This is the accessible text file for GAO report number GAO-11-290 
entitled 'Intercity Passenger and Freight Rail: Better Data and 
Communication of Uncertainties Can Help Decision Makers Understand 
Benefits and Trade-offs of Programs and Policies' which was released 
on March 29, 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 Congressional Committees: 

February 2011: 

Intercity Passenger and Freight Rail: 

Better Data and Communication of Uncertainties Can Help Decision 
Makers Understand Benefits and Trade-offs of Programs and Policies: 

GAO-11-290: 

GAO Highlights: 

Highlights of GAO-11-290, a report to congressional committees. 

Why GAO Did This Study: 

Concerns about the weak economy, congestion in the transportation 
system, and the potentially harmful effects of air emissions generated 
by the transportation sector have raised awareness of the potential 
benefits and costs of intercity passenger and freight rail relative to 
other transportation modes such as highways. GAO was asked to review 
(1) the extent to which transportation policy tools that provide 
incentives to shift passenger and freight traffic to rail may generate 
emissions, congestion, and economic development benefits and (2) how 
project benefits and costs are assessed for investment in intercity 
passenger and freight rail and how the strengths and limitations of 
these assessments impact federal decision making. GAO reviewed 
studies; interviewed federal, state, local, and other stakeholders 
regarding methods to assess benefit and cost information; assessed 
information on project benefits and costs included in rail grant 
applications; and conducted case studies of selected policies and 
programs in the United Kingdom and Germany to learn more about their 
policies designed to provide incentives to shift traffic to rail. 

What GAO Found: 

Although implementing policies designed to shift traffic to rail from 
other modes may generate benefits, and selected European countries’ 
experiences suggest that some benefits can be achieved through these 
types of policies; many factors will affect whether traffic shifts. 
The extent to which rail can generate sufficient demand to draw 
traffic from other modes to achieve the desired level of net benefits 
will depend on numerous factors. Some passenger or freight traffic may 
not be substitutable or practical to move by a different mode. For 
example, certain freight shipments may be time-sensitive and thus 
cannot go by rail. Another key factor will be the extent to which 
sufficient capacity exists or is being planned to accommodate shifts 
in traffic from other modes. How transport markets respond to a given 
policy—-such as one that changes the relative price of road transport—-
will also affect the level of benefits generated by that policy. 
Experiences in selected countries suggest that varying amounts of mode 
shift and some benefits were attained where decision makers 
implemented policies to move traffic from other modes to rail. For 
example, a road freight pricing policy in Germany resulted in 
environmental and efficiency improvements, and freight rail grants in 
the United Kingdom led to congestion relief at the country’s largest 
port. Pursuing policies to encourage traffic to shift to rail is one 
potential way to generate benefits, and other policies may be 
implemented to generate specific benefits at a lower cost. 

Information on the benefits and costs of intercity passenger and 
freight rail is assessed to varying degrees by those seeking federal 
funding for investment in rail projects; however, data limitations and 
other factors reduce the usefulness of such assessments for federal 
decision makers. Applicants to two discretionary federal grant 
programs—the Transportation Investment Generating Economic Recovery 
program and the High-Speed Intercity Passenger Rail program—provided 
assessments of potential project benefits and costs that were 
generally not comprehensive. For instance, applications varied widely 
in the extent to which they quantified and monetized some categories 
of benefits. In addition, GAO’s assessment of selected applications 
found that most applicants did not provide key information recommended 
in federal guidance for such assessments, including information 
related to uncertainty in projections, data limitations, or the 
assumptions underlying their models. Applicants, industry experts, and 
Department of Transportation (DOT) officials GAO spoke with reported 
that many challenges impacted their ability to produce useful 
assessments of project benefits and costs, including: short time 
frames in which to prepare the assessments, limited resources and 
expertise for performing assessments, poor data quality, lack of 
access to data, and lack of standard values for monetizing some 
benefits. As a result, while information on project benefits and costs 
was considered as one of many factors in the decision-making process, 
according to DOT officials, the varying quality and focus of 
assessments resulted in additional work, and the information provided 
was of limited usefulness to DOT decision makers. 

What GAO Recommends: 

GAO recommends DOT conduct a data needs assessment to improve the 
effectiveness of modeling and analysis for rail and provide consistent 
requirements for assessing rail project benefits and costs. DOT, 
Amtrak and EPA provided technical comments, and DOT agreed to consider 
the recommendations. 

View [hyperlink, http://www.gao.gov/products/GAO-11-290] or key 
components. For more information, contact Susan Fleming at (202) 512-
4431 or flemings@gao.gov. 

[End of section] 

Contents: 

Letter: 

Background: 

Shifting Traffic to Rail from Other Modes May Generate Benefits, but 
Many Factors Will Affect Whether Traffic Shifts, and Policies Abroad 
Have Produced Mixed Results: 

Grant Applicants' Assessments of Project Benefits and Costs Are of 
Varying Quality and Usefulness to Decision Makers: 

Conclusions: 

Recommendations for Executive Action: 

Agency Comments and Our Evaluation: 

Appendix I: Objectives, Scope, and Methodology: 

Appendix II: International Case Study Summaries: The United Kingdom 
and Germany: 

Appendix III: HSIPR and TIGER Discretionary Grant Program Information: 

Appendix IV: Computer Simulations of Freight Diversion from Truck to 
Rail: 

Appendix V: GAO Contact and Staff Acknowledgments: 

Tables: 

Table 1: Key Elements for Benefit-Cost Analysis from Presidential 
Exec. Order No. 12893 and OMB Circulars Nos. A-94 and A-4: 

Table 2: Federal Discretionary Transportation Grant Program 
Requirements for Assessments of Project Benefits and Costs: 

Table 3: Interviews: 

Table 4: High-Speed Intercity Passenger Rail Program Funding Tracks: 

Table 5: Extent of Data and Assumptions Underlying Intermodal 
Transportation Inventory Cost Model Inputs and Calculations: 

Table 6: Inputs to the ITIC Model: 

Table 7: ITIC Calculations: 

Figures: 

Figure 1: 2008 Estimates of U.S. Greenhouse Gas Emissions by 
Transportation Mode: 

Figure 2: Competition among Passenger Transportation Modes: 

Figure 3: Competition among Freight Transportation Modes: 

Figure 4: DOT Assessment of Usefulness of Benefit-Cost Analyses from 
Forwarded Rail-Related TIGER Applications: 

Figure 5: Selected Polices to Benefit Intercity Passenger and Freight 
Rail in the United Kingdom: 

Figure 6: Selected Polices to Benefit Intercity Passenger and Freight 
Rail in Germany: 

Figure 7: Impact of Increased Per-Mile Truck Rates on Vehicle Miles 
Traveled (VMT) by Trucks under Two Scenarios: 

Figure 8: Intermodal Transportation and Inventory Cost (ITIC) Model 
Process: 

Abbreviations: 

AAR: Association of American Railroads: 

BTS: Bureau of Transportation Statistics: 

DOT: Department of Transportation: 

EPA: Environmental Protection Agency: 

FAA: Federal Aviation Administration: 

FAF: Freight Analysis Framework: 

FHWA: Federal Highway Administration: 

FRA: Federal Railroad Administration: 

HGV: Heavy Goods Vehicle: 

HSIPR: High-Speed Intercity Passenger Rail Program: 

ITIC: Intermodal Transportation and Inventory Cost Model: 

NHTSA: National Highway Traffic Safety Administration: 

OMB: Office of Management and Budget: 

PRIIA: Passenger Rail Investment and Improvement Act of 2008: 

Recovery Act: American Recovery and Reinvestment Act of 2009: 

RRIF: Railroad Rehabilitation and Improvement Financing: 

SAFETEA-LU: Safe, Accountable, Flexible, Efficient Transportation 
Equity Act: A Legacy for Users: 

TIFIA: Transportation Infrastructure Finance and Innovation Act: 

TIGER: Transportation Investment Generating Economic Recovery: 

VMT: vehicle miles traveled: 

[End of section] 

United States Government Accountability Office: 
Washington, DC 20548: 

February 24, 2011: 

The Honorable John D. Rockefeller IV: 
Chairman: 
The Honorable Kay Bailey Hutchison: 
Ranking Member: 
Committee on Commerce, Science, and Transportation: 
United States Senate: 

The Honorable Frank R. Lautenberg: 
Chairman: 
The Honorable John Thune: 
Ranking Member: 
Subcommittee on Surface Transportation and Merchant Marine 
Infrastructure, Safety and Security: 
Committee on Commerce, Science, and Transportation: 
United States Senate: 

Concerns about the weak economy, congestion in the transportation 
system, and the potentially harmful effects of greenhouse gases and 
airborne pollutants from transportation have raised awareness of the 
potential benefits and costs of intercity passenger and freight rail 
relative to other transportation modes. The U.S. economy and its 
competitive position in the global economy depend in part on the 
nation's transportation networks working efficiently. In addition, 
factors such as cost and time can impact passengers' and shippers' 
demand for a particular transportation mode. Congestion delays that 
significantly constrain both passenger and freight mobility can result 
in increased economic costs to passengers, shippers and also to the 
nation. According to the Texas Transportation Institute, in 2009 the 
yearly peak-period delay per auto commuter was 34 hours, with a total 
cost of $115 billion.[Footnote 1] Continued development and efficient 
management of the nation's transportation system is essential to 
accommodate the anticipated future growth of the nation's passenger 
and freight mobility demands. For example, the Department of 
Transportation (DOT) forecasts that between 2010 and 2035 the freight 
transportation system will experience a 22 percent increase in total 
freight tonnage moved in the United States, from 12.5 billion to 15.3 
billion tons.[Footnote 2] In addition, the transportation industry 
continues to be one of the biggest energy users and contributors to 
greenhouse gas emissions. According to the Environmental Protection 
Agency (EPA), for 2008 the transportation sector accounts for 27 
percent of the nation's greenhouse gas emissions.[Footnote 3] Because 
shifting intercity passenger and freight traffic to rail from other 
more energy-intensive modes is seen as a potential option to address 
some of these concerns, there is a growing interest in investing in 
and enhancing rail capacity and implementing policies that will 
encourage more traffic by rail.[Footnote 4] 

The Passenger Rail Investment and Improvement Act (PRIIA), enacted in 
October 2008,[Footnote 5] authorized over $3.7 billion for three 
different federal programs for high-speed rail,[Footnote 6] intercity 
passenger rail congestion,[Footnote 7] and intercity passenger rail 
service corridor capital grants.[Footnote 8] The American Recovery and 
Reinvestment Act of 2009 (Recovery Act), enacted in February 2009, 
appropriated $8 billion for the three PRIIA-established intercity 
passenger rail programs. In addition, the Recovery Act authorized new 
discretionary grants under the Transportation Investment Generating 
Economic Recovery (TIGER) program.[Footnote 9] 

PRIIA and the Recovery Act created new responsibilities for Federal 
Railroad Administration (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, that 
outlined the requirements and procedures for obtaining federal 
funds.[Footnote 10] Moreover, in 2010 DOT awarded over $2 billion in 
TIGER and $10 billion in HSIPR grants. Both programs required 
applications to include information on the costs and benefits of 
proposed projects, including information on such benefits as reducing 
environmental impacts and congestion and encouraging economic 
development. 

One of the many considerations that can help inform transportation 
decision making is determining which investment or set of policies 
will yield the greatest net benefit (that is, benefits minus costs). 
While there is some debate around the extent to which investment in 
rail or policies that encourage shifting traffic to rail from other 
modes can help address problems, such as congestion and greenhouse gas 
emissions, there are a variety of analytical approaches, such as 
benefit-cost analysis and others, that may be employed to help 
evaluate proposed transportation investments. Tools such as these can 
provide decision makers with information on the benefits and costs of 
alternative investments and policy choices needed to make informed 
decisions. Given your interest in the potential net benefits of 
intercity passenger and freight rail policies and programs, we 
examined (1) the extent to which transportation policy tools that 
provide incentives to shift passenger and freight traffic to rail may 
generate emissions, congestion, and economic development benefits and 
(2) how project benefits and costs are assessed for investment in 
intercity passenger and freight rail and how the strengths and 
limitations of these assessments impact federal decision making. 

To address our objectives, we reviewed our prior work on rail and 
transportation investment decision making and documentation from an 
array of sources, as well as interviewing officials and various 
stakeholders regarding methods to assess the benefits and costs of 
transportation investments. Our interviews included discussions with 
officials from DOT, EPA, and the National Railroad Passenger 
Corporation (Amtrak); representatives from transportation coalitions 
and associations, metropolitan planning organizations, and state DOTs; 
and other transportation stakeholders. We also reviewed and assessed 
information on potential project benefits and costs included in 40 
rail-related applications submitted to the HSIPR and TIGER grant 
programs--20 from each program. We selected a random sample of 
applications that was weighted to ensure approximately proportional 
representation of the range of applications submitted to each program. 
Two GAO analysts independently reviewed each selected application 
based on Office of Management and Budget (OMB) guidelines on benefit-
cost analysis,[Footnote 11] with input from GAO economists and 
methodologists. We conducted an extensive literature search to 
identify studies analyzing potential mode shift and the impact of mode 
shift on selected benefits for intercity passenger or freight rail 
projects and policies. We used the studies and information we reviewed 
to inform our work and relied on multiple sources of additional 
information, including testimonial evidence, interviews, and case 
studies. We conducted case studies of selected policies and programs 
designed to provide incentives to shift passenger and freight traffic 
from other modes to rail in the United Kingdom and Germany to learn 
more about their experiences with efforts to shift traffic to rail in 
order to generate benefits. These two countries were chosen based on a 
number of criteria, including their experience in implementing such 
policies. While European intercity passenger and freight rail systems 
are very different in size, structure, and scope than the U.S. rail 
system, the experiences of countries such as the United Kingdom and 
Germany provide illustrative examples of other countries' experiences 
with policy tools that provide incentives to shift traffic to rail. 
[Footnote 12] Finally, we conducted our own computer simulation of 
transportation scenarios on mode choice for freight shipments. See 
appendix IV for a discussion of the simulation and appendix I for a 
detailed discussion of our scope and methodology. 

We conducted this performance audit from December 2009 to February 
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 the evidence obtained provides a reasonable basis for our 
findings and conclusions based on our audit objectives. 

Background: 

Passenger and freight rail are part of a complex national 
transportation system for transporting people and goods. Currently, 
there are seven Class I railroads and over 500 short line and regional 
railroads operating in the United States.[Footnote 13] These railroads 
operate the nation's freight rail system and own the majority of rail 
infrastructure in the United States. Railroads are the primary mode of 
transportation for many products, especially for such bulk commodities 
as coal and grain. In addition, railroads are carrying increasing 
levels of intermodal freight (e.g., containers and trailers), which 
travel on multiple modes and typically require faster delivery than 
bulk commodities. According to the Association of American Railroads 
(AAR), based on ton-miles, freight railroads carried about 43 percent 
of domestic intercity freight volume in 2009. In addition, according 
to DOT, the amount of freight rail is expected to continue to grow 
with a projected increase of nearly 22 percent by 2035. Intercity 
passenger rail service is primarily provided by Amtrak. Amtrak 
operates a 21,000-mile network, which provides service to 46 states 
and Washington, D.C., primarily over tracks owned by freight 
railroads. Federal law requires that freight railroads give Amtrak 
trains preference over freight transportation and, in general, charge 
Amtrak the incremental cost--rather than an apportioned cost--
associated with the use of their tracks.[Footnote 14] Amtrak also owns 
about 650 route miles of track, primarily on the Northeast Corridor, 
which runs between Boston, Massachusetts, and Washington, D.C. 

Transportation may impose a variety of "external" costs that can 
result in impacts such as health and environmental damage caused by 
pollution.[Footnote 15] For example, in choosing to drive to work, a 
commuter may not take into account the car emissions' contribution to 
local pollution, which may damage property or the health of others. 
Following are some negative effects of transportation: 

* Greenhouse gas emissions, nitrogen oxide (NOX) and fine particulate 
matter, and other pollutants: Based on estimated data from the EPA, 
from 1990 through 2008, transportation greenhouse gas emissions 
increased 22 percent. Carbon dioxide (CO2) is the primary greenhouse 
gas associated with the combustion of diesel (and other fossil fuels) 
and accounted for over 95.5 percent of the transportation sector's 
greenhouse gas emissions.[Footnote 16] Based on 2008 data from the 
EPA, cars, light trucks, and freight trucks together contributed over 
80 percent of the transportation sector greenhouse gas emissions (see 
figure 1).[Footnote 17] 

Figure 1: 2008 Estimates of U.S. Greenhouse Gas Emissions by 
Transportation Mode: 

[Refer to PDF for image: pie-chart] 

Light duty vehicles: 62.6%; 
Freight trucks: 21.2%; 
Aircraft: 8.3%; 
Rail: 2.7%; 
Marine: 2.0%; 
Other: 3.1%. 

Source: GAO analysis of EPA data. 

Note: "Light duty vehicles" includes passenger cars and light duty 
trucks. "Freight trucks" includes heavy and medium trucks. "Marine" 
includes ships and boats. For Marine, the source indicates that 
emission estimates reflect data collection problems. "Other" includes 
motorcycles, lubricants, buses, and pipelines. 

[End of figure] 

In addition, NOX and fine particulate matter with a diameter of 2.5 
microns or less (PM2.5) contribute to air pollution. Both of these 
pollutants are emitted through high temperature combustion and 
activities such as burning fossil fuels. For 2002, based on our 
analysis of EPA data, it was estimated that trucks emitted 3.02 tons 
of NOX and .12 tons of PM2.5 per million ton-miles.[Footnote 18] 

* Congestion: While congestion is geographically concentrated in 
metropolitan areas, international trade gateways, and on some 
intercity trade routes, congestion is a serious problem, contributing 
to longer and more unpredictable transit times and resulting in 
increased transportation costs. The Texas Transportation Institute 
estimates that for 439 domestic urban areas, congestion costs in 2009 
alone were $115 billion and accounted for a total of 3.9 billion 
gallons of gasoline consumption. For freight, congestion delays that 
significantly constrain freight mobility could result in increased 
economic costs for the nation. The Federal Highway Administration 
(FHWA) has calculated that delays caused by highway bottlenecks cost 
the trucking industry alone more than $8 billion a year. Similarly, we 
have previously reported on the significant level of congestion that 
exists, and is expected to grow, at airports in large urban areas 
throughout the country. The Federal Aviation Administration (FAA) 
predicts that, by 2025, the number of airline passengers will increase 
57 percent--from about 700 million to about 1.1 billion per year--and 
the number of daily flights will increase from about 80,000 to more 
than 95,000. Today's air transportation system will be strained to 
meet this growth in air traffic.[Footnote 19] 

* Accidents: Each year, there are tens of thousands of truck and 
vehicle accidents that result in injury or fatality. Based on National 
Highway Traffic Safety Administration (NHTSA) data, there were 33,808 
fatal motor vehicle crashes in the United States in 2009. This 
resulted in a national motor vehicle death rate of 1.13 deaths per 100 
million vehicle miles traveled (VMT). For freight, preliminary data 
from DOT for 2009 shows the rate of fatalities involving large trucks 
and buses was 0.121 per 100 million VMT. A portion of motor vehicle 
crash costs are not covered by private insurance. According to a 2000 
NHTSA report, approximately $21 billion, or 9 percent of all costs are 
borne by public sources. Similarly, we estimated truck external 
accident costs of $8,000 per million ton-miles that are not passed on 
to consumers.[Footnote 20] 

Investment and Policy Tools to Attain Benefits through Rail: 

While there are multiple approaches to address externalities in 
transportation, policies that provide incentives to shift traffic to 
rail can be appealing because they offer an option to address multiple 
externalities simultaneously by changing behavior to favor rail over 
other modes. For example, market-based policies that change the 
relative prices of the modes are likely going to be the most cost- 
effective. Policies such as increasing fuel taxes, imposing new fees 
such as a vehicle mile travel fee or a congestion charge, investing in 
increased capacity in one mode, or subsidizing travel in one mode can 
provide incentives to users to switch travel from one mode to another, 
and can achieve both a reduction in greenhouse gas emissions and 
alleviate congestion.[Footnote 21] Some stakeholders also believe that 
investing in rail may help to stimulate economic development. In order 
to obtain similar benefits without the goal of shifting traffic to 
rail, it might be necessary to introduce a suite of policies, each 
more directly targeted at a specific externality.[Footnote 22] For 
example, a congestion pricing policy may reduce traffic during peak 
travel times, but if it shifts traffic to nonpeak times, it may have a 
limited impact on overall emissions. Conversely, providing incentives 
to purchase more fuel efficient truck engines may do nothing to 
improve congestion or economic development. 

With respect to direct investment, the federal government typically 
has not provided extensive funding for freight rail or for intercity 
passenger rail outside of the Northeast Corridor between Boston, 
Massachusetts and Washington, D.C. In addition, according to Amtrak 
officials, funding has not been predictable, consistent, or sustained. 
However, recent legislation has increased the federal role and funding 
available for investment in intercity passenger and freight rail 
infrastructure. In 2008, PRIIA authorized the HSIPR program.[Footnote 
23] The program is administered through DOT's FRA, which has 
responsibility for planning, awarding, and overseeing the use of 
federal funds for the development of high-speed and intercity 
passenger rail.[Footnote 24] As of 2010, over $10 billion had been 
awarded through the HSIPR program to fund high-speed rail projects. 
[Footnote 25] Moreover, through the Recovery Act, Congress authorized 
the TIGER Discretionary Grant Program for investment in a variety of 
transportation areas, including freight and passenger rail.[Footnote 
26] In 2010, DOT awarded over $2 billion in TIGER funding. The TIGER 
program was designed to preserve and create jobs and to promote 
economic recovery and investment in transportation infrastructure that 
will provide long-term economic benefits and assist those most 
affected by the current economic downturn. The TIGER grants are 
multimodal, and criteria were developed for a framework to assess 
projects across various modes. For more information on the HSIPR and 
TIGER programs, see appendix III.[Footnote 27] 

Assessment of Benefits and Costs in Decision Making: 

Decision makers may consider a number of factors in deciding between 
various alternative investments or policies. These factors may include 
the objective or goal of the proposed actions--for example, preserving 
and creating jobs or promoting economic recovery or reducing an 
environmental externality. Other factors, such as the benefits and 
costs of alternatives, are also important to consider in decision 
making. Some benefits are associated with reducing an externality and 
are part of the assessment of whether policy alternatives for 
addressing the externality can be justified on economic principles. 
[Footnote 28] Costs should also be accounted for when considering 
various investment or policy alternatives. For example, there are 
direct costs, such as construction, maintenance, and operations, and 
less obvious types of costs, such as delays and pollution generated 
during construction. 

There are tools that can be employed in evaluating proposed 
transportation alternatives, including benefit-cost analysis[Footnote 
29] and economic impact analysis. Benefit-cost analysis is designed to 
identify the alternative with the greatest net benefit by comparing 
the monetary value of benefits and costs of each alternative with a 
baseline. Benefit-cost analysis provides for a comparison of 
alternatives based on economic efficiency, that is, which investment 
or policy would provide the greatest net benefit (i.e., greater 
benefits than costs). As we have previously reported, while benefit-
cost analysis may not be the most important decision-making factor--
but rather, one of many tools that decision makers may use to 
organize, evaluate, and determine trade-offs of various alternatives--
the increased use of systematic analytical tools such as benefit-cost 
analysis can provide important additional information that can lead to 
better informed transportation decision making.[Footnote 30] Economic 
impact analysis is a tool for assessing how the benefits and costs of 
transportation alternatives would be distributed throughout the 
economy and for identifying groups in society (for example, by region, 
income, or race) that are likely to gain from, or bear the costs of, a 
policy. 

The use of benefit-cost analysis information is not consistent across 
modes or types of programs that provide funding to transportation 
projects. Competitive programs such as TIGER and HSIPR and loan 
guarantee programs such as TIFIA and RRIF require information on 
benefits and costs.[Footnote 31] Formula programs (such as the Federal-
Aid Highway Program)[Footnote 32] do not necessarily require benefit- 
cost information. Federal guidance exists for conducting benefit-cost 
analyses, including OMB Circular No. A-94, OMB Circular No. A-4, and 
Executive Order No. 12893. The directive and related OMB guidance 
outline a number of key elements that should be included in the 
assessment of benefits and costs in decision making, as described in 
table 1. 

Table 1: Key Elements for Benefit-Cost Analysis from Presidential 
Exec. Order No. 12893 and OMB Circulars Nos. A-94 and A-4: 

Comparison to base case and alternatives: 
* Establish a base case for comparison; 
* Identify alternative projects--benefits and costs should be defined 
in comparison with a clearly stated alternative. 

Analysis of benefits and costs: 
* Define a time frame for analysis; 
* Quantify and monetize impacts as benefits and costs to the maximum 
extent possible, but consider qualitative measures reflecting values 
that are not readily quantified; 
* Measure and discount benefits and costs over the full life cycle of 
the project and identify the year in which dollars are presented. 

Transparency of information and treatment of uncertainty: 
* Clearly state all assumptions underlying the analysis of benefits 
and costs; 
* Assess the sensitivity of the analysis to changes in assumptions and 
forecasted inputs and recognize uncertainty through appropriate 
quantitative and qualitative assessments. 

Sources: GAO analysis of Presidential Exec. Order No. 12893 and OMB 
Circulars Nos. A-94 and A-4. 

[End of table] 

Specifically Executive Order No. 12893 and OMB Circulars Nos. A-94 and 
A-4 indicate that benefit and cost information shall be used in 
decision making, and the level of uncertainty in estimates of benefits 
and costs shall be disclosed.[Footnote 33] Other aspects of the 
benefit-cost analysis should be completed to the extent possible. For 
example, while the guidance suggests that impacts should be quantified 
and monetized, to the extent that this is not possible, qualitative 
assessments should be provided for those impacts that are not readily 
quantifiable. As we have previously reported in our work on transit 
investments, qualitative information can help ensure that project 
impacts that cannot be easily quantified are considered in decision 
making.[Footnote 34] 

Both the HSIPR and TIGER grant programs required applicants to provide 
information on proposed project benefits and costs. The type of 
information required, however, differed between the two programs and, 
for the TIGER program, depended on the level of federal funding 
sought, as described in table 2. In addition, while requirements for 
assessment of project benefits and costs were more specific for TIGER 
than for the HSIPR program, officials for both programs considered 
whether project benefits were likely to exceed project costs as part 
of their respective application assessment processes. 

Table 2: Federal Discretionary Transportation Grant Program 
Requirements for Assessments of Project Benefits and Costs: 

Grant program: TIGER; 
Size of grant sought: Less than $20 million; 
Program requirements for assessment of benefits and costs: Information 
on project benefits and costs not required; 
Benefit-cost guidance referred to in the Federal Register: Exec. Order 
No.12893 and OMB Circular Nos. A-94 and A-4. 

Grant program: TIGER; 
Size of grant sought: Between $20 million and $100 million; 
Program requirements for assessment of benefits and costs: Required to 
include estimates of the projects' expected benefits in five long-term 
outcome categories: (1) state of good repair, (2) economic 
competitiveness, (3) livability, (4) sustainability, and (5) safety[A]; 
Benefit-cost guidance referred to in the Federal Register: Exec. Order 
No.12893 and OMB Circular Nos. A-94 and A-4. 

Grant program: TIGER; 
Size of grant sought: More than $100 million; 
Program requirements for assessment of benefits and costs: Required to 
provide a "well developed" analysis of expected benefits and costs, 
including calculation of net benefits and a description of input data 
and methodological standards used for the analysis; 
Benefit-cost guidance referred to in the Federal Register: Exec. Order 
No.12893 and OMB Circular Nos. A-94 and A-4. 

Grant program: HSIPR; 
Size of grant sought: Any amount; 
Program requirements for assessment of benefits and costs: Required to 
provide information on public return on investment in three 
categories: (1) transportation benefits, (2) economic recovery 
benefits, and (3) public benefits, which include energy independence 
and efficiency, environmental quality, and livable communities. 
Applications to HSIPR were divided into four groups, each of which 
required assessments of public return on investment in these 
categories. However, importance of benefit categories varied across 
these groups (see appendix III); 
Benefit-cost guidance referred to in the Federal Register: Exec. Order 
No.12893. 

Source: GAO analysis of TIGER and HSIPR Federal Register notices. 

Note: For TIGER II--the second round of TIGER funding that DOT awarded 
in October 2010--DOT required benefit-cost analyses from all 
applicants, regardless of the amount of funding requested. 

[A] According to DOT, projects that contribute to a state of good 
repair by improving the condition of existing facilities and 
transportation systems; projects that meet economic competitiveness 
criteria contribute to the economic competitiveness of the United 
States over the medium-term and long-term; projects that meet 
livability criteria improve the quality of living and working 
environments and experience for people in communities across the 
United States; projects that meet sustainability criteria improve 
energy efficiency, reduce dependence on oil, reduce greenhouse gas 
emissions, and benefit the environment; and projects that meet safety 
criteria improve the safety of U.S. transportation facilities and 
systems. 

[End of table] 

Shifting Traffic to Rail from Other Modes May Generate Benefits, but 
Many Factors Will Affect Whether Traffic Shifts, and Policies Abroad 
Have Produced Mixed Results: 

Determining the Extent of Benefits That Can Be Achieved through Rail 
Is Complicated by Numerous Factors: 

In order to generate benefits--such as a decrease in the harmful 
effects of transportation-related pollution--through mode shift, a 
policy first has to attract sufficient rail ridership or rail freight 
demand from other modes that have higher harmful emissions. In 
practice, the extent to which rail can generate sufficient demand to 
draw traffic from other modes and generate net benefits will depend on 
numerous factors.[Footnote 35] In addition to mode shift, policies 
that produce price changes can prompt other economic responses in the 
short run, such as the use of lighter-weight materials or a shift 
toward more fuel-efficient vehicles; over the longer term, there is 
greater potential for responses that will shape the overall 
distribution and use of freight and passenger transportation services. 
[Footnote 36] 

For intercity passenger rail, factors such as high levels of 
population density, expected population growth along a corridor, and 
strong business and cultural ties between cities can lead to a higher 
demand for intercity passenger travel. In order for rail to be 
competitive with other transportation modes, it needs to be time-and 
price-competitive and have favorable service characteristics related 
to frequency, reliability, and safety. Further, high-speed rail has 
more potential to attract riders in corridors experiencing heavy 
intercity travel on existing modes of transportation--particularly 
where air transportation has high traffic levels and a large share of 
the market over relatively short distances--and where there is, or is 
projected to be, growth in congestion and constraints on the capacity 
of existing systems. For example, rail traffic in the densely 
populated Northeast Corridor is highly competitive with other modes, 
and Amtrak now has a 65 percent share of the air-rail market between 
Washington, D.C. and New York and a 52 percent share between New York 
and Boston.[Footnote 37] The potential for network effects are also an 
important factor in the level of traffic that may shift to rail, as 
more riders are attracted when the line is located where it can carry 
traffic to a wide number of destinations or connect to other modes. 
For example, local transit systems can serve as feeders to the success 
of intercity passenger rail operations.[Footnote 38] Passenger modes 
can also work as complements, if, for example, passenger rail service 
delivers passengers to airports. DOT has indicated where passenger 
rail generally competes with other modes. For example, for intercity 
distances of 100-600 miles, in corridors with moderate population 
densities, high-speed rail competes with auto and bus and at high 
population densities competes with air, as shown in figure 2.[Footnote 
39] 

Figure 2: Competition among Passenger Transportation Modes: 

[Refer to PDF for image: illustrated table] 

Intercity distance: 0-100 miles[A]; 
Population density: Light: Auto; Bus; 
Population density: Moderate: Auto; Bus; Commuter rail; 
Population density: High: Auto; Bus; Commuter rail. 

Intercity distance: 100-600 miles[A]; 
Population density: Light: Auto; Bus; Conventional rail; 
Population density: Moderate: Auto; Bus; High-speed rail; 
Population density: High: High-speed rail; Air. 

Intercity distance: 600-3,000 miles[A]; 
Population density: Light: Auto; Bus; Air; 
Population density: Moderate: Auto; Bus; Air; 
Population density: High: Air. 

Source: GAO analysis of DOT information. 

[A] In certain corridors in high-density areas, conventional rail also 
competes with other modes (e.g., New York to Philadelphia). 

[End of figure] 

In freight markets, one mode may have a distinct comparative advantage 
over another for certain types of shipments, thereby limiting the 
potential for traffic to shift to rail. For example, carriage of bulk 
commodities (e.g., coal) relies almost entirely on rail and waterways, 
while carriage of high-value and very time-sensitive commodities is 
dominated by truck and aviation. Conversely, modes often work as 
complements to complete a shipment. Intermodal freight is designed to 
move on multiple modes, using a container that can be moved from a 
truck to a train to a ship without handling any of the freight itself 
when changing modes. In other cases, the modes may be substitutable 
for certain types of trips and will compete directly for shipments or 
for segments of shipments based on price and performance. For example, 
some long-haul trucking and rail shipments may be substitutable. DOT 
has produced some basic parameters that influence competition across 
the modes for freight, as shown in figure 3. 

Figure 3: Competition among Freight Transportation Modes: 

[Refer to PDF for image: illustrated table] 

Freight intercity distance: 0-250 miles; 
Transportation modes that compete for different types of freight: 
Retail goods: Truck; 
Consumer durables and other manufactured goods: Truck; Rail; 
Bulk goods: Truck; Rail; Water. 

Freight intercity distance: 250-500 miles; 
Transportation modes that compete for different types of freight: 
Retail goods: Truck; 
Consumer durables and other manufactured goods: Truck; Rail 
intermodal; Rail; 
Bulk goods: Truck; Rail; Water. 

Freight intercity distance: 500-1,000 miles; 
Transportation modes that compete for different types of freight: 
Retail goods: Truck; Rail intermodal; 
Consumer durables and other manufactured goods: Truck; Rail 
intermodal; Rail; 
Bulk goods: Rail; Water. 

Freight intercity distance: More than 1,000 miles; 
Transportation modes that compete for different types of freight: 
Retail goods: Truck; Rail intermodal; 
Consumer durables and other manufactured goods: Truck; Rail 
intermodal; Rail; 
Bulk goods: Rail; Water. 

Source: GAO analysis of DOT information. 

[End of figure] 

The extent to which mode-shifting is possible in the United States is 
difficult to estimate and will largely be determined by the types of 
parameters discussed above, such as whether shipping is feasible by 
another mode (e.g., rail lines may not be available for some routes), 
or practical (e.g., sending heavy coal shipments long distance by 
truck or time-sensitive shipments by rail may not be practical), and 
by the relative prices and other service characteristics of shipping 
by different modes. 

To further explore the potential for mode shift, we used a computer 
model developed by DOT[Footnote 40] to simulate the short-term change 
in VMT resulting from a 50-cent increase in per-mile truck rates. We 
simulated two scenarios: one using the model's default assumptions and 
one in which the assumptions pertaining to truck speed, reliability, 
and loss and damage were adjusted to make truck relatively more costly 
than rail.[Footnote 41] Under both scenarios, the 50-cent increase in 
truck rates (an increase of roughly 30 percent) resulted in less than 
a 1 percent decrease in truck VMT. Although both the default scenario 
and the alternative scenario produced similar estimates, these 
simulations are only suggestive, rather than definitive, of the impact 
that an increase in per-mile truck rates might have on VMT reduction. 
While the results of our simulation suggest that a 50-cent increase in 
per-mile truck rates would have a limited impact on diversion of 
freight from truck to rail, data limitations prevent us from making 
precise predictions with a high level of confidence. See appendix IV 
for a more detailed description of our modeling efforts, data quality 
issues, and a full list of assumptions in the model. 

In both the United States and in other countries we visited--where 
freight and passenger traffic generally share the same rail 
infrastructure--the potential benefits of a policy designed to shift 
freight traffic to rail are also affected by the amount of capacity 
available or planned on the rail network to accommodate a shift in 
traffic, as well as the capacity available or planned on competing 
transportation modes. For example, freight rail officials we met with 
in the United States indicated that in heavily congested corridors, 
such as in the Northeast, there is limited capacity available to 
accommodate both planned freight rail projects and proposed intercity 
passenger rail traffic, because the rail line is already congested. 
Plans for new dedicated high-speed rail lines would eliminate some of 
these capacity sharing issues and could potentially create the 
capacity needed to accommodate both freight and improved or expanded 
passenger service but must be weighed against the costs associated 
with constructing and maintaining new equipment and infrastructure, as 
well as acquiring rights of way for the track.[Footnote 42] 
Furthermore, significant investment and improvements to operations for 
highway infrastructure or airport infrastructure could offset the 
impact of policies designed to shift passenger or freight traffic to 
rail. For example, the FAA is currently pursuing modernization of the 
air transportation system to create additional capacity and 
efficiencies. If, as a result, flights become more efficient and 
travel times decrease, then travelers originally expected to shift to 
rail as a result of the implemented policy may not do so. In contrast, 
the existence of other policies in place concurrently may also be a 
contributing factor to improvements in environmental or congestion 
benefits, as separate policies may work together and lead to greater 
cumulative benefits. In either case, it can be difficult to 
distinguish the impact of a given policy due to these other factors. 

Following are descriptions of how shifting traffic to rail can address 
externalities and produce benefits, as well as some of the factors 
that affect the extent to which those benefits may materialize: 

Reduced greenhouse gas emissions and increased fuel efficiency: Rail 
emits fewer air emissions and is generally more fuel efficient than 
trucks. For example, a report by the American Association of State 
Highway and Transportation Officials (AASHTO) cites that the American 
Society of Mechanical Engineers estimates 2.5 million fewer tons of 
carbon dioxide would be emitted into the air annually if 10 percent of 
intercity freight now moving by highway were shifted to rail, if such 
traffic has the potential to shift.[Footnote 43] A recent study 
conducted by FRA comparing the fuel efficiency of rail to freight 
trucks calculated that rail had fuel efficiencies ranging from 156 to 
512 ton-miles per gallon, while trucks had fuel efficiencies ranging 
from 68 to 133 ton-miles per gallon.[Footnote 44] According to Amtrak 
officials, their intercity passenger rail service has also been shown 
to be more energy efficient than air or passenger vehicle traffic. 
[Footnote 45] In addition, passenger and freight rail can be 
electrified to eliminate even current emissions generated by rail 
transport, as alternative power (e.g., hydro or nuclear) may be used 
to generate electric propulsion. For example, many of the routes in 
the United Kingdom are electrified, and efforts are under way to 
continue to electrify additional segments of the rail network in order 
to reduce emissions. While rail generally provides favorable emissions 
attributes and fuel efficiency in comparison with highway and air 
travel, there are many factors that could affect the extent to which 
environmental benefits are achieved. These factors may include the 
type of train equipment, the mix of commodities being transported, the 
length of the rail route versus the truck route for a given shipment, 
traffic volume, and capacity. In addition, if the current 
transportation system is not designed to facilitate rail transport, it 
may be necessary to invest in additional capital infrastructure or 
build new rail yards closer to urban areas, which could have 
additional environmental costs and may diminish the extent of 
potential net benefits. Furthermore, how transport system users 
respond to a given policy will also impact the extent to which the 
policy generates any benefits. For example, a policy that changes the 
price of road transport by tolling could result in a freight hauler 
responding by changing the load factor of existing road shipments by 
consolidating shipments or increasing return loads to decrease the 
number of empty return trips. A similar policy could also lead to 
reduced transport volumes due to reduced demand for the product being 
shipped. According to DOT officials, correctly pricing usage of the 
transportation system is an ongoing challenge, as incorrect pricing 
can lead to inefficiencies and misallocation of resources beyond what 
market conditions would otherwise allow. Other policies aside from 
mode shift can more directly target environmental externalities. More 
targeted policies--such as increasing fuel taxes or implementing a 
carbon pricing scheme--may encourage drivers to purchase more fuel- 
efficient vehicles or make fewer vehicle trips, without shifting 
significant traffic to rail. 

Congestion: Where passenger or freight rail service provides a less 
costly alternative to other modes--through more timely or reliable 
transport--individuals and shippers can shift out of more congested 
modes and onto rail, thus alleviating congestion. For certain goods, a 
train can generally carry the freight of 280 or more trucks, relieving 
congestion by removing freight trucks from the highways.[Footnote 46] 
Similarly, an intercity passenger train can carry many times more 
people than the typical passenger vehicle.[Footnote 47] Consequently, 
if fewer vehicle miles are traveled, then there is less wear and tear 
on the highways and less cost to the public for related repairs and 
maintenance. However, congestion relief will vary based on specific 
locations, times of day, types of trips being diverted to another 
mode, and the conditions of the corridors and areas where trips are 
being diverted. For freight, long-haul shipments might have the most 
potential to shift to rail, but diversion of these trips to rail, 
while removing trucks from certain stretches of highway, may do little 
to address problems at the most congested bottlenecks in urban areas. 
Similarly, Amtrak officials noted that aviation can provide travelers' 
alternative options for travel in high-density corridors which may 
help relieve congestion at capacity-constrained airports. If high-
speed rail can divert travelers from making an intercity trip through 
congested highway bottlenecks or airports at peak travel times, then 
there may be a noticeable effect on traffic. However, any trips on a 
congested highway corridor that are diverted to another mode of 
travel, such as rail, may at least partially be replaced by other 
trips through induced demand. For example, since congestion has been 
reduced on a highway, making it easier to travel, more people may 
respond by choosing to drive on that highway where faster travel times 
are available, limiting the relief in the long-run. Other policies can 
be implemented that are designed to more directly address congestion 
where it is most acute, such as congestion pricing (e.g., converting 
high-occupancy vehicle lanes to high-occupancy toll lanes) or other 
demand management strategies. 

Safety: While safety has improved across all transportation modes over 
time, both passenger and freight rail may have a comparative advantage 
over other modes. Shippers and passengers who use rail in lieu of 
other modes may accrue measurable safety benefits because rail traffic 
is, for the most part, separated from other traffic. Because most rail 
accidents--both injuries and fatalities--involve traffic at limited 
locations such as grade crossings or on railroad property, safety 
benefits can be expected when more traffic is moved via rail. On a per-
mile basis, passenger and freight rail are substantially safer than 
cars or trucks. For example, according to Amtrak, there were 8 
passenger fatalities between 2003 and 2007. In addition, in 2007 most 
freight accidents occurred on highways--over 6 million--as compared 
with rail, which accounted for approximately 5,400 accidents. Between 
2003 and 2007, freight rail averaged 0.39 fatalities per billion ton- 
miles, compared with 2.54 fatalities per billion ton-miles for 
truck.[Footnote 48] There are a variety of policies and regulations 
that directly address safety concerns for each mode (e.g., safety 
standards and inspections for rail, vehicle safety features, etc.). 

Economic development: The recent economic downturn has spurred 
interest in developing opportunities to preserve and create jobs in 
order to help promote economic recovery. According to DOT, investment 
in intercity passenger and freight rail may aid in the short-term 
creation of jobs and potentially in the long-term development of 
higher density economic activity through concentrating retail and 
commercial business activity near rail lines or stations. Investment 
in intercity passenger and freight rail may be viewed as a potential 
avenue to generate economic development and produce wider economic 
impacts.[Footnote 49] Wider economic impacts associated with the 
investment in rail may include such things as added regional and 
national economic output and higher productivity and lower 
infrastructure costs. For example, investment in intercity high-speed 
passenger rail service could significantly influence the nature of 
regional economies beyond employment and income growth related to the 
investment in a rail system by spurring increases in business activity 
through travel efficiency gains. Moreover, the existence of new 
transport hubs and corridors creates the potential for economic 
development, as businesses may start to operate in the newly developed 
area in and around the rail corridor over the medium-term and the long-
term. However, in some cases, these types of impacts may reflect 
transfers of economic activity from one region to another and thus may 
not be viewed as benefits from a national perspective, or these 
impacts may already be accounted for through users' direct benefits. 
As such, there is much debate about achieving these wider economic 
impacts and a number of challenges associated with assessing these 
types of impacts. While high-speed rail may have wider economic 
impacts, the impact varies greatly from case to case and is difficult 
to predict. Estimates of benefits vary, as one study has suggested 
that wider economic benefits would not generally exceed 10 to 20 
percent of measured benefits, while an evaluation of another proposed 
high-speed rail line estimated these benefits to add 40 percent to 
direct benefits.[Footnote 50] There are a variety of other policies 
that could be implemented to help stimulate economic development 
without mode shift. 

In Selected European Countries, Experiences Suggest That Policies 
Intended to Produce Mode Shift May Lead to Varying Amounts of Mode 
Shift and Some Benefits: 

Based on experience in the United Kingdom and Germany where decision 
makers made a concerted effort to move traffic from other modes to 
rail through pricing policies, targeted grants, and infrastructure 
investments, these policies resulted in varying amounts of mode shift. 
[Footnote 51] The full extent of benefits generated from these 
policies is ultimately uncertain, though benefits realized included 
environmental and efficiency improvements or localized congestion 
relief. Foreign rail officials told us it was difficult to determine 
the full extent of the benefits due to complicating factors (as 
described throughout the previous section). While some benefits were 
attained through implementation of policies designed to shift traffic 
to rail, these benefits were not necessarily achieved in the manner 
originally anticipated or at the level originally estimated. 
Furthermore, it is uncertain whether the benefits attained were 
achieved in the most efficient manner, or whether similar benefits 
could have been attained through other policies at a lower cost. 

Road freight pricing policies: In 2005, the German government 
implemented a Heavy Goods Vehicle (HGV) tolling policy on motorways to 
generate revenue to further upgrade and maintain the transportation 
system and to introduce infrastructure charging based on the "user 
pays" principle by changing the relative price of road transport 
relative to rail. The HGV tolling policy was also designed to provide 
an incentive to shift approximately 10 percent of road freight traffic 
to rail and waterways in the interests of the environment and to 
deploy HGVs more efficiently. According to German Ministry of 
Transport officials, while the HGV toll policy did not result in the 
amount of mode shift originally anticipated, some level of 
environmental benefits and road freight industry efficiency 
improvements were realized. These benefits are attributed to a more 
fuel-efficient HGV fleet making fewer empty trips. For example, 
officials told us that, in response to the tolling policy, trucking 
companies purchased more lower emission vehicles, which were charged a 
lower per-mile rate in order to decrease their toll.[Footnote 52] For 
the most part, German freight shipments continued to be made primarily 
on trucks, and trucks' mode share has not changed appreciably since 
instituting the policy. Findings in a study conducted for the Ministry 
of Transport also indicated that transport on lower emission trucks 
has increased significantly, totaling 49 percent of all freight 
operations subject to tolls in 2009. According to German transport 
officials, the share of freight moved by rail has only slightly 
increased during the last decade. However, this increase cannot be 
clearly attributed to a particular policy tool, such as the HGV toll. 

Other countries have had similar experiences implementing pricing 
policies to provide incentives to shift traffic to rail. For example, 
the Swiss government implemented a HGV fee in 2001 on all roads to 
encourage freight traffic to shift from road to rail. This policy 
similarly resulted in improved efficiency because the trucking 
industry adapted its fleet and replaced some high emission vehicles 
with new lower emission vehicles. According to Swiss Federal Office of 
Transport documentation, HGV traffic through the Swiss Alps also 
decreased compared with what it would have been without introduction 
of the fee. However, to fully assess the magnitude of benefits of 
these types of tolling policies, these improvements would need to be 
weighed against the costs of implementing the policy, and this type of 
analysis has not been conducted. 

Freight rail operations and capital support: The United Kingdom's 
Department for Transport uses two grant programs providing financial 
support for specific rail freight projects to encourage mode shift and 
provide congestion relief, based on the view that road freight 
generally does not pay its share of the significant external costs 
that it creates. The department's Mode Shift Revenue Support scheme 
provides funding for operational expenses and the Freight Facilities 
Grant program supplements capital projects for freight infrastructure. 
The British government's experience with these policies--which draw 
from a relatively small pool of annual funding and are intentionally 
designed to serve a targeted market--led to localized benefits for 
particular segments of the freight transport market in specific 
geographic locations such as congested bottlenecks near major ports. 
An evaluation of the Freight Facilities Grants program found that the 
program funding played an important role in developing or retaining 
rail freight flows, traditionally focused on bulk commodities. 
[Footnote 53] According to officials we met with, the grants from the 
Mode Shift Revenue Support scheme encourage mode shift principally for 
the economically important and growing intermodal container market and 
have been successful in reducing congestion on specific road freight 
routes because the program focuses on container flows from major ports 
(in which rail now has a 25 percent market share). These officials 
noted that, out of approximately 800,000 truck journeys removed from 
the road as a result of the grants from the Mode Shift Revenue Support 
scheme, between 2009 and 2010, 450,000 trucks were removed from 
England's largest port--the Port of Felixstowe. Therefore, officials 
said the grants appear to have led to a decrease in truck traffic 
concentrated in specific locations for a particular segment of the 
freight transport industry. 

Intercity passenger rail infrastructure investments: Few 
postimplementation studies have been conducted to empirically assess 
the benefits resulting from investment in high-speed intercity 
passenger rail. Based on our previous work, some countries that have 
invested in new high-speed intercity passenger rail services have 
experienced discernable mode shift from air to rail where rail is trip-
time competitive. For example, the introduction of high-speed 
intercity rail lines in France and Spain led to a decrease in air 
travel with an increase in rail ridership, and Air France officials 
estimated that high-speed rail is likely to capture about 80 percent 
of the air-rail market when rail journey times are between 2 and 3 
hours.[Footnote 54] For example, with the introduction of the Madrid-
Barcelona high-speed rail line in February 2008, air travel dropped an 
estimated 30 percent. In France, high-speed rail has captured 90 
percent of the Paris-Lyon air-rail market. While discernible mode 
shift has been observed, the extent to which net benefits were 
achieved is unclear. Factors such as the proportion of traffic 
diverted from air or conventional rail versus newly generated traffic 
affect the extent of benefits. Furthermore, quantifying any resulting 
environmental benefits, such as reduced greenhouse gas emissions, or 
assessing the extent to which these benefits exceed the costs 
associated with developing these new high-speed rail routes is 
difficult. Some evaluations have been conducted in Spain and France 
and have indicated that net benefits were less than expected due to 
higher costs and lower than expected ridership, although, in France, 
the evaluations still found acceptable financial and social rates of 
return.[Footnote 55] 

Policies that provide incentives to shift passenger and freight 
traffic to rail offer the opportunity to attain a range of benefits 
simultaneously, but a variety of complicating factors can have a 
significant impact on the extent to which these benefits may be 
attained. In addition, if these policies are unable to generate the 
ridership or demand necessary to shift traffic from other modes to 
rail, the potential benefits may be further limited. While officials 
from some European countries we visited indicated that they have 
attained benefits from policies intended to shift traffic to rail, 
gains have been mixed, and the extent of benefits attained has 
depended on the specific context of policy implementation in each 
location, as the benefits realized are directly related to the 
particulars of each project. Furthermore, it is not always clear that 
the policy goals were feasible to begin with or that mode shift would 
have been the most cost-effective way to achieve the benefits sought. 
Some officials and stakeholders we met with told us that it is very 
difficult to attribute causation and draw conclusions regarding the 
effectiveness of transportation policy tools because so many factors 
are at play and may change simultaneously. In some cases, officials 
cannot determine the full extent of benefits or link impacts to a 
given policy with certainty, making it difficult for decision makers 
to know what to expect from future policies being considered or 
developed. 

In the next section, we look at two recent U.S. investment programs 
that awarded grant funding to freight and intercity passenger rail 
projects. Although neither of these programs were adopted for the 
specific purpose of shifting passenger or freight traffic to rail, 
both programs do seek to attain benefits, such an economic development 
and environmental benefits, by investing in rail. As previously noted, 
the degree to which benefits can be generated depends on a variety of 
factors, including the ability to attract riders or freight shipments 
either through mode shift or new demand. We discuss how applicants 
assessed the potential benefits and costs of their specific projects, 
based on the particular circumstances of each project, and the 
usefulness of those assessments for federal decision makers in making 
their investment decisions. 

Grant Applicants' Assessments of Project Benefits and Costs Are of 
Varying Quality and Usefulness to Decision Makers: 

Grant Applicants' Assessments of Project Benefits and Costs Were Not 
Comprehensive in Many Respects: 

According to DOT officials from both programs, as well as our 
assessment of 40 randomly selected rail-related TIGER and HSIPR 
applications,[Footnote 56] information on project benefits and costs 
submitted by applicants to the TIGER and HSIPR[Footnote 57] programs 
varied in both quality and comprehensiveness. While a small number of 
analyses of project benefits and costs were analytically strong--with 
sophisticated numerical projections of both benefits and costs and 
detailed information on their data and methodology--many others (1) 
did not quantify or monetize benefits to the extent possible, (2) did 
not appropriately account for benefits and costs, (3) omitted certain 
costs, and (4) did not include information on data limitations, 
methodologies for estimating benefits and costs, and uncertainties and 
assumptions underlying their analyses. 

First, the majority of applications we assessed contained primarily 
qualitative discussion of project benefits, such as potential 
reductions in emissions, fuel consumption, or roadway congestion, 
which could have been quantified and monetized. For instance, while 36 
of the 40 applications we assessed included qualitative information 
regarding potential reductions in congestion, 20 provided quantitative 
assessments of these benefits, and 13 provided monetary estimates. 
[Footnote 58] This pattern was consistent across categories of 
benefits we assessed; however, some categories of impacts, such as 
safety and economic development, were even less frequently quantified. 
While federal guidelines, including Executive Order No. 12893, allow 
for discussion of benefits in a qualitative manner, they note the 
importance of quantifying and monetizing benefits to the maximum 
extent practicable. However, in some cases, certain categories of 
impacts may be more difficult to quantify than others and qualitative 
information on potential benefits and costs can be useful to decision 
makers. 

Second, common issues identified by DOT economists in the applications 
they assessed[Footnote 59] included failure to discount future 
benefits and costs to present values or failure to use appropriate 
discount rates,[Footnote 60] double counting of benefits, and 
presenting costs only for the portion of the project accounted for in 
the application while presenting benefits for the full project. 
Similarly, 33 of the 40 applications we assessed did not use discount 
rates as recommended in OMB Circular No. A-94 and OMB Circular No. A-
4. Further, DOT economists who reviewed assessments of project 
benefits and costs contained in selected TIGER applications stated 
that many applicants submitted economic impact analyses--which are 
generally used to assess how economic impacts would be distributed 
throughout an economy but not for conducting benefit-cost analysis of 
policy alternatives. Economic impact analyses may contain information 
that does not factor into calculations of net benefits, such as tax 
revenue and induced jobs, and do not generally include information on 
other key benefits that would be accounted for in a benefit-cost 
analysis, such as emissions reduction or congestion relief. 
Applicants' focus on economic impacts in their assessments of project 
benefits may have stemmed from additional funding criteria that DOT 
identified for both programs related to job creation and economic 
stimulus, as well as decision makers' focus on these issues at the 
state and local levels. 

Third, important costs were often omitted from applications. In many 
cases, applicants would estimate a benefit, but not account for 
associated costs, such as increased noise, emissions, or potential 
additional accidents from new rail service. For instance, applicants 
often counted emission reduction benefits from mode shift to rail as a 
benefit but did not include corresponding increases in emissions from 
increased rail capacity and operation in their calculations of net 
benefits. Our assessments of TIGER and HSIPR applications found that 
of the applicants who projected potential safety or environmental 
benefits for their projects, only three applicants addressed potential 
safety costs, and only four applicants addressed potential 
environmental costs. 

Finally, we also found that analyses of benefits and costs in many 
applications consistently lacked other key data and methodological 
information that federal guidelines such as OMB Circular No. A-94 and 
OMB Circular No. A-4 recommend should be accounted for in analyses of 
project benefits and costs.[Footnote 61] Notably, the majority of the 
applications to the TIGER and HSIPR programs that we reviewed did not 
provide information related to uncertainty in projections, data 
limitations, and the assumptions underlying their models. While a 
small number of applications we assessed provided information in all 
of these areas, 31 out of 40 did not provide information on the 
uncertainty associated with their estimates of benefits and costs, 28 
out of 40 did not provide information on the models or other 
calculations used to arrive at estimates of benefits and costs, and 36 
out of 40 did not provide information on the strengths and limitations 
of data used in their projections. Furthermore, of those that did 
provide information in these areas, the information was generally not 
comprehensive in nature. For example, multiple applications provided 
information on the models or calculations used to quantify or monetize 
benefits, but did not do so for all the benefit and cost calculations 
included in their analysis. 

Short Time Frames, a Lack of Clear Standard Values, and Data 
Limitations Contributed to the Inconsistent Quality and Limited 
Usefulness of Assessments of Project Benefits and Costs: 

Applicants, industry experts, and DOT officials we spoke with reported 
that numerous challenges related to performing assessments of the 
benefits and costs of intercity passenger or freight rail projects can 
contribute to variation in the quality of assessments of project 
benefits and costs in applications to federal programs such as the 
TIGER and HSIPR programs. These challenges include (1) limited time, 
resources, and expertise for performing assessments of project 
benefits and costs; (2) a lack of clear guidance on standard values to 
use in the estimation of project benefits; and (3) limitations in data 
quality and access. These challenges impacted the usefulness of the 
information provided for decision makers, and, as a result, changes 
have been made or are being considered for future rounds of funding. 

Time, Resources, and Expertise: 

Performing a comprehensive assessment of a proposed project's 
potential benefits and costs is time and resource intensive and 
requires significant expertise. According to experts, a detailed and 
comprehensive benefit-cost analysis requires careful analysis and may 
call for specialized data collection in order to develop projections 
of benefits and costs. The short time frames for assembling 
applications for the TIGER and HSIPR programs--which were designed to 
award funds quickly in order to provide economic stimulus--may have 
contributed to the poor quality of many assessments. In addition, 
according to DOT officials, many applicants to the TIGER and HSIPR 
programs may not have understood what information to include in their 
analyses. The recent nature of federal requirements for state rail 
planning means that states are still building their capacity to 
perform complex analyses to assess rail projects and, in many cases, 
rail divisions within state departments of transportation are very 
small. State rail divisions often face funding and manpower issues 
since there is typically no dedicated state funding for rail services, 
and state transportation planning has historically focused more on 
highway projects.[Footnote 62] As a result, some applicants to 
competitive federal grant programs may have more capacity to perform 
assessments of project benefits and costs than others. For example, 
according to DOT officials, freight railroads have more resources to 
devote to developing models and estimating potential project benefits 
and costs. 

Valuing Benefits: 

Standard values to monetize some benefits are not yet fully 
established, which can create inconsistency in the values used by 
applicants in their projections. While DOT has published guidance on 
standard estimates for the value of travel time and the value of a 
statistical life--which can be used to estimate the value of 
congestion mitigation efforts and safety improvements, respectively--
values for other benefits are less clear. For instance, according to 
DOT officials, uncertainties associated with analyzing the value of 
time for freight shipment prevents DOT from issuing specific guidance 
in this area. In addition, there are substantial uncertainties 
associated with analyzing the value of many benefits, such as 
reduction in greenhouse gas emissions. While mode shift to rail may 
reduce pollution and greenhouse gas emissions, experts do not agree on 
the value to place on that benefit.[Footnote 63] DOT has issued 
guidance on values for use in calculating the social benefits of 
pollutant emissions, however according to modeling experts we 
interviewed, disagreement regarding how to value different benefits 
can lead some analysts to limit their assessments of benefits and 
costs to only that which can be monetized, while others may include 
all categories of benefits and costs in their assessment. As a result, 
some TIGER and HSIPR applicants may have used differing values to 
monetize projected benefits and costs, while others did not monetize 
benefits at all. Without clear guidance to applicants on preferred 
values for use in assessments of project benefits and costs, DOT 
decision makers may be hindered in their ability to compare the 
results of assessments of benefits and costs across projects or across 
modes. A standard set of values for key benefit categories may enable 
transportation officials to more readily compare projects and 
potentially place more weight on the results of assessments of project 
benefits and costs in their decision-making processes. 

Data Quality and Access: 

According to DOT officials, historically lower levels of state and 
federal funding for rail compared with other modes of transportation 
have contributed to data gaps that impact the ability of applicants to 
project benefits and costs for both intercity passenger rail and 
freight rail projects. For instance, lack of data on intercity 
passenger travel demand made it difficult for some applicants to the 
HSIPR program to quantify potential benefits for some new high-speed 
rail lines. The lack of data may be related to cuts to federal funding 
for the Bureau of Transportation Statistics resulting in a decreased 
emphasis on the collection of rail-related data. Multiple state and 
association officials stated that previous state and national surveys 
of travel behavior did not capture traveler purposes for intercity 
travel and did not have a sufficient number of intercity traveler 
responses for use in travel modeling. In addition, lack of access to 
proprietary data on goods movement made it challenging for some 
applicants to the TIGER program to quantify benefits that might be 
associated with freight rail. According to officials from the 
California Department of Transportation (Caltrans), when performing 
analyses to estimate project benefits and costs, Caltrans employees 
had to manually count freight trains for a 24-hour period in order to 
gather data for use in their analyses. Furthermore, state 
transportation officials we spoke with indicated that the quality of 
data available for use in projecting benefits and costs of a project 
is often inconsistent. Officials we interviewed stated that data 
included in assessments of project benefits and costs are often from 
different years, contain sampling error, and may be insufficient for 
their intended use. These limitations lessen the reliability of 
estimates produced to inform transportation decision-making, as 
available data provide critical inputs for travel models. 

Modeling and forecasting limitations also make it harder to project 
shifts in transportation demand and related benefits and costs 
accurately. Benefit-cost analyses of transportation projects depend on 
forecasts of projected levels of usage, such as passenger rail 
ridership or potential freight shipments, in order to inform 
calculation of benefits and costs. Limitations of current models and 
data make it difficult to predict changes in traveler behavior, 
changes in warehousing and shipper behaviors for businesses, land use, 
or usage of nearby roads or alternative travel options that may result 
from a rail project. Since transportation demand modeling depends on 
information on traveler or shipper preferences in order to inform 
predictions, the lack of good intercity traveler and shipper demand 
data greatly impacts the quality of projections, particularly for new 
intercity passenger or freight rail service where no prior data exists 
to inform demand projections. 

Usefulness of Assessments of Benefits and Costs: 

As a result of the limitations described above, DOT officials stated 
that the assessments of benefits and costs provided by TIGER and HSIPR 
applicants were less useful to decision makers than anticipated. In 
general, the majority of rail-related applications we reviewed that 
were forwarded for additional consideration for the TIGER program 
[Footnote 64] contained assessments of project benefits and costs that 
were either marginally useful or not useful to DOT officials in their 
efforts to determine whether project benefits were likely to exceed 
project costs.[Footnote 65] Overall, 62 percent of forwarded rail-
related applications had assessments of benefits and costs that were 
rated by DOT economists as "marginally useful" or "not useful," and 38 
percent had assessments that were rated as "very useful" or "useful" 
(see figure 4). However, DOT officials noted that railroads generally 
did a better job with their benefit-cost analyses in their 
applications than other modes. 

Figure 4: DOT Assessment of Usefulness of Benefit-Cost Analyses from 
Forwarded Rail-Related TIGER Applications: 

[Refer to PDF for image: pie-chart] 

Very useful: 21%; 
Useful: 17%; 
Marginally useful: 45%; 
Not useful: 17%. 

Source: GAO analysis of DOT data. 

Note: DOT economists assessed usefulness of benefit-cost information 
only for those applications that were forwarded by initial DOT review 
teams for additional consideration. 

[End of figure] 

While applicants to the HSIPR program were not required to conduct a 
benefit-cost analysis, the Federal Register notice for the program 
stated that information on benefits and costs provided by applicants 
would be used by DOT to conduct a comprehensive benefit-cost analysis 
for projects. However, according to FRA officials, the quality of the 
information provided prevented DOT from being able to use the 
information in this manner. 

While it is possible to offset the impact of the limitations described 
above and improve the usefulness of assessments of benefits and costs 
to decision makers by providing clear information on assumptions and 
uncertainty within analyses, as we stated above, very few TIGER and 
HSIPR applicants did so. Without information on projection 
methodologies and assumptions, DOT officials were not able to 
consistently determine how demand and benefit-cost projections were 
developed and whether the projections were reasonable. As a result, 
officials for both programs focused on simply determining whether 
project benefits were likely to exceed project costs, rather than a 
more detailed assessment of the magnitude of projects' benefits and 
costs in relation to one another. See appendix IV for a discussion of 
the challenges related to assumptions and uncertainty we encountered 
during our attempt to use a model to predict freight mode shift from 
truck to rail. 

The varying quality and focus of assessments of project benefits and 
costs included in both TIGER and HSIPR applications resulted in 
additional work for DOT officials in order for DOT to be able to 
determine whether project benefits were likely to exceed project 
costs. For example, DOT officials stated that DOT economists for the 
TIGER program spent 3 to 4 hours per application examining whether it 
contained any improper analysis techniques or other weaknesses, 
seeking missing information, and resolving issues in the analyses. For 
the HSIPR program, a DOT economist with subject matter expertise 
reviewed the demand forecasts provided by selected Track 2 applicants, 
[Footnote 66] devoting significant time to assess the level of risk 
the uncertainty in these projections was likely to pose to the 
ultimate success of the project. 

Recent Changes and Improvements to Program Guidance: 

In order to improve the quality of applicant assessments of project 
benefits and costs, DOT economists identified limitations of the 
benefit-cost analyses submitted during TIGER I and used that 
information to develop guidance for TIGER II.[Footnote 67] In the 
Federal Register notice for TIGER II, DOT provided additional 
information to applicants regarding what should be included in 
assessments of project benefits and costs.[Footnote 68] This guidance 
included information on the differences between benefit-cost analysis 
and economic impact analysis, assessment of alternatives in relation 
to a baseline, discounting, forecasting, transparency and 
reproducibility of calculations, and methods of calculating various 
benefits and costs. As part of its guidance on assessing costs, DOT 
noted that applicants should use life-cycle cost analysis in 
estimating the costs of projects.[Footnote 69] For example, DOT 
guidance states that external costs, such as noise, increased 
congestion, and environmental pollutants resulting from construction 
or other project activities, should be included as costs in 
applicants' analyses. Furthermore, applicants should include, to the 
extent possible, other costs caused during construction, such as 
delays and increased vehicle operating costs. 

FRA also plans to alter HSIPR requirements in order to increase the 
quality of information on project benefits and costs provided by 
future applicants. According to FRA officials, while applicants to the 
second round of HSIPR funding were presented with similar guidelines 
for assessing project benefits and costs as those provided in the 
first round, future HSIPR applicants will be required to provide more 
rigorous projections of ridership, benefits, and costs and to revise 
their assessments of project benefits and costs based on their 
improved ridership projections. Officials noted, however, that the 
process will be iterative and anticipated that models for the high-
speed rail program will improve as domestic historical data on 
ridership becomes available over time. In addition, officials stated 
that FRA plans to take steps to encourage consistency in the 
methodologies grant applicants use to project demand, benefits, and 
costs. For instance, FRA is currently in the preliminary stages of 
developing a benefit-cost framework for states and localities, which 
represent the majority of applicants to programs such as TIGER and 
HSIPR, to use in assessing rail projects. Officials stated that FRA 
plans to issue guidance on performing assessments of benefits and 
costs for passenger rail projects when the framework is fully 
developed but did not provide a timeline for its development. 

While DOT officials for both programs have taken steps to improve the 
quality of benefit-cost information and associated analyses in the 
short term, other steps are necessary to improve quality over time. 
Some of these additional steps, such as developing historical data for 
intercity passenger rail demand, making improvements to forecasting 
and modeling, and increasing accessibility and quality of key data, 
may take more time. Nonetheless, improving the quality of benefit and 
cost information considered for programs such as TIGER and HSIPR could 
simplify the decision-making process and lend more credence to the 
merit of the projects ultimately selected for funding. 

Conclusions: 

Difficult and persistent problems face the U.S. transportation system 
today. Our system is largely powered by vehicles that use fossil fuels 
that produce harmful air emissions and contribute to climate change. 
Our existing infrastructure is aging and, in many places, is in a poor 
state of repair. Demand for freight and passenger travel will continue 
to grow, and the growing congestion in urban areas and at key 
bottlenecks in the system costs Americans billions of dollars in 
wasted time, fuel, and productivity each year. Adding to these 
problems, expanding or improving the efficiency of our existing road 
and air transportation networks has proven difficult, costly, and time-
consuming. Both the HSIPR and TIGER programs provided a new 
opportunity to invest in rail--a mode that has historically been 
underrepresented in the U.S. transportation funding framework. Some 
see investment in rail infrastructure, along with other policies 
designed to shift traffic to rail, as important to addressing these 
problems, pointing to rail's advantages over cars and freight trucks 
in terms of energy efficiency, safety, and lower emissions. While 
investments in rail or policies designed to shift traffic to rail may 
generate some benefits--as occurred to some degree in the United 
Kingdom and Germany--benefits must be weighed against direct project 
costs and other costs (e.g., noise) to determine whether an investment 
or policy produces overall net benefits. Further, close attention must 
be paid to the extent to which freight and passenger travel can 
actually shift to rail from other modes, given the choices available 
to, and the preferences of, travelers and shippers. 

While an assessment of benefits and costs is only one factor among 
many in decision making regarding these investments and policies, a 
decision maker's ability to weigh information depends on the quality 
of benefit and cost information provided by project sponsors--
regardless of whether this information is provided in a benefit-cost 
analysis or a more general discussion or enumeration of benefits and 
costs. We found that many TIGER and HSIPR applicants struggled to 
provide the benefit-cost information requested or to use appropriate 
values designated for their respective program. The lack of 
consistency and completeness in the benefit-cost information provided 
makes it more difficult for decision makers to conduct direct project 
comparisons or to fully understand the extent to which benefits are 
achievable and the trade-offs involved. While the shortened time 
frames of the programs and resource limitations among project sponsors 
were key causes of the varying quality of analyses, data limitations 
(including a lack of historical data--particularly with respect to 
high-speed rail), data inconsistencies, and data unavailability also 
accounted for some limitations in applicants' benefit-cost information 
and will continue to impact these analyses in future funding rounds. 
Until data quality, data gaps, and access issues are addressed for the 
data inputs needed for rail modeling and analysis, projections of rail 
benefits will continue to be of limited use. In addition, almost no 
applicants discussed limitations in their analysis, including the 
assumptions made and levels of uncertainty in their projections. Only 
when assumptions and uncertainty are conveyed in assessments of 
benefits and costs can decision makers determine the appropriate 
weight to give to certain projections. 

To its credit, DOT has provided more explicit guidance to TIGER 
applicants in its second round of grant applications on how to meet 
federal benefit-cost analysis guidelines. While such guidance should 
result in improved quality of benefit-cost information provided for 
this program, this guidance neither ensures consistency across 
analyses in terms of common data sources, values, and models, nor will 
it have any impact on how benefits and costs are evaluated across 
programs that invest in other modes (such as the Federal-Aid Highway 
Program) which do not have a benefit-cost analysis requirement. 
Providing more standardized values for calculating project benefits 
and costs and developing a more consistent approach to assessing 
project benefits and costs so that proposed projects across modes may 
be more easily compared with one another can have numerous benefits. 
For instance, standardized values and a consistent approach allow for 
more confidence that projects and policies chosen will produce the 
greatest benefits relative to other alternatives, give more credence 
to investment decisions across programs and modes, and limit DOT 
officials' need to invest time and resources in order to use the 
information as part of the decision making process. If benefit-cost 
considerations are ever to play a greater role, DOT will need to look 
at ways it can improve the quality and consistency of the data 
available to project sponsors. 

Recommendations for Executive Action: 

To improve the data available to the Department of Transportation and 
rail project sponsors, we recommend that the Secretary of 
Transportation, in consultation with Congress and other stakeholders, 
take the following two actions: 

* Conduct a data needs assessment and identify which data are needed 
to conduct cost-effective modeling and analysis for intercity rail, 
determine limitations to the data used for inputs, and develop a 
strategy to address these limitations. In doing so, DOT should 
identify barriers to accessing existing data, consider whether 
authorization for additional data collection for intercity rail travel 
is warranted, and determine which entities shall be responsible for 
generating or collecting needed data. 

* Encourage effective decision making and enhance the usefulness of 
assessments of benefits and costs, for both intercity passenger and 
freight rail projects by providing ongoing guidance and training on 
developing benefit and cost information for rail projects and by 
providing more direct and consistent requirements for assessing 
benefits and costs across transportation funding programs. In doing 
so, DOT should: 

- Direct applicants to follow federal guidance outlined in both the 
Presidential Executive Order 12893 and OMB Circulars Nos. A-94 and A-4 
in developing benefit and cost information. 

- Require applicants to clearly communicate their methodology for 
calculating project benefits and costs including information on 
assumptions underlying calculations, strengths and limitations of data 
used, and the level of uncertainty in estimates of project benefits 
and costs. 

- Ensure that applicants receive clear and consistent guidance on 
values to apply for key assumptions used to estimate potential project 
benefits and costs. 

Agency Comments and Our Evaluation: 

We provided copies of our draft report to DOT, Amtrak and EPA for 
their review and comment. DOT provided technical comments and agreed 
to consider the recommendations. Amtrak and EPA provided technical 
comments, which we incorporated as appropriate. 

As agreed with your offices, unless you publicly announce the contents 
of this report earlier, we plan no further distribution until 30 days 
from the report date. At that time, we will send copies of this report 
to the appropriate congressional committees, the Secretary of 
Transportation, the Administrator of the Federal Railroad 
Administration, Amtrak, EPA, the Director of the Office of Management 
and Budget, and other interested parties. The report also will be 
available at no charge on the GAO Web site at [hyperlink, 
http://www.gao.gov]. 

If you or your staff members have any questions about 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 major 
contributions to this report are listed in appendix V. 

Signed by: 

Susan A. Fleming: 
Director, Physical Infrastructure Issues: 

[End of section] 

Appendix I: Objectives, Scope, and Methodology: 

To better understand the potential net benefits of intercity passenger 
and freight rail, we examined (1) the extent to which transportation 
policy tools that provide incentives to shift passenger and freight 
traffic to rail may generate emissions, congestion, and economic 
development benefits and (2) how project benefits and costs are 
assessed for investment in intercity passenger and freight rail and 
how the strengths and limitations of these assessments impact federal 
decision making. 

Interviews: 

We conducted interviews with the Department of Transportation (DOT), 
the Environmental Protection Agency (EPA), and Amtrak. We also 
interviewed representatives from transportation coalitions and 
associations, metropolitan planning organizations, state DOTs, and 
transportation consultants. Interviews with officials were in regards 
to methods to assess the benefits and costs of transportation 
investments and the limitations and challenges to assessing benefits. 
We also conducted interviews with officials from the High-Speed 
Intercity Passenger Rail (HSIPR), the Transportation Investment 
Generating Economic Recovery (TIGER), and Transportation 
Infrastructure Finance and Innovation Act (TIFIA) programs to gather 
insights into the usefulness of the cost-benefit study information in 
decision making. In addition to interviews with agency officials, 
interviews were conducted in three rail corridors (California, 
Midwest, and the Northeast) to ascertain additional information on 
challenges associated with conducting and communicating findings from 
benefit and cost assessments to decision makers. These interviews 
involved applicants and other corridor stakeholders who had applied to 
either or both the HSIPR and TIGER grant programs. Similarly, some of 
our interviews with organizations in the rail corridors included 
consultants such as Cambridge Systematics and Parsons Brinckerhoff 
which were involved in the development of studies for corridors. 
Following is table 3 with a list of selected organizations whose 
officials and representatives we interviewed. 

Table 3: Interviews: 

Federal Agencies and Entities: 
* Amtrak: 
* DOT: 
- Bureau of Transportation Statistics; 
- Federal Highway Administration; 
- Federal Railroad Administration; 
- Inspector General; 
- Office of the Secretary; 
* EPA: 
* Industry associations: 
- American Association of State Highway and Transportation Officials; 
- American Public Transportation Association; 
- Association of American Railroads. 

United States: 
* California Department of Transportation; 
* California High Speed Rail Authority; 
* CSX; 
* I-95 Corridor Coalition; 
* Illinois Department Of Transportation; 
* Metropolitan Transportation Commission; 
* Northern New England Passenger Rail Authority; 
* Ohio Department of Transportation; 
* Ohio Rail Development Commission; 
* Pennsylvania Department of Transportation; 
* San Diego Association of Governments. 

United Kingdom: 
* Department for Transport; 
* Greengauge 21; 
* National Audit Office; 
* Network Rail; 
* Rail Freight Group. 

Germany: 
* Deustche Bahn; 
* Federal Ministry of Environment, Nature Conservation and Nuclear 
Safety; 
* Federal Ministry of Finance; 
* Federal Ministry of Transport, Building and Urban Development. 

Other: 
* Louis Thompson, Thompson, Galenson and Associates; 
* Organization for Economic Cooperation and Development (OECD), 
International Transport Forum; 
* University of Leeds, Institute for Transport Studies. 

Source: GAO. 

[End of table] 

Study Review: 

We reviewed our prior reports and documentation from an array of 
sources, including the DOT Inspector General, Congressional Research 
Service, and Congressional Budget Office. In addition, we identified 
studies through our interviews with stakeholders and conducted an 
extensive systematic search of literature published in the last 15 
years. We reviewed this information to identify studies that analyzed 
the benefits and costs of intercity passenger and freight rail, mode 
shift to intercity passenger or freight rail, or the potential net 
benefits that could be attained through mode shift. In general, we did 
not find a sufficient number of available studies that adequately 
addressed our researchable questions, had an appropriate scope, or 
utilized empirically reliable methodologies. As a result, we used the 
studies and information we reviewed to inform the engagement as a 
whole and provided examples and illustrations of the potential costs 
and benefits that may be attained from policies that provide 
incentives to shift traffic to rail. In addition, we conducted case 
studies in the United Kingdom and Germany and asked officials to 
synthesize their experiences based on their professional judgment and 
data. Officials we met with also confirmed that it is difficult to 
causally link policy interventions to specific outcomes. 

Assessment of HSIPR and TIGER Applications' Cost and Benefit 
Information: 

We reviewed and assessed information on potential project benefits and 
costs included in selected applications to the HSIPR grant program and 
the Grants for TIGER grant program--20 applications from each grant 
program. We selected a nongeneralizable random sample of 40 
applications from a larger pool of HSIPR and TIGER applications that 
we identified as including components related to intercity passenger 
rail or freight rail. For HSIPR, we included all applications 
submitted under Track 2 of the program, which focused on intercity 
passenger rail projects, in our selection pool, while for TIGER, we 
included all applications requesting more than $20 million that 
included components related to intercity passenger rail or freight 
rail in project descriptions provided by DOT. Twenty applications from 
each grant program were randomly selected for our review. The random 
sample of applications was weighted to ensure approximately 
proportional representation of applications from both programs that 
were awarded funding by DOT to those that were not awarded funding by 
DOT and, for the TIGER program, weighted to ensure approximately 
proportional representation of applications that were selected by DOT 
for additional review during DOT's application review process to those 
that were not selected by DOT for additional review. 

Information pertaining to project benefits and costs in each of the 40 
randomly selected applications was independently reviewed by two of 
our analysts based on Office of Management and Budget (OMB) guidelines 
for benefit-cost analysis and input from our economists and 
methodologists. Application information assessed by our analysts 
included whether benefits and costs related to congestion mitigation, 
emissions reduction, and economic development were assessed 
qualitatively, quantitatively, or were monetized. In addition, 
analysts identified whether applications included information on a 
number of key methodological elements identified by OMB and in our 
prior work. Any discrepancies in findings by the two analysts were 
reconciled for the final assessment. 

International Case Studies: The United Kingdom and Germany: 

We conducted case studies of selected policies and programs in the 
United Kingdom and Germany to learn more about policies to address 
concerns about emissions, congestion and economic development. These 
two countries were chosen based on a number of criteria, including 
experience in implementing capacity enhancing and demand management 
policy tools in order to encourage mode shift to rail and attain 
potential benefits. We reviewed studies and reports on policy tools 
used in these countries and in the European Union. We interviewed 
officials from the United Kingdom's Department for Transport and 
Germany's Ministry of Transport, Building and Urban Development. In 
addition, we interviewed officials in the German Federal Ministry of 
Finance and Ministry for the Environment, Nature Conservation and 
Nuclear Safety, as well as the United Kingdom's National Audit Office. 
We also met with representatives from rail industry organizations and 
rail companies and stakeholder groups from these countries. For more 
information, see appendix II. 

Computer Simulations of Freight Diversion from Truck to Rail: 

We conducted our own simulation of transportation policy scenarios on 
mode choice for freight shipments. Disaggregated data from the Freight 
Analysis Framework (FAF)[Footnote 70] was analyzed to obtain the 
distance traveled for shipments across commodity and truck types. Then 
this data from FAF, along with aggregated data on underlying 
assumptions, were used as inputs into the Intermodal Transportation 
and Inventory Cost Model (ITIC).[Footnote 71] This model estimates 
mode choices for each shipment under baseline conditions and various 
policy scenarios. See appendix IV for additional discussion of the 
simulations. 

We reviewed technical documentation associated with both of these 
models. We also conducted interviews with officials at DOT to better 
understand any data limitations or reliability issues with the model 
and data inputs. For more information see appendix IV. 

[End of section] 

Appendix II: International Case Study Summaries: The United Kingdom 
and Germany: 

The United Kingdom: 

Background: 

The United Kingdom's Department for Transport sets the strategic 
direction for the railways and Network Rail owns and operates 
Britain's rail infrastructure. Network Rail is a private corporation 
run by a board of directors and composed of approximately 100 members--
some rail industry stakeholders and some members of the general 
public. Freight and passenger operators pay access charges to Network 
Rail for access to the rail tracks. In the United Kingdom, freight and 
passenger rail share many of the same tracks. The system is open to 
competition through passenger rail franchises and through "open 
access" provisions for freight and other new passenger services. 

Transportation Project Planning Process: 

The Department for Transport's current approach to transportation 
policy planning emphasizes the assessment of a range of options driven 
by the desire to push transportation as a means to improve general 
economic performance, as well as environmental and societal goals. The 
Department for Transport plans and develops freight and intercity 
passenger rail projects based on a 5-year planning cycle, referred to 
as a Control Period. The last Control Period covering 2009-2014 
resulted in plans to invest £6.6 billion (at 2010/2011 prices) in 
capacity enhancements for the passenger and freight rail system and 
strategic rail freight network. The 5-year cycle is intended to 
identify, develop, and prioritize policy interventions and investment 
decisions, reflecting the long-term nature of the transportation 
sector. The Department for Transport publishes High Level Output 
Specifications and Statements of Funds Available, reflecting what 
types of rail projects the government wants to buy based on the 
government's transport goals and objectives and how much money it has 
to spend on those projects. Network Rail selects and implements 
projects to meet the High Level Output Specifications and outlines 
planned projects in a detailed delivery plan. All potential United 
Kingdom transportation projects are required to undergo standardized 
assessment processes to evaluate benefits and costs through the Web-
based Transport Appraisal Guidance, which includes guidance on benefit-
cost analysis for major transportation projects, including information 
on comparisons of proposed projects to alternatives, data sources for 
use in analyses, and methods for quantifying benefits and costs and 
performing sensitivity analysis. 

Selected Policy Tools: 

The Department for Transport has developed and implemented a range of 
policies to encourage a shift to rail transport. We explored some of 
these policies--in figure 5 below--during our site visits in the 
United Kingdom. 

Figure 5: Selected Polices to Benefit Intercity Passenger and Freight 
Rail in the United Kingdom: 

[Refer to PDF for image: illustrated table] 

Country: United Kingdom [map of Europe, depicting U.K]; 

Selected Policies: Recent and planned high-speed rail projects; 
Type of rail: Intercity passenger. 

Selected Policies: Mode shift revenue support scheme; 
Type of rail: Freight. 

Selected Policies: Freight facilities grants; 
Type of rail: Freight. 

Sources: GAO and Map Resources (map). 

[End of figure] 

Recent and planned high-speed rail projects (HS1 and HS2)--The Channel 
Tunnel Rail Link--referred to as HS1--is the United Kingdom portion of 
the route used by the Eurostar services from London to Paris and 
Brussels and was completed in 2007. The 109-kilometer Channel Tunnel 
Rail Link was the first major new railway to be constructed in the 
United Kingdom for over a century and the first high-speed railway. In 
2009, the government began to develop plans for a new dedicated high- 
speed passenger rail line--HS2. The current government plans to begin 
a formal consultation process in 2011 and hopes to begin construction 
on the new high-speed line by 2015. 

Mode shift revenue support scheme--This program provides funding to 
companies for operating costs associated with shipping via rail or 
inland water freight instead of road. It is intended to facilitate and 
support modal shift, as well as generating environmental and wider 
social benefits from having fewer freight shipments on Britain's roads. 

Freight facilities grants--These grants provide support for freight 
infrastructure capital projects such as rail sidings or loading and 
unloading equipment. Funding is granted on the principle that if the 
facilities were not provided, the freight in question would go by 
road. Applicants must predict the type and quantity of goods that will 
use the proposed facility and demonstrate that the freight facility 
will secure the removal of freight trucks from specific routes. The 
program has been available since the 1970s, and it has a long history 
of providing funding for capital infrastructure. 

Germany: 

Background: 

In Germany, the Federal Ministry of Transport, Building and Urban 
Development (Ministry of Transport) is responsible for financing the 
development and maintenance of the country's intercity passenger and 
freight rail network. Germany has the largest rail network in Europe, 
and both the intercity passenger and freight rail systems are open to 
competition. The majority of the rail system in Germany is managed by 
a single infrastructure provider--Deutsche Bahn.[Footnote 72] The 
German government provides Deutsche Bahn with approximately €3.9 
billion a year in investment grants for infrastructure renewal, 
upgrades, and new projects; freight and passenger operators pay access 
charges to Deutsche Bahn for access to the rail tracks. In addition to 
serving as the railway infrastructure provider, Deutsche Bahn also 
provides much of the intercity passenger and freight logistics service 
in Germany. Passenger and freight rail usually share the same track in 
Germany which, according to German transport officials, can enhance 
the efficiency of the network. However, sharing the same network also 
impacts the overall capacity available to accommodate new passenger or 
freight traffic. 

