Funding Options for Freight Transportation Projects (2009)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

2Freight System Performance and Infrastructure Finance As Chapter 1 explained, the committee’s charge arose from the concerns of public- and private-sector participants that present finance arrange- ments for freight infrastructure are inadequate and that the performance of the freight transportation system is suffering as a consequence. This chapter examines trends in system performance and evidence that per- formance is related to finance arrangements. The first section below cites industry participants’ descriptions of failings in system performance and in finance practices, for freight transportation and for U.S. transportation as a whole, to illustrate per- ceptions of the problems confronting the system. The section also reviews the conclusions on the relation of system performance to finance in the Transportation Research Board (TRB) committee report Freight Capacity for the 21st Century (2003). The second section surveys the limited avail- able data on trends in freight transportation system performance, that is, the reliability, speed, and cost of goods movement. The section also describes trends in freight infrastructure spending, the stock of infra- structure, and freight traffic. These measures—in particular, the gap between the rates of growth of traffic and of infrastructure—often are cited in support of proposals to increase public investment. The third section considers the extent to which any of the problems and chal- lenges facing the freight system are related to finance arrangements. The final section is a summary. Throughout the chapter, a recurring difficulty is the near absence of systematic information about the per- formance of public-sector transportation facilities or about returns on public investments. 43

44 Funding Options for Freight Transportation Projects VIEWS ON FINANCE PROBLEMS Arepresentative statement of the infrastructure problem (referring to pas- senger as well as freight facilities) is from a statement of the U.S. Chamber of Commerce (U.S. Chamber of Commerce 2007): The nation’s transportation infrastructure is in crisis. Without significant repairs and new construction, our aging roads, bridges, and transit cannot begin to handle the growing transportation needs that commuters, emer- gency responders, truckers and delivery drivers, and law enforcement require on a daily basis. To begin facing this enormous challenge, we need to com- mit adequate resources while finding new and creative ways of financing the new construction and repair of existing roads, bridges, and transit as quickly as possible. Policy makers in Congress and the Administration need to rec- ognize the enormous impact that our deteriorating transportation infra- structure is having on the economy and the health and safety of our citizens. We need their strong commitment to address this growing crisis if we are to make the necessary and often difficult financial decisions that will resolve this transportation crisis. An earlier Chamber of Commerce–sponsored study (Cambridge System- atics 2005a, 1) described a “funding shortfall that immediately threatens national mobility” and identified a variety of possible revenue enhance- ments to support increased federal and state government spending. A similar characterization of transportation finance inadequacies is found in recent testimony of a finance sector executive before a House committee (Florian 2008): Thenation’s transportation system is in a crisis because current funding sources and financing tools are insufficient to maintain and improve this country’s highways, public transportation systems, and intermodal connectors. . . . [T]he continued availability of abundant and efficient transportation infrastruc- ture is critical to the economic growth and prosperity of our economy, and to the quality of life of individual Americans. I believe that this problem can be expressed in several key observations: Demands on our transportation system are outpacing investment in it. For example,VehicleMiles Traveled . . . onU.S. highways have doubled in the last 25 years, but capacity onour highway system is up only 3 percent. Maintenance costs of existing transportation assets are competing for the same fundsneeded to expandour transportation system. . . . Construction inflation has accelerated, up 40 percent cumulatively in the last 3 years. . . . The fuel tax . . . is no longer sufficient to meet the large and grow- ing needs for transportation infrastructure development in the United States.

Freight System Performance and Infrastructure Finance 45 The same concerns and diagnoses are expressed by public officials. A statement of the American Association of State Highway and Transporta- tionOfficials, prepared to advise theNational Surface TransportationPol- icy and Revenue Commission created in the Safe, Accountable, Flexible, Efficient Transportation Equity Act: A Legacy for Users, describes freight problems as follows (AASHTO 2007, 58): The nation is entering the early stages of a freight transportation capacity cri- sis. All systems are aging and stretched to capacity.Highways, railroads, ports, waterways, and airports all require investment well beyond current levels to maintain, much less improve, their performance. Projections of freight vol- ume increases reveal that the nation is unprepared and is not preparing fast enough for the freight increase. A recent report forecasts a four-fold increase of container volumes in Los Angeles, Houston, and Savannah, near tripling of volumes at the ports of New York/New Jersey, Charleston, and Virginia, and greater than doubling at the ports ofMiami, Tacoma, andOakland. These volumes will overwhelm the ports and the surface freight system in each of these metropolitan areas. The characterization of the General Accounting Office (GAO, later renamed theGovernmentAccountabilityOffice) (GAO2003), in a review of proposals for freight finance provisions in the federal surface trans- portation aid program, was that “increasing congestion within the freight transportation system poses a threat to the efficient flow of the nation’s goods” and that “congestion delays that significantly constrain freight mobility [at ports, border crossings, anddistributionhubs] . . . could result in serious economic implications for the nation” (GAO 2008, 1). GAO does not cite estimates of the spending gap between outlays and needs but states that among the “challenges” faced by government planners in advancing freight projects is “limited and restricted availability of public funds available for freight transportation” (GAO 2008, Highlights page). GAO observes that “there is a growing concern that the current fund- ing structure [of federal programs for transportation infrastructure] is not well suited to advancing freight improvements and that additional actionmight be needed to better allocate federal funds in order to address impediments to freight mobility” (GAO 2008, 2). The two failings of the present structure that GAO cites are bias in project selection in favor of passenger-oriented projects (“state and local planners are more likely to fund projects that directly benefit passengers in their localities rather

than freight traffic that moves through the region”) and that “[federal] funding sources have remained largely tied to individual modes” (GAO 2008, 2), apparently referring to restrictions in the federal-aid highway program that prevent states from using federal grants for rail projects (GAO 2008, 37). The consequence of failure to act to redress the capacity deficiency, in the view of a federal official, would be to risk a general failure: “[The]U.S. is exposed to a major breakdown in the flow of commerce that would sig- nificantly impact the U.S. economy” (Shane 2006). In summary, the consistent points in the frequently expressed criticism is that the transportation system suffers from a gap between the rate of spending that would allow service to be maintained and improved and the spending that the public and private sectors are willing to undertake. The gap is widening to the point of crisis; the evidence of the crisis is grow- ing congestion and physical deterioration. In the public sector, the gap is the result of bias in spending decisions against projects important for freight; arbitrary restrictions on project eligibility in funding programs, especially the federal-aid program; and unwillingness of elected officials to increase the special taxes that fund most government transportation spending. Statements of this criticism usually refer to general, systemwide levels of investment, rather than to specific bottlenecks requiring solutions, or else cite examples from the small group of most prominent bottlenecks (e.g., the Southern California ports and the Chicago rail hub). Overall trends in traffic growth are the primary evidence of need cited; as will be described below, data that would directly demonstrate declining freight performance are not available. The criticism is not new; essentially the same concerns prompted two TRB policy studies concerning freight capacity (TRB 1993; TRB 1998). A comparison of the criticism summarized above with results of a sur- vey of industry participants reveals a divergence of views. The 2005 survey of 500 shipper executives, carrier executives, and government administra- tors by theMassachusetts Institute of Technology (MIT)Center forTrans- portation andLogistics found that “perceptions in the causes and remedies of the congestion crisis differ between the government andprivate sectors” (Caplice and Blanco 2006, 5). When asked to rank a list of 29 “potential 46 Funding Options for Freight Transportation Projects

Freight System Performance and Infrastructure Finance 47 root causes” of the current (in 2005) freight congestion situation, govern- ment respondents ranked “lack of funding for freight infrastructure by the state governments” second and “lack of funding for freight infrastructure by the federal government” third; carriers ranked these two potential causes 11th and ninth, respectively; and shippers ranked them 14th and 10th. Carriers’ top three root causes were “West Coast port congestion,” “growth of international imports,” and “highway congestion near metro- politan areas”; shippers’ top three were “driver shortages in the long-haul trucking industry,” “growth of international imports,” and “West Coast port congestion.” Government respondents’ top choice was “highway congestion near metropolitan areas” (Caplice and Blanco 2006, 7–8) (see Table 2-1). While it may be circular to cite congestion as the cause of “the current congestion crisis,” one interpretation of the shippers’ and carriers’ responses is that they are emphasizing the need to correct problems at TABLE 2-1 Root Causes of Freight Congestion According to Shipper, Carrier, and Government Survey Respondents Rank Shipper Carrier Government Driver shortages in the long-haul trucking industry 1 5 11 Growth of international imports 2 2 5 West Coast port congestion 3 1 6 Highway congestion near metropolitan areas 4 3 1 Volatility of fuel prices 5 16 12 Lack of 24/7 operations in West Coast ports 7 4 9 Lack of funding for freight infrastructure by the 10 9 3 federal government Lack of funding and investment in high-priority 12 8 4 projects by the railroads Lack of funding for freight infrastructure by the 14 11 2 state governments NOTE: The survey question was as follows: Freight transportation congestion across all modes has been listed as the most critical issue for shippers and carriers alike for the last 18 months. What do you think are the root causes of the current situation? 1. Please rate each of the potential root causes listed below as to its significance in causing the cur- rent congestion crisis. No Impact Moderate Impact Significant Impact Very Significant Impact [followed by a list of 29 potential root causes]. SOURCE: Caplice and Blanco 2006.

48 Funding Options for Freight Transportation Projects specific bottlenecks. The responses indicate that the shippers and carriers do not see chronic systemwide infrastructure underfunding as among the most important root causes of their congestion problems. The MIT survey findings parallel conclusions that the TRB Freight Capacity committee drew froma series of informal interviewswith freight transportation industry executives. Interviewees frequently cited capac- ity constraints other than infrastructure, including labor supply shortages and regulatory factors (truck size and weight regulations and interna- tional border delays), and noted that continual change in the character- istics of freight demand is a fundamental source of mismatches between capacity and demand (TRB 2003, 102–103, 116–118). In its conclusions, the Freight Capacity committee offered an alterna- tive perspective to the criticism outlined above. The committee found that because governments do not measure returns on infrastructure investments, evidence is lacking to support claims of underinvestment or of systematic misallocation in government investment decisions in favor of projects primarily serving passengers over projects of particular importance to freight. The committee endorsed higher spending in the federal surface transportation aid program. However, it concluded that increased spending alone would be unable to sustain freight transporta- tion productivity growth; greater reliance on pricing and market forces to manage facilities and guide investment decisions would be essential as well. In the absence of finance reform, the committee foresaw protracted decline rather than collapse (TRB 2003, 111–112): The trends show evidence of the emergence of social and political forces that will influence freight transportation development in new ways in the next decades. Increasing population density, urbanization, and wealth ensure that conflicts between freight and passenger traffic; conflicts between freight transportation and residential, recreational, and other competing land uses; and requirements to control pollution will increase. These forces will tend to increase the cost of expanding capacity and add to the risk of investment. As congestionworsens, demand for increased public spendingwill appear. The United States has ample resources for expanding the transportation sys- tem; in most regions, the densities of population, employment, transporta- tion networks, and traffic are low in comparison with the Northeast or with western Europe. If capacity additiondoes lag traffic growth and congestionworsens, as seems likely, the long-run consequence will not be massive breakdown. Freight

Freight System Performance and Infrastructure Finance 49 markets have self-correcting capabilities. Users will make numerous adjust- ments over time to accommodate or avoid congestion. Shippers will change logistics practices, for example, by shipping more in bulk and holding larger inventories than theywould if freight transport were cheaper. From a close-up perspective, these long-term repercussions of congestion may not be evident. One of themost important copingmechanismswill be changes in landuse and in the location of activities: workplaces and residences will move away from congestion within metropolitan areas and from more-congested to less- congested regionswithin theUnited States. Such adjustment has been themost important means of accommodating growth throughout U.S. history. Con- gestion will be a constraint on the growth of some urban areas. Some produc- tion will move from the United States to other countries if congestion costs cause the United States to lose comparative advantage in some industries. Therefore, one plausible course of development is that the nation will continue to accommodate growing freight traffic volumes by increasing capital spending on infrastructure, accepting more congestion, altering pro- duction and logistics practices, and moving away from the most congested locations. This resolution might be tolerable, but will certainly be far from the economic optimum, for two reasons. First, the available capacity will continue to be used poorly on those parts of the system where users do not pay prices that reflect costs and where operators lack incentives to be respon- sive to user costs and preferences. Second, obstacles exist to effective target- ing of capital expenditures, particularly in the public sector. Public capital spending will dissipate much of its impact because some high-payoff projects are passed by and some low-payoff ones are carried out. Private-sector cap- ital expenditure may not be efficient if a suboptimally small number of firms dominate a market, hindering competition. The potential future costs of delay and other direct consequences of con- gestion canbe estimated, but other costs of this “business-as-usual” scenario— for example, the costs of distortion of land use and regional development patterns—are difficult to observe or predict. Changes in government pol- icy that would allow the nation to make better use of existing capacity and [better] investment decisions, compared with this scenario, would have important economic benefits. In short, the obstacles to reducing the cost of freight transportation are rising input costs (from growth in population, income, and environ- mental awareness) and continued tolerance of inefficiencies that result from the established practices for selecting investments and charging for use of publicly provided facilities. This alternative diagnosis of the fundamental weakness in transporta- tion finance arrangements—emphasizing inefficient operations and poor

targeting of investment rather than a gap between spending and needs— has been repeatedmore recently by other observers. For example, the Sec- retary of Transportation in 2007 described highway finance failings as follows (Peters 2007): We have an increasingly flawed investment model and a system performance crisis. . . . The underperformance in the highway sector is fundamental, not incremental. In other words, an increase in federal taxes and spending would likely do little, if anything, without a more basic change in how we analyze competing spending options andmanage existing systemsmore efficiently. . . . The degree to which one capital investment generates more returns than a competing investment is themost basic question asked in virtually every other capital intensive sector. . . . Yet, when it comes to . . . highways . . . , there is virtually no analysis of this question. The recession that began in December 2007 reduced traffic and con- gestion throughout the transportation system; therefore, the immediate conditions that gave rise to the statements above no longer pertain. Per- formance problems that were evident during the previous economic expansion may reemerge with economic recovery, but the recent expe- rience highlights the necessity of distinguishing long-term trends from normal cyclical patterns and unique events in diagnosing freight system problems. Regardless of how the performance and finance trends of the past two decades are characterized, broad agreement exists that historical methods of developing freight infrastructure will become increasingly inadequate in the future. Finance reformwill be a necessary element of any national program to build and manage infrastructure more efficiently. TRENDS IN FREIGHT TRANSPORTATION SYSTEM PERFORMANCE The purpose of any proposal to reform freight infrastructure finance will be to improve the future performance of the freight transportation sys- tem. That is, the goal is a system that provides diverse services to ship- pers at lower public cost (including shippers’ transportation and logistics costs, external impacts like pollution, and the cost to taxpayers if they pay for public infrastructure programs). The cost of freight transportation will depend on the investments that are made in the system, the operat- 50 Funding Options for Freight Transportation Projects

Freight System Performance and Infrastructure Finance 51 ing efficiency of facilities, and application of new technology, as well as on a variety of external factors. Finance proposals usually are prefaced with observations about inadequate performance at present or to be expected in the future. This section reviews evidence on characteristics of the freight transportation system that are related to performance. It summarizes and updates the examination of freight performance in the report of the TRB Freight Capacity committee (TRB 2003, 50–103). The first two subsections below examine congestion, the primary physical indicator of performance, and shipper and infrastructure costs, the eco- nomic measure. Data are not yet available to show systematically the effects of the recession that began in December 2007 on these measures. The final two subsections address trends that are frequently cited in sup- port of funding proposals: the rate of investment and capital stock growth compared with traffic growth, and projections of future traffic levels that the system will be serving. Congestion Data are presented below concerning congestion on highways, freight railroads, and other modes. Congestion by itself is not an adequate mea- sure of the performance of a transportation system. Even the most effi- ciently operated facility will experience occasional congestion because transportation demand varies with time and because public benefits and (in the case of private-sector facilities) operators’ profits usually are maximized by allowing some degradation of service quality during peak periods rather than by restricting use to the point that traffic flow is never hindered. Efficient operation of a congested facility serving commercial freight traffic requires (a) avoiding gridlock conditions (that is, a cir- cumstance where throughput nearly stops because too much traffic has been allowed to enter the facility) and (b) giving priority to the freight movers who will gain the most from use of the facility. For example, at a congested port, shipments of high-valued or time-sensitive goods should be able to gain priority over shipments of low-value commodities. Highways Highways are the nation’smajor freight transportation infrastructure sys- tem. Freight and passenger traffic share the same facilities, and with few

52 Funding Options for Freight Transportation Projects exceptions, the worst bottlenecks for highway passenger transportation are also major freight bottlenecks. Therefore, highway congestion is an important indicator of freight system performance. The Texas Transportation Institute’s (TTI’s)UrbanMobility reports (Schrank and Lomax 2007) have become the accepted authority on urbanhighway congestion trends. TheU.S.Department of Transportation (USDOT) nowuses the TTI estimates as its principal congestion indicator in its biennial report to Congress on the conditions and performance of the U.S. highway system (USDOT 2007, 4-2–4-14). They are cited as evi- dence of underinvestment by GAO (GAO 2008, 12) and are the primary source on congestion referred to by the National Surface Transportation Policy and Revenue Study Commission (2007, 3-13). According to the TTI estimates, congestion delay in 437 U.S. urban areas grew from an average of 14 hours annually per peak period traveler in 1982 to 38 hours in 2005, a 170 percent increase, and the costs of con- gestion (including the value of lost time to passengers and commercial traffic and extra fuel costs) grew from $15 billion in 1982 to $78 billion in 2005 in 2005 dollars, a 420 percent increase (Schrank and Lomax 2007, 1). The study’s assumptions about traffic volumes and time costs imply that costs borne by operators of commercial vehicles account for about 20 percent of the total, or $16 billion annually, equal to 2 to 3 per- cent of shippers’ total annual expenditures for truck freight transporta- tion (Council of Supply Chain Management Professionals 2008). The most congested metropolitan areas are estimated to have delays 50 to 90 percent worse than the national average: 72 hours per traveler per year in Los Angeles and 60 hours in San Francisco and Washington, D.C. Fur- thermore, congestion is estimated to be growing much faster in certain metropolitan areas than the national average, with 1995–2005 increases of 24 hours per traveler per year in Dallas and Houston, 22 hours in San Diego, and 17 hours in Portland, Oregon, compared with an average of 7 hours for the decade for all urban areas (Schrank and Lomax 2007, 78). The derivation of the TTI congestion estimates does not rely on metro- politan area speed data or data on trends in highway speeds. Instead, the estimates are derived from data on traffic volume [annual average daily traffic (AADT)] for a sample of road segments in each metropolitan area in the Highway Performance Monitoring System database maintained by the Federal Highway Administration (FHWA) and from national average

Freight System Performance and Infrastructure Finance 53 relationships between AADT and peak volume and between volume and speed (TTI n.d., 6-11). The documentation of the TTI estimates does not report any comparison with direct measurements of congestion or travel times; therefore, the validity of the estimates is unknown. USDOT ceased collection of nationwide data on highway speed in 1994, and few states compile systematic and timely speed data (OECD and ECMT 2006, 255). Therefore, no information derived from obser- vation exists about the nationwide extent, characteristics, or time trends in highway congestion. An indicator of congestion based on actual reported travel times is found in the decennial censuses (Bureau of the Census n.d. c) and the annual American Community Survey (Bureau of the Census n.d. a; Bureau of the Census n.d. b). The average travel times to work for workers outside the home, from survey responses, are as follows: Year Travel Time, OneWay (minutes) 2006 25.0 2000 24.4 1990 22.4 1980 21.7 Of course, the travel time changes depend not only on changes in the speed of travel but also on a variety of factors including changes in use of transit and in the location of residences and workplaces over the period. However, location changes are driven in part by travel costs, so any added time required for commuting over the period as a result of workers mov- ing their residences farther away from their jobs is to an extent a reflection of congestion impacts. The census data imply that the average commuter spent 27 hours more per year traveling to and from work in 2006 than in 1980, a 15 percent increase in 26 years. The census data are not necessarily inconsistentwith the TTI estimates, but they offer a contrasting perspective on trends in congestion and travel delay: Item Percentage Change TTI annual delay per peak traveler, 1982–2005 +171 TTI annual congestion cost, 1982–2005 +425 Census average travel time to work, 1980–2006 +15

54 Funding Options for Freight Transportation Projects The percentage changes appear different because of differences in the base from which each change is computed. TTI reports changes in excess travel time, starting from a year in which excess travel time is estimated to have been relatively small; the percentage change from the census data is the change in total travel time. In an exceptional actual measurement of highway system travel times and speeds, FHWA,with trucking industry cooperation, analyzed data on truck movements over a 1-year period (calendar year 2005) collected from 250,000 commercial trucks equipped with automatic vehicle loca- tion devices, for travel on five major Interstate corridors totaling 7,000 miles and accounting for one-fourth of all commodity-carrying truck vehiclemiles (Mallett et al. 2006). The corridors, average speeds observed, and buffer indices (a measure of travel time reliability) were as follows: Corridor Average Speed (mph) Buffer Index (percent) I-5, California–Washington 50 19 I-10, California–Florida 56 21 I-45, Texas 54 31 I-65, Alabama–Indiana 58 7 I-70, Utah–Maryland 54 11 The analysis excludes time for rest, refueling, or delivery stops. The buffer index is an estimate of how much earlier a truck would need to depart from an origin at one end of the corridor (expressed as a percentage of the average trip time) to have a 95 percent chance of arriving at a destination at the other end within the average trip time. The I-45 corridor appears the least reliable by this measure because it is the shortest corridor (times for shorter trips will almost always be more variable, in percentage terms, than times for longer trips because delays tend to average out over a long itinerary) and because a high percentage of its length is through urban areas. I-5 has the lowest average speed in part because the truck speed limit over a large portion of the route is 55 mph. The FHWA data indicate that intercity truck traffic operates fairly unimpeded over much of the highway system. Trucks encounter sub- stantial delays in certain urban areas, but not in all. Trucks on I-5 were significantly delayed at SanDiego, LosAngeles, Portland, Seattle, and near theCanadian andMexican borders, but on I-65, trucks encountered only

Freight System Performance and Infrastructure Finance 55 minor or no delays at Mobile, Montgomery, Birmingham, Nashville, Louisville, Indianapolis, and Gary. This observation is consistent with estimates of truck congestion costs by a TRB committee that studied the social costs of freight transportation. The TRB estimates indicate that nearly all congestion costs of long-distance truck travel are incurred in urban areas and that these costs are highly sensitive to time of day. For example, if a truck can time its departure so that it passes through inter- mediate urban areas at night,most congestion costs are avoided (although any cost of altering operating practices to avoid congestion is itself a cost of congestion) (TRB 1996, 94–95, 109). The FHWA truck travel time measurements are an excellent example of the kind of monitoring that would provide useful information about freight systemperformance. Although the report of the 2005 results states FHWA’s intent to continue the monitoring, no further data have been published. Another FHWA study highlights the concentration of highway freight capacity problems at a small number of urban bottlenecks. Using the same method for estimating congestion that is used in theUrbanMobility reports, the study identified locations that account for large travel time losses for freight trucks. The map in Figure 2-1 shows expressway inter- change bottlenecks. They account for 80 percent of expressway recurring congestion delay for trucks, according to the study estimates, and 20 per- cent of all truck congestion delay [the remainder is primarily nonrecur- ring congestion (congestion caused by crashes, breakdowns, weather, and construction), which accounts for 60 percent of the total]. The study estimates the direct user cost of recurring congestion delay of trucks to be $7.8 billion annually. The map shows congestion concentrated in urban areas, and particularly in a few urbanized regions: the Northeast corri- dor, Southern California, and Chicago (Cambridge Systematics 2005b, ES-1–ES-4). This finding is consistent with the FHWA motor carrier travel time study results. In summary, growth in highway traffic volume at a rate higher than the rate of addition of capacity is not sustainable indefinitely. Eventually, congestionwill severely curtail traffic growth if it is not relieved bydemand management or capacity expansion. Also, because existing traffic control measures are of limited effectiveness and road users are not accountable

56 Funding Options for Freight Transportation Projects for the delay costs they impose on others, it would be possible, through better management, to reduce delay costs greatly without reducing total benefits to travelers. However, highway congestion is not systematically measured, and therefore the scope and costs of the problem, and how system users are coping with it, are poorly understood. The census travel survey data and the FHWA truck travel timemeasurements do not appear to lend strong support to the characterization of the congestion problem nationwide at present as a crisis, although the cost of congestion in cer- tain metropolitan areas appears extremely high. Rail Congestion The economic significance of congestion in the freight rail system differs from that of highway congestion. The rail operator has an incentive to offer the quality of service that its customers are willing to pay for. Oper- ational mistakes occur, and lack of competition in some freight markets can harm performance, but overall, the rail system can be expected to FIGURE 2-1 Freight bottlenecks at expressway interchanges. (SOURCE: Cambridge Systematics 2005b.)

Freight System Performance and Infrastructure Finance 57 maintain the economically optimum balance between the value of speed and reliability to shippers and the cost to the carrier of providing the service. Highways do not maintain this balance because the market mechanism that regulates rail use does not function for highways, and nonmarket traffic management techniques do not work very well. Con- gestion costs on railroads can be reduced only by technological break- throughs (including advances in management techniques) that allow a given service quality to be provided at lower cost or by reducing non- economic barriers that add to costs and delay (e.g., conflicts between freight and passenger traffic). Average train speed for Class I railroads is an available performance measure related to freight congestion. Average speed rose 30 percent from 1980, at the time that economic regulation of the railroads was curtailed, to 1992, then declined through 1999. The latter period was characterized by rapid traffic growth and rail service disruptions that followed the 1996 Union Pacific–Southern Pacific merger and the 1999 purchase and division of the Consolidated Railroad Corporation (Conrail) by the Norfolk Southern and CSX Railroads. Speeds increased during the 2000–2001 economic downturn but have declined again since (AAR 2007, 37, 38; TRB 2003, 61). The trend in speed reflects changes in the composition of freight traffic as well as changes in speed on services, but it suggests growing congestionduring a periodof steady traffic growth. The rail service disruptions of 1997 and 1999 focused attention on the freight transportation system. Service was seriously degraded over large areas of the country for durations of several months. The Freight Capac- ity committee accepted the view that the disruptions were extraordinary events triggered by industry restructuring (TRB 2003, 63–65). In 2004, during the fall peak traffic season, the Southern California ports experi- enced severe congestion in part because rail capacity for distributing the arriving containers was strained (CBO 2005). A few published studies have measured railroad system performance by analyzing samples of actual freight movements. A review of four such studies conducted between 1977 and 2006 found remarkably constant average trip times and reliabilities for similar categories of movements. The authors concluded that “despite continuing advances in technol- ogy for track, equipment, and control, the service provided to general

58 Funding Options for Freight Transportation Projects merchandise freight has been relatively unchanged formore than 30 years” (Martland and Alpert 2007, 53). The Surface Transportation Board (STB) commissioned a detailed review of railroad industry economic conditions, which was published in 2008. With regard to capacity and performance, the study concludes as follows (Laurits R. Christensen Associates 2008, ES-28): [T]here currently do not appear to be global or networkwide rail capacity con- straints. Rather, as often occurs in network industries, congestion at various points or corridors in railroad networks appears to be the major culprit in capacity-related performance issues over the last ten years. . . . These results [of the STB study] are consistent with the conclusion reached by a number of economic researchers that the railroad industry still has a considerable amount of overall excess capacity. With regard to the trend in capacity, the study concludes as follows (Laurits R. Christensen Associates 2008, ES-30): [R]ecent increases in railroad capital spending, combined with a relatively weak economy, indicate that any capacity tightness that may have existed at the beginning of this decade has likely loosened in recent years. Thus, with the caveat that congestion issues are likely to continue to exist at localized points and cause service performance issues, near-term systemwide railroad capacity constraints are not likely to be a major issue. Regarding the longer- term forecasts of capacity constraints, . . . it is our assessment that one must treat these forecasts of future capacity needs as tentative, at best, particularly given the current economic climate [in] the U.S. Another frequently cited recent analysis of rail capacity, commis- sioned by the Association of American Railroads (AAR), concluded that $148 billion (in 2007 dollars) of infrastructure investmentwill be required to accommodate projected freight traffic growth and maintain service on the railroads from 2007 to 2035 and that without this investment, 30 per- cent of the rail miles in the nation’s major freight corridors would be operating above capacity by 2035, causing severe congestion (Cambridge Systematics 2007, ES-1–ES-2). The author of the STB-commissioned study compares his results with those of the AAR-commissioned study and concludes that the two are consistent with regard to the present state of capacity: although local capacity bottlenecks exist, nearly all major

Freight System Performance and Infrastructure Finance 59 corridormileage is operating below capacity.He cautions that the 30-year projections in the AAR-sponsored study are highly uncertain because of their sensitivity to assumptions about future economic activity, technol- ogy, plant location decisions, and availability of other transportation modes (Laurits R. Christensen Associates 2008, ES-28–ES-31). Congestion at Ports and Other Intermodal Terminals and at Border Crossings Traffic through ports and at border crossings rose dramatically with the growth of international trade over the past two decades. Congestion in port regions and at crossings has been one of the most visible indications of stresses on the freight system, and port interests have been prominent in public debates over freight infrastructure funding. Port congestion can arise from shortages of berths for docking ships; cargo-handling equipment; space for handling and storing cargoes and containers at the port; or capacity on the regional access network of high- ways, rail lines, and terminals that handle cargo to and from the port. Access route congestion is exacerbated by conflicts between freight and passenger traffic. As with other components of freight infrastructure, data that would give a systematic view of port congestion delays and costs are lacking. The TRB Freight Capacity committee reviewed the available evidence of port congestion and concluded that “while growth of some categories of freight at certain ports has been remarkable, there is some evidence that the waterside facilities of the U.S. port system as a whole are at pres- ent not capacity constrained” (TRB 2003, 67). With regard to landside access capacity, the TRB committee cited only a 2000 congressionally commissioned USDOT study of roads connecting freight terminals, including ports, to the National Highway System (NHS), which found that data are insufficient to judge the adequacy of funding levels devoted to improving these roads (TRB 2003, 66). Traffic trends at the Ports of Los Angeles and Long Beach, which together handle more than one-third of U.S. marine container traffic, illustrate the source of congestion problems. Container traffic at the two ports [measured innumbers of 20-foot container-equivalent units (TEUs) passing through the ports annually] grew 153 percent (9.7 percent per

60 Funding Options for Freight Transportation Projects year) from 1991 to 2001 and 62 percent (8.4 percent per year) from 2001 to 2007. The volume was 15.7 million TEUs in 2007, compared with 3.8 million in 1991 (AAPA 2008). In the 2004 fall peak season, congestion delays at Los Angeles–Long Beach were severe, and traffic was diverted to other ports. However, the disruptions have not recurred. By the next year, ports, shippers, and car- riers had begun instituting operational changes to manage or avoid con- gestion, and by 2007 and 2008, traffic was declining at the ports as a result of general economic conditions. The TRB Freight Capacity committee identified growing congestion at intermodal terminals as among the trends that have shaped industry concerns about freight systemperformance. Important intermodal nodes include the seaports and inland rail hubswhere containers are transferred between railcars and trucks. Only fragmentary data are available on per- formance of intermodal terminals. Problems with intermodal operations at Chicago, the most important inland hub, are the topic of the case study on the CREATE project summarized in Chapter 3. Data on the magnitude and costs of delays at land border crossings likewise are sparse. U.S. trade with its North American Free Trade Agree- ment partners Canada and Mexico grew at 6 percent annually from 2001 to 2007, reaching $909 billion in 2007 (Gallagher and Cassidy 2008). USDOT’s most recent published measurements of delays are for 2001, before the 2001 terrorist attacks that occasioned greatly increased border security precautions. The average delay for trucks was less than 20 min- utes at 11 of the 14 traffic flows observed (two directions at each of seven crossings), but variability of delay was large, with 5 percent of all trucks entering the United States waiting an hour or more (TTI and Battelle Memorial Institute 2002, v–vi). More recently, the Ontario Chamber of Commerce has conducted shipper surveys and estimated costs of border delays. The Ontario report identifies infrastructure deficiencies as a source of delays but emphasizes administrative delays: “Numerous and complex regulations governing the Ontario–U.S. border is the top ranked border concern from a 2006 survey of Ontario Chamber of Commerce members” (Ontario Chamber of Commerce 2007, 19). The report claims that “at least 44 different Canadian and U.S. agencies have jurisdiction over border operations”

Freight System Performance and Infrastructure Finance 61 (Ontario Chamber of Commerce 2007, 2) and identifies lack of coopera- tionbetweenU.S. andCanadian agencies as a primary obstacle to improve- ment. It cites aCanadian government estimate that the cost ofU.S. security measures to Canadian truck operators is C$250 million to C$500 million per year and a somewhat more speculative estimate that the total added logistics cost to U.S. and Canadian businesses from border delays is C$13.6 billion per year (Ontario Chamber of Commerce 2007, 10, 19). Freight Rates and Freight Industry Productivity Freight rates are an economic measure of freight system performance in a competitivemarket. If freight transportation productivity is improving, shippers in a competitive freight market will be able to buy better service (i.e., more reliable and faster transport) at a given price or pay a lower price for a given level of service. Conversely, if productivity is declining, rising carrier operating costs will cause rates to rise. Capacity constraints (which may be shortages of labor, equipment, or infrastructure) also will cause rates to rise. When capacity is tight, carriers have the opportunity to increase profits by raising rates. Railroads can increase margins when rail infrastructure begins to be congested; profits in these circumstances allow the railroads to recover their full costs and provide funds for replacement and expansionof infrastructure. Truck operators experience highway con- gestion as an increased cost, which nonetheless affects rates. The TRB Freight Capacity committee summarized freight railroad rate and productivity trends through the 1990s. Constant-dollar average rev- enue per ton-mile fell nearly continuously from at least the 1950s until 2002 andmore rapidly after 1980, whenmost economic regulation of rail- roads was ended. Constant-dollar average revenue in 2002 was one-third the level of 1960 (TRB 2003, 61–62; AAR 2007, 31). The report cautions that ton-miles is an imperfectmeasure of freight industry output and that trends in rail costs and average rates depend on changes in themix of traf- fic among bulk commodities, containers, and other lines of business. Labor productivity (quality-adjusted output per worker hour) grew from 1980 to 1998by 109percent in the railroad industry comparedwith 19per- cent in all U.S. business (TRB 2003, 74). The trend of continuously falling average rates was interrupted after 2002, as constant-dollar average revenue per ton-mile rose 15 percent

62 Funding Options for Freight Transportation Projects from 2002 to 2007 (AAR 2008, 31). The Producer Price Index for the rail transportation industry (which is adjusted for changes in the mix of outputs within the industry) indicates that constant-dollar rail freight rates rose 27 percent from 2002 to 2008 (BLS 2009). The industry prob- ably took advantage of tight capacity to increase margins in this period, although a large share of the average rate increase is attributable to the price of fuel. Fuel expense rose from 9 percent of total industry operat- ing expenses in 2002 to 21 percent in 2007, the result of a tripling in fuel price (AAR 2008, 61). Revenue, output, and productivity data are less available for truck- ing than for railroads. USDOT estimates indicate that constant-dollar average expenses per mile for for-hire truck transportation declined 11 percent from 1990 to 2000 (FHWA 2002). Labor productivity in truck- ing increased 38 percent between 1980 and 1998, more slowly than in the rail industry but faster than in all U.S. business. The Producer Price Index for the truck transportation industry (BLS 2009), available only for 2003–2008, indicates that constant-dollar truck rates rose 5 percent during the period. Rising fuel prices would more than account for this increase. Data are not available that would allow systematic examination of trends in port services charges or port productivity. By onemeasure of cap- ital productivity, annual containermovements per terminal acre,U.S. con- tainer port productivity appears low by international standards. If ports at which transshipment is not a major share of activity are compared, the major East Asian and European container ports were about 60 percent more productive by thismeasure thanwereU.S. ports in 2000 (Vickerman 2003). The difference was mostly attributable to longer operating hours at the foreign ports. SomeU.S. ports have increased their hours of operation, so this gap may be closing. A case study analysis of U.S. port finance and investment in the 1990s concluded that productivity was not improving and that most new terminals did not exhibit increased productivity com- pared with older facilities. Landside access bottlenecks did not appear to be the cause of stagnant productivity (Ricklefs 2000). In summary, throughout most of the past 25 years, U.S. freight ship- pers have experienced declining constant-dollar rates and (judging from the physical performance data described in the preceding section) more or less constant service quality. From this aggregate, nationwide per-

Freight System Performance and Infrastructure Finance 63 spective, the available data on costs and performance do not reveal any critical deterioration in freight system performance. Severe congestion and service disruptions occur but for the most part have been episodic and localized. The trend of declining rates has been interrupted by the rise in fuel price of the past 5 years. The future course of the price of fuel will have fundamental influence on the form of development of freight transportation. Publicly provided infrastructure has not achieved pro- ductivity gains comparable with those of the freight carrier industries. Port productivity has been stagnant or growing slowly. Data described in the next section show that highway infrastructure productivity improve- ment also has been slow. Trends in Investment and Capital Stock Versus Traffic Growth The disparity between the rate of growth of highway traffic and the rate of growth of constant-dollar highway infrastructure capital expendi- tures is sometimes cited as evidence that increased spending is justified. From 1971 to 2004, annual constant-dollar highway capital expenditures increased by 50 percent while vehicle miles of travel (VMT) increased 150 percent (TRB 2006, 28). The TRB Freight Capacity committee noted that this comparison is not directly relevant to performance because a constant rate of investment can produce growth in the stock of infra- structure if assets have long lifetimes. A more meaningful question is whether the stock of infrastructure is growing along with traffic (TRB 2003, 52). Comparisons between travel growth and simple physical mea- sures of system size (e.g., highway lane miles, as in Figure 2-2, or railroad track miles) also are inconclusive, because most types of capital improve- ments increase capacity although many do not increase lane miles. For example, improving highway pavement quality, straightening alignment, installing and upgrading signals, widening lanes, and reconstructing intersections all increase highway capacity. Similarly, rebuilding track to increase load-bearing capacity and upgrading signal and communications systems increase railroad capacity. Both the Bureau of Economic Analysis (BEA) and FHWA have devel- oped economic measures of the capital stock of highways, and BEA pub- lishes a measure of railroad infrastructure capital stock. These measures

64 Funding Options for Freight Transportation Projects are derived from data on all past capital expenditures for infrastructure and estimates of the rates of depreciation of facilities. A capital stock measure combines the stocks of different kinds of facilities in a single index of capacity by weighting each facility according to the cost of pro- viding it (TRB 2006, 55–56). It is an imperfect measure of capacity, as any single index of a multidimensional characteristic must be, but it is more comprehensive than the lane-mile or track-mile measures. Between 1990 and 2003, the BEA estimate of the net capital stock of highways and streets increased 27 percent (BEA 2007b, Tables 7.2A, 7.2B), somewhat slower than VMT increased in the period (35 percent) but nine times greater than the increase in lane miles. In other words, BEA estimates that governments are investing in highways faster than they are wearing out and that the stock of roads is therefore increasing. The TRB Fuel Tax committee pointed out that it is typical in the pri- vate sector for output to increase faster than the stock of infrastructure, as a result of productivity growth. It reported the following changes in capital–output ratios for highways, railroads, and another capital- intensive network industry, electric utilities, from 1959 to 1995: 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Lane-miles PMT VMT Truck Ton-miles Year 1.60 1.50 1.40 1.30 1.20 1.10 1.00 In de x( 19 90 =1 .0 0) FIGURE 2-2 Growth of passenger miles traveled, vehicle miles traveled, truck ton-miles, and highway lane miles, 1990–2003 (VMT = annual vehicle miles of travel; PMT = annual passenger miles of travel). (SOURCE: FHWA n.d.)

