BTS Home > Publications > Transportation Statistics Annual Report > 2005

Chapter 2
Transportation Indicators

Introduction

The Intermodal Surface Transportation Efficiency Act of 1991¹ and subsequent authorizing legislation charged the Bureau of Transportation Statistics (BTS)-now a part of the Research and Innovative Technology Administration-with compiling, analyzing, and publishing a comprehensive set of transportation statistics, including information on a specified list of topics.

In this chapter, each of these topics is represented by a series of key indicators. Data tables supporting all the indicators are in appendix B at the end of the report. Appendix table numbers correspond to the figure numbers in the chapter. The chapter is organized thematically rather than in the order the topics are presented in the legislation (table 1). As in the two previous annual reports, BTS includes three topics that are not on the congressional list.

About the Data in the Report

For consistency, most trend indicator data are shown over at least a 10-year period. Because of the differing availability of data among all the indicators included, it has not been possible to use the same 10-year span for each indicator without sacrificing timeliness. Instead, the data span a decade up to the year of most recent data available when this report was prepared. There are some instances where less than 10 years of data are presented-either because the data are not comparable over the period or are not ­available.

With a few exceptions, trend data involving costs were converted to 2000 chained ("real") dollars to eliminate the effect of inflation over time. Appendix B provides both 2000 chained dollar and current dollar value tables. Throughout the text in the report, results of most percent calculations have been rounded up or down, as appropriate, to a whole number. If the percent value is less than 5, data are presented with one decimal point because rounding these data can mask differences when making comparisons. Annual growth rate calculations are made using a logarithmic formula to account for compounding over time.² A reader may not obtain the same percentage or other calculation presented in this report using the tabulated data in appendix B because of the rounding of data on the tables.

Data in this report come from a variety of sources, principally from BTS and operating administrations of the U.S. Department of Transportation. However, other sources are federal government agencies, such as the U.S. Census Bureau, the Bureau of Economic Analysis, the U.S. Environmental Protection Agency, the U.S. Coast Guard, and the Energy Information Administration. To supplement government sources, the report occasionally uses data and information from trade associations, such as the Association of American Railroads and the American Public Transportation Association. Data from any of these sources may be subject to omissions and errors in reporting, recording, and processing. Sampling data are subject to sampling variability. Documents cited as sources in this report often provide detailed information about definitions, methodologies, and statistical reliability.

Source information in the report details where BTS obtained data used (e.g., from a printed document, website, or by direct communication with an individual). The same data BTS obtained from websites and used in this report may not be available to readers because of frequent changes in such postings. However, the day and month of the BTS download is included in the source information, along with the website address (url) at that time.

¹ See 49 U.S. Code 111(c)(1). As this report was nearing completion in 2005, the U.S. Congress enacted the Safe, Accountable, Flexible, Efficient Transportation Equity Act-A Legacy for Users (SAFETEA-LU, Public Law 109-59). This legislation amended section 111(c)(1). These amendments are discussed in this report's chapter 3, The State of Transportation Statistics.

² The formula is: average annual rate = Exp [(lnY-lnX)/(n-m)] -1, where Y is the end year value, X is the initial year value, n is the end year, and m is the initial year.

Section 1: Traffic Flows

Passenger-Miles of Travel

Estimated U.S. passenger-miles of travel (pmt) increased 27 percent between 1992 and 2002 to total 5.0 trillion in 2002, an average of about 17,000 miles for every man, woman, and child (box 1-A) [2].

Almost 87 percent of pmt in 2002 was in personal vehicles (passenger cars and light trucks, including sport utility vehicles, pickups, and minivans) (figure 1-1). Most of the balance (10 percent) occurred by air. Passenger travel in light trucks accounted for one-third of all pmt. Bus was nearly 3 percent of pmt in 2002, while transit-excluding bus-made up less than 1 ­percent.

The growth in pmt between 1992 and 2002 varied by mode and vehicle type. While pmt by light trucks grew 39 percent, passenger car pmt rose 19 percent (figure 1-2). Air carrier pmt grew at 36 percent despite a decline in passenger traffic between 2000 and 2002, which most likely occurred because of the economic downturn at the time and the terrorist attacks in 2001. Pmt by intercity train (Amtrak) declined, although there has been modest growth since 1996. Transit pmt has grown since the mid-1990s.

The increase in pmt between 1992 and 2002 occurred for a variety of reasons. While the U.S. resident population grew less (12 percent) than pmt over this period, the economy grew appreciably. Gross Domestic Product (GDP) increased 37 percent¹ and GDP per capita grew 22 percent between 1992 and 2002 (figure 1-3) [1, 2].

Sources

1. U.S. Department of Commerce, Bureau of Economic Analysis, National Income and Product Accounts, summary GDP table, available at http://www.bea.doc/, as of January 2005.

2. U.S. Department of Commerce, U.S. Census Bureau, Statistical Abstract of the United States, section 1, table 2, available at http://www.census.gov/, as of May 2005.

¹ Calculation is based on chained 2000 dollars.

Passenger Border Crossings

There were approximately 312 million passenger crossings into the United States by land from Canada and Mexico in 2004, an increase of 2.5 percent from the 304 million crossings in 1995¹ [1]. These crossings were made in personal vehicles, buses, and trains, and by pedestrians at U.S. border gateways.² The majority of crossings (82 percent), however, were in personal vehicles.

Crossings from Mexico accounted for more than three-quarters of the total (242 million) in 2004, or an average of 660,000 per day, up from an average of 558,000 per day in 1995. From Canada there were almost 70 million passenger crossings in 2004, about 191,000 a day, a decrease of 31 percent since 1995.

In general, the number of crossings by personal vehicle from Canada have been declining since 1996 (figure 1-4). From Mexico , however, passenger crossings by personal vehicle rose 43 percent between 1995 and 1999 and then fell 21 percent (to 191 million) by 2004. Over the 1995 to 2004 period, the largest one-year decline (13 percent) occurred between 2000 and 2001, the year of the terrorist attacks in the United States .

The differences between crossings from Canada and Mexico are most evident for pedestrians (figure 1-5). Almost 20 percent of passenger crossings into the United States from Mexico in 2004 were made on foot, while from Canada only 1.2 percent were. While the number of pedestrian crossings from Mexico fluctuated between 1994 and 2004, they declined 7 percent between 2001 and 2004. Conversely, pedestrian crossings from Canada grew 10 percent between 2001 and 2004 and were the highest (1.1 million) in 2002 for the entire 1994 to 2004 period.

Mexico and Canada had similar numbers of passenger crossings by bus in 2004 (3.4 million and 3.9 million, respectively). Bus crossings constituted 1.4 percent of crossings from Mexico and 6 percent of those from Canada in 2004. In recent years, between 2002 and 2004, bus crossings from Canada declined. Bus crossings from Mexico rose to their highest level in 2002 (3.9 million) and then also declined (figure 1-6).

Considerably more people arrive by train from Canada than Mexico (figure 1-7). In 2004, for instance, over 220,000 people arrived from Canada by train, while only about 13,000 did from Mexico . However, arrivals by train constituted less than 1 percent of all crossings from both Canada and Mexico in 2004.

Source

1. U.S. Department of Transportation, Research and Innovative Technology Administration, Bureau of Transportation Statistics, using data from U.S. Department of Homeland Security, U.S. Customs and Border Protection, Office of Management Reporting, Data Warehouse CD-ROM, May 2005.

¹ 1994 data for passenger crossings by personal vehicle are not available for both Mexico and Canada .

² See, "Surface Border Wait Times" in section 5 for specific information on U.S.-Canada and U.S.-Mexico gateways.

Amtrak Station Boardings

Amtrak ridership increased 18 percent, between fiscal years 1994 and 2004, from 21.2 million riders to 25.1 million riders [1, 4]. The number of riders in fiscal year 2004, about 68,800 per day on average, was the largest ever on the Amtrak system [2].

In numbers of passengers boarded, the top five Amtrak stations in fiscal year 2004 were New York; Washington, DC; Philadelphia; Chicago; and Newark. Almost 40 percent of all passengers boarded at these stations. Over 79 percent of ridership volume is accounted for by Amtrak's top 50 stations [5] (figure 1-8).

Amtrak ridership is heavily concentrated in the Northeast Corridor from Washington, DC, to Boston and to a lesser extent, along the Pacific coast. Among Amtrak's top 50 stations, 19 are located in areas served by Amtrak's Northeast Corridor service.¹ Almost 13.0 million passengers boarded trains at these stations, accounting for almost 52 percent of the entire system's passenger volume in fiscal year 2004. Twenty-one of Amtrak's top 50 stations are located along the Pacific coast. These 21 stations accounted for nearly 18 percent of Amtrak's ridership in fiscal year 2004. The remaining 10 top 50 stations are in Florida, Illinois, Louisiana, Massachusetts, New York, Virginia, and Wisconsin.

Nationally, Amtrak operates 523 rail stations serving 46 states [2, 3]. Of these, 74 are owned by Amtrak, 204 are privately owned, and 245 are owned by a public entity [3]. According to an analysis by the Bureau of Transportation Statistics, Amtrak is accessible to about 35 million rural residents (42 percent of all rural residents). For approximately 300,000 rural residents, Amtrak is the only public intercity transportation available [5].²

Sources

1. Amtrak, Amtrak Annual Report, Statistical Appendix (Washington, DC: 2002).

2. ______. Amtrak Facts, available at http://www.amtrak.com/, as of May 2005.

3. ______. Amtrak Strategic Plan: FY 2005-2009 (Washington, DC: June 29, 2004).

4. ______. Annual Report to Congress, Feb. 17, 2005, available at http://www.amtrak.com/, as of May 2005.

5. U.S. Department of Transportation, Research and Innovative Technology Administration, Bureau of Transportation Statistics, Scheduled Intercity Transportation: Rural Service Areas in the United States, available at http://www.bts.gov/, as of March 2005.

¹ For purposes of this report, Amtrak's Northeast Corridor (NEC) service includes the Boston-Washington mainline plus the Springfield, MA-New Haven, CT and Harrisburg, PA-Philadelphia, PA branch lines.  In recent years, Amtrak's former Northeast Corridor Strategic Business Unit also considered the Boston, MA-Portland, ME; New York, NY-Niagara Falls, NY; and Washington, DC-Newport News, VA routes to be part of the NEC.

² See, "Scheduled Intercity Transportation in Rural America" in section 4 (Variables Influencing Traveling Behavior).

Domestic Freight Ton-Miles

All modes of freight transportation, combined, generated 4.4 trillion domestic ton-miles in 2002, 18 percent more than in 1992 (box 1-B). This represents a growth rate of 1.7 percent per year during the period.

Domestic ton-miles for all modes, except water, grew during most of the 1992 to 2002 period (figure 1-9). Rail grew the fastest (46 percent), closely followed by truck (40 percent) and air (23 percent). Rail and truck accounted for the majority of domestic ton-miles at 37 and 29 percent, respectively, in 2002 (figure 1-10). Truck data, however, do not include retail and government shipments and some imports and, therefore, understate total truck traffic.

Water transportation and oil and natural gas pipelines accounted for 14 and 20 percent of domestic ton-miles, respectively, in 2002. Although domestic waterborne ton-miles decreased 29 percent between 1992 and 2002, waterborne vessels continued to play a prominent role in international trade [2]. U.S. waterborne imports and exports, valued at $728 million, totaled 1.1 billion metric tons in 2002 [1]. Oil and natural gas pipeline combined ton-miles, which grew 7 percent between 1992 and 1996, were stagnant or declining through the rest of the period.

Air freight declined between 2000 and 2001, from 15.8 billion ton-miles to 13.3 billion ton-miles, reflecting the economic downturn at the time, the impact of the terrorist attacks of September 11, 2001, and perhaps restrictions placed on the air transport of U.S. mail packages as a security precaution in late 2001. However, air freight rose again, reaching 13.6 billion ton-miles in 2002.

Sources

1. U.S. Department of Transportation, Maritime Administration, Office of Statistical and Economic Analysis, U.S. Foreign Waterborne Transportation Statistics, available at http://www.marad.dot.gov/, as of February 2005.

2. U.S. Department of Transportation, Research and Innovative Technology Administration, Bureau of Transportation Statistics, U.S. International Trade and Freight Transportation Trends (Washington, DC: 2003).

Commercial Freight Activity

The nation's freight transportation system, all modes combined, carried 15.8 billion tons of raw materials and finished goods in 2002, up 18 percent from 13.4 billion tons in 1993 (figure 1-11).¹ The 2002 freight activity also represented 4,506 billion ton-miles at a value of $10,460 billion (in chained 2000 dollars²). Ton-miles have grown 24 percent since 1993, while value rose 45 percent (figure 1-12 and figure 1-13).

Trucking moved the majority of freight by tonnage and by shipment value in 2002: 9.2 billon tons (58 percent of the total tonnage) and $6,660 billion (64 percent of the total value). Multimodal shipments-a combination of more than one mode-were second by value at 11 percent ($1,111 billion), while waterborne carried 15 percent by weight (2.3 billion tons). Trucking and rail were responsible for 32 and 28 percent, respectively, of the total ton-miles.

These total commercial freight data were calculated by the Bureau of Transportation Statistics, using data from the Commodity Flow Survey (CFS) conducted in 1993, 1997, and 2002 and estimates of activity not covered by CFS (box 1-C). While these total estimates provide the most complete commercial freight picture for all modes of transportation, they exclude most shipments by the retail sector and governments (e.g., goods for defense operations and the collection of municipal solid waste). The estimate also excludes shipments by nongoods-producing sectors (e.g., services, construction, household goods movers, and transportation service ­providers).

¹ All 2002 total commercial freight data here and in the accompanying figures and tables are preliminary.  Although final 2002 Commodity Flow Survey data were available at the time this report was prepared, final 2002 supplemental estimates were still forthcoming.

² All dollar amounts are expressed in chained 2000 dollars, unless otherwise specified. Current dollar amounts were adjusted to eliminate the effects of inflation over time.

Geography of Freight Flows by Mode

The geography of freight flows by mode is determined, for the most part, by the distribution of population and industry and availability of transportation infrastructure. The effect of transportation infrastructure is especially pronounced with waterborne shipments, which rely on inland waterways, including the Great Lakes and the Mississippi River system, and coastal ports (figure 1-14). Some of the leaders in waterborne shipments, for instance California and Washington, are states with large coastal ports. Others, such as West Virginia and Indiana, ship or receive large amounts of freight via the inland waterway system. Some, like Louisiana, ship and receive freight through coastal ports and the inland waterway system.

With the ubiquity of the highway network, the amount of freight moving to and from each state by truck is closely related to population size (figure 1-15). Thus, 8 of the 10 most populated states (California, Florida, Georgia , Illinois, Michigan, Ohio, Pennsylvania, and Texas) are leaders in both inbound and outbound truck shipments.

States producing or consuming large amounts of coal are often the leaders in shipments of goods originating or terminating by rail (figure 1-16). For instance, Wyoming, West Virginia, Kentucky, and Pennsylvania are the four largest producers of coal in the United States. Coal shipments to Georgia , Missouri, Indiana, Wisconsin, and Ohio place these states among the leaders of inbound rail shipments. However, the top commodity originating and terminating in California by rail is mixed freight and the top commodity originating in Minnesota is metallic ores. Texas leads in both inbound and outbound chemical shipments [1].

The amount of inbound and outbound shipments by air, like trucking, is closely related to state population (figure 1-17). A major exception is Hawaii, which, as an island state, is a leader in inbound air freight shipments despite its relatively low population. The Commodity Flow Survey,¹ the source of the data for trucking and air shipments, captures the state origin and destination of shipments but not in-transit shipments. Hence, states with airports that are major air freight sorting and distribution facilities, such as the FedEx facility in Memphis, Tennessee, may not register as leaders.

Source

1. Association of American Railroads, Railroads and States 2002 (Washington, DC: 2004).

¹ See Commercial Freight Activity, especially box 1-C.

Freight Border Crossings

The number of trucks entering the United States from Canada and Mexico rose from 7.7 million in 1994 to 11.4 million in 2004 (figure 1-18). While this resulted in annual growth of almost 4 percent per year, the number of trucks crossing into the United States declined in 2001 and 2003, compared with the previous year. For instance, the number of trucks entering from Canada fell by 3.8 percent and from Mexico by 4.9 percent in 2001. Truck entries in 2003 declined at 52 of the 72 U.S.-Canada ports of entry and 14 of the 22 U.S.-Mexico ports [1].

Between 1996 and 2004, the number of full rail containers entering from Canada increased 350 percent, without declining in 2001 (figure 1-19). From Mexico , the number of these rail containers rose 115 percent during the same period; however, most of the increase occurred between 1996 and 2000. Since 2000, growth has been slight. Rail crossings are also measured in number of trains (figure 1-20). These data show a different pattern, with uneven growth for both Canada and Mexico between 1994 and 2004. Total train crossings hit a low of 38,949 in 1999 and a high of 41,911 in 2003.

Trucks accounted for 64 percent ($453 billion) of total trade in 2004 between the United States and its two largest trading partners, Canada and Mexico . When rail is included, surface transportation carried 89 percent of this trade. The other 11 percent of cross border trade was transported by maritime vessels ($46 billion) and aircraft ($32 billion). Over $32 billion of the vessel trade was with Mexico and $23 billion of the air transported trade was with Canada [2]. Data are not available on the numbers of vessels and aircraft entering the United States from Canada and Mexico , however, as they are for surface transportation.

Sources

1. U.S. Department of Transportation, Research and Innovative Technology Administration, Bureau of Transportation Statistics, using data from U.S. Department of Homeland Security, U.S. Customs and Border Protection, Office of Management Reporting, Data Warehouse CD-ROM, May 2005.

2. U.S. Department of Transportation, Research and Innovative Technology Administration, Bureau of Transportation Statistics, U.S. -North American Trade and Freight Transportation Highlights Transborder Freight Data (Washington, DC: 2005).

Passenger and Freight Vehicle-Miles of Travel

Annual highway vehicle-miles of travel (vmt) amounted to 2.9 trillion in 2003, rising by 26 percent since 1993 [1], an annual 2.3 percent rate of change. Vmt per capita rose by 13 percent during the same period.

In recent years, the makeup and use of the highway vehicle fleet in the United States has changed, altering the share of vmt by vehicle type (figure 1-21). With the increasing popularity of sport utility vehicles and other light trucks, this class of vehicles registered the fastest passenger vmt growth (34 percent) between 1993 and 2003. During the same period, freight vehicle vmt for single-unit and combination trucks grew 35 percent, outpacing total passenger vehicle vmt growth (25 percent). Nevertheless, in 2003, passenger vehicles accounted for more than 90 percent of highway vmt.¹

Vehicle travel has also generally increased in other modes of transportation including freight and passenger rail, air, and transit rail.² Vehicle-miles by rail (measured in train-miles and excluding transit rail) grew 26 percent between 1993 and 2003. Freight train-miles made up over 90 percent of all rail vehicle travel in 2003. This share increased slightly between 1993 and 2003 as freight rail vehicle movements outpaced those of passenger rail over the period (figure 1-22).

Domestic service air carrier aircraft vmt increased by 46 percent between 1993 and 2003. Air carrier aircraft vmt reached 5.7 million in 2000, falling back to 5.5 million in 2001, mainly because of the terrorist attacks that year. Aircraft vmt has grown since then, reaching 6.1 million in 2003 [2].

The biggest change in transit rail between 1993 and 2003 was a doubling of light rail vmt as existing systems were expanded and new systems were built (e.g., in Baltimore, Dallas, Denver, St. Louis, and Salt Lake City). Commuter rail vehicle-miles were up 28 percent over this period and heavy rail vehicle-miles, 21 percent (figure 1-23).

Source

1. U.S. Department of Transportation, Federal Highway Administration, Highway Statistics 2003 (Washington, DC: 2004), table VM-1.

2. U.S. Department of Transportation, Research and Innovative Technology Administration, Bureau of Transportation Statistics, Air Carrier Traffic Statistics (Washington, DC: Annual December issues).

¹ Here, passenger vehicles includes passenger car, light truck, bus, and motorcycle vmt. Passenger cars alone accounted for 57 percent of highway vmt. See table 1-21b for detailed data.

² A vehicle-mile of travel (1 vehicle traveling 1 mile) is a concept that is more easily applied to highway vehicles than to other modes of transportation. For instance, rail can be measured in car-miles (1 car, 1 mile) or in train-miles, which include any number of cars but may be more comparable to highway vmt. For air transportation, vmt is synonymous with an aircraft-mile of travel (1 aircraft, 1 mile).

Section 2: Condition of the Transportation System

Transportation Capital Stock

Highway-related capital stock (public highways and streets, consumer motor vehicles, and commercial truck transportation) represented the majority of the nation's transportation capital stock, $2,917 billion in 2003 (in chained 2000 dollars¹). Public highways and streets constituted the majority (52 percent) of highway-related capital stock in 2003, as well as the largest portion (33 percent) of all transportation capital stock (figure 2-1). The combined value of capital stocks for other nonhighway-related modes of the transportation system, including rail, water, air, pipeline, and other publicly or privately owned transportation, is less than the value of consumer motor vehicles alone (figure 2-2).²

All transportation capital stocks, except for railroads, increased between 1993 and 2003. Highway-related capital stocks were not the fastest growing, however. The most rapid growth occurred in air transportation, which doubled over the period. In-house transportation, which can involve several modes, increased 84 percent. Consumer motor vehicles grew 64 percent; truck transportation, 52 percent; private ground passenger transportation, 38 percent; pipeline transportation, 32 percent; and water transportation, 22 percent [1].

Public highways and streets grew 21 percent, and other publicly owned transportation, which includes publicly owned airway, waterway, and transit structures, grew 25 percent over the period for which data were available. Other privately owned transportation, which includes sightseeing, couriers and messengers, and transportation support activities, grew by 4 percent from 1993 to 2003, while railroad transportation declined by 6 percent over the period.