Transportation Project Planning Process: 

The Ministry of Transport develops a Federal Transport Infrastructure 
Master Plan approximately every 10 years to set the long-term 
strategic policy direction for both passenger and freight 
transportation. These infrastructure plans describe projects required 
to cope with the forecast traffic development. The goals and 
objectives of these long-term plans are then translated into 5-year 
plans--Federal Transport Infrastructure Action Plans--which are then 
used to develop new projects. After determining short-term 
transportation priorities and developing action plans intended to 
align with long-term goals, all potential rail projects undergo 
standardized assessment processes to evaluate benefits and costs. As 
the primary infrastructure manager for the rail network in Germany, 
Deutsche Bahn maintains rail data sets that allow officials to 
generate consistent estimates of project benefits and costs with 
confidence, facilitated by centralized data collection. The rail 
infrastructure planning process is currently under way, and officials 
at the Ministry of Transport have just reviewed requirement plans for 
rail infrastructures projects--a process that occurs every 5 years--in 
order to complete and release an updated Action Plan. 

Selected Policy Tools: 

Germany's Ministry of Transport has developed and implemented a range 
of policies that may encourage a shift to rail transport. We explored 
some of these policies--in figure 6 below--during our site visits in 
Germany. 

Figure 6: Selected Polices to Benefit Intercity Passenger and Freight 
Rail in Germany: 

[Refer to PDF for image: illustrated table] 

Country: Germany [map of Europe, depicting Germany]; 

Selected Policies: Upgrade and maintain high-speed rail network; 
Type of rail: Intercity passenger. 

Selected Policies: Vehicle mineral oil (fuel) tax; 
Type of rail: Intercity passenger; Freight. 

Selected Policies: Heavy Goods Vehicle tolls; 
Type of rail: Freight. 

Sources: GAO and Map Resources (map). 

[End of figure] 

Upgrade and maintain the rail network--The German government has 
committed to investing annually in projects to upgrade and renew the 
existing high-speed and passenger rail network. Each year, the German 
government invests approximately €3.9 billion to renew the existing 
rail infrastructure and to construct, upgrade, or extend rail 
infrastructure. 

Vehicle mineral oil (fuel) tax--Between 1999 and 2003, the German 
government began to implement routine, annual increases in the vehicle 
fuel tax for the explicit purpose of curbing car use and promoting the 
purchase of more fuel-efficient vehicles. Diesel is now taxed at 
approximately 47 euro cents a liter, and gas is taxed at 65 euro cents 
a liter, generating approximately €39 billion in revenue in 2009 for 
the general tax fund. 

Heavy Goods Vehicle (HGV) tolls--Germany implemented a distance-based 
HGV toll in 2005, in part to support an explicit goal of shifting a 
portion of freight traffic to rail. The policy generated approximately €
4.4 billion revenue in 2009, which was primarily used to maintain and 
upgrade the road network.[Footnote 73] This policy was viewed as 
imposing additional costs on the business community, and the new 
government has said it will not raise the toll rates or expand the tax 
to passenger vehicles in this legislative period. 

[End of section] 

Appendix III: HSIPR and TIGER Discretionary Grant Program Information: 

HSIPR: 

The American Reinvestment and Recovery Act of 2009 (Recovery Act) 
[Footnote 74] provided $8 billion to develop high-speed and intercity 
passenger rail service, funding the Passenger Rail Investments and 
Improvement Act (PRIIA), which was enacted in October 2008.[Footnote 
75] The funding made available is significantly more money than 
Congress provided to fund rail in recent years. The Federal Railroad 
Administration (FRA) launched the high-speed and intercity passenger 
rail (HSIPR) program in June 2009 with the issuance of a notice of 
funding availability and interim program guidance, which outlined the 
requirements and procedures for obtaining federal funds.[Footnote 76] 
Congress appropriated an additional $2.5 billion for high-speed rail 
for fiscal year 2010,[Footnote 77] and in January 2010 FRA announced 
the selection of 62 projects in 23 states and the District of Columbia. 

FRA allowed applicants to the HSIPR program to submit applications to 
be evaluated under four funding tracks.[Footnote 78] See table 4 below. 

Table 4: High-Speed Intercity Passenger Rail Program Funding Tracks: 

Track 1: Applications aimed at addressing the economic recovery goals 
of the Recovery Act through construction of ready-to-go intercity 
passenger rail projects, including projects to relieve congestion. 

Track 2: Applications that included projects either to develop new 
high-speed rail corridors and intercity passenger rail services or 
substantially upgrade existing corridor services, excluding intercity 
passenger rail congestion projects. 

Track 3: Applications that focused on service planning activities. 
These projects are aimed at establishing a pipeline of future high- 
speed rail and intercity passenger rail projects and service 
development programs by advancing planning activities for applicants 
at earlier stages of the development process. 

Track 4: Provides an alternative for projects that would otherwise fit 
under Track 1, but applicants must offer at least a 50% nonfederal 
share of financing. Applicants have up to 5 years (as opposed to 2 
years) to complete projects. 

Source: HSIPR Federal Register notice. 

[End of table] 

Applications were evaluated by technical evaluation panels against 
three categories of criteria: (1) public return on investment across 
categories of benefits including transportation benefits, economic 
recovery benefits, and other public benefits; (2) project success 
factors, such as project management approach and sustainability of 
benefits, as assessed by adequacy of engineering, proposed project 
schedule, National Environmental Policy Act compliance, and 
thoroughness of management plan; and (3) other attributes, such as 
timeliness of project completion. Projects were rated on a scale of 1 
point to 5 points, with 1 point being the lowest, and 5 points being 
the highest, based on the fulfillment of objectives for each separate 
criterion. 

Using the best available tools, applicants were required to include 
benefit and cost information for the following three general 
categories of benefits: 

* Transportation benefits, which include improved intercity passenger 
service, improved transportation network integration, and safety 
benefits; 

* Economic recovery, which includes preserving and creating jobs 
(particularly in economically distressed areas); and: 

* Other public benefits, such as environmental quality, energy 
efficiency, and livable communities. 

Final project selections were made by the FRA Administrator building 
upon the work of the technical evaluation panels and applying four 
selection criteria specified in the Federal Register notice: (1) 
region/location, including regional balance across the country and 
balance among large and small population centers; (2) innovation, 
including pursuit of new technology and promotion of domestic 
manufacturing; (3) partnerships, including multistate agreements; and 
(4) tracks and round timing, including project schedules and costs. 

TIGER: 

The Recovery Act also appropriated $1.5 billion for discretionary 
grants to be administered by DOT for capital investments in the 
nation's surface transportation infrastructure.[Footnote 79] These 
grants were available on a competitive basis to fund transportation 
projects that would preserve and create jobs and provide long-term 
benefits, as well as incorporate innovation and promote public-private 
or other partnership approaches. In making awards, the legislation 
required DOT to address several statutory priorities, including 
achieving an equitable geographic distribution of the funds, balancing 
the needs of urban and rural communities, prioritizing projects for 
which a TIGER grant would complete a package of funding, and others. 
[Footnote 80] In December 2009 Congress appropriated $600 million to 
DOT for a "TIGER II" discretionary grant program, which was similar to 
the TIGER program's structure and objectives.[Footnote 81] 

Eligible projects included highway or bridge projects, public 
transportation, passenger and freight rail projects, and port 
infrastructure projects. The TIGER program established three 
categories of project applications based on the amount of federal 
funding sought[Footnote 82] and three sets of criteria to determine 
grant awards in each project application category: 

* Primary selection criteria: Long-term outcomes, such as state of 
good repair, evidence of long-term benefits, livability, 
sustainability, safety, and job creation and economic stimulus. 

* Secondary selection criteria: Priority to projects that use 
innovative strategies to pursue long-term outcomes and those that 
demonstrate strong collaboration among a broad range of participants. 
Secondary selection criteria were weighted less than primary selection 
criteria in the application review process. 

* Program-specific criteria: Program-specific information was used as 
a tie breaker to differentiate between similar projects. This 
information was only applied to projects in the following categories: 
bridge replacement, transit projects, TIGER-TIFIA payment projects, 
and port infrastructure projects. 

[End of section] 

Appendix IV: Computer Simulations of Freight Diversion from Truck to 
Rail: 

In general, quantifying benefits that may be attained through rail can 
be challenging, in part, because of data limitations. In order to both 
estimate the extent to which freight shipments might be diverted from 
truck to rail under various scenarios and identify challenges related 
to making such estimates, we conducted simulations using a computer 
model developed by DOT. We sought to estimate the number of diverted 
truck freight shipments under scenarios that increased the price or 
decreased the speed of freight shipments by truck as compared with 
rail. 

ITIC Model: 

The Intermodal Transportation Inventory Cost (ITIC) model is a 
computer model for calculating the costs associated with shipping 
freight via alternative modes, namely truck and rail. The model can be 
used to perform policy analysis of issues concerning long-haul freight 
movement, such as diversion of freight shipments from truck to rail. 
[Footnote 83] DOT provides the ITIC model framework as a useful tool 
for ongoing policy studies, and shares the model, along with some 
internally developed data, for this purpose. We chose to use the ITIC 
model to simulate mode shift from truck to rail because of its federal 
origins and its direct applicability to freight shipments.[Footnote 84] 

The ITIC model--of which we used the highway freight to rail 
intermodal version--predicts diversion from truck to rail by assuming 
that shippers will select the mode of transportation with lower total 
shipment cost. The model replicates the decision-making trade-offs 
made by shippers in selecting which transportation mode to use for 
freight shipments. The model estimates the total cost--including both 
transportation and logistics costs--required to ship freight by both 
truck and rail for a given type of commodity and a given county-to- 
county route. Transportation costs include the costs associated with 
the actual movement of commodities, such as loading and unloading 
freight, and logistics costs represent a range of other costs, such as 
loss and damage of the freight, safety stock carrying cost, and 
capital cost on claims (see figure 8 for the components of these 
costs). 

In order to estimate diversions of freight shipments from truck to 
rail, the ITIC model runs in two steps. First, the model establishes a 
baseline that can be used for comparison against each of the simulated 
scenarios. To do this, the ITIC model requires input data on actual 
truck freight shipments that it uses to calculate total cost to ship 
each type of commodity for each county-to-county pair for both truck 
and rail. After generating a base case, diversion of freight from 
truck to rail can be estimated for various scenarios by changing the 
input assumptions to the model. As these assumptions are changed, the 
model reestimates the transportation and logistics costs for both 
truck and rail and determines whether these estimated changes have 
made rail a lower cost option for any of the shipments that were 
originally sent by truck. The model assumes that shipments will switch 
from truck to rail if the total cost for making a shipment by rail is 
lower than the total costs for making a shipment by truck. 

Reliability of Model Inputs: 

A lack of reliable data for a number of major ITIC model inputs at the 
national level prevented us from fully assessing the uncertainty 
associated with estimates of freight diversion from truck to rail. As 
a result, we are unable to report on the confidence levels of the 
results of our simulations. The ITIC model is based on 26 inputs (see 
table 6 for a complete list of ITIC model inputs). For our national 
analysis, empirical data were available for 9 of the inputs; 
accordingly, we had to rely on the preprogrammed model assumptions for 
the remaining 17 inputs.[Footnote 85] Using these 26 inputs, the model 
made 24 calculations (see table 7 for full list of ITIC model 
calculations), 22 of which relied on at least one of the model's 17 
default assumptions (see table 5 below). 

Table 5: Extent of Data and Assumptions Underlying Intermodal 
Transportation Inventory Cost Model Inputs and Calculations: 

ITIC model inputs: 
ITIC components: Assumptions; 
Number: 17. 

ITIC model inputs:
ITIC components: Data; 
Number: Total: 9. 

ITIC model inputs: Total; 
Number: 26. 

ITIC model calculations: 
ITIC components: Calculated from assumptions; 
Number: 22. 

ITIC model calculations: 
ITIC components: Calculated only from data; 
Number: Total: 2. 

ITIC model calculations: Total; 
Number: 24. 

Source: GAO analysis of ITIC model. 

[End of table] 

To determine whether the available data and model assumptions were 
reliable for our purposes, we considered some important factors for 
assessing data reliability, including their relevance, completeness, 
accuracy, validity, and consistency.[Footnote 86] We found that the 
data and the basis for assumptions used in the ITIC model vary in 
terms of the following factors. 

* Relevance: The 26 ITIC model inputs are relevant for the purposes of 
determining total transportation and logistics costs. These inputs 
have been shown to be conceptually important because they reflect 
economic theory underlying shipper choices, include a range of factors 
specified in the literature on freight shipments, and provide default 
assumptions based on theory and professional expertise. 

* Completeness: Completeness refers to the extent that relevant 
records are present and the fields in each record are populated 
appropriately. We were unable to obtain complete national data for 20 
ITIC model inputs. Of these 20 inputs, partial data were available for 
3.[Footnote 87] For the remaining 17 inputs, we were unable to obtain 
any empirical data and consequently relied on the default assumptions 
that are provided in the model itself. However, without a reliable 
source of available data against which to judge the accuracy and 
validity of these assumed values, we could not determine how much 
uncertainty the assumptions added to any estimates produced by the 
model. 

* Accuracy: Accuracy refers to the extent that recorded data reflect 
the actual underlying information. Of the 26 ITIC model inputs, we 
were unable to verify the accuracy for 20, including all 17 
assumptions, as well as available truck rate data and 2 inputs (weight 
per cubic foot and value per pound of each commodity group) provided 
by FRA. FRA officials stated that they originally generated these 
input values using empirical data, but were unable to provide 
documentation of their analysis. We were therefore unable to judge the 
accuracy of the resulting data, or the level of uncertainty associated 
with estimates produced from FRA's data. 

* Validity: Validity refers to an input correctly representing what it 
is supposed to measure. Of the 26 ITIC model inputs, we were unable to 
verify the validity for 18, including all 17 default assumptions and 
available truck rate data. For the latter, we used the source of data 
previously used by the Federal Highway Administration, a proprietary 
collection of truck rates from 2006 for 120 city pairs. Documentation 
of the collection methods was unavailable, and we were not able to 
validate or assess the data for reliability, and thus could not 
estimate the uncertainty associated with per-mile truck rates. Because 
this value is a primary driver of total transportation and logistics 
costs, the uncertain reliability of truck rate data was a major 
limitation to using the model's estimates. 

* Consistency: Consistency is a subcategory of accuracy and refers to 
the need to obtain and use data that are clear and well defined enough 
to yield similar results in similar analyses. Of the 26 ITIC model 
inputs, we identified consistency issues for 7 data inputs. For 
example, truck rate data were collected in 2006, and data on truck 
shipments were from 2002, making it problematic to compare these 
figures. For the other 6 inputs, we encountered different levels of 
data aggregation for data that we had otherwise deemed reliable. For 
example, the FAF collects regional data, while the FRA lookup tables 
for certain truck and rail origin and destination miles are collected 
at a county level. In order to use both sources of data, the FAF data 
had to be disaggregated for use at the county level, and our 
disaggregation method adds additional uncertainty to our estimates. 

Reliability of Model Estimates: 

In order to better understand the impact of uncertainty in the ITIC 
model's estimates caused by use of assumptions and data of 
questionable reliability, we examined how the model's estimates change 
when key underlying assumptions were varied. In particular, we used 
the model to simulate the impact that a 50-cent increase in per-mile 
truck rates would have on vehicle miles traveled (VMT) under two 
scenarios: the first scenario uses the model's default values for all 
assumptions, including truck speeds of 50 miles per hour, freight loss 
and damage as a percentage of gross revenue equal to 0.07 percent, and 
a reliability factor equal to 0.4;[Footnote 88] the second scenario 
changes these three assumptions to respective values of 40 miles per 
hour, 0.10 percent freight loss and damage, and reliability factor 
equal to 0.5. Each of these changes creates a higher total cost for 
trucks, potentially leading the model to predict some additional 
diversion to rail. However, for these sensitivity analyses, we are 
more concerned with the impact of changing truck rates under the 
alternative scenarios than we are with the individual impacts of 
changing assumptions. 

For a 50-cent increase (approximately 30 percent of per mile truck 
rates) in the first scenario, the model estimates a reduction in VMT 
of about 1.02 percent. For the same reduction in rates in the second 
scenario, the model estimates a reduction in VMT of about 1.04 
percent. Figure 7 shows the estimated percentage reduction in VMT 
associated with increased per-mile truck rates for the two scenarios. 
Under either scenario, the impact of increasing per-mile truck rates 
by approximately 30 percent results in decreases of roughly 1 percent 
of VMT. This result suggests that we can have some degree of 
confidence that the model will consistently predict that changing per-
mile truck rates will have a minor impact on total VMT traveled. 

Figure 7: Impact of Increased Per-Mile Truck Rates on Vehicle Miles 
Traveled (VMT) by Trucks under Two Scenarios: 

[Refer to PDF for image: multiple line graph] 

Truck rate (dollars per mile): $1.55; 
Percentage reduction in VMT: Defaults: 0%; 	
Percentage reduction in VMT: Modified assumptions: 0%. 

Truck rate (dollars per mile): $1.8; 
Percentage reduction in VMT: Defaults: -0.11%; 	
Percentage reduction in VMT: Modified assumptions: -0.12%; 

Truck rate (dollars per mile): $2.05; 
Percentage reduction in VMT: Defaults: -0.29%; 
Percentage reduction in VMT: Modified assumptions: -0.32%. 

Source: GAO. 

