As Burbank (2009) points out, transportation data needed to support policies or programs that would be designed to affect travel behavior have major deficiencies. The Center for Clean Air Policy (2009) identified the difficulties that the poor data at the state and local levels would cause in measuring the effects of strategies to reduce travel. The group calls for substantial improvements in data collection and in the models that use these data to help make progress in reducing transportation’s greenhouse gas (GHG) emissions.
Current information about passenger travel depends to a great extent on the episodic conduct of the National Household Travel Survey (NHTS), which is intended to be a nationally representative, cross-sectional survey of household vehicle ownership, trip-making behavior, and travel distances with considerable demographic and household information. Many compromises have been made in this survey over the years to deal with uncertain administrative support and inadequate resources. The NHTS is itself a combination of the former Nationwide Personal Transportation Survey and the American Travel Survey; the latter was the only survey of long-distance (intercity) trip making. The NHTS has had a checkered history in terms of funding. Conduct of the survey has been postponed in previous cycles because of lack of funding. In 2008, the Bureau of Transportation Statistics (BTS) concluded that it would be unable to support the 2009–2010 NHTS; the Federal Highway Administration assumed responsibility for conducting the survey despite lack of
authorized funds to pay for it (other U.S. Department of Transportation administrations, including BTS, contribute to the survey cost).
Many metropolitan planning organizations (MPOs) and states supplement the survey by paying for expanded sampling in their regions. Even these data, however, have gaps. The survey is conducted irregularly, is only cross-sectional, does not provide state- or metropolitan-level detail for jurisdictions that cannot afford to participate, and suffers from the low response rates that characterize such travel surveys. Detailed information about work trips has been provided historically by the decennial census. That source has shifted to the American Community Survey, a rolling survey. To protect privacy, the American Community Survey will not provide the level of geographic detail that regional travel models demand. The consensus study on travel data described in Chapter 3 would provide guidance on what data should be collected to meet the needs of federal, state, regional, and local policy makers.
If legislation that requires reductions in vehicle miles of travel and implementation of strategies to change travel behavior is enacted, much more extensive data collection will be required. States and regions would need a much better baseline estimate of total travel than is currently available. Such information also would be helpful in analyzing the options available to them for achieving the goals of the legislation.
Information about the movement of cargo by rail and water is available from the Association of American Railroads and the U.S. Army Corps of Engineers, respectively. However, information about freight movement by truck is sorely lacking because of its proprietary nature. Trucks are critical to origin-to-destination trips for almost all freight movement, and a substantial share of ton-miles is moved by truck only. Therefore, the lack of information is a serious gap for public-sector officials trying to determine whether to invest in other modes to divert truck traffic.
Most publicly available information about commodity movements depends on the Commodity Flow Survey (CFS), a survey of domestic shippers conducted by the Census Bureau and funded by BTS. This irregularly conducted survey has suffered from inadequate funding, which in turn has resulted in declining sample sizes that have compromised analysis at
levels of geography below the nation as a whole (TRB 2003a; TRB 2003b). Thus, states and regions trying to understand current and future truck movements have great difficulty in analyzing options and forecasting demand in their jurisdictions. Furthermore, the CFS does not provide information about imported commodities or agricultural goods (a large share of total shipments), and it does not indicate the routes taken from origin to destination. Route information must be imputed on the basis of estimates that rely on other data sources. States obviously need route-specific information when they make long-range investments in freight capacity. Special Report 276 (TRB 2003a) provides one broad-based strategy to collect and provide information about freight movements. The consensus study on travel data described in Chapter 3 would provide guidance on the collection of data that would be helpful in meeting the needs of federal, state, regional, and local policy makers.
Burbank (2009) estimates an annual cost of about $300 million to collect data to inform decisions about the best mitigation policies for federal, state, and regional authorities to implement. The major part of this cost is based on a “bottoms up” estimate of what some MPOs currently spend to collect household travel data for metropolitan area travel models. Much better data and models than in common use would be required to develop MPOs’ capabilities to analyze alternatives to meet GHG emission reduction goals and measure progress toward meeting those goals (TRB 2007). Data costs for MPOs with better-than-average data collection and modeling programs are roughly $0.70 to $0.75 per capita per year, for a total of $210 million to $225 million (Burbank 2009). To this cost need to be added (a) the cost to states and regions of conducting surveys of transit users and freight movements on roads and highways and (b) the cost of national-level surveys of households, shippers, and owners of transportation vehicles. These costs could easily add up to another $0.25 per capita per year, resulting in Burbank’s estimate of $1 per capita or, in round figures, about $300 million per year.
Winkelman (2009a; 2009b), building on the work of the Center for Clean Air Policy, estimates that the necessary data collection, model improvements, implementation of model improvements at metropolitan
planning agencies, research, and evaluation could cost as much as $1 billion per year over the next authorization. According to his “top down” estimate, to make an overall investment of $500 billion in infrastructure, as recommended in the 2009 surface transportation reauthorization proposal of the House Transportation and Infrastructure Committee, the nation ought to be willing to invest 1 percent of that amount in ensuring that the investments are appropriate in moving toward national and regional transportation and environmental goals.
Winkelman’s estimate includes more costs than Burbank’s, because his estimate incorporates research and evaluation along with data collection, whereas Burbank’s is only for data collection. If one nets out of Winkelman’s $1 billion estimate the current cost of highway and transit research programs—on the order of $730 million (TRB 2008, Table 2-2)—$270 million is left for data collection, which is roughly comparable with Burbank’s $300 million estimate.
The travel data needed for improved modeling and analysis to meet goals to reduce transportation GHG emissions and conserve energy would be useful for other applications at the federal, state, and local levels. In particular, the data would be helpful for general state and regional transportation planning purposes and for establishing compliance with Clean Air Act mandates in regional and state transportation capital plans. Whereas climate change imperatives might provide the impetus for collecting such data, a portion of the extra cost of data collection should be attributed to these other applications.
TRB Transportation Research Board
Burbank, C. J. 2009. Greenhouse Gas (GHG) and Energy Mitigation for the Transportation Sector: Recommended Research and Evaluation Program. Transportation Research Board of the National Academies, Washington, D.C.
Center for Clean Air Policy. 2009. CCAP Travel Data and Modeling Recommendations to Support Climate Policy and Performance-Based Transportation Policy. http://www.ccap.org/docs/resources/613/CCAP%20Travel%20Data%20Recommendations%20%28Final%201%2030%2009%29.pdf. Accessed Aug. 4, 2009.
TRB. 2003a. Special Report 276: A Concept for a National Freight Data Program. National Academies, Washington, D.C.
TRB. 2003b. Special Report 277: Measuring Personal Travel and Goods Movement: A Review of the Bureau of Transportation Statistics’ Surveys. National Academies, Washington, D.C.
TRB. 2007. Special Report 288: Metropolitan Travel Forecasting: Current Practice and Future Direction. National Academies, Washington, D.C.
TRB. 2008. Special Report 295: The Federal Investment in Highway Research 2006–2009: Strengths and Weaknesses. National Academies, Washington, D.C.
Winkelman, S. 2009a. Do. Measure. Learn. (Repeat). Presented at the 12th Asilomar Transportation and Energy Conference, July 20. http://www.its.ucdavis.edu/events/outreachevents/asilomar2009/presentations/Session%205/Winkelman_Asilomar_2009.pdf.
Winkelman, S. 2009b. Testimony Before the Subcommittee on Technology and Innovation, House Committee on Science and Technology: The Role of Research in Addressing Climate Change in Transportation Infrastructure. March 31. http://www.ccap.org/docs/resources/612/Winkelman%20testimony%20(3%2031%2009).pdf. Accessed Aug. 5, 2009.