This section examines the current distribution of freight traffic by mode, examining the roles that each mode plays in the freight system, and how those roles have evolved in recent years. The first subsection discusses the data sources the analysis, uses, and their limitations. The following subsections analyze mode shares over the last 20 years, and projected in terms of tons, ton-miles, value, distance traveled and commodity carried.
The Freight Analysis Framework (Federal Highway Administration 2014) version three (FAF3) is a freight-transportation data program designed to provide a comprehensive picture of freight movement in the United States. Analysts often think of it as a model because of the traffic assignment procedure that FAF uses. The database that makes such a traffic assignment possible is the result of an enormous data compilation that represents the only publicly available comprehensive data for the entire US freight transportation system. FAF (Federal Highway Administration 2014) is further described, below.
The Commodity Flow Survey (CFS) that the Census Bureau conducts as part of the economic census is the largest source of input data, and provides the majority of domestic freight movement data. However, the Census conducts this survey only every five years, so FAF also develops estimates for the most recent years, as well as forecasts, based on trends from specific sources, international trade, and economic data. FAF also uses the origin destination matrix to conduct network assignment and regional flows, which allows the model to display trucks trips, tonnage, and congestion on the National Highway network.
FAF3 provides estimates for tonnage, value, and domestic ton-miles by region of origin and destination, commodity type, and mode for 2007, the most recent CFS year, as well as forecasts through 2040. Also included are state-to-state flows for these years, plus 1997 and 2002, summary statistics, and flows by truck assigned to the highway network for 2007 and 2040. FAF has eight domestic modes, including truck, rail, water, air, multiple modes and mail, pipeline, and other/unknown (for which the mode cannot be determined). The eighth mode is a no-domestic mode category, i.e. there is no movement inside the US. For example, if there is a shipment of crude petroleum that goes directly to the refinery from the port of entry, without any domestic transportation, it would be included in the no-domestic mode category. FAF also has seven foreign modes that match the domestic modes minus the no-domestic mode category. FAF has 123 domestic regions taken from the regions used in the CFS, and eight international regions, which are Canada, Mexico individually, and six groupings based on United Nations definitions. The commodity classifications are from the Commodity Flow Survey, that include 43, 2-digit Standard Classification of Transported Goods (SCTG) codes.
The CFS collects data on all shipments from the surveyed establishment for an entire week in each of the four quarters of the census year. It reaches a subset of all American establishments, and participation is mandatory. In 2007, the CFS collected data from almost twice as many establishments as in 2002, resulting in about 100,000 samples. Even with this larger sample, the CFS is missing about 30 percent of the total FAF in volume, and 32 percent in tons.
However, at this point the data do not include flows that are Out-of-Scope (OOS), or undercounted in the CFS. The OOS commodities include Farm-based agricultural shipments, Fishery, Crude Petroleum, Natural Gas, Municipal Solid Waste, Forestry, Construction, Retail, Services, Household and Businesses Moves, and Imports. FAF assigns commodity and mode-specific OD flows to these commodities using other methods and many different data sources. In addition, the CFS only captures outbound shipments from domestic firms. Therefore, as it does not include movements of foreign imports, FAF adds foreign trade data, including land crossings between North American countries.
Mode shares in tons
As described in the introduction, modal shares vary considerably depending on the metric (tons, ton-miles, value) that is the focus of the analysis. Figure 2 provides estimates of the number of tons moved by each freight mode for the four historical CFS years (1997, 2002, 2007 and 2012) and four forecast years (2015, 2020, 2025, 2030). The figure clearly shows the domination of trucks in terms of total tonnage moved. The high level of growth in truck tonnage between 2002 and 2007 slowed considerably from 2007 to 2012, and FAF forecasts continued sluggish growth through 2015 followed by robust growth in future years. Rail is the next largest mode in terms of tonnage, followed by pipelines. Rail exhibited tonnage growth between 1997 and 2002, but has stagnated since, while pipeline has moved a stable amount of tonnage during that period.
Figure 3 provides the same set of tonnage data, but in terms of percentages. Historically, truck tonnage has hovered around 70%, and FAF projects the truck share to remain in that range. Meanwhile, rail has a share just over 10%, while pipelines are just below 10%. From 1997, rail has slightly increased its share, while the pipeline share has dipped slightly.
Mode shares in ton-miles
The distribution by mode is significantly different when analyzed by ton-miles, as average rail and pipeline movements travel longer distances. Figure 4 provides estimates of the number of ton-miles moved by each freight mode for the four historical CFS years (1997, 2002, 2007 and 2012), and four forecast years (2015, 2020, 2025, 2030). While truck is still the dominant mode, rail and pipeline capture much larger shares. While truck has steadily increased ton-miles from 1997 to 2012, rail has had a larger increase in ton-miles. Pipelines lost ton-miles between 1997 and 2002, but have stabilized its share since.
Figure 5 provides the same set of ton-mile data, but in terms of percentages. Historically, truck tonnage has hovered around 40%, but has lost share slightly over the last 10 years. FAF projects the truck share to grow to 45% over the next two decades. Historically, rail has increased its share of ton-miles by over five percent, but FAF forecasts that share to slip in the future. Pipelines have been losing share, and FAF projects that trend to continue.
Mode shares in values
The distribution by mode is again significantly different when analyzed by the value of the goods carried. Figure 6 provides estimates of the value of goods moved by each freight mode for the four historical CFS years (1997, 2002, 2007 and 2012), and four forecast years (2015, 2020, 2025, 2030). Truck still is by far the dominant mode, when analyzed by value, but rail and pipeline are replaced as the next tier of modes by air and multiple mode (intermodal) shipments.
Figure 7 provides the same set of value of goods data, but in terms of percentages. FAF projects the truck share to slip slightly over the next two decades, with an increase in the share of multiple mode shipments.
Mode share by distance
Mode shares vary considerably by the distance that the freight moves. Figure 8 depicts freight mode shares for ton-miles by distance for 2007. Trucks carry over 80 percent of the ton-miles for shipments of less than 100 miles, and for shipments from 100 to 249 miles. For shipments under 750 miles, trucks remain the leader in modal share. Rail is the dominant mode, followed by pipelines for goods moved distances greater than 750 miles, and fewer than 2,000 miles. For shipments over 2000 miles, there is a fairly even split between the five major freight modes, including truck, rail, water, multiple modes and pipelines, with rail leading the way.