Capital stock is a commonly used economic measure of the capacity of the transportation system. It combines the capabilities of modes, components, and owners into a single measure of capacity in dollar value. This measure takes into account both the quantity of each component (through initial investment) and its condition (through depreciation and retirements). The Bureau of Transportation Statistics has been developing data on airports, waterways, and transit systems that will enhance the available data on publicly owned capital stock.

Sources

1. U.S. Department of Commerce, Bureau of Economic Analysis, Fixed Assets and Consumer Durable Goods in the United States, tables 3.1ES, 3.2ES, 7.1, 7.2, 8.1, and 8.2, available at http://www.bea.gov/, as of May 2005.

¹ All dollar amounts are expressed in chained 2000 dollars, unless otherwise specified. Current dollar amounts (available in appendix B of this report) were adjusted to eliminate the effects of inflation over time.

² Because the Bureau of Economic Analysis has recategorized capital stock data, the time-series data in this report differ from the capital stock data in previous editions of the Transportation Statistics Annual Report.

Highway Condition

The condition of roads in the United States improved between 1993 and 2003. For instance, the percentage of rural Interstate mileage in poor or mediocre condition declined from 35 percent in 1993 to 11 percent in 2003 (figure 2-3). Poor or mediocre urban Interstate mileage decreased from 42 to 27 percent over this period (figure 2-4).

However, while all classes of rural roads (box 2-A) have improved in recent years, the condition of urban collectors and minor arterials has declined. For instance, 28 percent of urban minor arterial mileage and 34 percent of collector mileage were rated poor or mediocre in 2003, rising from 18 percent and 21 percent, respectively, in 1998.

Just under 41 percent of all U.S. urban and rural roads were in good or very good condition in 2003, while approximately 18 percent were in poor or mediocre condition. The rest were in fair condition.1 In general, rural roads are in better condition than urban roads. In 2003, for instance, 30 percent of urban road-miles were classified as poor or mediocre compared with only 14 percent of rural-miles [1].

Source

1. U.S. Department of Transportation, Federal Highway Administration, Highway Statistics 2003 (Washington, DC: 2004), table HM-64.

¹ These percentages include all classes of roads except local roads or minor collector roads.

Bridge Condition

The condition of bridges nationwide has improved markedly since the early 1990s. Of the 590,853 roadway bridges in 2003, the Federal Highway Administration found that 14 percent were structurally deficient and 14 percent were functionally obsolete. About 33 percent of all bridges in 1993 were either structurally deficient or functionally obsolete [1].

Structurally deficient bridges are those that are restricted to light vehicles, require immediate rehabilitation to remain open, or are closed. Functionally obsolete bridges are those with deck geometry (e.g., lane width), load carrying capacity, clearance, or approach roadway alignment that no longer meet the criteria for the system of which the bridge is a part.¹ While the number of structurally deficient bridges steadily declined between 1993 and 2003, the number of functionally obsolete bridges remained constant (figure 2-5).

In general, bridges in rural areas suffer more from structural deficiencies than functional obsolescence (particularly on local roads), whereas the reverse is true for bridges on roads in urban areas (figure 2-6 and figure 2-7). A large number of problem bridges nationwide are those supporting local rural roads: 118,381 of the 160,659 deficient and obsolete bridges in 2003 (74 percent) were rural local bridges. Problems are much less prevalent on other parts of the highway network. Nevertheless, in 2003, 26 percent of rural Interstate bridges and 16 percent of urban Interstate bridges were deficient or obsolete.

Source

1. U.S. Department of Transportation, Federal Highway Administration, Office of Engineering, Bridge Division, National Bridge Inventory database, available at http://www.fhwa.dot.gov/bridge/, as of January 2005.

¹ Structurally deficient bridges are counted separately from functionally obsolete bridges even though most structurally deficient bridges are, by definition, functionally obsolete.

Airport Runway Conditions

Airport runway conditions stayed about the same at the nation's major public-use airports (box 2-B) between 1997 and 2004¹ [1, 2]. At the nation's commercial service airports, pavement in poor condition remained at 2 percent from 1997 through 2004 (figure 2-8). At the larger group of National Plan of Integrated Airport Systems (NPIAS) airports, the Federal Aviation Administration (FAA) found poor conditions on 4 percent of runways in 2004, down from 5 percent in 1997 (figure 2-9).

FAA inspects runways at public-use airports and classifies runway condition as good, fair, or poor. A runway is classified as good if all cracks and joints are sealed. Fair condition means there is mild surface cracking, unsealed joints, and slab edge spalling.² Runways are in poor condition if there are large open cracks, surface and edge spalling, and/or vegetation growing through cracks and joints [2].

Sources

1. U.S. Department of Transportation, Federal Aviation Administration, National Planning Division, personal communication, February 2005.

2. U.S. Department of Transportation, Research and Innovative Technology Administration, Bureau of Transportation Statistics, National Transportation Statistics 2004, table 1-24, available at http://www.bts.gov/, as of January 2005.

¹ Data on airport runway conditions do not exist for 1994 to 1996 or for 1998.

² Spalling refers to chips, scales, or slabs breaking off of surface pavement.

Age of Highway and Transit Fleet Vehicles

The median age of the automobile fleet in the United States increased, by 19 percent, from 7.5 years in 1994 to 8.9 years in 2004. The median age of the truck fleet,¹ by contrast, began to increase in the early 1990s but has declined since 1997 as the purchase of light trucks increased (figure 2-10). As a result, the truck median age of 6.6 years in 2004 is less than its 7.5 years in 1994 [1].

The age of transit vehicle fleets varies by transit and vehicle type and tends to fluctuate (figure 2-11). The average age of heavy-rail passenger cars and ferryboats increased 7 percent and 10 percent, respectively, between 1993 and 2003.  By contrast, the average age of full-size transit buses decreased 14 percent [1].

The age of fleets as a measure of condition is not very precise. Because of the different characteristics of vehicle fleets across the modes-some serving freight and others passengers, some owned predominantly by businesses, and others by individuals-the measure varies widely.

Source

1. U.S. Department of Transportation, Research and Innovative Technology Administration, Bureau of Transportation Statistics, National Transportation Statistics 2005, tables 1-25 and 1-28, available at http://www.bts.gov/, as of June 2005.

¹ This includes all truck categories: light, heavy, and heavy-heavy.

Age of Rail, Aircraft, and Maritime Vessel Fleets

The average age of Amtrak locomotives and passenger train cars fluctuated in a narrow range for most of the 1990s (figure 2-12). The average age of locomotives was 14 years in fiscal year 2001, up 8 percent from 13 years in fiscal year 1991. Meanwhile, the age of Amtrak railcars dropped from 21 to 19 years over this period. Of the 20,744 Class I freight locomotives in service in 2003, 33 percent were built before 1980, 17 percent between 1980 and 1989, and 50 percent from 1990 onwards [1].

Over 32 percent of the U.S.-flag vessel fleet (almost 13,000 vessels) was 25 years old or more in 2003, up from 19 percent (over 7,500 vessels) in 1993 [2]. However, during the same period, the percentage of the fleet less than 6 years old grew from 11 percent (more than 4,300 vessels) to 16 percent (nearly 6,400 vessels). Of the various components of the fleet, the offshore support fleet was one of the youngest in 2003 with 20 percent of its vessels under 6 years old and 24 percent over 25 years old. The towboat fleet had the highest proportion of older ships (60 percent) in 2003 (figure 2-13).

The average age of U.S. commercial aircraft was 12 years in 2002, up from 11 years in 1992 (figure 2-14). Commercial airlines are air carriers providing scheduled or nonscheduled passenger or freight service, including commuter and air taxi on-demand services. Major airlines-those with $1 billion or more in annual revenues-accounted for 78 percent of commercial aircraft in 2002 [3]. These aircraft were approximately one year younger on average than all commercial aircraft during the 1990s, but the gap narrowed in 2001, and by 2002 the average age of both categories was the same (12 years).

Sources

1. Association of American Railroads, Railroad Facts 2004 (Washington, DC: 2004), pp. 49-50.

2. U.S. Department of Transportation (USDOT), Research and Innovative Technology Administration, Bureau of Transportation Statistics (BTS), National Transportation Statistics 2005, table 1-31, available at http://www.bts.gov/, as of June 2005.

3. ______. calculations using data from USDOT, BTS, Form 41, Schedule B-43, 1992-2002.

Section 3: Accidents

Transportation Fatality Rates

There were about 44,900 fatalities related to transportation in 2003-15.4 fatalities per 100,000 U.S. residents.¹ This is the same rate as in 1993, when there were about 42,800 deaths [1, 3]. Approximately 95 percent of all transportation fatalities in 2003 were highway-related. Most of these people who died were occupants of passenger cars or light trucks (including pickup trucks, sport utility vehicles, and minivans). Air, rail, transit, water, and pipeline transportation result in comparatively few deaths per capita (box 3-A). For instance, railroad incidents resulted in 0.3 deaths per 100,000 residents in 2003² (figure 3-1).

Overall, highway safety remained about the same between 1993 and 2003 when compared with the size of the population. There were 14.7 fatalities per 100,000 residents each year over the entire period. Fatality rates declined 19 percent for occupants of passenger cars but increased 31 percent for occupants of light trucks between 1993 and 2003 (figure 3-2). (This is a period during which the number of registered light trucks increased from 60 million to 87 million [2].) Motorcyclist fatalities per 100,000 residents have been rising since 1998. Pedestrian and pedalcyclist fatality rates (at 1.6 and 0.2, respectively in 2003) have declined the most (down 25 percent and 32 percent, respectively) since 1993.

Similar trends in highway fatality rates are apparent when the rate is based on vehicle-miles of travel (vmt). Passenger car occupant fatalities per 100 million vmt declined 25 percent between 1993 and 2003, while light-truck occupant fatalities per 100 million vmt rose 9 percent (figure 3-3). The motorcyclist fatality rate grew 55 percent during the period. After falling from 25 fatalities per 100 million vmt in 1993 to 21 fatalities per 100 million vmt in 1997, motorcyclist fatalities grew to 38 per 100 million vmt in 2003.³

Sources

1. U.S. Department of Commerce, U.S. Census Bureau, Monthly Population Estimates for the United States, available at http://www.census.gov/, as of December 2004.

2. U.S. Department of Transportation, Federal Highway Administration, Highway Statistics Summary to 1995 and Highway Statistics 2003 (Washington DC: 1997 and 2004), tables VM-201A and VM-1.

3. U.S. Department of Transportation, Research and Innovative Technology Administration, Bureau of Transportation Statistics, National Transportation Statistics 2005, table 2-1, available at http://www.bts.gov, as of August 2005.

¹ This total fatality rate has not been adjusted for double counting across modes because detailed data needed to do so were not available at the time this report was prepared. See table 3-1 in appendix B for further information on double-counting impacts.

² This calculation includes fatalities occurring at highway-rail grade crossings.

³ Because of their magnitude, these motorcycle data are not shown in figure 3-3 (see table 3-3 in appendix B).