[End of figure] 

In spite of the results of our two scenarios, the estimates of VMT 
diversion based on the ITIC model are still subject to limitations. As 
a result, these estimates are only suggestive, rather than conclusive, 
of the impact that an increase in per-mile truck rates might have on 
VMT reduction in actual policy scenarios. First, the issues of 
completeness, accuracy, validity, and consistency of our data 
negatively impact their reliability and increase the uncertainty of 
our estimates. Second, because of resource constraints, our analysis 
only varies 3 of the 17 default ITIC model assumptions and considers 
only one change in these values, instead of varying a larger number of 
assumptions for a wider range of scenarios (see table 6 for a full 
list of assumptions). Therefore, we cannot conclude that the model 
results are robust to all plausible variations in all of the model 
assumptions. Therefore, while the results of our simulation suggest 
that a 50-cent increase in per-mile truck rates would have a limited 
impact on diversion of freight from truck to rail in the short-term, 
we do not have enough confidence in the quality of data inputs to make 
precise predictions that would be reliable enough to inform 
policymaking decisions. Reliable data for model inputs would be 
necessary in order to produce estimates of changes in VMT with 
confidence. 

Implications for Future Simulations: 

Sufficiently reliable data were not readily available for producing 
national estimates of mode shift under specific policy scenarios. As a 
result, it was necessary to rely on assumptions and data of 
undetermined reliability when conducting national simulations, which 
may result in unreliable estimates of freight diversion and an 
inability to fully quantify the uncertainty of the estimates produced. 
Our simulations suggest that a large increase (approximately 30 
percent) in per-mile truck rates results could result in a relatively 
small (approximately 1 percent) decrease in VMT, even when multiple 
assumptions related to truck freight cost are changed. Despite this, 
limitations in the reliability of our data and ability to conduct 
further sensitivity analyses reduce our confidence in these estimates. 
While reliable data may be available at state and local levels for use 
in simulations of mode shift, the importance of communicating the 
uncertainty underlying projections to decision makers remains. 
Assessments of data reliability and assumptions, along with 
quantification of uncertainty, are necessary to enable the comparison 
of the risk of inaccurate results against the potential value of the 
estimates produced and would improve decision makers ability to 
reliably interpret these estimates and compare estimates across 
projects. In order to accomplish this and produce reliable estimates 
of freight diversion and uncertainty at the national level, it would 
be necessary to obtain complete, accurate, and valid data that are 
collected consistently for the model's relevant inputs. 

Figure 8: Intermodal Transportation and Inventory Cost (ITIC) Model 
Process: 

[Refer to PDF for image: illustrated flow chart] 

Assumptions: 
Service percent; 
Reliability; 
Wait time; 
Inventory carrying costs percentage; 
Truck mph; 
Dray mph; 
Rail mph; 
Pickup charges per shipment; 
Delivery charges per shipment; 
Pickup charge per mile; 
Delivery charge per mile; 
Inventory carrying costs percentage; 
Interest; 
Dunnage; 
Order cost; 
Hourly wage; 
Load and unload hours; 
Loss and damage as a percentage of gross revenue. 

From generally reliable data: 
Line-haul miles; 
Rail miles; 
Pickup miles; 
Delivery miles; 
Size–maximum weight; 
Weight. 

From questionable data: 
Dollars per pound; 
Pounds per cubic foot; 
Truck rate per mile. 

Calculated from questionable data: 
Shipments; 
Mileage costs (truck). 

Calculated from assumptions: 
Standard deviation of lead time; 
Mean lead time; 
Standard deviation of demand during lead time; 
Mean demand during lead time; 
Alpha; 
Beta; 
Reorder point; 
Safety stock; 
Safety stock carrying costs; 
Transit time; 
Drayage costs (rail); 
Cycle stock carrying cost; 
Load and unload cost; 
Order costs; 
Mileage costs (rail); 
Rail rate per cwt; 
Transportation costs; 
Transit stock carrying cost; 
Capital cost on claims; 
Loss and damage claims; 
Logistics costs; 
Total costs. 

Source: GAO. 

[End of figure] 

Table 6: Inputs to the ITIC Model: 

Input: Truck rate per mile; 
Source: Proprietary data; 
Reliability: Undetermined; 
Definition: Per-mile cost of using a truck for shipping (2006). 

Input: Line haul miles; 
Source: FRA lookup table; 
Reliability: Reliable for our purposes; 
Definition: Distance traveled by truck. 

Input: Pickup miles; 
Source: FRA lookup table; 
Reliability: Reliable for our purposes; 
Definition: Length of rail drayage at origin. 

Input: Delivery miles; 
Source: FRA lookup table; 
Reliability: Reliable for our purposes; 
Definition: Length of rail drayage at destination. 

Input: Rail miles; 
Source: FRA lookup table; 
Reliability: Reliable for our purposes; 
Definition: Distance traveled on rail. 

Input: Dollars per pound; 
Source: FRA commodity attribute table; 
Reliability: Undetermined; 
Definition: Average value of a given commodity class. 

Input: Pounds per cubic foot; 
Source: FRA commodity attribute table; 
Reliability: Undetermined; 
Definition: Weight per cubic foot of a given commodity class. 

Input: Commodity type; 
Source: Freight Analysis Framework (FAF); 
Reliable for our purposes; 
Definition: Two-digit STGC code for commodity being carried in the 
shipment. 

Input: Trailer size/max weight factor; 
Source: FRA; 
Reliability: Reliable for our purposes; 
Definition: Factor for ensuring that weight per shipment is not over 
legal limits or cubic footage of truck trailer or COFC. 

Input: Weight; 
Source: FAF; 
Reliability: Reliable for our purposes; 
Definition: Total annual weight of a given commodity transported 
between regions (2002). 

Input: Interest; 
Source: Assumption; 
Reliability: Undetermined; 
Definition: Cost of capital during transit and during loss and damage 
claims. 

Input: Inventory carrying cost percentage; 
Source: Assumption; 
Reliability: Undetermined; 
Definition: Costs associated with possession of a commodity, including 
capital cost, insurance, taxes, obsolescence, pilferage, transfer, 
handling, and storage. 

Input: Load and unload hours; 
Source: Assumption; 
Reliability: Undetermined; 
Definition: Amount of time needed to load or unload a truck trailer or 
COFC. 

Input: Hourly wage; 
Source: Assumption; 
Undetermined; 
Amount paid to workers loading and unloading a truck trailer or 
container on a flat car (COFC). 

Input: Pickup charges per shipment; 
Source: Assumption; 
Reliability: Undetermined; 
Definition: Flat fee charged to pick up shipments by rail. 

Input: Delivery charges per shipment; 
Source: Assumption; 
Reliability: Undetermined; 
Definition: Flat fee charged to deliver shipments by rail. 

Input: Pickup charge per mile; 
Source: Assumption; 
Reliability: Undetermined; 
Definition: Per-mile charge for rail drayage over 30 miles at origin. 

Input: Delivery charge per mile; 
Source: Assumption; 
Reliability: Undetermined; 
Definition: Per-mile charge for rail drayage over 30 miles at 
destination. 

Input: Reliability; 
Source: Assumption; 
Reliability: Undetermined; 
Definition: A factor used to represent the skewness of the transit 
time distribution for truck and rail to represent likelihood that 
transit time will be the predicted value. 

Input: Loss and damage as a percentage of gross revenue; 
Source: Assumption; 
Reliability: Undetermined; 
Definition: Ratio of loss and damage costs to commodities over the 
gross revenue from shipping. 

Input: Order cost; 
Source: Assumption; 
Reliability: Undetermined; 
Definition: Cost of placing an order to be shipped. 

Input: Dunnage; 
Source: Assumption; 
Reliability: Undetermined; 
Definition: extra charge (assumed $50) to rail orders. 

Input: Truck mph; 
Source: Assumption; 
Reliability: Undetermined; 
Definition: Average truck speed. 

Input: Rail mph; 
Source: Assumption; 
Reliability: Undetermined; 
Definition: Average rail speed. 

Input: Dray mph; 
Source: Assumption; 
Reliability: Undetermined; 
Definition: Average speed of drayage to/from rail. 

Input: Wait time; 
Source: Assumption; 
Reliability: Undetermined; 
Definition: Number of days before a shipment can be transported. 

Input: Service percent; 
Source: Assumption; 
Reliability: Undetermined; 
Definition: Probability of no stock out (inventory) during the 
replenishment cycle. 

Disaggregation: Origin region; 
Source: FAF; 
Reliability: Reliable for our purposes; 
Definition: FAF-defined regions. 

Disaggregation: Destination region; 
Source: FAF; 
Reliability: Reliable for our purposes; 
Definition: FAF-defined regions. 

Disaggregation: County establishments; 
Source: QCEW; 
Reliability: Reliable for our purposes; 
Definition: Proxy for the share of economic activity an individual 
county within a FAF region is responsible for (2006). 

[End of table] 

Source: GAO analysis of ITIC model. 

Table 7: ITIC Calculations: 

Calculated values: Shipments; 
Definitions: Number of shipments per year needed to transport total 
annual weight. 

Calculated values: Transit time; 
Definitions: Average amount of time (in days) from origin to 
destination. 

Calculated values: Mean lead time; 
Definitions: Average amount of time in advance a shipper needs to 
order to receive a commodity on time. 

Calculated values: Standard deviation of lead time; 
Definitions: Error associated with average lead time. 

Calculated values: Mean demand during lead time; 
Definitions: Average demand (in tons) of a commodity during lead time. 

Calculated values: Standard deviation of demand during lead time; 
Definitions: Error associated with average demand during lead time. 

Calculated values: Alpha; 
Definitions: Measure of the variance of demand during lead time. 

Calculated values: Beta; 
Definitions: Measure of the skewness of the distribution of demand 
during lead time. 

Calculated values: Reorder point; 
Definitions: Amount of commodity (in tons) remaining in a shippers 
stock when they should reorder. 

Calculated values: Safety stock; 
Definitions: Amount of commodity (in pounds) a shippers needs to 
maintain in stock to insure they won't stock out. 

Calculated values: Rail cost per one hundred pounds; 
Definitions: Cost of transporting 100 pounds of a commodity by rail. 

Calculated values: Cycle stock carrying cost; 
Definitions: Cost of holding inventory of a commodity. 

Calculated values: Safety stock carrying cost; 
Definitions: Cost associated with carrying additional inventory of a 
commodity to prevent stock out. 

Calculated values: Capital cost on claims; 
Definitions: Cost incurred through interest paid while filling loss 
and damage claims. 

Calculated values: Loss and damage claims; 
Definitions: Cost incurred through loss and damage to commodities 
during transit. 

Calculated values: Order costs; 
Definitions: Total cost (order cost plus dunnage) to place an order. 

Calculated values: In-transit stock carrying cost; 
Definitions: Cost incurred through interest accrued while stock is in 
transit. 

Calculated values: Mileage costs: truck; 
Definitions: Cost for a shipment to move from origin to destination by 
truck. 

Calculated values: Mileage costs: rail; 
Definitions: Cost for a shipment to move from origin rail junction to 
destination rail junction. 

Calculated values: Drayage costs: rail; 
Definitions: Cost for a shipment to move between origin/destination 
and rail junctions. 

Calculated values: Load and unload cost; 
Definitions: Cost to load and unload a truck or container on a flat 
car. 

Calculated values: Logistics costs; 
Definitions: Costs associated with possession of the commodity. 

Calculated values: Transportation costs; 
Definitions: Costs associated with movement of the commodity. 

Calculated values: Total costs; 
Definitions: Sum of transportation and logistics costs. 

Source: GAO analysis of ITIC model. 

[End of table] 

[End of section] 

Appendix V: GAO Contact and Staff Acknowledgments: 

GAO Contact: 

Susan Fleming, (202) 512-4431, flemings@gao.gov: 

Staff Acknowledgments: 

In addition to the individual named above, Andrew Von Ah, Assistant 
Director; Mark Braza; Caroline Epley; Tim Guinane; Bert Japikse; 
Delwen Jones; Brooke Leary; Steven Putansu; Max Sawicky; Sharon Silas; 
and Maria Wallace made key contributions to this report. 

[End of section] 

Footnotes: 

[1] Data are based on a review of 439 urban areas in the United States 
and includes both highways and principal arterials. Yearly delay per 
auto commuter is the extra time spent traveling at congested speeds 
rather than free-flow speeds by private vehicle drivers and passengers 
who typically travel in peak periods. The value of travel time delay 
is estimated at $16 per hour of person travel and $106 per hour of 
truck time. Texas Transportation Institute Urban Mobility Report 2010. 

[2] Forecast is based on an analysis of the Commodity Flow Survey 
(CFS) which is developed in partnership by the Census Bureau and the 
Bureau of Transportation Statistics (BTS). 

[3] The primary greenhouse gasses produced by the transportation 
sector are carbon dioxide (CO2), methane (CH4), nitrous oxide (N20), 
and hydrofluorocarbon (HFC). 

[4] While outside the scope of this study--the potential benefits of 
rail are not solely limited to emissions, congestion, and economic 
development benefits that result from modal shift. Improved or 
expanded rail service may simply increase the desire and ability of 
people to travel or engage in trade, and to enjoy the subsequent 
benefits that flow from that enhancement in mobility and access. 

[5] Pub. L. No. 110-432, Div. B, 122 Stat. 4907 (October 2008). 

[6] 49 U.S.C. § 26106. 

[7] 49 U.S.C. § 24105. 

[8] 49 U.S.C. § 24402. 

[9] Pub. L. No. 111-5, Title XII, 123 Stat. 115 (2009). 

[10] 74 Fed. Reg. 29900 (June 23, 2009). 

[11] OMB, Guidelines and Discount Rates for Benefit-Cost Analysis of 
Federal Programs, Circular No. A-94 (Oct. 29, 1992), as revised 
through Dec. 8, 2009. OMB, Principles for Federal Infrastructure 
Investments, Exec. Order No. 12893 (Jan. 26, 1994). Regulatory 
Analysis, Circular No. A-4 (Sept. 17, 2003). 

[12] For example, the European rail system is focused primarily on 
passenger operations, while the U.S. rail network is predominantly a 
freight transport system. 

[13] As defined by revenue, for 2009, Class I railroads are freight 
rail carriers having annual operating revenues of $379 million or 
more. [49 C.F.R. 1201-1]. The railroads include CSX Transportation 
(CSX), BNSF Railway Company (BNSF), Union Pacific Railroad Company 
(Union Pacific), Norfolk Southern, Kansas City Southern Railway 
Company, Canadian National Railway, and Canadian Pacific Railway. 
Regional and short line railroads are medium-sized and small 
railroads, respectively, and are categorized based on operating 
revenues and mileage. Generally, for 2009, regional railroads are 
Class II railroads (carrier having annual operating revenues greater 
than $30 million but less than 379 million) and short line railroads 
are Class III railroads (carriers having annual operating revenues of 
$30 million or less). 

[14] 49 U.S.C. § 24308. 

[15] An externality is an unintended side effect (negative or 
positive) of an activity of one individual or firm on the well-being 
of others. 

[16] For 2008, HFCs accounted for 3 percent, and CH4 and N20 together 
accounted for about 1.5 percent of the transportation total greenhouse 
gas emissions. N20 and CH4 gasses are released during fuel 
consumption, although in much smaller quantities than CO2, and are 
also affected by vehicle emissions control technologies. U.S. DOT, 
Transportation's Role in Reducing U.S. Greenhouse Gas Emissions, 
volume 1, Synthesis Report, P. 2-5, April 2010. 

[17] Data are based on "tailpipe" emissions and do not include other 
processes that also produce additional greenhouse gas emissions. These 
include the production and distribution of fuel, the manufacture of 
vehicles, and the construction and maintenance of transportation 
infrastructure. These supporting processes--known as the fuel, vehicle 
manufacture, and infrastructure cycles--generally are not included in 
U.S. transportation sector greenhouse gas estimates. 

[18] GAO, Surface Freight Transportation: A Comparison of the Costs of 
Road, Rail, and Waterways Freight Shipments That Are Not Passed on to 
Consumers, [hyperlink, http://www.gao.gov/products/GAO-11-134] 
(Washington D.C.: Jan. 26, 2011). Estimates are based on the most 
current data available. Estimated emissions were obtained directly 
from EPA and are based on the current MOVES2010 model for estimating 
on-road vehicle emissions. Estimates assume that nearly all on-road 
diesel emissions are freight-related, and 15 percent of gasoline 
powered vehicle emissions are freight-related. 

[19] GAO, Nextgen Air Transportation System: FAA's Metrics Can Be Used 
to Report on Status of Individual Programs, but Not of Overall NextGen 
Implementation or Outcomes, [hyperlink, 
http://www.gao.gov/products/GAO-10-629] (Washington, D.C.: July 27, 
2010). 

[20] Estimates are in 2010 dollars. To obtain an estimate in accident 
costs we included the number of fatalities multiplied by the latest 
value for human life used by DOT in guidance for its own analysts, and 
then assumed that carriers are already compensated for 50 percent of 
those costs. The economic costs of transportation accidents reflect 
the value assigned to the loss of a human life and the reduced 
productive life and pain and suffering related to serious injuries. 
[hyperlink, http://www.gao.gov/products/GAO-11-134]. 

[21] A recent legislative proposal has put forth potential policy 
goals for transportation that include such things as reducing delays, 
improving safety, reducing greenhouse gas emissions, and shifting 10 
percent of freight traffic in the United States off of highways and 
onto other modes. See S.1036, 11TH Cong. (2009). 