Freight System Performance and Infrastructure Finance 65 Industry and Ratio Change (%) Highways [annual VMT/(productive capital stock)] +16 Railroads [annual ton-miles/(structures net capital stock)] +360 Electric utilities [annual electric energy consumption/(utility net capital stock)] +200 The earlier TRB committee observed that, rather than examinewhy high- way capital has not grown as fast as VMT, highway administrators might more constructively examine why productivity growth in the highway industry has been slower than growth in comparable industries (TRB 2006, 56). [The measure of highway infrastructure productivity above is not ameasure of the productivity of the highway freight industry (i.e., the trucking industry).] The increasing ratio of traffic volume to capital stock in the railroad industry and (to a lesser extent) in highway transportation does not demonstrate improved productivity if the cost of declining service qual- ity because of rising congestion has offset the savings from higher capac- ity utilization. The data on rail congestion summarized above suggest that railroad service quality, for at least some kinds of freight, has been fairly constant in recent decades. Highway congestion has been worsen- ing, so highway capital productivity may be declining. Traffic Projections An examination of aggregate traffic trends is relevant as an indication of the magnitude of the task of providing transportation services and infra- structure to sustain commerce in the coming decades. From1995 to 2006, truck and rail traffic grew more slowly than did gross domestic product (GDP), air cargo somewhat faster, and port container traffic twice as fast (Table 2-2). Growth of truck and of port container traffic in the past decadewas slower than in the preceding 20 years. These rates will not nec- essarily be sustained, but if continued until 2020 they would result in increases of 31 percent in truck and 47 percent in rail traffic, more than a doubling of air cargo, and a 140 percent increase in port container traffic. A 2006 analysis of demographic and travel behavior patterns that influ- ence VMT growth (Polzin 2006) projected that household VMT, which grew by 150 percent between 1977 and 2001, will grow by only 50 to 60 percent between 2001 and 2025, less than half the historical annual

66 Funding Options for Freight Transportation Projects rate, because of slower growth in population, in annual trips per person, and in other factors. The analysis does not consider fuel price explicitly. The results of other research suggest that even large increases in fuel prices might not have a major impact on household VMT, because motorists mainly accommodate higher fuel prices by buying more fuel-efficient vehicles rather than by reducing the amount they travel (Small and Van Dender 2007, 17). The most recent published comprehensive freight traffic forecasts are those of the U.S. Department of Energy (DOE) in itsAnnual EnergyOut- look. DOE’s Reference Case projections are for a 22 percent increase in truck VMT from 2007 to 2020 (1.6 percent per year) and an 11 percent increase in rail ton-miles in the same period (0.8 percent per year) (EIA 2008). Both projections are for much slower growth than historically: from 1990 to 2006, annual growth rates were 2.6 percent for truck VMT and 3.4 percent for rail ton-miles. The projections assume high and ris- ing energy prices over the period [petroleum at $116 per barrel (in 2007 dollars) in 2020] and moderate GDP growth. It was noted above that the growth of imports of containerized mer- chandise from Asia through the Southern California Ports of Los Angeles and Long Beach over the past 15 years strongly influenced perceptions TABLE 2-2 Trends in U.S. Domestic Freight Traffic Projected Growth, 1995– 1975– 2006 to 2020, at 1975 1985 1995 2006 2006 1995 1995–2006 Rate (%) Combination truck 47 77 115 143 1.9 4.6 31 vehicle miles (billions) Rail ton-miles (billions) 754 877 1,306 1,772 2.8 2.8 47 Port container TEUs 11.7 22.3 44.4 6.4 6.7a 140 (millions) Air cargo ton-miles, 23.2 39.7 5.0b 98 U.S. carriers (billions) GDP ($2000 trillions) 4.3 6.1 8.0 11.3 3.2 3.2 56 a1985–1995. bChanges in reporting requirements may have affected apparent growth. SOURCES: AAR 2007, 27; FHWA 2008, Table VM-1; AAPA 2007; BEA 2007a, Table 2A; FAA 2000, Tables 17 and 18; FAA 2007, Table 19. Growth (%/year)

Freight System Performance and Infrastructure Finance 67 of freight transportation investment needs. Accommodating the growth required expansion of the ports, major improvements in port access in the Los Angeles metropolitan area, and expansion of the transcontinen- tal capacity of the railroads, which distribute a major share of the goods nationwide. From 1994 to 2006, container traffic through the two ports more than tripled, from 5.1 million to 15.8 million TEUs per year. In the same period, container traffic increased 136 percent at all West Coast U.S. ports and 117 percent at all U.S. ports (AAPA 2008). The growth in port container volumes closely parallels the growth in the value of total U.S. imports of goods in the period (Figure 2-3). Container transport capacity requirements are essentially determined by the containerized import volume, since imports exceed exports and containers must be returned whether loaded or empty. There are grounds for expecting that the historical rates and patterns of trade growth of recent decades will not continue unaltered. Some import merchandise traffic is diverting from Southern California to less congested West Coast and East Coast ports, and railroads are adjusting their expansion plans accordingly (Boyd 2008; Mongelluzzo 2008). Growth in the total volume of merchandise imports will at some point begin to moderate as domestic markets for particular categories of goods become saturated. It is generally recognized that the present level of the U.S. trade deficit is not sustainable because the deficit is causing U.S. debt to rise faster than income; the eventual correction will involve a decline in imports and a rise in exports (CBO 2007, 1–3). Goods make up a smaller share of U.S. exports than of imports and some growth in goods exports can be accommodated with existing landside capacity, so the adjustment will tend to relieve pressure on freight infrastructure serving international trade. Growth in U.S. exports of noncontainerized bulk commodities (e.g., coal and grain) would eventually necessitate land transport capacity expansion. Summary: Performance Trends The TRB Freight Capacity committee characterized the past performance of the freight transportation system as follows (TRB 2003, 15): Thehistory of freight transportation in theUnited States has beenone of nearly continuous, often dramatic, productivity improvement. The performance of

0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 19 67 19 69 19 71 19 73 19 75 19 77 19 79 19 81 19 83 19 85 19 87 19 89 19 91 19 93 19 95 19 97 19 99 20 01 20 03 20 05 20 07 Year Pe rc en t 0 200 400 600 800 1000 1200 1400 1600 1800 19 90 19 91 19 92 19 93 19 94 19 95 19 96 19 97 19 98 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 - I Year Bi llio ns of 20 00 do lla rs (a) (b) (c) 0 5,000,000 10,000,000 15,000,000 20,000,000 25,000,000 20 07 20 06 20 05 20 04 20 03 20 02 20 01 20 00 19 99 19 98 19 97 19 96 19 95 19 94 19 93 19 92 19 91 19 90 Year An nu al co nt ain er TE Us FIGURE 2-3 Selected U.S. goods import trends: (a) U.S. imports of nonpetroleum goods as percentage of GDP, 1967–2008 (first quarter) (BEA 2008, Tables 1.1.5, 4.2.5, 4.2.6); (b) real imports of nonpetroleum goods (BEA 2008, Tables 1.1.5, 4.2.5, 4.2.6); (c) U.S. Pacific Coast port container traffic (AAPA 2008). 68 Funding Options for Freight Transportation Projects

Freight System Performance and Infrastructure Finance 69 the freight transportation sector has been instrumental in allowing the United States to become the world’s largest integrated market and to partic- ipate successfully in global trade. The sector has in this way advanced eco- nomic welfare, and interruption of the historical trend of productivity improvement would be a substantial loss. The updated trends summarized in this section do not reveal any fun- damental reversal in this historical performance record. Providers of freight infrastructure and services recently have shown adaptability in responding to pronounced shifts in the patterns of demand caused by international trade growth and to disruptive swings in fuel prices. Freight system users (and passengers, on facilities shared by passen- gers and freight) do not pay all the costs of the services they receive; therefore, performance of the system from the standpoint of users is not a complete test. The fees and special taxes that truck operators and port and waterway users pay do not fully cover the cost to the government of providing infrastructure for them. In part because fees for public infra- structure do not fully reflect costs, excessive congestion hampers opera- tions at certain major nodes of the freight system. The economic benefit of relieving these bottlenecks probably would be substantial. Also, all the freight modes have environmental costs for which their users are not responsible. Cost-effective regulations or other measures to reduce these costs would improve freight system performance from the standpoint of the public as a whole. In the past 40 years the industry has benefited from a series of major stimuli to productivity growth: the construction of the major part of the Interstate highway system in the 1960s and 1970s, economic deregula- tion in the late 1970s, and applications of information technology to improve resource utilization from the 1980s to the present. New break- throughs will be needed if the industry is to match its historical record in the future. Along with the entire economy, the industry faces great uncertainty over the next decade with regard to energy prices, environ- mental regulation, and change in patterns of trade and consumption. Finance reforms that promoted efficient investment and operations could make an important contribution to maintaining industry perfor- mance in this period.

70 Funding Options for Freight Transportation Projects RELATION OF FINANCE TO PERFORMANCE Finance arrangements in public infrastructure programs affect project selection decisions, the distribution of costs and benefits, and the oper- ating efficiency of facilities. For example, the decision of Congress in the Federal-Aid Highway Act of 1956 to fund the federal highway program through dedicated revenue from taxes on users has influenced the devel- opment of the highway system (TRB 2006, 80–81). If instead Congress had decided to fund the program from general revenue or from tolls, the evolution and performance of the system probably would have differed greatly from those of the highway system as it actually has developed. Any new government programs to build or pay for infrastructure should be designed with the connection between finance arrangements and out- comes in mind. The choice among funding from user charges (e.g., the tolls on a toll road or the charges a railroad collects from its customers); from a spe- cial tax related in some way to use, like the motor fuel taxes paid into fed- eral and state highway trust funds; or from other public revenue (general tax revenue or other broad-based taxes) is the most consequential deci- sion in designing the finance arrangements of such programs.When con- struction and operation of a facility must be covered by revenue from fees charged to users, only facilities for which there is a strong expecta- tion of sufficient revenue will be built. If a facility is built and operated with user charge revenue, benefits to users are certain to exceed its con- struction and operating costs. From the perspective of national economic well-being, the important benefits of a freight infrastructure improve- ment are the reduced cost of transporting the freight that traverses the improved facility; therefore, when user charges are relied upon, the risk of poor investments in low-value facilities is reduced. Reliance on user charges does not guarantee optimum investment. In some circumstances, a worthwhile facility cannot be fully funded by such charges; an example is a toll road that must compete with an untolled parallel route. In addition, when the provider of infrastructure is a pub- lic or private monopoly, user charge funding does not guarantee that the provider will build facilities of the design and scale that would be most beneficial to users. Nonetheless, compared with the alternative of fund- ing from general tax revenue (for either publicly provided facilities or

Freight System Performance and Infrastructure Finance 71 subsidies to private-sector providers), user charge funding of freight facilities is likely to lead to more disciplined investment decisions and therefore to a higher return on the investment program. In addition to influencing which facilities are built, the choice between user charges and general revenue funding affects the efficiency of use of existing facilities. If charges are related to the cost of serving each indi- vidual user, then users for whom the value of the service is less than the cost of providing it are priced off the facility. If congestion charging is employed, peak usage can be regulated so as to maximize the total eco- nomic benefit that the facility yields. The TRB Fuel Tax committee, in its examination of problems of applying road pricing, warned of the risk of poorly designed charges: “user fees can be harmful if the charges that travelers incur for many trips exceed the added cost to the public of providing those trips. User fees can promote efficient use of facilities if they bear at least a degree of corre- spondence to the public’s costs of providing the facilities. Then users will make decisions . . . that take these costs into account. Fees set too low allow wasteful use of facilities, and fees set too high needlessly discour- age travel” (TRB 2006, 85). The committee observed that local govern- ments controlling road pricing might conceivably attempt to divert undesirable traffic to neighboring communities by high charges or, alter- natively, might depress charges below costs to subsidize favored users. It concluded that “because of inexperience, highway agencies do not now have the competence to set mileage fees that maximize the benefits of the transportation system or to use the information provided by fee revenues to improve the payoffs from capacity expansions. Improper pricing prac- tices could degrade system performance and harm the public welfare” (TRB 2006, 148). The committee recommended a program to develop the necessary competencies. TheFuel Tax report addressed charging for roads, but similar concerns may be justified with regard to charges at other government-operated facilities, such as ports. As the previous section described, past TRB committees concluded that public transportation infrastructure programs today are characterized by poor targeting of investments to the highest-payoff projects and by ineffi- cient use of existing facilities. The two subsections below review evidence on the efficiency of operating practices and investment targeting. As noted in the introduction to this chapter, the available evidence is fragmentary.

72 Funding Options for Freight Transportation Projects Operating Efficiency Most publicly provided transportation facilities, including roads, the inland waterways, and airways and airports, experience excessive levels of congestion. That is, with better management of use, many of these facili- ties could carry more traffic than they do today with lower total operating and time delay costs, and the investment required to serve traffic growth would be reduced. The most effective means of managing use in the pub- lic interest would be through pricing. The problem of inefficient levels of congestion is much less severe on the private-sector components of the freight transportation system. From the point of view of the owners of private infrastructure, most congestion costs are internal; that is, the owners have an economic incentive to man- age these costs efficiently. (An exception is the road congestion caused by rail traffic at grade crossings.) USDOT submits biennial reports to Congress that project the effects of alternative levels of highway capital spending (for all levels of government) on highway performance and highway user costs (travel time, vehicle operating costs, and accident costs). The 2006 study, with projections for the period 2005–2024, for the first time includes an estimate of how road congestion pricing would alter the benefits of increased highway capital expenditures. For a scenario that assumed application of optimal conges- tionpricing on all congested roads,USDOTestimated that the annual cap- ital cost to maintain present physical and operating conditions on the highway system over the forecast period would be reduced by $22 billion per year (from$79 billion to $57 billion in 2004 dollars). The road charges would cause some users to give up or alter some trips, but this loss would be offset by higher travel speeds for other users, and the net user benefit of the roads would not decrease. USDOT notes that this estimate represents an upper bound on potential savings, because of the practical difficulties (at least currently) of pricing, but nonetheless indicates the great cost of present inefficient road operating practices (USDOT 2007, 10-3–10-6). The USDOT report does not estimate how congestion charging would affect travel volume or average speeds, but the practice would avoid grid- lock conditions and increase off-peak utilization of capacity. Charging reduces the investment required to maintain the present level of service by increasing effective capacity on existing roads and by eliminating some

Freight System Performance and Infrastructure Finance 73 trips that are worth less to the traveler than the cost they impose on the road authority and other users. Excessive congestion is not the only inefficiency in road operation that could be ameliorated by better pricing. Opportunities exist for consid- erable economies in the management and regulation of truck traffic on highways. Roughly 80 percent of all U.S. freight transportation expendi- tures are for trucking services; therefore, incremental improvements in trucking industry efficiency have high value. USDOT and TRB studies have concluded that changes in truck fees accompanied by changes in highway agencies’ pavement and bridge management could reduce the infrastructure cost of truck traffic, reduce average vehicle operating costs per ton-mile, and at the same time eliminate the implicit subsidy that some trucks probably are receiving because road user taxes they pay do not cover the cost of providing roads for them. This subsidy biases com- petition among trucks, railroads, and waterways. The fee changes would be designed to align payments more closely with the highway agency’s cost for serving individual trucks, which depends on design and weight of the truck, its mileage, and the roads used. The new fees would give truck operators an incentive to consider highway costs in selecting oper- ating practices and truck designs. Estimates of potential net savings are on the order of several percent of annual truck transportation expendi- tures (TRB 2003, 47–80; TRB 2006, 67–68). The users of the inland waterway system (particularly users of the locks on the upper Mississippi River) and of the air traffic control system also bear needlessly high costs from congestion delay, and the effective capacity of these systems is reduced, on account of inefficient demand management. On neither system are users charged directly (although both systems are funded in part through dedicated user tax revenues). User charges would provide a practical and effective means of managing demand so as to increase the public benefit derived from these systems. The revenue would be a source of funding for operations and capital improvements (TRB 2003, 42–43, 85–89; TRB 1991, 231). Investment Decisions The Fuel Tax and Freight Capacity committees identified weaknesses in investment decisionmaking in the public sector—especially dissipation of

74 Funding Options for Freight Transportation Projects available funds on projects with low returns—as a fault of present trans- portation infrastructure finance arrangements that could be repaired by reforms (TRB 2006, 186–187; TRB 2003, 112). Both committees con- cluded that getting higher public returns on investments of available funds will be essential in improving transportation system performance. The states and the federal government have long recognized the need for improved capital programming, and governments have promulgated rules and guidelines calling for adoption of rational evaluation methods (OMB 1997; Neumann 1997). However, the investment decision- making problem is not simply one of inadequate planning techniques in public agencies. Government capital investment decisions are inevitably more difficult than decisions in the private sector. Private-sector deci- sions are guided by markets; that is, revenue and cost data provide man- agers with information about which facilities and lines of business are most likely to generate profit if expanded, and firms that make consis- tently poor investment decisions cease to operate. In the public sector, for facilities that are paid for with tax revenue, no such direct feedback from consumers guides decisions. Public officials may rely to some extent on formal analyses of costs and benefits but for the most part are guided by officials’ judgments regarding constituents’ preferences and equity impacts (TRB 2003, 39). The two earlier TRB committees also concluded that public-sector capabilities for evaluating transportation investments are deficient. On account of the lack of evaluation, it is impossible to measure the extent of misallocation of public capital expenditures (with respect to the alloca- tion that would yield the greatest return). The Freight Capacity commit- tee reviewed case studies of decision making on freight projects and capital programs for intercity highways, a multimodal corridor proposal, inland waterways, and port access improvements. It found that decisions consistently were made without good estimates of benefits or systematic exploration of alternatives, were sometimes inconsistent with the eco- nomic information that was available, and sometimes appeared to be biased in favor of capital-intensive solutions over operational improve- ments (TRB 2003, 113–117, 137–138). The committee also concluded that government research and development spending (a form of capital investment) has not established a record of success in developing and fos-

Freight System Performance and Infrastructure Finance 75 tering implementation of new transportation system technologies (TRB 2003, 40–41). The three subsections below review the limited available information on the quality of public transportation investment decisions for high- ways and other modes and the question of antifreight bias in public- sector investment decisions. The observation that public-sector projects sometimes have low returns is relevant to the committee’s study in two respects. First, any new federal or state freight infrastructure programs should be designed to maximize economic returns on investments. Therefore, it will be necessary to understand the sources of poor invest- ment decisions and to provide for avoiding these risks in future pro- grams. It was argued above that the surest way to avoid investing in low-return projects is to require project-level funding by means of user charge revenue; that is, finance arrangements will influence how well investment dollars in any new program are spent. Second, the incidence of projects with low return (or of projects motivated by considerations other than national economic benefits) provides a test of the merits of proposals to increase investment in freight infrastructure with public funding. If infrastructure is under- funded, then a backlog of projects for which users would willingly pay should exist, once institutional obstacles to finance and construction were removed. Instead, in several of the most prominent freight proj- ects and proposals of recent years (including some of the case studies described in Chapter 3), it is acknowledged that users would not be willing to pay the full costs. If underfunding is a crisis, as asserted by several observers cited at the beginning of this chapter, then it should not be difficult to find examples of a backlog of projects with very high returns, and such projects should be crowding out most low-return projects. The data are insufficient to allow the committee to judge the size of the backlog of high-return projects. However, it would be prudent to verify the existence of a backlog before enacting new spending programs. In this section, as throughout this report, references to economic ben- efits or to returns on investment aremeant to encompass bothmarket and nonmarket benefits. Themarket benefits of a project that improves freight infrastructure include reduced prices of goods and services resulting from

76 Funding Options for Freight Transportation Projects lower transportation costs. Nonmarket benefits may include reduced pollution and congestion costs of freight transportation. Highway Investments The TRB Fuel Tax committee reviewed evidence of the return earned by U.S. highway investments in recent decades, includingUSDOT’s biennial reports to Congress on the conditions and performance of the U.S. high- way system, and four econometric studies from the academic research literature (TRB 2006, 68–73). The USDOT conditions and performance reports are derived from a sample database of U.S. road conditions and a model of benefits and costs of road improvement projects. The 2006 report’s (USDOT 2007) projections (similar to those in the earlier report reviewed by the 2003 TRB committee), for the period 2004–2024, indicate that if all high- way capital projects nationwide with a benefit–cost ratio greater than 1 were carried out, annual capital spending over the period would aver- age 87 percent higher than actual 2004 spending. Most of the incremen- tal benefit of this optimal program probably could be attained with a much smaller increment in spending (CBO 1998, 21). The 2006 USDOT study did not report rates of return. However, the 2000 study reported that if all projects with benefit–cost ratio greater than 1 were carried out over the 20-year period 1998–2017, the average benefit–cost ratio would be 3.7 (USDOT 2000). The USDOT studies thus indicate that the direct benefits to highway users of additional capital spending for highway system preservation and expansion would exceed the cost to the gov- ernment and that spending would have to expand to a level substantially greater than present spending before highway agencies ran out of worth- while projects, provided that projects are carried out in the order of the benefit–cost ratios. The Fuel Tax committee concluded that the evidence, although highly incomplete, indicates that the nation has earned a positive return on the investment in the highway system, that historical annual expenditure levels have been justified by the incremental benefits received, and that the highway finance system—in particular, reliance on user fees as the primary source of funds—has contributed to this success. The fees act (imperfectly) as prices, discouraging low-value trips, and the constraint

Freight System Performance and Infrastructure Finance 77 of dependence on revenue generated from users reduces the risk that overall spending will exceed economically justified limits. The commit- tee also concluded, however, that the highway finance system does not provide a strong internal check that individual projects are economically justified (as would a system under which more projects had to rely on funds they generated themselves through tolls) (TRB 2006, 185–186). The Fuel Tax committee noted that each of the economic research studies reviewed measured a decline over time in the benefits of incre- mental investment in the highway system (although the timing of the decline was inconsistent among the studies). The studies reported esti- mates of the effects of past increases in the supply of highways on truck transportation costs, inventory holding costs, and productivity in in- dividual industry sectors (TRB 2006, 204–207). Two of the studies concluded that the most recent incremental investments were earning returns below the opportunity cost of the funds invested. With regard to the cause of the apparent downward trend in returns, the authors of one of the studies speculated as follows: “It is also possible that inefficient highway pricing and investment policies have undermined the benefits from government spending. The inefficiencies associated with such policies include but are not limited to wasteful pork barrel spending, poor responses to demographic changes, and suboptimal maintenance of the road system. . . . [L]engthy delays and large cost overruns . . . have postponed service improvements and lowered returns” (Shirley and Winston 2004, 212–213). In addition, the period analyzed included the initial construction of the Interstate system, the first completed com- ponents of which produced large returns, while successive incremental expansionswould be expected to yield declining returns.However, declin- ing returns on transportation investments is by no means inevitable. In a healthy, growing industry, technological progress and changing patterns of demand (as population and economic activity grow and relocate and consumers’ preferences for transportation services evolve) should con- tinually create new opportunities for high-return investment. Unusual insight into the capital programming practices of a state highway program was provided by a 2007 performance audit of conges- tion management and reduction efforts of the Washington State Depart- ment of Transportation (WSDOT) in the Seattle region, conducted at

78 Funding Options for Freight Transportation Projects the direction of the state legislature. The audit concluded that, although WSDOT is a national leader in measurement of congestion and in apply- ing congestion management techniques, the goal of congestion reduc- tion plays no explicit role in the state’s capital programming practices. Agency priorities direct investment toward physical preservation of the system, and programming aims to minimize agency costs rather than the sum of agency and user costs. Analysis of congestion impacts usually is not performed until after projects are selected. The audit recommended that the legislature require WSDOT, in forming its program, to project congestion impacts of projects and to evaluate impacts in terms of hours of delay avoided per dollar spent (Sonntag 2007, 51, 55–58). A review of bridge improvement programming in three states (Marshall et al. 2000) lends support to similar conclusions: that statesmaybe overlooking lower- cost alternatives to capital expenditures (i.e., they tend to favor replace- ment over repair) and that they tend to give greater weight to agency costs than to user costs in planning construction programs. The findings of the Washington audit report are consistent with the conclusions of a GAO survey of state highway agencies that asked what factors agencies considered in forming their capital programs. The sur- vey responses indicated that formal analysis comparing benefitswith costs plays almost no role in investment decisions. Among 10 projects whose files were examined in detail, GAO found none for which a benefit–cost analysis had been completed. Expected safety and environmental impacts of projects often were quantified, but other benefits were rarely if ever quantified. Evaluation of the outcomes of completed projects appeared to be rare, and often data on costs and usage that would be needed for such evaluations were not available (GAO 2005, 23–36). A 1998 Congressional Budget Office (CBO) report examined specifi- cally the evidence on the economic effects of federal capital programs, including the federal-aid highway program (CBO 1998). The review of infrastructure programs was updated in 2008 (CBO 2008). For its analy- sis, CBO reviewed economic studies of returns to public investment and data provided by FHWA and the Federal Aviation Administration. CBO concluded that “many federal investment projects yield economic ben- efits that are small, or even negative. Others yield high returns that would

Freight System Performance and Infrastructure Finance 79 be forgone in the absence of federal involvement, but the number of such projects appears to be limited. . . . Increases in federal investment spend- ing that are not targeted toward cost-beneficial projects can reduce growth” (CBO 2008, 2). Low-return projects are common because the award of federal grants is influenced by goals other than maximizing national income. The goals often are primarily distributional: either to aid particular groups or regions or to ensure that all regions receive proportional shares of funds regardless of the relative merits of local projects. These may be legitimate federal policy goals, but the weight that they are given in the direction of federal investment programs reduces the effectiveness of these programs as means of promoting economic growth because fed- erally directed investment will not flow to the projects with the highest economic returns. CBO points out that the net benefit of federal investments funded by general taxes is reduced as a consequence of the structure of federal taxes. The federal tax system distorts relative prices in the private sector and thus reduces economic efficiency. CBO cites estimates that the opportu- nity cost of an extra dollar of federal tax revenue is on the order of $1.20 to $1.60 (CBO 1998, 13). Investments in Other Modes Reviews by National Research Council (NRC) committees and by GAO of U.S. Army Corps of Engineers evaluations of its civil works projects, including improvements to the inland waterways and harbor channels, have reached similar conclusions. A GAO review of Corps of Engineers projects and actions (including one transportation project, the Delaware River channel-deepening project, which is described in Chapter 3) found that “generally, . . . the Corps’ studies understated costs and overstated benefits, and therefore did not provide a reasonable basis for decision- making” (Mittal 2006, Highlights page). For the Delaware River project, GAO found that most of the projected benefits lacked substantiation. The NRC review of planning for reconstruction of the locks on the upper Mississippi River (NRC 2001) similarly concluded that the plan- ning for the project probably overestimated benefits. More problematic,

80 Funding Options for Freight Transportation Projects in the NRC committee’s view, was the superficial consideration given to noncapital alternatives, that is, demand management to reduce con- gestion costs. The committee also noted that no consideration had been given to the consequences of alternative stagings of improvements (e.g., delaying some or all improvements until a portion of projected traffic growth had materialized). When traffic projections are highly uncertain and the contemplated capital improvements have high cost and long life, delaying construction can be a prudent means of reduc- ing the risk of the project. Federal Aviation Administration data on projected costs and benefits of 18 airport improvement projects were reviewed by CBO (CBO 1998, 20–21). CBO cautions that the sample cannot be taken as representative, but the data illustrate how variable returns may be in a single program. The benefit–cost ratio was less than 1 for four of the projects, while for three the ratio exceeded 10. The absence of credible evaluations of benefits of a project does not demonstrate that the benefits are in fact small. However, without evalu- ation, success in targeting investment funds to the highest-return proj- ects will be unlikely. The Question of Antifreight Bias in Capital Programs A frequent complaint of freight industry participants concerning public investment priorities is that transportation agency decisions are biased in favor of passengers over freight (USDOT 2000, 5). Much evidence points to poor public–private coordination in identifying and designing projects that would have freight mobility benefits, including the case studies in Chapter 3 of this report and the case studies and industry interviews con- ducted by the Freight Capacity committee (TRB 2003, 116–118). The 2005 survey of shippers, carriers, and government officials by the MIT Cen- ter for Transportation and Logistics found little public–private collab- oration on freight matters and divergent perceptions of the root causes of congestion—officials citing insufficient infrastructure investment and private respondents citing primarily operational issues (Caplice and Blanco 2006, 5). Nonetheless, the Freight Capacity committee was unable to verify the claim of antifreight bias in programming because the project evaluations and alternatives analyses needed to reveal any bias have never

Freight System Performance and Infrastructure Finance 81 been carried out (TRB 2003, 65–66, 116–117). The committee noted that its case studies demonstrate that major bottlenecks are simultaneously freight and passenger mobility problems. USDOT’s 2000 study of roads that serve as intermodal freight con- nectors and are not part of the NHS reached a similar conclusion: that because states do not systematically evaluate the freight or passenger user costs and benefits of alternative road improvements, it is not possible to judge whether higher spending on these roads would be justifiable. USDOT found that average capital spending per mile on the connector roads was greater than spending on comparable NHS roads of the same classes, although much of the reported spending was concentrated on a small percentage of the connector routes (USDOT 2000, 5, 26, 32). The weaknesses of capital programming practices are more funda- mental than a bias in favor of passengers over freight. The complaints of freight industry participants about poor communication with gov- ernment officials on investment priorities, the lack of information about congestion available to transportation agencies, and the obser- vations that investment decisions often do not take into account con- gestion and other user cost impacts together suggest a more general inefficiency in public transportation programs: that administrators are not proficient in allocating funds to the projects that would yield the greatest user benefits or in budgeting between capital and maintenance expenditures. SUMMARY Government and industry observers have described the primary challenge facing present transportation finance arrangements as a funding gap: that the current rate of investment in capacity is inadequate to accommodate expected future traffic. In this view, the necessary response is an increase in public-sector capital spending, which may be drawn from general tax revenue, special user taxes, or user charges. However, freight transportation system performance, especially on the publicly owned parts of the infrastructure—roads, waterways, and airports and airways—is hindered by operating practices that lead to needlessly high congestion costs and by investment decision making

82 Funding Options for Freight Transportation Projects that often directs capital spending according to noneconomic factors. Considerations related to the distribution of funds and project benefits among regions and constituencies will inevitably influence public-sector investment choices; nonetheless, this multiplicity of objectives weakens the effectiveness of public spending programs for promoting economic growth because available funds are not concentrated on the capital proj- ects that would yield the highest returns. Efforts to maintain freight system performance and expand freight infrastructure in the future will confront social, economic, and institu- tional challenges, regardless of the level of funding. The costs of building infrastructure will continue to rise because of increasing population den- sity and competition for uses of land and because of the greater value placed on environmental quality. Also, in many parts of the country, regional institutions lack capabilities for carrying out complex multi- modal infrastructure projects. The ability of the managers of public infrastructure to identify invest- ment priorities is hindered by lack of information. Public debate over transportation policy is similarly hampered. No nationwide systematic measurement of the performance of public infrastructure is conducted. Commonly cited measures of physical performance (i.e., trends in con- gestion and capacity) are of unknown validity, and no effort is made to measure economic performance of public infrastructure (i.e., in terms of productivity trends and return on investment). Public investment proposals are rarely subject to credible quantitative economic evalua- tion. Outcomes of completed projects are almost never evaluated. The lack of information makes poor investment decisions unavoidable. Correcting the failings of operations management and investment targeting in public infrastructure programs would improve current performance and increase the cost-effectiveness of future investments in maintaining or improving performance. Conversely,major new spending programs that tolerated these failingswould be ineffectual. The failings are linked to the present finance arrangements for public transportation infra- structure. When users of facilities are not responsible for the cost of pro- viding service to them, accountability for investment decisions is weak. Furthermore, when facility use charges are not in place, pricing, the most effective tool for optimizing facility performance, is not available.

Freight System Performance and Infrastructure Finance 83 REFERENCES Abbreviations AAPA American Association of Port Authorities AAR Association of American Railroads AASHTO American Association of State Highway and Transportation Officials BEA Bureau of Economic Analysis BLS Bureau of Labor Statistics CBO Congressional Budget Office ECMT European Conference of Ministers of Transport EIA Energy Information Administration FAA Federal Aviation Administration FHWA Federal Highway Administration GAO General Accounting Office; Government Accountability Office NRC National Research Council OECD Organisation for Economic Co-operation and Development OMB Office of Management and Budget TRB Transportation Research Board TTI Texas Transportation Institute USDOT U.S. Department of Transportation AAPA. 2007. U.S./Canada Container Traffic in TEUs (1980–2006). May 1. AAPA. 2008. U.S./Canada Container Traffic in TEUs (1990–2007). Updated April 22. www.aapa-ports.org/home.cfm. AAR. 2007. Railroad Facts: 2007 Edition.Washington, D.C., Nov. AAR. 2008. Railroad Facts: 2008 Edition.Washington, D.C., Oct. AASHTO. 2007. Transportation: Invest in Our Future: A New Vision for the 21st Century. July. BEA. 2007a. GDP and Other Major NIPA Series for 1929–2006: II. Survey of Current Business, Aug., pp. 168–192. BEA. 2007b. Government Fixed Assets. Aug. 8. www.bea.gov/national/FA2004/Select Table.asp. BEA. 2008. National Economic Accounts, revised June 26. www.bea.gov/national/ nipaweb/Index.asp. BLS. 2009. Producer Price Index IndustryData. Extracted Jan. 14, 2009. http://data.bls.gov/ PDQ/outside.jsp?survey=pc. Boyd, J. D. 2008. Boxes Go East. Traffic World, Jan. 21, pp. 18–21. Bureau of the Census. n.d. a. Commuting Characteristics by Sex; Data Set: 2006 American Community Survey. http://factfinder.census.gov/home/saff/main.html?_lang=en.

84 Funding Options for Freight Transportation Projects Bureau of the Census. n.d. b. Ranking Tables 2000: Average Travel Time to Work of Workers 16Years andOverWhoDidNotWork atHome (Minutes). www.census.gov/ acs/www/Products/Ranking/2000/R04T040.htm. Bureau of the Census. n.d. c. Travel Time to Work for the United States: 1990 and 1980 Census. www.census.gov/population/socdemo/journey/ustime.txt. Cambridge Systematics. 2005a. Executive Summary: Future Highway and Public Trans- portation Financing: Study Release Event. www.uschamber.com/NR/rdonlyres/ ehezkyo2b7sm5xv33grprwecajd44h3xuwzliwbdyy64725l5q3gfmdc7ycypwsvbbhxdagf 5swmoj7gz3b4r3xkqsb/ExecSummary.pdf. Cambridge Systematics (in association with Battelle Memorial Institute). 2005b. An Initial Assessment of Freight Bottlenecks on Highways. Prepared for Federal Highway Administration, Oct. Cambridge Systematics. 2007. National Rail Freight Infrastructure Capacity and Invest- ment Study. Sept. Caplice, C., and E. Blanco. 2006. Freight Transportation Infrastructure Survey: Causes and Solutions to the Current Capacity Crisis. Center for Transportation and Logistics, Massachusetts Institute of Technology, Cambridge. CBO. 1998. The Economic Effects of Federal Spending on Infrastructure and Other Invest- ments.Washington, D.C. CBO. 2005. Freight Rail Transportation: A Review of the 2004 Experience.Washington, D.C., May. CBO. 2007. Will the U.S. Current Account Have a Hard or Soft Landing? Economic and Budget Issue Brief. Washington, D.C., June 11. CBO. 2008. Issues and Options in Infrastructure Investment.Washington, D.C., May. Council of Supply Chain Management Professionals. 2008. 19th Annual State of Logis- tics Report. Cited by A. Field, Spotlight on Efficiency, Journal of Commerce, June 30, pp. 30–31. EIA. 2008. Annual Energy Outlook 2009 Early Release. U.S. Department of Energy, Dec. www.eia.doe.gov/oiaf/aeo/aeoref_tab.html. FAA. 2000. FAA Aerospace Forecasts FY 2000–2011. FAA. 2007. FAA Aerospace Forecasts FY 2007–2020. FHWA. 2002. Expenses per Mile for the Motor Carrier Industry: 1990 Through 2000 and Forecasts Through 2005. http://ops.fhwa.dot.gov/freight/documents/bts.pdf. FHWA. 2008.Highway Statistics 2006. FHWA. n.d. Relationships Between Asset Management and Travel Demand: Findings and Recommendations from Four State DOT Site Visits.www.fhwa.dot.gov/infrastructure/ asstmgmt/vmt.pdf.

Freight System Performance and Infrastructure Finance 85 Florian, M. 2008. Testimony Before the House Committee on Transportation and Infra- structure, Hearing on Investment in Transportation Infrastructure. June 10. Gallagher, J., andW. B.Cassidy. 2008.Difficult Crossing.TrafficWorld,Oct. 13, pp. 18–22. GAO. 2003. Freight Transportation: Strategies Needed to Address Planning and Financing Limitations.Dec. GAO. 2005. Highway and Transit Investments: Options for Improving Information on Projects’ Benefits and Costs and Increasing Accountability for Results. Jan. GAO. 2008. National Policy and Strategies Can Help Improve Freight Mobility. Jan. Laurits R. Christensen Associates. 2008. A Study of Competition in the U.S. Freight Rail- road Industry and Analysis of Proposals ThatMight Enhance Competition. Prepared for the Surface Transportation Board, Nov. Mallett, W., C. Jones, J. Sedor, and J. Short. 2006. Freight Performance Measurement: Travel Time in Freight-Significant Corridors. Federal Highway Administration, Dec. Marshall, A., W. Robert, K. G. Anderson, R. L. Floyd, and F. Corso, Jr. 2000. Compari- son of Pontis Bridge Project Recommendations to Programmed Work for Three U.S. Transportation Agencies. In Transportation Research Circular 498: Presentations from the 8th International Bridge Management Conference, Vol. 2, TRB, National Research Council, Washington, D.C., June, pp. I-1/1–I-1/12. Martland, C. D., and S. Alpert. 2007. Origin-to-Destination Performance for General Merchandise Traffic Moving to or from Short Line Railroads. Journal of the Trans- portation Research Forum, Vol. 46, No. 7, Fall, pp. 51–67. Mittal, A. 2006. Testimony Before the Subcommittee on Energy and Resources, Com- mittee onGovernment Reform,House of Representatives: Corps of Engineers: Obser- vations on Planning and Project Management Processes for the Civil Works Program. Government Accountability Office, March 15. Mongelluzzo, B. 2008. ChangingTimes for Intermodal. Journal of Commerce,Feb. 25, p. 18. National Surface Transportation Policy and Revenue Study Commission. 2007. Report of the National Surface Transportation Policy and Revenue Study Commission: Trans- portation for Tomorrow.Dec. Neumann, L. A. 1997. Synthesis ofHighway Practice 243:Methods for Capital Programming and Project Selection. Transportation Research Board, National Research Council, Washington, D.C. NRC. 2001. Inland Navigation System Planning: The Upper Mississippi River–Illinois Waterway.National Academy Press, Washington, D.C. OECD and ECMT. 2006. Speed Management. Working Group of the Joint Transport Research Centre. OMB. 1997.Capital ProgrammingGuide. July. www.whitehouse.gov/omb/circulars/a11/ cpgtoc.html.

86 Funding Options for Freight Transportation Projects Ontario Chamber of Commerce. 2007. Easing the Checkpoints: A Plan for an Efficient Canada–US Border. Aug. Peters, M. 2007. Statement Before the House Committee on Transportation and Infra- structure. Sept. 5. Polzin, S. E. 2006. The Case for Moderate Growth in Vehicle Miles of Travel: A Critical Juncture in U.S. Travel Behavior Trends.U.S. Department of Transportation, April. Ricklefs, J. E. 2000. The Self-Sufficiency of U.S. Ports and the Role of State Subsidies. In Toward Improved Intermodal Freight Transport Between Europe and the United States: Report of the Third EU–USForum,EnoTransportation Foundation,Washington,D.C. Schrank, D., and T. Lomax. 2007.The 2007UrbanMobility Report.Texas Transportation Institute, College Station, Sept. Shane, J. N. 2006. The State of the U.S. Freight System; Trends and Challenges, 2000–2020. Presentation to the National Surface Transportation Policy and Revenue Study Commission, June 26. Shirley, C., and C. Winston. 2004. Firm Inventory Behavior and the Returns from Highway Infrastructure Investments. Journal of Urban Economics, Vol. 55, March, pp. 398–415. Small, K., and K. Van Dender. 2007. Long Run Trends in Transport Demand, Fuel Price Elasticities, and Implications of the Oil Outlook for Transport Policy.Organisation for Economic Co-operation and Development–International Transport Forum Joint Transport Research Centre Discussion Paper 2007-16, Dec. www.international transportforum.org/jtrc/DiscussionPapers/DiscussionPaper16.pdf. Sonntag, B. 2007. Performance Audit Report: Washington State Department of Trans- portation: Managing and Reducing Congestion in Puget Sound. Washington State Auditor, Oct. 10. TRB. 1991. Special Report 230:Winds of Change:Domestic Air Transport SinceDeregulation. National Research Council, Washington, D.C. TRB. 1993. Special Report 238: Landside Access to U.S. Ports.National Research Council, Washington, D.C. TRB. 1996. Special Report 246: Paying Our Way: Estimating Marginal Social Costs of Freight Transportation.National Research Council, Washington, D.C. TRB. 1998. Special Report 252: Policy Options for Intermodal Freight Transportation. National Research Council, Washington, D.C. TRB. 2003. Special Report 271: Freight Capacity for the 21st Century.National Academies, Washington, D.C. TRB. 2006. Special Report 285: The Fuel Tax and Alternatives for Transportation Funding. National Academies, Washington, D.C.

Freight System Performance and Infrastructure Finance 87 TTI. n.d. Appendix A: Methodology for 2007 Annual Report. http://mobility.tamu.edu/ ums/report/methodology.stm. TTI and Battelle Memorial Institute. 2002. International Border Crossing Truck Travel for 2001. Federal Highway Administration, April 17. U.S. Chamber of Commerce. 2007. Transportation Infrastructure in Crisis—Talking Points. Aug. 3. www.uschamber.com/issues/index/transportation/070803talking points.htm. USDOT. 2000. 1999 Status of the Nation’s Highways, Bridges, and Transit: Conditions and Performance: Report to Congress. USDOT. 2007. 2006 Status of the Nation’s Highways, Bridges, and Transit: Conditions and Performance. Jan. 22. Vickerman, J. 2003. America’s Ports: Falling Further Behind? Presented at Council of Logistics Management Annual Conference. http://transportation.northwestern.edu/ docs/0000/vickerman_CLM03.pdf.