistic regressions: 5.89[A]; 
Multivariate logistic regressions: 1.93. 

Variable category: Amount requested: $100,000-999,999; 
Bivariate logistic regressions: reference; 
Multivariate logistic regressions: reference. 

Variable category: Amount requested: $1,000,000-9,999,999; 
Bivariate logistic regressions: 0.56; 
Multivariate logistic regressions: 1.35. 

Variable category: Amount requested: $10,000,000-49,999,999; 
Bivariate logistic regressions: 0.26[A]; 
Multivariate logistic regressions: 0.37. 

Variable category: Amount requested: $50,000,000 or more; 
Bivariate logistic regressions: 0.86; 
Multivariate logistic regressions: 0.79. 

Variable category: Applications submitted by state and state group: 
States with 1 to 3 applications; 
Bivariate logistic regressions: 3.94[A]; 
Multivariate logistic regressions: 2.94. 

Variable category: Applications submitted by state and state group: 
States with 4 to 9 applications; 
Bivariate logistic regressions: 1.14; 
Multivariate logistic regressions: 0.88. 

Variable category: Applications submitted by state and state group: 
Missouri; 
Bivariate logistic regressions: 5.06[A]; 
Multivariate logistic regressions: 9.12[A]. 

Variable category: Applications submitted by state and state group: 
Washington; 
Bivariate logistic regressions: 0.08[A]; 
Multivariate logistic regressions: 0.07[A]. 

Variable category: Applications submitted by state and state group: 
New York; 
Bivariate logistic regressions: 0.38; 
Multivariate logistic regressions: 0.37. 

Variable category: Applications submitted by state and state group: 
California; 
Bivariate logistic regressions: reference; 
Multivariate logistic regressions: reference. 

Source: GAO analysis of FRA data. 

[A] Ratios which are significant at P < 0.05. 

[End of table] 

The multivariate model estimates the net effects of these different 
variables on the likelihood of applications being selected for 
funding, or the effects of each variable when the effects of other 
variables are considered simultaneously, rather than one at a time. 
Our results indicate that the differences between tracks and amount 
requested categories are rendered insignificant when technical review 
scores and state and state group are taken into account, while the 
effect of technical review score and the differences between state and 
state group remain sizable and in most cases significant. 
Specifically, we found that, when we accounted for all four variables, 
applications receiving a technical review score of 3, 4, or 5 were 
about seven to eight times more likely to be selected for funding than 
applications which scored 1 or 2. In addition, applications from 
states that submitted one to three applications, and applications from 
Missouri, were three and nine times as likely, respectively, to be 
selected as those from California, while those from Washington state 
were less than one-tenth as likely as those from California to be 
selected. The remaining variable categories were not significant in 
the multivariate model and, therefore, do not provide a statistical 
explanation for why applications were more or less likely to be 
selected for an award. 

[End of section] 

Footnotes: 

[1] Pub. L. No. 111-5, 123 Stat. 115 (Feb. 17, 2009). 

[2] On December 9, 2010, the department redirected $1.195 billion in 
intercity passenger rail funds originally designated for Ohio and 
Wisconsin to 13 other states, which were selected for Recovery Act 
awards in January 2010. At the time of this announcement, our audit 
work was substantially complete and, therefore, we did not assess 
FRA's approach to making these funding decisions. 

[3] GAO, High Speed Rail: Learning From Start-ups, Prospects for 
Increased Industry Investment, and Federal Oversight Plans, 
[hyperlink, http://www.gao.gov/products/GAO-10-625] (Washington, D.C.: 
June 17, 2010). 

[4] On October 28, 2010, the department announced 54 additional awards 
totaling $2.4 billion. These awards will be funded through the 
department's annual appropriation for fiscal years 2009 and 2010, 
which remain available until expended. The department also requested 
$1 billion for intercity passenger rail in fiscal year 2011. 

[5] We did not assess whether the selected applications will achieve 
benefits and costs stated in the applications submitted to FRA. We 
have reported that applicants for major infrastructure projects, such 
as high speed rail projects, often overstate benefits, such as the 
number of likely riders. See GAO, High Speed Passenger Rail: Future 
Development Will Depend on Addressing Financial and Other Challenges 
and Establishing a Clear Federal Role, [hyperlink, 
http://www.gao.gov/products/GAO-09-317] (Washington, D.C.: Mar. 19, 
2009). 

[6] At this time FRA also used the same approach to assess 
applications for planning grants using up to $9.54 million in fiscal 
year 2008 and 2009 funds (called "track 3") and for final design and 
construction projects using at least $82.3 million of fiscal year 2008 
and 2009 funds (called "track 4"). Application for tracks 3 and 4 were 
assessed by the same technical review panels and at the same time as 
the track 1a and 1b applications. Unless otherwise noted, this report 
deals with only track 1a, 1b, and 2 applications. 

[7] Pub. L. No. 110-432, 122 Stat. 4848 (Oct. 16, 2008). 

[8] FRA, Preliminary National Rail Plan (Washington, D.C., October 
2009) and FRA, National Rail Plan-Moving Forward: A Progress Report 
(Washington, D.C., September 2010). 

[9] By comparison, the fiscal years 2008 and 2009 appropriations for 
the department included $30 million and $90 million, respectively, for 
intercity passenger rail grants to states. 

[10] An obligation is a commitment that creates a legal liability of 
the government for the payment of goods or services ordered or 
received. 

[11] Department of Transportation, Vision for High-Speed Rail in 
America (Washington, D.C., April 2009). 

[12] 1274 Fed. Reg. 29900 (June 23, 2009). 

[13] Only states, groups of states, interstate compacts, public 
agencies, and Amtrak were eligible to apply for funding. Amtrak did 
not independently submit any applications, but was included in a 
number of other applications as the anticipated service operator. 

[14] The number of applications and amount of requested funds 
submitted by applicants includes several duplicate projects. For 
example, Washington state submitted three track 2 applications in 
which 11 of the same projects were contained in each application. In 
tallying the number of applications and amounts requested, we did not 
double count duplicate applications. 

[15] Some of the criteria for the eligibility, technical, and 
selection reviews were derived from requirements in PRIIA. For 
example, PRIIA directed FRA to select projects that encourage 
intermodal connectivity, which is covered under the technical review 
criteria of transportation benefits. In addition, according to an FRA 
official, the technical review criteria were based on the Recovery Act 
and the department's general goals for transportation projects 
identified in the funding announcement (e.g. developing livable 
communities and encouraging environmental benefits). 

[16] Final application scores were derived from individual panelists' 
scores for each technical review criterion, which were weighted based 
on the track under which they were submitted. For example, track 1 
application scores were weighted to emphasize the reviewers' technical 
review scores for transportation benefits, economic recovery benefits, 
and project management approach, and place less weight on the scores 
for other public benefits. In contrast, track 2 application scores 
were weighted to emphasize the scores for transportation benefits and 
other public benefits. 

[17] Standardized scores, called z-scores, were applied only to tracks 
1a, 1b, 3, and 4 technical review scores as an internal control to 
ensure interrater reliability across the 12 review panels. 
Standardization was not required for track 2 projects because the 
technical review was conducted by a single panel. 

[18] For a list of applications and the extent to which they made it 
through eligibility determination, and selection see [hyperlink, 
http://www.dot.govirecovery/docs/hsiprapplist.pdf]. 

[19] There were three instances in which the criteria in the funding 
announcement did not completely align with the guidebooks. For 
example, the funding announcement includes a technical review 
criterion that projects create an integrated intercity passenger rail 
network, including allowance for and support of future network 
expansion. The guidebooks discuss the connection of the proposed 
project to other intercity rail services, but do not indicate that 
panelists should consider future network expansion. We viewed these 
instances as minor and, therefore, concluded that the funding 
announcement and technical review guidebooks generally align. 

[20] GAO, Standards for Internal Control in the Federal Government, 
[hyperlink, http://www.gao.gov/products/GAO/AIMD-00-21.3.1] 
(Washington, D.C.: November 1999). 

[21] U.S. Department of Transportation, Financial Assistance Guidance 
Manual (Washington, D.C., March 2009). 

[22] These amounts include track 1 and 2 awards, and may change as FRA 
finalizes awards. 

[23] On October 28, 2010, FRA announced that Florida was selected for 
an additional $808 million from the department's annual appropriations 
for the Tampa to Orlando high speed rail project. In addition, on 
December 9, 2010, FRA announced that it had redistributed an 
additional $342.3 million to Florida originally designated for Ohio 
and Wisconsin. On February 16, 2011, Florida announced that they will 
turn down $2.4 billion in funding awarded for the Tampa to Orlando 
high speed rail project. As of late February, FRA had not announced 
how this funding would be used. 

[24] On October 28, 2010, FRA announced that Washington state was 
selected for an additional $31 million from department's annual 
appropriations. In addition, on December 9, 2010, FRA announced that 
it had redistributed an additional $161.5 million to Washington state 
originally designated for Ohio and Wisconsin. We did not ask 
Washington state officials how these additional funds will affect 
their ability to complete the 16 proposed projects. 

[25] Officials also noted that they were adjusting award amounts and 
the scope of the projects as they negotiate cooperative agreements 
with each state, which will serve as documentation of the final award 
decisions. 

[26] The statistical tests we ran were bivariate and multivariate 
logistic regression models, which included the following factors: (1) 
technical review scores, (2) number of applications submitted per 
state, (3) application track, and (4) the amount of project funding 
requested. The third and fourth factors were not statistically 
significant and, therefore, do not provide a statistical explanation 
for why projects were more or less likely to be selected for an award. 
For more information on our methodology and additional analyses see 
appendix IV. 

[27] The federal agency guidance we examined came from the Departments 
of Commerce, Education, Labor, and Transportation. For more 
information on our methodology for developing recommended practices 
for this review see appendix II. 

[28] Race to the Top, which is part of the Recovery Act's State Fiscal 
Stabilization Fund, is a competitive grant program which provides 
funds to states to encourage educational reform that will result in 
improved academic performance. 

[29] The Recovery Act contains a number of provisions related to 
transparency, notably the requirement that recipients of these funds 
report quarterly on a number of things, such as the purpose and 
expected outcomes of their awards and on jobs created. These reports 
are available on the administration's Web site at [hyperlink, 
http://www.recovery.gov]. See GAO, Recovery Act: Increasing the 
Public's Understanding of What Funds Are Being Spent on and What 
Outcomes Are Expected, [hyperlink, 
http://www.gao.gov/products/GAO-10-581] (Washington, D.C.: May 27, 
2010). 

[30] For a list of programs we examined and more information on our 
methodology for selecting these programs see appendix II. 

[31] [hyperlink, http://www.gao.gov/products/GA0-10-625]. 

[32] FRA identified provisional amounts of funding to applicants, 
subject to negotiation. Further, FRA did not specify funding amounts 
for each project in its notification letters to applicants, providing 
the flexibility for states to allocate funding across FRA-selected 
applications. As a result, we were unable to determine the amount of 
funding awarded for each category of high speed rail applications. 

[33] Given that FRA was provided discretion to determine the number of 
high speed rail applications to select during its funding competition, 
we did not evaluate whether it selected an appropriate number of 
projects for funding. 

[34] In contrast, several other capital grants and investment programs 
have relied on existing program structures, such as the department's 
Highway Infrastructure Investment and Transit Capital Assistance 
programs, and only authorized agencies to obligate funds through 
September 30, 2010. The Highway Infrastructure Investment program, 
administered by the Federal Highway Administration provides funding to 
states for restoration, repair, and construction of highways among 
other things. The Transit Capital Assistance program, administered by 
the Federal Transit Administration, provides grants for facility 
renovation or construction, vehicle replacements, preventive 
maintenance, and other related activities. 

[35] The states were Arizona, California, Florida, Iowa, Kansas, 
Louisiana, New York, Oregon, Pennsylvania, and South Carolina. 

[36] We selected Ohio and Washington because FRA proposed to provide 
these states with $164 million and $386 million, respectively, less 
than requested. 

[37] To identify these practices, we reviewed prior work on 
discretionary grants to compile an initial list of grants manuals from 
a number of federal agencies. We then verified and added to this 
listing through a separate search and review of government agency Web 
sites. In addition, we consulted with GAO staff who have expertise on 
federal discretionary grant practices. These practices reflect our 
review of a judgmental sample of discretionary grant guidance and may 
not include all recommended practices used by federal agencies.	 

[38] We typically reviewed information on each program's agency Web 
site as part of our review of these three sources. 

[39] Ohio is slated to receive $15 million for preliminary engineering 
and environmental analysis work conducted on its track 2 project. 
According to FRA, $30 million remains obligated to Wisconsin for costs 
incurred on its track 2 project, and the state is scheduled to receive 
$14 million for the two previously selected track 1a projects. 

[40] The technical review scores were averaged and weighted based on 
the priorities listed in the funding announcement. 

[41] The inclusion of these additional data points slightly affected 
the final technical review scores for other applications because FRA 
used the average score across applications when it used a 
standardization procedure, called a z-score, to correct for potential 
differences in the ways applications were scored across track 1, 3, 
and 4 panels. 

[42] The bivariate models estimate differences in the odds on 
selection across groups or categories of applications when the other 
variables are ignored. The multivariate models estimate differences in 
the odds on selection across categories of each variable when the 
other variables are taken into account, or controlled statistically. 

[43] For example, a category within the technical review score 
variable might include applications receiving a score of 3, as in 
table 11. 

[44] We have omitted this expanded table to save space. The 
differences across the 34 states that submitted an application are 
statistically significant given the likelihood ratio chi-square 
statistic calculated to test the independence of the state from which 
the application was submitted and the likelihood of selection (L2 = 
67.40 with 33 df, P < .01). 

[45] When we grouped the states submitting lower number of 
applications, two-thirds of the variability from the individual state 
analysis was retained, and the variability lost as a result of the 
grouping was statistically insignificant. 

[End of section] 

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 GAO Reports and Testimony: 

The fastest and easiest way to obtain copies of GAO documents at no 
cost is through GAO's Web site [hyperlink, http://www.gao.gov]. Each 
weekday, GAO posts newly released reports, testimony, and 
correspondence on its Web site. To have GAO e-mail you a list of newly 
posted products every afternoon, go to [hyperlink, http://www.gao.gov] 
and select "E-mail Updates." 

Order by Phone: 

The price of each GAO publication reflects GAO’s actual cost of
production and distribution and depends on the number of pages in the
publication and whether the publication is printed in color or black and
white. Pricing and ordering information is posted on GAO’s Web site, 
[hyperlink, http://www.gao.gov/ordering.htm]. 

Place orders by calling (202) 512-6000, toll free (866) 801-7077, or
TDD (202) 512-2537. 

Orders may be paid for using American Express, Discover Card,
MasterCard, Visa, check, or money order. Call for additional 
information. 

To Report Fraud, Waste, and Abuse in Federal Programs: 

Contact: 

Web site: [hyperlink, http://www.gao.gov/fraudnet/fraudnet.htm]: 
E-mail: fraudnet@gao.gov: 
Automated answering system: (800) 424-5454 or (202) 512-7470: 

Congressional Relations: 

Ralph Dawn, Managing Director, dawnr@gao.gov: 
(202) 512-4400: 
U.S. Government Accountability Office: 
441 G Street NW, Room 7125: 
Washington, D.C. 20548: 

Public Affairs: 

Chuck Young, Managing Director, youngc1@gao.gov: 
(202) 512-4800: 
U.S. Government Accountability Office: 
441 G Street NW, Room 7149: 
Washington, D.C. 20548: