Executive Summary
In 2003 Congress directed the Bureau of Transportation Statistics (BTS) to carry out a study: “(1) to measure the ton-miles and value-miles of international trade traffic carried by highway for each state; (2) to evaluate the accuracy and reliability of such measures for use in the formula for highway apportionments…; (3) to evaluate the accuracy and reliability of the use of diesel fuel data as a measure of international trade traffic by state; and (4) to identify needed improvements in long-term data collection programs….” The legislation also directed BTS to request that the Committee on National Statistics convene a panel study to review the findings and recommendations of the BTS study and to explore additional data sources and methods for providing improved estimates for use in apportionment formulas. A key purpose of this work is to find appropriate formulas for allocating highway funds to states on the basis of international trade traffic.
There are no direct measures of ton-miles and value-miles of international trade traffic carried by highway in each state. Therefore, BTS had to use models to produce estimates, using as input data from a multitude of public and private sources, none of which were specifically designed for the task. In addition, as directed by the legislation, BTS examined the possibility of using diesel fuel data to directly predict international trade traffic by highway.
The BTS report concluded that its estimates should not be used as input into fund allocation formula programs because of the poor quality of
the estimates. To support its conclusion, BTS identified a number of data deficiencies and the need to rely on strong assumptions in the models. BTS also concluded that the relationship between diesel fuel data and international trade traffic carried by highway was not sufficiently strong to adequately differentiate between states in modeling ton-miles or value-miles of international trade traffic. The BTS report included some suggestions on how to address the various data deficiencies, including the introduction of new surveys and the possibility of increased access to nonpublic or currently protected public data sources.
The panel agrees in general with all of the major findings and conclusions in the BTS study:
-
The panel concurs with the central conclusion that the present state of available freight data does not permit determination of international trade traffic by state.
-
The panel concurs that the untested assumptions relied on by the various models used to calculate the estimates are unlikely to be valid and therefore the resulting estimates are of unknown accuracy.
-
The panel concurs that using diesel fuel data to model ton-miles and value-miles of international trade traffic by highway at the state level is unlikely to produce sufficiently reliable estimates for forecasting highway use.
In terms of the underlying purpose of the BTS study, the panel questions the use of value-miles as a useful measure in addition to ton-miles. The panel also questions the separation of estimates of domestic and international trade traffic as being inconsistent with the objective of allocating funds to address needed maintenance for and new construction of the nation’s highways. The panel was limited in its evaluation of the utility of the estimates of ton-miles and value-miles for purposes of fund allocation given that such an evaluation necessarily depends on the application with respect to a specific formula, and to date, no formula has been proposed for this purpose.
The panel examined the advantages and disadvantages of various possibilities for additional data collection, including new surveys, additional data collected from administrative records, greater access to existing records, and passive means for data collection using new technologies. Several of these hold promise for both near-term and long-term improvement of the measurement of ton-miles of international trade traffic, but none can now provide reliable estimates of such traffic by state.