My name is Robert Poole, and I'm the Director of Transportation Policy at the Reason Foundation, a nonprofit public policy think tank. After receiving two engineering degrees from MIT, I've spent nearly three decades studying both air and ground transportation and have advised a number of state DOTs as well as the Federal Highway Administration, the Federal Transit Administration, and the Federal Aviation Administration. Although Reason's policy work focuses on the United States, we draw on transportation research and experience from all developed countries, and I have participated in a number of international conferences on these subjects.
To summarize my perspective on U.S. high speed rail proposals, it's not clear to me what problem or problems high-speed rail is intended to solve. Before a state embarks on what is likely to be a very costly project, I believe it is incumbent on state legislators and other officials to first figure out what the high-speed rail project is supposed to accomplish. After that, the critically important next step is to assess whether there are ways of accomplishing those objectives that involve lower cost and risk to the state's taxpayers.
What Is High-Speed Rail Supposed to Accomplish?
As I read the various reports from public agencies and speeches by high-speed rail advocates, four rationales seem to predominate. They are as follows:
- To reduce greenhouse gases (GHGs) due to inter-city passenger travel;
- To reduce the oil dependency of the U.S. economy;
- To catch up with Europe and Japan;
- To give Americans more travel choices.
Let's look briefly at each of these.
Greenhouse Gas Reduction
Some of the most detailed work on the greenhouse gas (GHG) impacts of various air and surface transportation modes has been carried out recently by Mikhail Chester and Arpad Horvath at UC Berkeley. Their 2009 paper demonstrated that accurate assessment of the GHG impact of transportation modes must include life-cycle GHG emissions, not just those from vehicle operations. For air transportation, a life-cycle analysis adds 31% to the carbon footprint, while for road-based modes, the addition is 63%. But for rail modes, the addition is 155%. That dramatically affects the carbon footprint comparisons among modes, to the detriment of high-speed rail (HSR).
In a 2010 paper, Chester and Horvath apply their assessment methodology to the proposed California HSR project, comparing it with conventional rail, automobile, and airline travel north-south in California. Their assessment makes it clear that load factor (the percentage of seats occupied) makes a major difference to the carbon footprint. We have pretty good data on non-commuter auto occupancy and airline load factors, but HSR load factor is a major unknown. For the scenario with the average inter-city auto trip having 2.25 occupants, airlines having today's 85% load factor, and HSR achieving 50% load factor, it would take 71 years for HSR to achieve GHG "break-even"-i.e., for its GHG savings to offset the GHGs released by its construction.
Nearly all such calculations assume today's level of vehicle fuel efficiency for the non-rail modes. For example, a much-cited report on HSR by the Center for Clean Air Policy assumes that cars and airliners in 2025 will be only slightly more fuel-efficient than those in use today. Yet recent history and current projections show significant improvements in fuel-efficiency for both those modes. By contrast, HSR projects built over the next decade will use today's HSR technology, and their long service lives ensure that the same motive power will be in use in 2025, 2035, and likely beyond. Thus, the comparative carbon footprints of autos, airliners, and HSR are likely to be more favorable to the non-HSR modes than even Chester and Horvath estimate.
Similar data are available from overseas. In 2004, Roger Kemp of Lancaster University in the U.K. compared energy consumption (which is directly proportional to CO2 emissions) for airlines and European HSR. On a per-passenger-mile basis, Kemp concluded that an A-321 jetliner had a similar carbon footprint to the French TGV and the German ICE. In 2006, the highly respected Eddington report concluded that "high level analysis of the potential carbon benefits from modal shift from air to high speed rail suggests that these benefits would be small relative to the very high cost of constructing and operating such a scheme, and that under current assumptions a high speed line connecting London to Scotland is unlikely to be a cost-effective policy for achieving reductions in carbon emissions compared to other policy measures."
Sweden's Expert Group for Environmental Studies in 2009 commissioned a social benefit/cost analysis of a proposed HSR line in Sweden. It found that the net social benefit would be only 80% of project cost and that it would eliminate only one percent of the transport sector's carbon emissions, at a very high cost per ton. (This study did not include the much higher carbon emissions from HSR construction.) And a 2009 paper for the OECD's International Transport Forum concluded that "Environmental benefits [from HSR] are unlikely to be a significant part of the case for HSR when all relevant factors are considered."
Reducing Oil Dependence
In most of the world, HSR denotes trains operating at speeds in excess of 150 mph. Such speeds require considerably more energy than conventional passenger rail, to overcome frictional air drag. They also require broader curves and shallower grades and, for safety reasons, complete physical separation from intersecting roadways. Thus, true HSR requires dedicated rights of way and complete grade separation. True HSR is everywhere powered by electricity, almost always via an overhead catenary system. To the extent that such HSR projects shift travelers from petroleum-based modes (auto, bus, conventional rail, airline), they will reduce petroleum use to some extent. Yet even major investments in true HSR are unlikely to produce large mode shifts in the United States. In Europe, where conditions are generally more favorable for HSR than in the United States, the average citizen uses HSR for only a very small fraction of trips.
More important for the question of oil dependence, the large majority of U.S. projects proposed under the rubric of "HSR" are more accurately termed "higher-speed rail." Of the 10 projects selected for federal funding in January 2010, eight are for diesel-powered service on existing Amtrak and freight rail corridors; only the California and Florida projects are for electrically powered HSR on exclusive right of way. Only a few of the diesel projects will produce speeds in excess of 79 mph. In a major report on HSR in 2009, the GAO included a table of all current U.S. "HSR" proposals. Of the 44 projects, 35 would be diesel-powered.
Intercity rail currently handles 0.1 percent of all passenger travel in the United States, compared with 5.8% in Europe. If very large-scale investment in HSR increased the U.S. share ten-fold, it would still account for only one percent of U.S. passenger travel. The current energy intensity of diesel-powered intercity rail is about 2700 BTUs per passenger mile, while current airliners average just over 3000 and the auto fleet averages about 3400. By 2035, diesel-powered trains will be about the same, but cars are projected to average 2500 BTUs per passenger mile, while airliners will likely be close to 2700.
Thus, in terms of what is actually being proposed as HSR in the United States, the potential for reducing petroleum dependence is very small. Rail is unlikely to shift a large fraction of inter-city trips from driving and flying, and most of the rail projects will be diesel-powered in any case.
Catching Up with Europe and Japan
Many HSR proponents argue that misguided public policy has kept the United States from enjoying the benefits of this mode of travel and that this country is in some sense "backward" by not having a national HSR network. Yet this sentiment ignores some important differences between the United States and countries where true HSR has captured the largest fraction of intercity travel-such as Japan, France, Spain, Italy, and Germany. As the Congressional Research Service has pointed out, "[C]ompared with the United States, countries with HSR have higher population densities, smaller land areas, lower per capita levels of car ownership, higher gas prices, lower levels of car use (measured by both number of trips per day and average distance per trip), and higher levels of public transportation availability and use."
For example, gasoline is taxed at eight to ten times US levels in most of Europe and Japan, and most of the intercity highway networks are toll roads. Hence, the cost of intercity car travel is much higher in those countries than here. Similarly, domestic air service in Japan has never been deregulated, and such deregulation is only a recent phenomenon in Europe. Hence, air fares are also relatively higher there than here. In short, the two modes from which HSR proponents expect most HSR passengers to come-driving and flying-cost considerably less in the United States than in HSR countries.
Another factor that differs is that European countries and Japan are not only higher in density and extent of urbanization than the United States. Their urban areas are also far more centralized. In 21st century America, the "central business district" is no longer the location of the majority of jobs in the metro area, even for New York City. The suburbanization of jobs has taken place over the last 60 years, leading to "edge cities" and later to "edgeless cities." The far more centralized metro areas of Europe and Japan are more viable as origins and destinations for HSR services, since a much larger fraction of desired trips originate or terminate in a downtown area. Moreover, those centralized downtowns are relatively easier to serve via mass transit. In the United States, the suburban locations of most airports may be easier for the majority of inter-city travelers to get to than a downtown HSR station.
With all the inherent advantages for HSR in Europe (compared with the United States), one might expect that HSR would be a major factor in inter-city travel there. Yet between 1980 and 2000-a time period that saw the introduction of HSR in France, Germany, and other countries-the mode share of inter-city rail in Europe declined from 8.2% to 6.3%. Belated airline deregulation thanks to the EU's "open skies" policy led the share of inter-city travel by air to increase from 2.5% in 1980 to 5.8% in 2000. More recently, air has climbed to 8.0% while rail has declined further to 5.8%.
Giving Americans More Travel Choices
It's hard to argue with the idea that it's good to have choices for inter-city travel, just as for urban travel. People have different purposes for trips, different values of time, and different levels of affluence. For most business trips, only the speed of air travel will do, though for downtown-to-downtown trips in the New York to Washington corridor, Amtrak's Acela can provide a competitive alternative. In that same corridor, however, those of more modest means increasingly prefer the new generation of bus travel. And for those whose origins and destinations are not in or near a major city along that route, the flexibility of driving is likely to be the better choice.
From a public policy standpoint, I think the key question on this factor is who pays for these various choices? Some of those advocating HSR point to the various amounts of federal spending on highway, airline, and rail modes to make the argument that rail (currently meaning Amtrak) does not get its fair share. But gross dollars are not the important point. What counts is the extent to which the various modes are supported by their users versus being subsidized by general taxpayers. Fortunately, we have good data on this question. In December 2004 the U.S. DOT's Bureau of Transportation Statistics published a careful analysis on this subject. For each mode, they compared annual federal spending with the amounts received in user taxes from that mode. The difference, which is made up from general federal revenues (i.e., all taxpayers) is the net federal subsidy. For comparative purposes, the annual net federal subsidy (in dollars) was divided by annual passenger miles carried by the mode in question. The results were presented as annual averages over the 12-year period 1990-2002. For intercity modes, they were as follows:
- Intercity rail (Amtrak) $186/thousand passenger miles
- Airlines $6/thousand passenger miles
- Highways -$2/thousand passenger miles
The highway number is negative because highway users pay slightly more in federal highway user taxes each year than the federal government spends on highways. The airline number is low because the various aviation user taxes (on tickets, fuel, international travelers, etc.) cover nearly all of the federal government's spending on airport grants and operation of the air traffic control system.
As will be discussed below, at best only two of the world's HSR lines-Tokyo-Osaka and Paris-Lyon-claim to have covered both their capital costs and operating costs from passenger revenues. All the others have had some or all of their capital costs paid for by the national government, out of general tax revenues. Many also receive part of their operating budgets from the taxpayers, though the best corridors are self-supporting on an operating-cost basis.
What this means is that those requesting-or demanding-the new choice of HSR corridors are not proposing to pay for this choice themselves, as do those who choose to travel inter-city by plane, car, or bus. Instead, they are asking federal and state taxpayers to pay for the right of way, track, bridges, stations, and rolling stock-and possibly also a portion of the operating and maintenance cost. This makes HSR profoundly different from all the other choices.
What are the Alternatives?
The previous discussion has argued that the usual rationales for HSR are weak, which suggests to this analyst that HSR is a solution looking for a problem to solve. But before going on to consider the costs and risks to state governments, it's worth taking a closer look at the viability of alternatives. In this case, the alternatives are future versions of the current inter-city passenger modes: bus, car, and airline.
Many people have an out-of-date image of the inter-city bus (motorcoach) sector. According to a 2008 report by Nathan Associates, as of that year scheduled inter-city bus was available at more than 3,000 locations nationwide, which compares with just 505 Amtrak stations. The market niche of most motorcoach service is lower-income people (with 54% earning $25,000 or less). So if the argument for spending general tax money is to provide more affordable inter-city passenger service, the motorcoach industry is doing that, and providing service covering far more cities and towns than Amtrak. In recent years, a new generation of express bus services has been launched into various markets, especially in the Northeast. Carriers such as Megabus and BoltBus compete for the business of young professionals, offering free on-board wi-fi and other services.
Similar unsubsidized motorcoach services exist in Europe, serving essentially the same demographic segments as in this country. They are most common in Britain, Italy, Norway, Poland, Spain, and Sweden. France and Germany, by contrast, have not permitted such services, for fear of losing passengers from their subsidized inter-city rail services.
When it comes to carbon footprint, inter-city bus is surprisingly low in GHG emissions. A recent report from the Union of Concerned Scientists, as a guide for consumers, compares the carbon footprint of a vacation trip made by five alternative modes: motorcoach, train, car, SUV, and airplane. The motorcoach beats inter-city rail by a very large margin, for every trip distance from 100 miles to 2,500 miles, and for trips by one, two, or four travelers.
Tomorrow's Cars on Tomorrow's Highways
In making transportation policy, the relevant comparison is not with historical conditions or even today's. Rather, when considering whether to make investments of tens or hundreds of billions of dollars in long-lived infrastructure, we need to consider conditions decades from now, when that infrastructure would be in service.
Most HSR corridors being planned today would not likely be in operation until a decade or more, and will then operate for 30 years or more with their initial motive power. By contrast, the motor vehicle fleet is turning over continually, with complete replacement approximately every 20 years. Hence, better technology in propulsion will be an important background factor during whatever planning period we consider-and certainly over the next three decades. The new federal fuel-efficiency (CAFÉ) standards recently finalized by the federal government require the average new vehicle in 2016 to achieve 35.5 mpg. The EPA projects that this will lead to a 21% reduction in GHG emissions from cars by 2030, as the vehicle fleet turns over. Assuming future improvements in fuel economy continue at that rate, by 2035-2040 personal motor vehicles are likely to have a lower carbon footprint than inter-city rail.
What about congestion? To be sure, traffic congestion is a major U.S. problem, but it is mostly a problem in urban areas, and mostly during weekday peak travel periods. HSR serves inter-city, not urban travel markets, so is not really relevant to urban traffic congestion (except to the extent that, operating at slower speeds, it can add to the existing commuter rail capacity on one or two corridors in large urban areas).
One of the rationales for both the California and Florida HSR projects is to relieve congestion on, respectively, I-5 and I-4, both of which are projected to need widening over the next decade or two. The Orlando-Tampa (I-4) corridor allows us to do a comparison of highway and HSR alternatives. The official cost estimate for this 84-mile HSR project is $2.6 billion, but if it ends up with a cost/mile that's the average of recent European HSR projects ($51 million/mile in 2008 dollars, per GAO-09-317), the cost would be $4.3 billion. Using Federal Highway Administration HERS cost figures for adding lanes to existing Interstates, the cost of adding one lane each direction in the Tampa-Orlando corridor would be about $715 million. Unlike the HSR dollars, which would all come from federal and state taxpayers, the highway lane additions would be paid for out of highway user taxes-or if the lanes were built as express lanes, the funding could come partly or entirely from tolls. And with variable (value-priced) tolls, the congestion relief on I-4 would be sustainable long-term.
Likewise, in California, there is no inherent problem with widening I-5, most of which goes through open country with a wide central median. On the southern end, both I-5 and the planned HSR line must cross the Tehachapi mountains, making the cost/mile for either considerably higher than in the flat terrain of Florida.
Short-Haul Airline Service
In Europe and Japan, HSR has captured significant mode share from airlines on a small number of routes, including Tokyo-Osaka, London-Paris, Paris-Lyon, and Madrid-Barcelona. In these high-density corridors of 200 to 400 miles between centralized metro areas, HSR can be competitive with air travel, especially if the capital costs are not covered by the HSR fares. Amtrak's popular Acela service in the Northeast Corridor has captured significant share from airline shuttles, but Acela fares likewise have to cover only operating and maintenance costs, unlike the full cost recovery included in airline fares.
HSR advocates argue that airport and airway congestion makes it problematical to increase short-haul airline service in coming decades. Yet that claim has not been substantiated. Several factors suggest that the airline/airport/air traffic control system can adapt so as to accommodate continued growth in short-haul demand.
One factor is the coming NextGen air traffic control system, which will increase both runway throughput and the capacity of intercity airways thanks to more precise navigation technology and automation of routine aspects of aircraft spacing and separation. A second factor is "up-gauging" of the aircraft used in short-haul service-e.g., substituting 737s for regional jets, to provide larger passenger capacity at the same frequency of service. A third factor is the further development of a multi-airport strategy. Most of the largest U.S. metro areas already have multiple airports, including Los Angeles, San Francisco, Dallas, Houston, Chicago, Miami, Washington/Baltimore, and New York. A proposal to add short-haul airline service to the Gwinnett County airport in metro Atlanta is currently getting serious local support.
Short-haul air service in California provides an important contrast with the proposed HSR system. Currently, north-south service is offered between every combination of four northern (San Francisco, San Jose, Oakland, and Sacramento) and six southern (Los Angeles, Burbank, Long Beach, Orange County, Ontario, and San Diego) airports. All those hundreds of daily flights are nonstop. The HSR system will also serve all of those points, and dozens in between. Only a relatively small number of HSR trips will be nonstop, however, with most of those leaving from "downtown" rail stations. Given the decentralized land-use patterns in California, for most north-south travelers a non-downtown airport location will be more convenient to their actual origin and/or destination than a downtown railroad station. Southern California has the potential of considerably more airport capacity in the form of Palmdale airport. If access to that airport from the major population centers to the south is eventually improved (such as via the proposed Glendale-Palmdale tunnel), Palmdale could become a viable alternative.
Future aviation technology could lead to quieter short-haul turboprop aircraft, able to land and take off from very short runways. This could make possible new airline service from what are now airports serving only small private planes (general aviation). An example that already exists is Porter Airlines, operating from Toronto City Center Airport on an island directly adjacent to downtown Toronto. As of late 2009, Porter was providing multiple daily flights between Toronto and six Canadian and three U.S. cities (Boston, Chicago, and Newark), using Bombardier Q400 twin turboprop aircraft. The flexibility of this approach contrasts sharply with that of HSR. Porter and similar airlines can test a city-pair market at very low cost, and if the demand is not robust, it can shift its assets to other markets. With HSR, if a route fails to generate much traffic, there is an enormous, unrecoverable sunk cost.
The Cost and Risk to States
The data are very clear that HSR is not a self-supporting business venture (such as airlines and motorcoach services) anywhere in the world. In its 2009 report on HSR, the GAO notes that "Passenger fares are generally insufficient to finance the capital and operating costs of a high speed rail system." In writing about the Japanese HSR experience, the 2009 CRS report notes that after the initial Tokyo-Osaka line, "None of the additional lines generated enough passenger revenue to even cover their operating costs." In summarizing its findings on overseas HSR, the CRS report states that "Typically, governments have paid the construction costs, and in many cases have subsidized the operating costs, as well."
A special report commissioned by the OECD's International Transport Forum, Prof. Gines de Rus of the University of Las Palmas (Spain) concluded the following: "High proportion of fixed and sunk costs, indivisibilities, long life and asset specificity make this [HSR] public investment risky, with a very wide range of values for the average cost per passenger trip." And he adds, "Even in the case of particularly favorable conditions, the net present value of HSR investment has to be compared with other 'do something' alternatives [such] as road or airport pricing and/or investment, upgrading of conventional trains, etc."
The risks involved in HSR are formidable. A 2008 "due diligence" assessment of the proposed California HSR project raised serious questions about its projected capital costs, ridership, load factor, operating profits, and GHG impact. A 2010 assessment by the California Legislative Analyst's Office confirmed most of these findings, concluding that the revised HSR plan was "on the fast track to fiscal failure," as summarized by the California Taxpayers Association. Although the revised business plan increased the projected capital cost and reduced projected ridership, the agency still claims the system will cover its operating and maintenance cost from the farebox, thanks to doubling the projected fares compared to its 2008 figures.
There is much talk, in California and elsewhere, about private-sector investment in HSR, in some form of public-private partnership. In Europe and Japan, private companies have played roles in some of the more recent HSR projects. But those companies do not put their own funds at risk. In Japan, the railroad system was "privatized" in 1987, at which point the government relieved the former state railway of more than $300 billion in accumulated debt from building HSR lines. "The national government and local governments still assume the financial risk of constructing a new high speed rail line, investing two-thirds and one-third of the construction costs, respectively," according to the GAO report. Then, "with the government's financial commitment, the private operating companies undertake the operational risk and rely on ticket revenues to cover operating and maintenance costs."
The California LAO report notes that although the $9 billion HSR bond measure approved by the voters in 2008 "explicitly prohibits any public operating subsidy," the HSR Authority's current business plan "assumes some form of revenue guarantee from the public sector to attract private investment. This generally means some public entity promises to pay the contractor the difference between projected and realized revenues if necessary. The plan does not explain how the guarantee could be structured so as not to violate the law."
It's clear from the global experience with HSR that in nearly every case, the general taxpayers must pay for the very large capital cost of building the systems. In very high-demand corridors that can attract significant mode share from short-haul air service, passenger fares may be able to cover operating and maintenance costs, though these often require taxpayer subsidy as well. Therefore, we must consider what kind of financial obligations states will be taking on by spending federal grant funds on HSR systems.
It's clear that advocates of HSR envision the initial federal projects as just the starting point. The widely quoted Gil Carmichael, formerly head of the Federal Railroad Administration and later chairman of the Amtrak Reform Council, has repeatedly called for a 20,000 to 30,000 mile national HSR system, which he likens to an "Interstate 2.0." Even a 10,000-mile system, at the GAO's average capital cost of $51 million per mile, would cost state and federal taxpayers $510 billion. On top of that would be operating subsidies for many of those routes.
The initial $13 billion committed by the Administration ($8 billion in ARRA funds plus $1 billion per year for five years) would cover only a small fraction of the estimated cost of those initial projects. The feds committed only $2.3 billion to the California project, while the HSR Authority's current estimate of capital costs for the 520-mile Phase 1 (of the 800-mile project) is now $42.6 billion. Between the $9 billion available from the HSR bond issue and the federal grant, the state is "only" $31.3 billion short. Florida likewise faces a nominal shortfall of $1.35 billion ($2.6 billion official cost minus $1.25 billion federal grant), but if the cost ends up being close to GAO's $51 billion per mile average, the real cost could be $4.3 billion-meaning the state will need to find $3 billion just to build the 84-mile system. And if ridership ends up significantly lower than projected, there will need to be operating subsidies, as well.
State budgets are in poor shape to take on such new obligations. It is not just the current recession, though that is forcing unprecedented budget cuts in most states. More important is that states, like the federal government, are on an unsustainable fiscal path, due to unfunded liabilities for health-care programs and employee pensions. To take on huge new funding commitments this year, or this decade, strikes me as not being responsible fiscal management.
Given the near-certainty that 100% of the capital costs of HSR must come from general taxpayers, and the risk that many of these projects, if built, will require ongoing operating subsidies, state legislators and other officials must think through where the funds to support those expenditures could come from. And in doing so, they must weigh the projected benefits of HSR against those costs.
Economists make use of a concept called "opportunity cost." In simple terms, it means that the real cost of doing something is what you have to give up to do that something. In the severely constrained fiscal environment in which state governments will find themselves over the next several decades, the opportunity cost of shifting billions of dollars in general tax monies to HSR means those same billions will not be available for other public purposes-such as properly funding public pension systems, schools, highways, etc.
Given the very modest benefits of HSR, in terms of GHG reduction, energy savings, and added choices for a small fraction of travelers, it seems like a poor use of many billions of dollars in state tax money.
 Mikhail V. Chester and Arpad Horvath, "Environmental Assessment of Passenger Transportation Should Include Infrastructure and Supply Chains," Environmental Research Letters, Vol. 4, 2009.
 Mikhail Chester and Arpad Horvath, "Life-Cycle Assessment of High-Speed Rail: the Case of California," Environmental Research Letters, Vol. 5, 2010.
 Center for Clean Air Policy and Center for Neighborhood Technology, "High Speed Rail and Greenhouse Gas Emissions in the U.S," Washington, DC, 2006.
 HM Treasury and Department for Transport, The Eddington Transport Study: Main Report, Volume 3, London, 2006.
 Chris Nash, "When to Invest in High-Speed Rail Links and Networks?" Discussion Paper No. 2009-16, OECD International Transport Forum, November 2009.
 Government Accountability Office, "High Speed Passenger Rail: Future Development Will Depend on Addressing Financial and Other Challenges and Establishing a Clear Federal Role," GAO-09-317, Washington, DC, March 2009
 "National Transportation Statistics," US DOT Bureau of Transportation Statistics, 2008, Table 1-37 and "Panorama of Transport," European Commission, 2007, p. 102.
 David Randall Peterman, et al, "High Speed Rail (HSR) in the United States," Congressional Research Service, 7-5700, R40973, December 8, 2009.
 European Commission, "Key Facts and Figures About the European Union," 2004, p. 52.
 Bureau of Transportation Statistics, "Federal Subsidies to Passenger Transportation," U.S. Department of Transportation, December 2004.
 Victoria Burnett, "Europe's High-Speed Trains Hold Lessons for U.S.," New York Times, May 29, 2009, cited in Peterman, Congressional Research Service, op cit.
 Robert Damuth, "The Economic Impacts and Social Benefits of the U.S. Motorcoach Industry," Nathan Associates, December 2008.
 Didier Van de Velde, "Long-Distance Bus Services in Europe: Concessions or Free Market?" Discussion Paper 2009-21, OECD International Transport Forum, 2009.
 Table 4 in "Getting There Greener: The Guide to Your Lower-Carbon Vacation," Union of Concerned Scientists,December 2008.
 Juliet Eilperin, "Emissions Limits, Greater Fuel Efficiency for Cars, Light Trucks Made Official," The Washington Post, April 2, 2010.
 Costs are from FHWA's 2006 HERS cost database, adjusted to 2008 dollars. Calculation based on 90% of the 84-mile route considered rural (at $2.75 million/lane-mile in 2006) and 10% urban (at $14.6 million per lane-mile in 2006 dollars). For 84 lane-miles each way (168 lane-miles total) at a weighted average cost/lane-mile of $3.935 million, the total is $661 in 2006 dollars, converted to $715 million in 2008 dollars.
 Interview with Porter CEO Robert Deluce, Airport Business, August 2009, p. 16.
 GAO-09-317, p. 30.
 CRS, p. 7
 Ibid, Summary page.
 Gines de Rus, "The Economic Effects of High Speed Rail Investment, Discussion Paper No. 2008-16 revised, OECD International Transport Forum, October 2008.
 Wendell Cox and Joseph Vranich, "The California High Speed Rail Proposal: A Due Diligence Report," Policy Study No. 370, Reason Foundation, September 2008.
 Legislative Analyst's Office, "The 2009 High-Speed Rail Business Plan," January 11, 2010.
 Gil Carmichael, "Interstate 2.0: Getting Rails on Track," The Journal of Commerce, July 27, 2009.
 Mary Williams Walsh, "State Debt Woes Grow Too big to Camouflage," The New York Times, March 29, 2010.