Transportation Injury Rates

Each year a far larger number of people are injured than killed in transportation-related accidents. Over 2.9 million people suffered some kind of injury involving passenger and freight transportation in 2003 (box 3-B). Most of these injuries, 99 percent, resulted from highway crashes¹ [1, 2].

Highway injury rates vary by the type of vehicle used (figure 3-4). In 2003, 67 passenger car occupants were injured per 100 million passenger-miles of travel (pmt) compared with 51 injured light-truck occupants. Occupants of large trucks and buses are less likely to sustain an injury per mile of travel. Motorcycle riders are, by far, the most likely to get hurt.

Injury rates for some highway modes declined between 1993 and 2003.2 However, rates for light-truck occupants rose 7 percent, from 48 per 100 million pmt in 1993 to 51 per 100 million pmt in 2003 (figure 3-5). Motorcycling became safer in terms of injuries per mile ridden until 1999; but since then, the injury rate increased from 429 per 100 million pmt to 554 per 100 million pmt by 2003. Bus injuries have fluctuated between 10 per 100 million pmt and 15 per 100 million pmt.

Sources

1. U.S. Department of Transportation, Federal Highway Administration, Highway Statistics 2003 (Washington DC: 2004), table VM-1.

2. U.S. Department of Transportation, Research and Innovative Technology Administration, Bureau of Transportation Statistics, National Transportation Statistics 2005, table 2-2, available at http://www.bts.gov/, as of August 2005.

¹ There is the potential for some double counting involving highway-rail grade-crossing and transit bus data.

² Bicycling, walking, and boating (including recreational boating) are excluded, because there are no national annual trend data estimates of pmt for these forms of transportation.

Motor Vehicle-Related Injuries

There were an estimated 3.6 million motor vehicle-related injuries in the United States in 2003, according to data reported to the U.S. Consumer Product Safety Commission (CPSC)1 (box 3-C) [1]. An estimated 3.3 million of these injuries involved motor vehicle occupants. The rest involved about 133,000 motorcyclists, 127,000 pedestrians, and 59,000 pedalcyclists.

More females than males were treated for minor injuries in 2003 across most age groups (figure 3-6). The 20 to 24 age group sustained almost 494,000 minor motor vehicle-related injuries, 53 percent of them by females. For serious injuries, more males than females were treated across all age groups up to about 75 years (figure 3-7). Again, serious injuries spiked at ages 20 to 24, but male injuries spiked substantially higher. This age group incurred over 41,000 serious injuries in 2003, 62 percent of which happened to males.

In summary, there were sharp peaks in injuries associated with youth: for motor vehicle occupants and motorcyclists, the peak spanned ages 15 to 24; for pedalcyclists and pedestrians, the peak spanned ages 10 to 14. Young males exhibited a substantially greater peak in serious injuries than young females. In addition, the percentage of injuries classified as serious was greater for motorcyclists (20 percent of all motorcyclist injuries were serious), pedestrians (18 percent), and pedalcyclists (13 percent) than it was for motor vehicle occupants (7 percent) (figure 3-8).

This analysis is the second update of a Bureau of Transportation Statistics comprehensive study originally conducted using 2001 data from the CPSC's National Electronic Injury Surveillance System.²

Sources

1. U.S. Consumer Product Safety Commission, National Electronic Injury Surveillance System (NEISS), available at http://www.cpsc.gov/about/clrnghse.html, as of February 2005.

2. U.S. Department of Transportation, National Highway Traffic Safety Administration, Traffic Safety Facts 2003, available at http://www.nhtsa.dot.gov/, as of March 2005.

¹ Because of methodological and other differences, motor vehicle-related injury data from CPSC differ from those estimated by the National Highway Traffic Safety Administration (NHTSA) of the U.S. Department of Transportation. For 2003, NHTSA reported an estimated 2.9 million highway injuries [2].

² For details on 2001 and 2002 motor vehicle-related injuries, see October 2003 and September 2004 editions of Transportation Statistics Annual Report, available at http://www.bts.gov/, as of March 2005.

Highway-Railroad Grade-Crossing Accidents

There were 3,045 collisions between trains and highway users in 2004, of which 319 involved at least one fatality (figure 3-9). These 319 fatal accidents resulted in 368 fatalities, 41 percent of the 896 rail-related fatalities in 2004¹ [2, 3]. The geographic distribution of fatal accidents, such as the cluster around Chicago, is associated with a high number of highway-railroad grade crossings.

Despite an increase in both motor vehicle traffic and rail traffic, safety at highway-railroad grade-crossings has improved markedly since the mid-1970s. Enhanced safety reflects grade-crossing improvements, such as gates and warning signals. The reduction in the number of accidents is also related to public education campaigns, better warning lights on trains, and fewer crossing opportunities. The number of highway-rail crossings declined by more than 30 percent between 1975 and 2004 as a result of grade separation projects, crossing consolidation, and railroad track abandonment [1].

Sources

1. Shannon Mok and Ian Savage, "Why has Safety Improved at Rail-Highway Grade Crossings?" Risk Analysis (forthcoming).

2. U.S. Department of Transportation, Federal Railroad Administration, Office of Safety Analysis, Highway-Rail Crossings, available at http://safetydata.fra.dot.gov/officeofsafety/, as of June 2005.

3. U.S. Department of Transportation, Research and Innovative Technology Administration, Bureau of Transportation Statistics, National Transportation Statistics 2005, table 2-1, available at http://www.bts.gov/, as of August 2005.

¹ At the time this report was prepared, these 2004 data were ­preliminary.

General Aviation Safety

There were 556 U.S. fatalities in 2004 caused by general aviation, amounting to 88 percent of all aviation fatalities in the United States [1]. However, general aviation has become safer between 1994 and 2004. Despite a 16 percent increase in general aviation flight hours during the period, fatalities declined by 24 percent (figure 3-10). In 1994, there were 3.3 general aviation fatalities for every 100,000 flight hours (figure 3-11). By 2004, that rate had fallen to 2.2 per 100,000 flight hours. The total number of general aviation accidents and fatal accidents declined from 1994 to 2004 by 20 and 23 percent, respectively (figure 3-12).

The National Transportation Safety Board (NTSB) often establishes more than one cause or factor to an aviation accident using three broad categories: personnel, environment, and aircraft. There were 1,758 general aviation accidents in 20001 for which NTSB has established causes. Personnel was cited as a cause or factor in 89 percent of those accidents, environment was cited in 45 percent, and the aircraft in 29 percent. Within the broad categories: the pilot was responsible in 95 percent of accidents where personnel was the cause or factor, weather was attributed to 47 percent of accidents where the environment was a factor,2 and in accidents where the aircraft was a factor, 47 percent of the time it could be attributed to the powerplant/propulsion system [2].

Runway incursions are another safety concern in general aviation. Of the 1,804 runway incursions between 1999 and 2003, just fewer than 75 percent of them involved general aviation aircraft. The rate of runway incursions involving general aviation aircraft per million operations increased from 6.0 in 1999, reaching a 5-year high in 2001 at 8.3 runway incursions per million operations. The rate fell back to 6.2 runway incursions per million operations in 2003 [4].

Sources

1. National Transportation Safety Board, Aviation Accident Statistics, tables 5, 8, 9, and 10, available at http://www.ntsb.gov/aviation/, as of July 2005.

2. ______. Aviation Statistical Reports, Annual Review of Aircraft Accident Data (Washington, DC: 2004), also available at http://www.ntsb.gov/, as of March 2005.

3. U.S. Department of Transportation, Federal Aviation Administration, NASDAC Review of NTSB Weather-Related Incidents, available at https://www.nasdac.faa.gov/, as of March 2005.

4. ______. Runway Safety Report (Washington, DC: Annual issues), also available at http://www.faa.gov/, as of March 2005.

¹ At the time this report was prepared, 2000 was the most recent year for which these data were available.

² NTSB specifically studied weather as a factor in general aviation accidents from 1991 to 2001. The board found that 21 percent of these accidents were weather related [3].

Section 4: Variables Influencing Traveling Behavior

Daily and Long-Distance Passenger Travel

According to the 2001 National Household Travel Survey, U.S. residents make, on average, about 4 one-way trips per person per day averaging 10 miles each and 9 roundtrip long-distance trips per person per year averaging about 520 miles each (box 4-A). This translates to annual travel per person of 14,500 miles on daily trips and 4,900 miles on long-distance trips¹ [1].

Shares by mode differ between long-distance and daily travel trips and miles traveled. In miles traveled, 89 percent of miles are made by personal vehicle on daily trips (figure 4-1), but only 56 percent by personal vehicle on long-distance trips (figure 4-2). Air transportation makes up 41 percent of long-distance travel miles. On a trip basis, nearly 90 percent of both daily and long-distance trips are accomplished by personal vehicle.2 Walking makes up most of the rest of daily trips, and air transportation makes up most of the rest of long-distance trips [1].

Source

1. U.S. Department of Transportation (USDOT), Research and Innovative Technology Administration, Bureau of Transportation Statistics and USDOT, Federal Highway Administration, 2001 National Household Travel Survey Data, CD-ROM, February 2004.

¹ These cannot be added together to get a total number because of double counting of daily trips of 50 miles or more from home and differing trip definitions.

² Personal vehicles are cars, vans, sport utility vehicles, pickup trucks, other trucks, recreational vehicles (not including watercraft), and motorcycles.

Vehicle Availability by Household

There were 9.3 million U.S. households without a car, truck, or van in 2003 (9 percent of all households), down from 9.8 million in 1993 (10 percent of households). The 4.6 percent decline in households without vehicles occurred while the number of households increased by 12 percent. The improvement in vehicle availability may be related to a variety of factors, such as better vehicle reliability and longevity, rising incomes, and suburbanization.

Black, Hispanic, poor, and elderly households are more likely to be without a car, van, or truck than the population as a whole (figure 4-3). Poor households are the least likely to have a vehicle. Nevertheless, the percentage of poor households without a vehicle dropped from 33 to 27 percent between 1993 and 2003 [1].

The geographic location of a household also affects vehicle ownership. Central city households are less likely than those in other areas to have at least one car, truck, or van (figure 4-4). This may be due, in part, to higher poverty rates found in central city areas. When data are aggregated on a regional basis, the heavily urban Northeast has the highest share of households without a vehicle (figure 4-5).

Source

1. U.S. Department of Housing and Urban Development and U.S. Department of Commerce, U.S. Census Bureau, American Housing Survey for the United States, H150 (Washington, DC: Biennial issues).

Daily Passenger Travel by Departure Time

On an annual basis, people in the United States make about the same number of trips on weekdays (56.3 billion) as they do on weekend days (62.7 billion)¹ [1]. However, trips made during the week are heavily concentrated in the morning and evening rush hour peaks (figure 4-6). Weekend trips, by contrast, are shifted more toward the middle of the day and peak later in the evening. One of the busiest hours of any day for trip starts is 6 p.m. to 7 p.m. on weekend days. The most common purposes for these trips are people going home from an activity and people going out (say, to eat) or to buy goods and services (e.g., groceries or video rental).

The large number of weekday trips beginning between 7 a.m. and 9 a.m. are predominantly people traveling to work and school (figure 4-7). A large number of trips during the afternoon peak are people returning home from work and school, but this is mixed in with people running errands (e.g., shopping) and making trips for social and recreational purposes. These patterns are linked with the modal pattern that shows that weekday transit trips are more concentrated than trips by personal vehicle, particularly in the morning rush period when work and school trips overlap and travelers are less likely to be making other types of trips [1].

Social and demographic characteristics are another influence on the distribution of trips throughout the day. For instance, weekday time of departure by age reflects the different opportunities and constraints of travelers. Those 20 and under have the most concentrated profile of trip times reflecting the beginning and ending of school and their heavy reliance on others for transportation. Those aged 66 and over are typically less constrained by work hours and thus make a large number of trips between the morning and evening rush periods (figure 4-8).

The concentration of trip-making at certain times of the day can often place a strain on transportation infrastructure. The morning and evening "rush hour" is the most obvious example. But when a trip is made varies with a range of factors including, among others, day of the week (weekend vs. weekday), transportation mode, purpose, and social and demographic ­characteristics.

Source

1. U.S. Department of Transportation (USDOT), Research and Innovative Technology Administration, Bureau of Transportation Statistics and USDOT, Federal Highway Administration, 2001 National Household Travel Survey, CD-ROM, February 2005.

¹ Standard error data are available in tables 4-6 through 4-8 in appendix B.

Commuting to Work

Nearly 9 out of 10 workers in 2003 traveled to work by car, truck, or van; and most of those who drove to work did so alone (figure 4-9). Between 1993 and 2003, the share of workers driving to work alone rose from 77 to 79 percent, while carpooling declined from 11 to 9 percent. Over this same period, transit's share of commuters hovered around 4 to 5 percent, and those working at home remained at about 3 percent. [1]

Poor workers are less likely to drive alone than workers as a whole. Their propensity to drive alone to work was the same in 2003 as it was in 1993, 64 percent (figure 4-10). Black workers, Hispanic workers, and workers over 65 are less likely than the average of all workers to drive alone to work, but the percentages for all three categories rose between 1993 and 2003.

In 2003, the median travel time from home to work was 21 minutes and the median distance was 11 miles. Overall, both median time and median distance are about the same as they were in 1993 [1]. More than a quarter of workers leave for work between 7 a.m. and 8 a.m., with nearly 20 percent leaving between 6 a.m. and 7 a.m., and another 20 percent leaving between 8 a.m. and 9 a.m. (figure 4-11).

Source

1. U.S. Department of Housing and Urban Development and U.S. Department of Commerce, U.S. Census Bureau, American Housing Survey for the United States, H150 (Washington, DC: Biennial issues).

Long-Distance Travel by Young Adults

Overall, the percentage of long-distance trips¹ made by young adults aged 18 to 29 (15.6 percent) was about the same as this age group's share of the U.S. population (16.4 percent). However, when the age group is broken down into two subgroups-ages 18 to 23 and ages 24 to 29-differences appear in travel patterns that may reflect the position of this age group between dependence on one side (going to school and living at home) and independence on the other (with a job and an independent income and place to live) (box 4-B).

For instance, those 18 to 23 years old make a smaller share of all long-distance trips than their share of the population, similar to those 5 to 17 years old² (figure 4-12). But trip-making increases for the 24 to 29 age group such that it begins to resemble the long-distance travel of the older 30 to 44 age group [1]. As young adults move from school to work, the reasons for long-distance travel change. For people aged 18 to 23 years, 11 percent of their long-distance trips are for commuting and 8 percent for business. For people aged 24 to 29 years, commuting and business shares of long-distance trip-making are 16 percent and 21 percent, respectively, about the same as those aged 30 to 44 years (figure 4-13).

The means of transportation for long-distance travel also varies by age, reflecting to some extent the changing reasons for traveling, widening choices (e.g., vehicle availability), and increasing income. All age groups make about 90 percent of their long-distance trips by personal vehicle, with larger variations occurring for air travel and other means (bus, train, and other) (figure 4-14). Those between 18 and 23 years of age make 92 percent of their long-distance trips by vehicle, 5 percent by air, and 3 percent by other means. The older young adults (ages 24 to 29) make 8 percent of their trips by air, reducing their vehicle usage to 89 percent.

Source

1. U.S. Department of Commerce, U.S. Census Bureau, National Estimates by Demographic Characteristics: Single Year of Age, Sex, Race, and Hispanic Origin, available at http://www.census.gov/, as of March 2005.

¹ Long-distance trips are defined as trips, originating from home, of 50 miles or more to the farthest destination and include the return component as well as any overnight stops and stops to change transportation mode.

² The standard errors of the data on this page are in tables 4-12 through 4-14 in appendix B.

Long-Distance Travel by Women

People in the United States took 2.6 billion long-distance trips¹ covering 1.4 trillion miles in 2001. Females made 43 percent of these trips (1.1 billion) while males made 57 percent of them (1.5 billion). Adult females (18 and over) take about two-thirds of the long-distance trips that adult males take (8 trips, on average per year, compared with 13 trips). However, the median distance per trip for women tends to be slightly longer than for men (216 and 201 miles, respectively) [1].

The largest differences in the number of long-distance trips taken by females and males occur in the working age group-typically defined as ages 25 to 64 (figure 4-15). Among those aged 35 to 44, for instance, men take 61 percent of all long-distance trips compared with 39 percent for women. This gap persists until people are 75 years and older; then women and men take approximately the same number of trips.

Trip purpose also varies between females and males (figure 4-16). Both make a similar number of trips for pleasure and personal business, but almost 8 out of 10 long-distance business and more than 8 out of 10 long-distance commuting trips are made by males [1]. While business travel accounts for 16 percent of all long-distance trips, it constitutes 21 percent of males' long-distance trips compared with 9 percent for females. Similarly, commuting accounts for 13 percent of all long-distance trips but 18 percent of males' and only 5 percent of females' long-distance trips.

Modal choice between males and females does not differ much-both use personal vehicles as their primary mode of transport, accounting for 90 percent of all long-distance trips. However, females make a slightly higher proportion of their long-distance trips by bus (2.7 percent) as compared to males (1.7 percent) (figure 4-17).

Source

1. Jonaki Bose, Lee Giesbrecht, Joy Sharp, Jeffery Memmott, Maha Khan, and Elizabeth Roberto, "A Picture of Long-Distance Travel Behavior of Americans Through Analysis of the 2001 National Household Travel Survey," paper presented at the National Household Travel Survey Conference: Understanding Our Nation's Travel, Nov. 1-2, 2004, available at http://www.trb.org/, as of March 2005.

¹ Long-distance trips are defined as trips, originating from home, of 50 miles or more to the farthest destination and include the return component as well as any overnight stops and stops to change transportation mode.

Scheduled Intercity Transportation in Rural America

Nearly 93 percent of the 82 million rural residents¹ in the United States lived within a 25-mile radius of an intercity rail station, an intercity bus or ferry terminal, or a nonhub or small hub² airport or within a 75-mile radius of a large or medium hub airport in April 2005 (figure 4-18). About 29 million rural residents (35 percent) were served by all three modes, while nearly 6 million lived outside this defined coverage area of any scheduled intercity transportation service [1].

These data result from an April 2005 update to a January 2003 geographic information system analysis conducted by the Bureau of Transportation Statistics (BTS) [1]. The results show that most rural residents can access scheduled transportation modes for long-distance intercity trips, based on the distance criteria BTS used. However, the analysis also shows that since the original study two years earlier about 1.1 million rural residents have lost access to intercity transportation. The most noteworthy change in the intercity network has been the elimination by Greyhound of bus service at over 400 locations as part of a system restructuring.³ Amtrak also discontinued part of a long-distance train route, eliminating service in three cities in Ohio and one in Indiana.

At the time of the April 2005 study, intercity buses reached nearly 73 million rural residents (89 percent) compared with nearly 75 million residents 2 years earlier. Scheduled airline service reached 58 million (71 percent), unchanged from 2003. Intercity rail (Amtrak and the Alaska Railroad) reached 35 million (42 percent), down by 300,000 from 2003. For 13 million residents in April 2005, bus was the sole mode providing service within 25 miles, air was the sole mode for 2.6 million rural residents, and rail was the only intercity mode for about 350,000 rural residents. The intercity ferries of the Alaska Marine Highway System, serving coastal Alaska communities as well as Bellingham, Washington, were accessible to 82,000 rural residents and provided the only intercity service to about 2,000 Alaska ­residents.

In April 2005, the United States had nearly 4,400 intercity passenger stations, terminals, and airports. Intercity bus served 72 percent of these facilities. Of the total, 278 of the stations, terminals, and airports were located in Hawaii and Alaska.

Source

1. U.S. Department of Transportation, Research and Innovative Technology Administration, Bureau of Transportation Statistics, Scheduled Intercity Transportation and the U.S. Rural Population, available at http://www.bts.gov/, as of June 2005.

¹ Rural residents are those who live outside of urbanized areas or urban clusters as defined by the U.S. Census Bureau.

² The term hub is used here within the context of individual airports rather than air traffic hubs, which can include more than one airport.

³ Replacement service for some of the locations discontinued by Greyhound was initiated by several regional bus lines.

Section 5: Travel Times

Urban Highway Travel Times

Highway travel times increased between 1993 and 2003 in all but 2 of the 85 urban areas studied by the Texas Transportation Institute. The average Travel Time Index (TTI) for the 85 areas in 2003 was 1.37, an increase from 1.28 in 1993 [2]. This means that in 2003 it took 37 percent longer, on average, to make a peak period trip in these urban areas compared with the time it would take if traffic flowed freely (box 5-A).

Travel times tend to deteriorate as urban area population increases (figure 5-1). For instance, Los Angeles, California, had the highest TTI (1.75) in 2003, while Corpus Christi, Texas and Anchorage, Alaska, had the lowest (1.05). Of the 30 urban areas with the highest index in 2003, only five had a population under 1 million: Austin, Texas (1.33); Tucson, Arizona and Charlotte, North Carolina-South Carolina (1.31 each); Bridgeport-Stamford, Connecticut-New York (1.29); and Salt Lake City, Utah (1.28). At the other end of the spectrum, urban areas of over 1 million people with low indexes include: Cleveland, Ohio (1.09); Buffalo, New York, Pittsburgh, Pennsylvania, and Oklahoma City, Oklahoma (all 1.10); and Kansas City, Missouri-Kansas (1.11).

Between 1993 and 2003, the greatest increases in TTI occurred in very large, large, and medium urban areas, while the increases were more moderate in small urban areas¹ (figure 5-2). Overall, the average index for very large urban areas increased by 10 points (from 1.38 to 1.48), while the index increased by 9 points in large areas (from 1.19 to 1.28) and by 7 points in medium areas (from 1.11 to 1.18). The TTI in small urban areas increased by 4 points (from 1.06 to 1.10).

In urban areas, where highway infrastructure is typically well developed, the principal factor affecting travel times is highway congestion resulting from recurring and nonrecurring events. Recurring delay is largely a phenomenon of the morning and evening commutes, although in some places congestion may occur all day and on weekends. National estimates, based on model simulations, of the effect of nonrecurring events on freeways and principal arterials suggest that about 50 percent are due to crashes, followed by work zones (27 percent), breakdowns (13 percent), and weather (10 percent) [1].

Sources

1. S.M Chin, O. Franzese, D.L. Greene, H.L. Hwang, and R. Gibson, "Temporary Losses of Highway Capacity and Impacts on Performance: Phase 2," Oak Ridge National Laboratory, 2004, table ES-1.

2. Texas A&M University, Texas Transportation Institute, 2005 Urban Mobility Report (College Station, TX: 2005).

¹ Very large urban areas have a population over 3 million; large urban areas, 1 million to 3 million; medium urban areas, 500,000 to 1 million; and small urban areas, less than 500,000.

Surface Border Wait Times

While there are over 75 land ports along the U.S.-Canadian border and over 25 along the U.S.-Mexican border, freight traffic crossings are heavily concentrated at a few major gateways. Commercial trucks crossing into the United States at the busiest gateways-Detroit, Michigan, and Laredo, Texas-generate heavy north-south truck traffic from Detroit through to Memphis, Tennessee, and from Laredo through to San Antonio, Texas. This concentration affects traffic and congestion at the border as well as the growth of major transportation corridors [1].

The average wait time in 2004 for commercial vehicles entering the United States from Canada was 8.5 minutes and 7.3 minutes for those entering from Mexico¹ (figure 5-3 and figure 5-4). There was, however, wide variation in the 2004 wait times for commercial traffic at individual surface gateways. The average wait time at Texas' Laredo World Trade Bridge, a gateway dedicated exclusively to commercial traffic, was the longest (21 minutes) on the Mexican border, while Michigan's Port Huron Bluewater Bridge had the longest average wait time (25 minutes) on the Canadian border.

In contrast to the flow of freight traffic, surface border personal vehicle² wait times are twice as long at U.S.-Mexican borders than at U.S.-Canadian borders. Mexican border crossings averaged about 14.5 minutes of delay in 2003 and 2004, and Canadian border crossings averaged 8 minutes of delay in 2003 and 6 minutes of delay in 2004 (figure 5-5). Passenger mode of choice also differed between those entering from Canada and Mexico . Personal vehicle was the most popular mode in which to cross the U.S. border in 2004 from Canada (64.8 million passengers) and Mexico (190.9 million passengers). However, over 48 million pedestrians entered from Mexico in 2004, making walking the second-most common way to enter the United States through Mexico gateways³ [2].

Sources

1. U.S. Department of Transportation, Research and Innovative Technology Administration, Bureau of Transportation Statistics, America's Freight Gateways, available at http://www.bts.gov/, as of April 2005.

2. U.S. Department of Transportation, Research and Innovative Technology Administration, Bureau of Transportation Statistics, using data from U.S. Department of Homeland Security, U.S. Customs and Border Protection, Data Warehouse CD-ROM, May 2005.

¹ Wait times for commercial vehicles (e.g., tractors pulling containers or beds, panel trucks, and pickup trucks and vans used for hauling commercial cargo) are recorded hourly for 16 surface border ports on the U.S.-Canadian border and for 17 surface border ports on the U.S.-Mexican border.

² Customs and Border Protection uses the term "private vehicles" and defines it as any vehicle of pickup truck size or smaller used for noncommercial purposes. This includes cars, sport utility vehicles, pickup trucks, and vans.

³ See "Passenger Border Crossings" in section 1 of this report.

U.S. Air Carrier On-Time Performance

About 78 percent of domestic air carrier scheduled flights arrived on time in 2004, compared with 79 percent in 1995. The number of on-time flights peaked in 2002 and 2003 (82 percent), after a low of 73 percent in 2000. The number of canceled and diverted flights declined to their lowest level in 2002 (less than 2 percent) (figure 5-6).

The total number of scheduled domestic passenger flights at the nation's airports rose 12 percent between 1995 and 2001 from 5.3 million to 5.9 million flights. After the shutdown of flight operations on September 11, 2001, the number of scheduled flights decreased 12 percent between 2001 and 2002 to 5.3 million flights. They then rose 23 percent to 6.5 million flights in 2003 and 10 percent to 7.1 millions flights the following year.

Air carriers with at least¹ percent of total domestic scheduled service passenger revenues have been required to report on-time performance data since 1987. As of mid-2003, the airlines began reporting data on the cause of delays as well.1 A flight has an on-time departure if the aircraft leaves the airport gate less than 15 minutes after its scheduled departure time, regardless of the time the aircraft actually lifts off from the runway. An arriving flight is counted as on time if it arrives less than 15 minutes after its scheduled gate arrival time.

On average in 2004, National Airspace System delays² had the most impact on airline schedules, accounting for almost 40 percent of all delays (figure 5-7). Another 26 percent of delays occurred, on average, because of circumstances within an airline's control (e.g., maintenance or crew problems), while 30 percent were caused by a previous flight arriving late. At 5.0 percent and 0.3 percent, respectively, extreme weather and airport security caused the fewest delays, on average, in 2004. The number of weather-related delays, however, varied by month and was highest in June 2004 (9,339 delayed flights) and lowest in April 2004 (3,129 delayed flights). By month in 2004, total delays ranged from 15 percent to 36 percent of all scheduled flights.

Source

1. U.S. Department of Transportation, Research and Innovative Technology Administration, Bureau of Transportation Statistics, Airline Service Quality Performance data, March 2005.

¹ See table 5-7 in appendix B for details on reporting carriers and detailed information on cause-of-delay categories.

² The reasons for National Airspace System delays include nonextreme weather conditions, airport operations, heavy traffic volume, and air traffic control.

Air Travel Time Index

Air travel times and the reliability of expected travel times are important determinants of customers' satisfaction, air system operating efficiency, and policymakers' success in meeting performance objectives. A major reason consumers choose to travel by air is that it is often the fastest way to travel long distances.

The Air Travel Time Index (ATTI) rose by 0.5 percent per year between 1990 and 2000 and then fell by 0.7 percent per year between 2000 and 2004 (figure 5-8). The ATTI measures average flight times of nonstop flights using the time elapsed between the scheduled departure and actual arrival, while controlling for different flight characteristics such as distance. In comparison, an index of the average scheduled travel time for nonstop flights in the United States rose by 0.2 percent per year between 1990 and 2000 and remained relatively unchanged between 2000 and 2004. The gap between the two measures widened from 8 minutes in 1990 to a maximum of 11 minutes in 2000 and then narrowed to 7 minutes in 2004.

The Air Travel Time Variability Index (ATTVI) rose by an average of 4 percent per year between 1990 and 2000 and then fell by 3 percent per year between 2000 and 2004 (figure 5-9). The ATTVI measures the variability of flight times of nonstop flights based on differences between travel times on individual flights and the average travel times for the same flight. Thus, not only did the travel time for a typical flight take longer between 1990 and 2000, but it also became more uncertain. However, between 2000 and 2004, both flight travel times and their variability improved despite an increase in the number of flight operations.¹

The Bureau of Transportation Statistics (BTS) research developing the ATTI and ATTVI is intended to improve the measurement of air travel time and reliability. Using data BTS collects from airlines (box 5-B), the ATTI enables analysis of changes in air travel time nationally, as well as by airport, carrier, time of day, and flight distance. For instance, from 1990 to 2004, most improvements occurred in flights departing in the evening offpeak (after 9:00 p.m.). The least improved were flights departing in the evening peak (between 3:00 p.m. and 9:00 p.m.). Grouped by distance, flights of more than 1,000 miles were approximately unchanged, while travel times of flights of 500 miles or less increased.

¹ Improvement occurs when the ATTI and ATTVI decrease.

Amtrak On-Time Performance

Seventy-one percent of Amtrak trains arrived at their final destination on time in 2004 [2]. This was below the system's performance peak of 76 percent in 2002 (figure 5-10). Amtrak counts a train as delayed if it arrives at least 10 to 30 minutes beyond the scheduled arrival time, depending on the distance the train has traveled.¹ In addition, Amtrak on-time data are based on a train's arrival at its final destination and do not include delay statistics for intermediate points.²

Over the years, short-distance Amtrak trains-those with runs of less than 400 miles (including all Northeast Corridor and Empire Service trains)-have consistently registered better on-time performance than long-distance trains-those with runs of 400 miles or more. Annual on-time performance for short-distance trains reached a high of 87 percent in 2002 but fell to 76 percent in 2004. Sixty-eight percent of long-distance trains arrived on time in 2004, up from 49 percent in 1994 but short of their high of 70 percent in 2001 and 2002.³

Amtrak also collects data on the cause and cumulative hours of delay for its trains, including delays at intermediate points, and attributes the cause of each delay to Amtrak, the host railroad, or "other" (figure 5-11). Delays assigned to Amtrak represented 30 percent of all delay hours in 2004. Delays ascribed to host railroads represented 64 percent, and other delays accounted for the remaining 6 percent.⁴ (Amtrak trains operate over tracks owned primarily by private freight railroads except in most of the Northeast Corridor, along a portion of the Detroit-Chicago route, and in a few other short stretches across the country [1].) Throughout the years, host railroad delays have consistently represented the largest share of delay hours. Between 2000 and 2004, host railroad and other delays increased each year. Amtrak-caused delay hours declined in both 2002 and 2003. However, delay hours in 2004 increased-accounting for the longest delay hours in four years.

Sources

1. National Passenger Railroad Corp., "Amtrak Facts," available at http://www.amtrak.com/, as of November 2003.

2. ______. personal communication, February 2005.

¹ Amtrak trips of up to 250 miles are considered on time if they arrive less than 10 minutes beyond the scheduled arrival time; 251-350 miles, 15 minutes; 351-450 miles, 20 minutes; 451-550 miles, 25 minutes; and greater than 550 miles, 30 minutes.

² Accordingly, a train traveling between Chicago and St. Louis (282 miles), for example, could arrive 15 minutes late at all intermediate points, yet arrive 12 minutes late at St. Louis and be reported as on time.

³ Amtrak revised its methodology for collecting and calculating on-time performance data in 2001.

⁴ In 2000, Amtrak revised the methodology for reporting delays by cause, which makes data beginning in 2000 not comparable to previous years.  The Bureau of Transportation Statistics presented Amtrak cause-of-delay data for 1990 through 1999 in its 2003 Transportation Statistics Annual Report.

Rail Freight Times

Class I rail freight line-haul speeds averaged 21.8 miles-per-hour in the first-quarter of 2005, a decrease of 1.5 percent from the previous quarter¹ (figure 5-12). Between the first quarter of 2002 and the first quarter of 2005, average line-haul speeds decreased 15 percent. This decrease followed a general upward trend in line-haul speeds since late 1999.

Line-haul speed is a shipper-related indicator of the performance of the railroad industry. To put the average speeds in perspective, revenue ton-miles totaled 416.7 billion in the first quarter of 2005 (figure 5-13). This represented an increase in revenue ton-miles of 18 percent from the first quarter of 2002 to first quarter of 2005, the same time period in which average line-haul speeds were declining.

Terminal dwell time, the time a train spends in terminals, is not included in line-haul speed data (box 5-C). It is, thus, a rail freight time indicator that supplements line-haul speeds. Terminal dwell time of Class I railroads averaged 24.2 hours in the first quarter of 2005, an increase of 0.7 percent compared with the previous quarter [1].

Source

1. U.S. Department of Transportation, Research and Innovative Technology Administration, Bureau of Transportation Statistics, calculations using Class I railroad data reported to the Association of American Railroads, available at http://www.railroadpm.org/.

¹ For the definition of Class I railroads, see the Glossary.

Section 6: Availability of Mass Transit and Number of Passengers Served

Transit Passenger-Miles of Travel

Transit passenger-miles of travel (pmt) grew 26 percent between 1993 and 2003, from 36.2 billion pmt to 45.6 billion pmt [2] (box 6-A). However, transit pmt declined 1.2 percent between 2001 and 2002, and it declined another 0.6 percent between 2002 and 2003. As they have historically, buses maintained the largest pmt share in 2003 (42 percent) while generating 19.1 billion pmt (figure 6-1). Also in 2003, heavy-rail pmt totaled 13.6 billion or 30 percent and commuter rail leveled off at 9.5 billion pmt, for a 21 percent share.

Light rail and demand-response¹ services had only 3.2 percent and 1.5 percent, respectively, of transit pmt shares in 2003. However, pmt for light rail more than doubled between 1993 and 2003 and nearly doubled for demand-response services (figure 6-2). In comparison, bus pmt grew 10 percent between 1993 and 2003.

The top 20 transit authorities (ranked by pmt) logged 32 billion passenger-miles in 2003 or 70 percent of all transit pmt that year. In 2003, people riding New York City Transit traveled 9.5 billion passenger-miles (or 21 percent of all transit pmt) and the Chicago Transit Authority generated 1.8 billion passenger-miles or 4 percent [1].

Sources

1. U.S. Department of Transportation, Federal Transit Administration, National Transit Database, 2003 Transit Profiles, available at http://www.ntdprogram.com/, as of April 2005.

2. ______. National Transit Summaries and Trends, available at http://www.ntdprogram.com/, as of April 2005.

¹ Demand-response transit operates on a nonfixed route and a nonfixed schedule in response to calls from passengers or their agents to the transit operator or dispatcher.

Transit Ridership by Trips

Transit ridership grew steadily from 1995 to 2002, reaching 9,017 million unlinked trips (box 6-B) in 2002, an increase of 20 percent. However, between 2002 and 2003, total transit ridership declined 1.6 percent as ridership in 2003 posted 8,876 million unlinked trips. This decline follows a slowing of growth in transit ridership between 2001 and 2002 (less than 1 percent) compared with ridership growth between 2000 and 2001 (3.3 percent) [1].

Bus ridership comprised the majority of unlinked trips in 2003 (5,147 million). After having grown 15 percent between 1995 and 2002, bus ridership declined 2.3 percent between 2002 and 2003 (figure 6-3). Rail transit ridership, with 3,414 million trips in 2003, posted strong growth from 1993 to 2003 (34 percent). Heavy rail grew 30 percent; light rail, 80 percent; and commuter rail, 28 percent (figure 6-4). However, among rail services only light-rail ridership grew between 2002 and 2003 (0.4 percent), while heavy-rail and commuter-rail ridership each declined 1 percent.

Heavy-rail ridership posted 2,667 million trips; commuter-rail, 410 million trips; and light-rail, 338 million trips in 2003. Other transit services, such as ferryboats and demand response, posted a combined 315 million trips.

Source

1. U.S. Department of Transportation, Federal Transit Administration, National Transit Summaries and Trends, annual reports, available at http://www.ntdprogram.com/, as of May 2005.

Transit Ridership by Transit Authority

Approximately 78 percent of all unlinked transit passenger trips in 2003 were made within the service area of just 30 transit authorities [1]. These 30 top authorities logged 6.9 billion unlinked trips in 2003¹ (figure 6-5). New York City Transit alone reported 2.6 billion or 38 percent of unlinked passenger trips for the top 30 authorities. The Chicago Transit Authority followed with 475 million or 7 percent of trips for the top 30 authorities.

The top 30 transit authorities served a population of about 125 million in 2003. All transit authorities reporting to the National Transit Database estimate the population they serve using definitions of bus and rail service in the Americans with Disabilities Act of 1990 and their own local criteria for other service, such as ferryboat and vanpool. Some double-counting of populations served occurs, especially among authorities operating in the largest metropolitan areas such as New York City, Los Angeles, Chicago, and San Francisco.

Source

1. U.S. Department of Transportation (USDOT), Research and Innovative Technology Administration, Bureau of Transportation Statistics, calculations using data from USDOT, Federal Transit Administration, National Transit Database, available at http://www.ntdprogram.com/, as of April 2005.

¹ In 2003, 622 transit agencies submitted reports to the Federal Transit Administration. Of these, 74 reporting agencies operated nine or fewer vehicles across all modes and types of service and received waivers from detailed reporting. Thus, 548 transit agencies are included in the 2003 database.

Accessible Rail Stations and Buses

Transit rail stations that are compliant with requirements under the Americans with Disability Act (ADA) (box 6-C) increased 178 percent from just 553 stations (out of 2,452) in 1993 to 1,537 stations (out of 2,799) in 2003 (figure 6-6). Yet, the rate at which compliance increased at commuter-rail, light-rail, and heavy-rail stations differed (figure 6-7).

The percentage of light-rail stations that are ADA accessible rose the fastest among the transit rail modes, from 24 percent compliant (92 stations) in 1993 to 76 percent (466 stations) in 2003 (figure 6-7). During the same time period, commuter-rail station accessibility grew from 23 percent (242 stations) to 56 percent (643 stations). Heavy-rail riders also experienced an increase in ADA-compliant stations, from 22 percent (217 stations) in 1993 to 41 percent (416 stations) in 2003.

Transit buses are also subject to ADA requirements. As of 2003, 95 percent of all transit buses were equipped with lifts or ramps to make them accessible to disabled riders.¹

Source

1. U.S. Department of Transportation, Federal Transit Administration, National Transit Database 2003, available at http://www.ntdprogram.com/, as of April 2005.

¹ For more information on accessible buses, see Transportation Statistics Annual Report, September 2004.

Section 7: Travel Costs of Intracity Commuting and Intercity Trips

Household Spending on Transportation

On average, households spent $7,681 (in chained 2000 dollars¹) on transportation in 2003. This represented 20 percent of all household expenditures that year. Only housing cost households more (31 percent)² [1].

Between 1993 and 2003, consumer spending on private transportation (mainly motor vehicles and related expenses) increased by 27 percent. On average, households spent $3,834 purchasing new and used motor vehicles in 2003, up 49 percent from $2,569 in 1993 (figure 7-1). Spending on other vehicle expenses (e.g., insurance, financing charges, maintenance, and repairs) also increased, from $1,806 to $2,216 (23 percent).

Meanwhile, gasoline and oil expenditures declined 1 percent, to $1,268 in 2003. This decline was largely because of a 7 percent drop in these expenditures between 2002 and 2003. On an annual basis, gasoline and oil expenditures declined 0.1 percent between 1993 and 2003. Other transportation, such as local transit and airplane and train trips, is the smallest category of household spending on transportation (4.7 percent of the total in 2003). On average, households spent $364 to pay for other transportation in 2003, a decrease of 1 percent between 1993 and 2003.

Source

1. U.S. Department of Labor, Bureau of Labor Statistics, Consumer Expenditure Survey, data query, available from http://www.bls.gov/, as of March 2005.

¹ All dollar amounts are expressed in chained 2000 dollars, unless otherwise specified. Current dollar amounts (which are available in appendix B of this report) were adjusted to eliminate the effects of inflation over time.

² The Bureau of Labor Statistics (BLS) collects these data. In its survey, BLS uses the term consumer units instead of households and public transportation rather than other transportation. There are an average of 2.5 persons in each consumer unit, according to BLS. Public transportation, according to BLS, includes both local transit, such as bus travel, and long-distance travel, such as airplane trips. (See complete definitions of these categories on figure 7-1 and table 7-1 in appendix B.)

Cost of Owning and Operating an Automobile

Driving an automobile 15,000 miles per year cost 53¢ per mile in 2003, or 20 percent more than it did in 1993 when total costs were 44¢ per mile (figure 7-2). These data, which are expressed in 2000 chained dollars,¹ include fixed costs (e.g., depreciation, insurance, finance charges, and license fees) and variable costs (e.g., gasoline and oil, maintenance, and tires). Between 1993 and 2003, fixed costs represented an average of 75 percent of total per-mile costs. Gasoline and oil, a component of variable costs, represented 13 percent of driving costs per mile in 2003, down from 18 percent in 1993 [1].

Annually, each person in the United States travels an average of 14,500 miles on daily trips [2]. About 89 percent of these trip-miles are by personal vehicle (e.g., cars, vans, sport utility vehicles, and light trucks). For the balance, people travel via public transportation or air, ride bicycles, walk, or travel by other means.

Sources

1. U.S. Department of Transportation, Research and Innovative Technology Administration, Bureau of Transportation Statistics, National Transportation Statistics 2004 (Washington, DC: 2005), table 3-14.

2. U.S. Department of Transportation, Research and Innovative Technology Administration, Bureau of Transportation Statistics and Federal Highway Administration, Highlights of the 2001 National Household Travel Survey, available at http://www.bts.gov/, as of August 2005.

¹ All dollar amounts are expressed in chained 2000 dollars, unless otherwise specified. To eliminate the effects of inflation over time, the Bureau of Transportation Statistics converted current dollars (which are available in appendix B of this report) to chained 2000 dollars.

Cost of Intercity Trips by Train and Bus

Amtrak collected an average of 23¢ per ­revenue passenger-mile in 2003 (in chained 2000 dollars¹), up 46 percent from 16¢ per revenue passenger-mile in 1993 (figure 7-3). During the 1990s, Amtrak shifted its focus to urban routes in the Northeast and West. When Amtrak reduced its number of route-miles by 3 percent in 1995, revenue per passenger-mile increased by 3 percent the following year. When track operational length was further reduced by 7 percent in 1999, revenue per passenger-mile increased 4 percent the following year [1]. Today, Northeast Corridor trains serve 13 million riders annually, representing about 60 percent of Amtrak's ticket revenues [2].

Average intercity Class I bus fares rose 23 percent, from $23 to $28 (in chained 2000 dollars), between 1992 and 2002² (figure 7-4). The average bus fare is based on total intercity passenger revenues and the number of intercity bus passenger trips. Because passenger-mile data are not reported, average bus fare per passenger-mile cannot be calculated and compared with similar Amtrak fare data.

Sources

1. Association of American Railroads, Railroad Facts (Washington, DC: 1994-2004 issues).

2. National Railroad Passenger Corp. (Amtrak), Amtrak Strategic Reform Initiatives and FY 06 Grant Request (Washington, DC: 2005).

¹ All dollar amounts are expressed in chained 2000 dollars, unless otherwise specified. To eliminate the effects of inflation over time, the Bureau of Transportation Statistics converted current dollars (which are available in appendix B of this report) to chained 2000 dollars.

² Intercity bus data through 2002 were reported by carriers to the Bureau of Transportation Statistics. These data are now reported to the U.S. Department of Transportation, Federal Motor Carrier Safety Administration, and data beyond 2002 were not available at the time this report was prepared.

Average Transit Fares

Transit fares remained relatively stable between 1993 and 2003 (figure 7-5). Increases in fares per passenger-mile for some types of transit service were offset by lower fares per passenger-mile for other types.

Local transit bus service, which accounted for 58 percent of public transportation ridership (by number of unlinked passenger trips¹) in 2003, cost the same (18¢ per passenger-mile) in 2003 as it did in 1993 (in chained 2000 dollars),² although it rose to 21¢ in 2000 (figure 7-6).

Demand-response transit³ fares rose the most between 1993 and 2003: from 19¢ to 23¢ per passenger-mile or 22 percent. These fares were at their highest point (33¢) in 1994. All rail transit fares declined during the 10-year period: commuter rail, -12 percent; heavy rail, -19 percent; and light rail, -17 percent. Rail transit, the second-most heavily used component of transit, accounted for 30 percent of unlinked passenger trips in 2003, while demand-response had less than 1 percent of the trips [1].

Source

1. U.S. Department of Transportation, Federal Transit Administration, National Transit Summaries and Trends, 2003 National Transit Profile, available at http://www.ntdprogram.com/, as of April 2005.

¹ See Transit Ridership in section 6, "Availability of Mass Transit" for a discussion of unlinked trips.

² All dollar amounts are expressed in chained 2000 dollars, unless otherwise specified. To eliminate the effects of inflation over time, the Bureau of Transportation Statistics converted current dollars (which are available in appendix B of this report) to chained 2000 dollars.

³ Demand-response transit operates on a nonfixed route and nonfixed schedule in response to calls from passengers or their agents to the transit operator or dispatcher.

Air Travel Price Index

Commercial airlines offer a variety of discount fares to fill their flights, but these special airfares, facilitated by Internet commerce and "frequent flyer" programs, complicate efforts to measure changes in the prices people pay for commercial air travel. To improve these measurements, the Bureau of Transportation Statistics (BTS), in consultation with the Bureau of Labor Statistics (BLS), developed an Air Travel Price Index (ATPI) (box 7-A).

ATPI data can be used to compare changes in prices among many cities. In a comparison of three medium-sized cities, for instance, a dip appears between 1995 and 1998 for flights originating in Colorado Springs, Colorado (figure 7-7). During this time, the discount carrier Western Pacific operated flights from Colorado Springs, bringing airfares down before it withdrew from the market. Fluctuations in the ATPI of the major U.S. cities of New York, Los Angeles, and Chicago varied less than in Colorado Springs and the other selected medium-sized cities. The ATPI of the three selected U.S. cities collectively peaked in the first quarter of 2001 and have since declined (figure 7-8). Between the first quarter of 2001 and the fourth quarter of 2004, Chicago's ATPI declined by 21 percent, New York's by 15 percent, and Los Angeles' by 13 percent.

A comparison of the U.S. Origin and the Foreign Origin national-level ATPI reveals a diverging trend.¹ While the "U.S. Origin Only" ATPI increased 2.2 percent from 1995 to 2004, the "Foreign Origin Only" ATPI decreased 9.8 percent over this same period (figure 7-9). Unlike the "U.S. Origin Only" ATPI, which peaked in the first quarter of 2001, the "Foreign Origin Only" ATPI has been trending downward since the third quarter of 1997, while maintaining its overall pattern of peaks in the third or fourth quarters followed by declines in other quarters.

¹ The U.S. Origin ATPI only includes itineraries originating in the United States whether the destinations are domestic or international. The Foreign Origin ATPI includes itineraries with a foreign origin and a U.S. destination.

Section 8: Productivity in the Transportation Sector

Labor Productivity in Transportation

Labor productivity (output per hour) in the for-hire transportation services industries increased by 18 percent from 1992 to 2002. This compares with an increase of 47 percent for all manufacturing and 24 percent for the overall business sector (figure 8-1). Labor productivity, a common and basic productivity measure, is calculated as the ratio of output to hours worked or to the number of full-time equivalent employees.

The growth of individual transportation subsector labor productivity between 1992 and 2002 varied¹ (figure 8-2). Compared with the overall business sector, rail labor productivity increased at a considerably higher rate (60 percent). Meanwhile, labor productivity in air transportation increased 27 percent, and long-distance trucking productivity grew 12 percent.

Comparing annual growth rates is another way to interpret changes in labor productivity over time. For overall business, labor productivity grew at an annual rate of 2.1 percent between 1992 and 2002. Labor productivity in rail transportation-where productivity has been affected by consolidation of companies, more efficient use of equipment and lines, increased ton-miles (output), and labor force reductions-increased by 4.6 percent annually. For long-distance trucking and air transportation, annual rates of growth were 1.1 percent and 2.2 percent, respectively.

¹ At the time this report was prepared, data were only available through 2000 for local trucking, petroleum pipeline, and bus carriers. See detailed notes on table 8-1 and table 8-2 for further information.

Multifactor Productivity

Multifactor productivity (MFP) in air transportation increased by 16 percent between 1991 and 2001 (an annual rate of 1.5 percent), while in the overall private business sector, MFP increased by 10 percent (just under 1 percent annually) (figure 8-3). Thus, the air transportation industry has contributed positively to increases in MFP in the business sector and to the U.S. economy over this period. Data are not available for the same period for rail transportation, but between 1991 and 1999, MFP in this industry increased by 26 percent (an annual rate of 3 percent).

While MFP measures are difficult to construct, they provide a much more comprehensive view of productivity than labor productivity measures. The conventional methodology for calculating multifactor productivity, which is used here, employs growth rates of inputs weighted by their share in total costs. This methodology has been developed and used by various academic researchers and government agencies, such as the Bureau of Labor Statistics.¹

Transportation MFP data are currently available from the Bureau of Labor Statistics for the rail and air transportation sectors only. The Bureau of Transportation Statistics is developing MFP measures for other transportation industries, such as trucking and pipelines. These data will provide more complete information on the relative importance of transportation in increasing the productivity of the U.S. economy and, hence, transportation's contribution to the economic growth of the country.

¹ See, for instance, the discussion on MFP by the Bureau of Labor Statistics in their Handbook of Methods, available at http://www.bls.gov/, as of August 2005.

Section 9: Transportation and Economic Growth

Transportation Services Index

The Transportation Services Index (TSI) rose to 112.6 in May 2005,¹ the highest level attained in the 15-year period beginning January 1990, and a 4.0 percent increase from its May 2004 level of 108.3 (figure 9-1). The TSI is an experimental, seasonally adjusted index of monthly changes in the output of services of the for-hire transportation industries, including railroad, air, truck, inland waterways, pipeline, and local transit [1].

The Bureau of Transportation Statistics (BTS), which produces the measure, calculates the TSI as a single transportation index and as separate indexes for its two components-freight and passenger transportation. The freight TSI rose to 113.1 in May 2005, 2.4 percent higher than May 2004 (110.5), and reached a record high for the 15-year period covered by the index. In May 2005, the passenger TSI was 111.2, an increase of 8.2 percent from 102.8 in May 2004.

BTS released the first TSI data (covering January 1990 through December 2003) in March 2004. The index is still under development as BTS works to refine the index data sources, methodologies, and interpretations. A prototype version of the TSI suggested a significant relationship with the economy, in particular, with cyclical downturns. To verify these linkages, however, more research is needed.

Source

1. U.S. Department of Transportation, Research and Innovative Technology Administration, Bureau of Transportation Statistics, Transportation Services Index, available at http://www.bts.gov/, as of August 2005.

¹ The TSI is a chained-type index where 2000 = 100.

Transportation-Related Final Demand

Total transportation-related final demand rose by 33 percent between 1993 and 2003 (in 2000 chained dollars¹) from $833.8 billion to $1,112.8 billion (figure 9-2). However, transportation-related final demand as a share of Gross Domestic Product (GDP) showed little change throughout the period. This implies that transportation-related final demand grew at about the same rate as GDP. In 2003, the share of transportation-related final demand in GDP was 10.7 percent, compared with 11.1 percent in 1993 [1].

Personal consumption of transportation-which includes household purchases of motor vehicles and parts, gasoline and oil, and transportation services-is the largest component of transportation-related final demand. It amounted to $911.8 billion in 2003 and accounted for 82 percent of the total transportation-related final demand (figure 9-3). Government purchases and private domestic investment commanded equal shares of transportation-related final demand in 1999. However, during the rest of the 1993 to 2003 period, government purchases held a greater share. Government purchases reached $199.8 billion in 2003 (an 18 percent share), while private investment totaled $127.3 billion (an 11 percent share).

The United States imported more transportation-related goods and services than it exported between 1993 and 2003. This gap has widened in recent years. In 1993, net exports were 3.9 percent of total transportation-related final demand. By 2003, net exports rose to 11 percent. Deficits in the trade of automobiles and other vehicles and parts have been the primary component of the deficit in transportation-related goods and services.

Transportation-related final demand is the total value of transportation-related goods and services purchased by consumers and government and by business as part of their investments.² Transportation-related final demand is part of GDP, and its share in GDP provides a direct measure of the importance of transportation in the economy from the demand side. The goods and services included in transportation-related final demand are diverse and extensive, ranging from automobiles and parts, fuel, maintenance, auto insurance, and so on, for user-operated transportation to various transportation services provided by for-hire transportation establishments.

Source

1. U.S. Department of Transportation, Research and Innovative Technology Administration, Bureau of Transportation Statistics, calculations based on data from U.S. Department of Commerce, Bureau of Economic Analysis, National Income and Product Account tables, available at http://www.bea.gov/, as of January 2005.

¹ To eliminate the effects of inflation over time, the Bureau of Transportation Statistics converted current dollars (which are available in appendix B of this report) to chained 2000 dollars.

² Also included are the net exports of these goods and services, because they represent spending by foreigners on transportation goods and services produced in the United States . Imports, however, are deducted because consumer, business, and government purchases include imported goods and services. Therefore, deducting imports ensures that total transportation-related spending reflects spending on domestic transportation goods and services.

For-Hire Transportation

For-hire transportation industries contributed $314.3 billion to the U.S. economy¹ in 2003, a less than 1 percent increase from $217.2 billion in 1993 (in 2000 chained dollars²) (figure 9-4). Over the same period, this segment's share in Gross Domestic Product (GDP) hovered around 3 percent. This suggests that the for-hire transportation segment of the economy has been growing at about the same rate as has GDP.

Among for-hire transportation industries, trucking, air, and the combined category of other transportation and support activities³