[22] It is difficult to assess whether the benefits associated with a 
policy that seeks to shift traffic to rail outweigh the various costs 
associated with these policies. In addition, there are also costs 
associated with alternative approaches that may affect which one or 
combination of policies would be most desirable for a given situation. 

[23] Pub. L. No. 110-432, Div. B, 122 Stat. 4907 (Oct. 16, 2008). 

[24] In addition to these responsibilities, FRA is also responsible 
for developing a national rail plan. The agency also has 
responsibility for railroad safety oversight, providing operational 
and capital grants to Amtrak, and approval for Railroad Rehabilitation 
and Improvement Financing loans and Rail Line Relocation and 
Improvement Capital Grants under the Transportation Infrastructure 
Finance and Innovation Act (TIFIA) (see 23 U.S.C. chapter 6) and 
Railroad Rehabilitation and Improvement Financing (RRIF) (at 45 U.S.C. 
chapter 17) programs. 

[25] Additional funding was provided through other FY 2009 and FY 2010 
appropriations to DOT. 

[26] Pub. L. No. 111-5, Title XII, 123 Stat. 115 (2009). A later, 
second round, known as TIGER II, was authorized and funds appropriated 
by the Consolidated Appropriations Act, 2010, Pub. L. No.111-117, Div. 
A, Title I, 123 Stat 3034, 3036 (Dec. 16, 2009). 

[27] In addition to the TIGER and HSIPR programs, other DOT programs 
that provide investment in rail projects also consider information on 
project benefits and costs as part of their application processes. The 
TIFIA and RRIF programs allow applicants to include information on 
economic, environmental, and safety benefits in their applications. 
However, neither program provides applicants with specific 
requirements for assessment of potential benefits and costs. 

[28] The benefits associated with policies to address external costs 
of transportation activities may include reductions in pollution, 
congestion, and improvements in safety (reducing accidents). The 
policies may also affect economic activity such as by increasing 
construction-related jobs. 

[29] For the purposes of this report, we use the term "assessment of 
benefits and costs" to mean a general evaluation of benefits and costs 
that may encompass a variety of types of analyses and "benefit-cost 
analysis" refers to a formalized analysis as it is strictly defined. 

[30] GAO, Highway And Transit Investments: Options for Improving 
Information on Projects' Benefits and Costs and Increasing 
Accountability for Results, [hyperlink, 
http://www.gao.gov/products/GAO-05-172] (Washington, D.C.: Jan. 24, 
2005) and GAO, Surface Transportation: Many Factors Affect Investment 
Decisions, [hyperlink, http://www.gao.gov/products/GAO-04-744] 
(Washington, D.C.: June 30, 2004). 

[31] See previous footnotes 24 and 27 for additional information on 
TIFIA and RRIF. 

[32] The Federal-Aid Highway Program provides federal financial 
resources and technical assistance to state and local governments for 
constructing, preserving, and improving the National Highway System. 
Funding is distributed to states through annual apportionments 
established by statutory formulas. 

[33] OMB Circular No. A-94 recognizes that "estimates of benefits and 
costs are typically uncertain because of imprecision in both 
underlying data and modeling assumptions." The type of information 
that would help decision makers understand the level of uncertainty 
associated with a benefit-cost analysis would include the key sources 
of uncertainty, the expected value estimates of outcomes, the 
sensitivity of results to important sources of uncertainty; and where 
possible, the probability distributions of benefits, costs, and net 
benefits. 

[34] GAO, Public Transportation: Improvements Are Needed to More Fully 
Assess Predicted Impacts of New Starts Projects, GAO-08-844 
(Washington, D.C.: July 25, 2008). 

[35] More generally, infrastructure policies should be assessed with 
respect to benefits and costs, as per Exec. Order No. 12893 and 
federal guidance. 

[36] [hyperlink, http://www.gao.gov/products/GAO-11-134]. 

[37] According to Amtrak officials, the Northeast Corridor has 
experienced a 37 percent increase in ridership between Washington, 
D.C. and New York and 20 percent between New York and Boston over the 
past 10 years. 

[38] A recently released report explored the relative ability of 
regional corridors to attract passengers based on factors that have 
contributed to rail ridership in other systems around the world. Petra 
Todorovich and Yoav Hagler, America 2050, High Speed Rail in America, 
January 2011. 

[39] DOT, National Rail Plan: Moving Forward, Progress Report, 
September 2010. 

[40] The Intermodal Transportation and Inventory Cost Model (ITIC) is 
a computer model for calculating the costs associated with shipping 
freight via alternative modes, namely truck and rail. The model can be 
used to perform policy analysis of issues concerning long-haul freight 
movement, such as diversion of freight shipments from truck to rail. 
DOT provides the ITIC model framework as a useful tool for ongoing 
policy studies, and shares the model, along with some internally 
developed data, for this purpose. We chose to use the ITIC model to 
simulate mode shift from truck to rail because of its federal origins 
and its direct applicability to freight shipments. 

[41] The model has 17 default assumptions. Because of resource 
constraints, our analysis only varied 3 of these assumptions and 
considered only one change in these values, instead of varying a 
larger number of assumptions for a wider range of scenarios. 
Therefore, we cannot conclude that the model results are robust to all 
plausible variations in all of the model assumptions. 

[42] Amtrak officials noted that dedicated high-speed rail lines make 
up a very small portion of worldwide rail mileage. 

[43] Cited from the American Society of Mechanical Engineers in 
"Transportation Invest in America: Freight-Rail Bottom Line Report," 
American Association of State Highway and Transportation Officials 
(2003). 

[44] Rail fuel efficiency was calculated in ton-miles per gallon to 
move commodity; truck fuel efficiency is calculated in lading ton-
miles per gallon. Federal Railroad Administration, "Comparative 
Evaluation of Rail and Truck Fuel Efficiency on Competitive Corridors" 
(Nov. 19, 2009). 

[45] Amtrak's relative fuel efficiency advantage is based on the 
available data in the Department of Energy's Transportation Energy 
Data Book [hyperlink, http://cta.ornl.gov/data/Index.shtml], table 
2.12. 

[46] For example, a major rail corridor for high-value, time-sensitive 
container freight exists between Los Angeles and Chicago. 

[47] According to Amtrak officials, intercity passenger trains also 
carry more passengers than the typical aircraft. 

[48] Freight moved by water between 2003 and 2007 averaged only .01 
fatalities per billion ton-miles. GAO has utilized ton-miles data from 
FHWA's Freight Analysis Framework (FAF)3 for these calculations, while 
the Bureau of Transportation Statistics uses a different estimate of 
ton-miles. 

[49] Certain effects, sometimes referred to as wider economic impacts, 
of investments in transportation infrastructure may not be captured in 
standard benefit-cost analysis. These impacts may include effects 
related to returns to scale and agglomeration. Because markets are 
often not perfect, such wider economic impacts--both positive and 
negative--may result from transportation investments. 

[50] Other studies have shown varying potential economic impacts. For 
example, a study of the Trans-European Transport Network suggested 
that it would not change regional GDP by more than 2 percent. 

[51] The European intercity passenger and freight rail systems are 
very different in size, structure, and scope than the U.S. rail 
system. For example, the European rail system is focused primarily on 
passenger operations, while the U.S. rail network is predominately a 
freight transport system. While the systems differ, the experiences of 
countries we visited, such as the United Kingdom and Germany provide 
illustrative examples of other countries experiences with policy tools 
that provide incentives to shift traffic to rail. 

[52] We did not analyze the costs associated with trucking companies' 
response to the HGV policy. Therefore, we cannot determine whether the 
costs associated with purchasing and transitioning to a more fuel- 
efficient fleet outweigh the policy's environmental and other 
benefits, including those from the increased fuel efficiency of the 
trucking fleet. 

[53] Allan Woodburn, "Evaluation of Rail Freight Facilities Grant 
Funding in Britain," School of Architecture and the Built Environment, 
Transport Reviews, 27 (3), pp. 311-326, May 2007. 

[54] 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). 

[55] Chris Nash, "Enhancing the Cost Benefit Analysis of High Speed 
Rail" (paper presented at the California High Speed Rail Symposium, 
Berkeley, Calif., Dec. 3, 2010). 

[56] We selected a nongeneralizable random sample of 20 applications 
from each program that included components of intercity passenger or 
freight rail and assessed the benefit and cost information contained 
in the applications based on OMB guidelines for benefit-cost analysis, 
with input from GAO economists and methodologists. For more 
information on the methodology of our study assessment, see appendix I. 

[57] FRA allowed applicants to the HSIPR program to submit 
applications under four different funding "Tracks." The pool of HSIPR 
applications from which we randomly selected projects for review were 
from Track 2, which included applicants evaluated under PRIIA §§301 
and 501, i.e., 49 U.S.C. §§ 24402 and 26106, which authorize grants to 
support intercity passenger rail service and development of high-speed 
intercity rail systems, respectively, excluding intercity passenger 
rail congestion projects and including only projects using Recovery 
Act funding. 74 Fed. Reg. 29909. Except as otherwise stated, our 
references to HSIPR in this portion of this report are to HSIPR Track 
2 as defined by FRA. See appendix III for more detail. 

[58] Our study assessment was limited to applications to TIGER and 
HSIPR that were required to include information on project benefits 
and costs. 

[59] Of the approximately 1,450 applications DOT received for the 
TIGER program, DOT officials selected 166 to be forwarded to review 
teams for additional consideration. These applications were selected 
based on criteria such as project readiness and potential for job 
creation. The benefit and cost information contained in these 166 
applications was reviewed by a team of DOT economists, who rated each 
evaluation for adequacy and value. For more information on these 
ratings, see below. 

[60] Benefits and costs expected to occur in future years are 
discounted to account for the time value of money. In general, 
discounting gives relatively less weight to benefits and costs 
expected to occur in the future. Not discounting or using an 
inappropriate discount rate can affect the results of a benefit-cost 
analysis. OMB provides guidance on choosing appropriate discount rates 
for different types of investments and recommends both 3 percent and 7 
percent discount rates for benefit-cost analyses of proposed 
investments. DOT asked applicants to the TIGER program to discount 
future benefits and costs using a discount rate of 7 percent and 
permitted them to provide an alternative analysis using a discount 
rate of 3 percent. However, HSIPR applicants were not required to 
perform benefit-cost analysis and were not provided information on 
discounting in the Federal Register notice for the program. 

[61] It is important to note that DOT did not specifically refer HSIPR 
applicants to this guidance. However, TIGER applicants were directed 
to this guidance through the federal "Notice of Funding Availability 
for Supplemental Discretionary Grants for Capital Investments in 
Surface Transportation Infrastructure" under the American Recovery and 
Reinvestment Act, 74 Fed. Reg. 28755 (June 17, 2009), which also 
directed applicants toward specific values to apply in assessing some 
categories of benefits. 

[62] [hyperlink, http://www.gao.gov/products/GAO-09-317]. 

[63] Economic research indicates that the value associated with 
reduction in greenhouse gas emissions can vary substantially depending 
on factors such as assumptions about future economic growth and 
discount rates. 

[64] We identified TIGER applications for projects that contained rail 
elements. Applications included those for projects that were rail-
only, as well as those that were multimodal in nature and included 
rail infrastructure improvements. Of these rail-related applications, 
DOT economists assessed the "usefulness" of benefit-cost information 
only for those applications that were forwarded by initial review 
teams for additional consideration. 

[65] DOT economists grouped benefit-cost analyses submitted by TIGER 
applicants into four categories of usefulness: (1) very useful 
assessments quantified and monetized the full range of costs and 
benefits for which such measures are reasonably available and provided 
a high degree of confidence that the benefits of the project will 
exceed the project's costs, (2) useful assessments quantified and 
monetized expected benefits and costs with some gaps and provided a 
sufficient degree of confidence that benefits of the project will 
exceed the project's costs, (3) marginally useful assessments had 
significant gaps in their analysis of project benefits and costs and 
were those for which DOT was uncertain whether the benefits of the 
project will exceed the project's costs, and (4) nonuseful assessments 
did not adequately quantify and monetize benefits and costs, did not 
provide sufficient confidence that the benefits of the project will 
exceed the project's costs, and demonstrated an unreasonable absence 
of data and analysis. 

[66] As mentioned earlier and discussed in more detail in appendix 
III, Track 2 applicants were selected under PRIIA §§301 and 501, i.e., 
49 U.S.C. §§ 24402 and 26106, which in turn authorized grants to 
support intercity passenger rail capital assistance and development of 
high-speed intercity rail systems, respectively, using Recovery Act 
funding, but excluding Track 1 projects. Track 1 included Recovery Act 
projects authorized under PRIIA §§301 (intercity passenger rail 
capital assistance projects) or 302 (projects to address intercity 
passenger rail congestion), imposed tighter time frames, but allowed 
applications from a broader range of applicants, including groups of 
states, public rail service providers, and entities established under 
Interstate Compacts. 49 U.S.C. §§ 24402, 24105. 74 Fed. Reg. 29900, 
29908-29917. 

[67] DOT announced the availability of $600 million in federal 
discretionary grant funding for transportation projects through the 
TIGER II program in June 2010 and announced TIGER II recipients in 
October 2010. 

[68] Notice of Funding Availability for the Department of 
Transportation's National Infrastructure Investments Under the 
Transportation, Housing and Urban Development, and Related Agencies 
Appropriations Act for 2010, 75 Fed. Reg. 30460 (June 1, 2010). 

[69] Life-cycle cost analysis can be used for the consideration of 
certain transportation investment decisions. In life-cycle cost 
analysis, all the relevant costs that occur throughout the life of a 
proposed project, not just the originating expenditures, are included. 
Costs accounted for in life-cycle cost analysis include the effects of 
construction and maintenance activities on users. 

[70] This is a data set maintained by the Federal Highway 
Administration (FHWA) that estimates commodity flows and related 
freight transportation activity among states, sub-state regions, and 
major international gateways. See [hyperlink, 
http://ops.fhwa.dot.gov/freight/freight_analysis/faf/index.htm]. FAF 
uses data from the Commodity Flow Survey, a nationally representative 
survey of freight shipments administered by the Bureau of 
Transportation Statistics (BTS). See [hyperlink, 
http://www.bts.gov/publications/commodity_flow_survey/index.html]. 

[71] U.S. Department of Transportation, Federal Railroad 
Administration, ITIC-IM Version 1.0: Intermodal Transportation and 
Inventory Cost Model Highway-to-Rail Intermodal User's Manual, March 
2005. 

[72] Approximately 34,000 kilometers of Germany's rail infrastructure 
are managed by Deutsche Bahn's DB Netz, while an additional 4,000 
kilometers are run by other infrastructure managers. 

[73] HGV toll revenue may also be used to maintain and upgrade the 
rail and waterway networks. 

[74] American Recovery and Reinvestment Act of 2009, Pub. L. No. 111-
5, Title XII, 123 Stat. 115 (2009) (Recovery Act). 

[75] Pub. L. No. 110-432, Div.B, 122 Stat. 4907 (October 2008). 

[76] 74 Fed. Reg. 29900 (June 23, 2009). 

[77] Consolidated Appropriations Act, 2010, Pub. L. 111-117, Div. A, 
Title I, 123 Stat 3034, 3056 (Dec.16, 2009). 

[78] Tracks 1 and 2 of the HSIPR program were funded from an $8 
billion appropriation of Recovery Act funds, while tracks 3 and 4 of 
the HSIPR program were funded from an appropriation of approximately 
$90 million from FY 2008 and FY 2009 Capital Grants to States-
Intercity Passenger Service DOT appropriations. Each track prioritized 
evaluation criteria differently. 

[79] Pub. L. No. 111-5, Title XII, 123 Stat. 115 (2009). 

[80] 74 Fed. Reg. 28755 (June 17, 2009). 

[81] Consolidated Appropriations Act, 2010, Pub. L. No. 111-117, Div. 
A. Title I, 123 Stat 3034 (Dec. 16, 2009). 

[82] Applicants requesting less than $20 million in federal funding 
were not required to submit a benefit-cost analysis for proposed 
projects, while those requesting between $20 million and $100 million 
in federal funding were required to include a basic benefit-cost 
analysis, and those requesting greater than $100 million were required 
to submit a more comprehensive benefit-cost analysis. 

[83] The ITIC model was first developed in 1995 under a joint effort 
by the U.S. Department of Transportation Office of the Secretary 
(OST), the Federal Railroad Administration (FRA), the Federal Highway 
Administration (FHWA) and the Bureau of Transportation Statistics 
(BTS). Since 1995, DOT has modified and updated the model, and used it 
in DOT's Comprehensive Truck Size and Weight Study, which was 
submitted to Congress in 2000. 

[84] The ITIC model is one tool of many that are available to aid in 
analysis, and its results should not be considered as the sole answer 
when making decisions or advancing a policy position. It should be 
used in concert with other models to build a framework for decision 
making. 

[85] Our choices of data sources were similar to data used in previous 
applications of the ITIC model by FRA and FHWA, but we selected more 
recent data when possible. We did not assess whether sufficiently 
reliable data were available at more disaggregate scales, such as 
single traffic corridors, individual states, or within regions. 

[86] GAO, Applied Research and Methods: Assessing the Reliability of 
Computer-Processed Data, [hyperlink, 
http://www.gao.gov/products/GAO-09-680G] (Washington D.C.: July 2009). 

[87] Specifically, partial data was available for truck rates, the 
weight per cubic foot, and the value per pound of particular 
commodities. 

[88] Reliability factor describes the shape of the reliability 
distribution, rather than a direct measure of truck reliability. 

[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: