Data for Evaluating Economic Development Consequences of Transportation Investments
Efforts to evaluate the economic development consequences of transportation investments frequently come up against a glaring gap -- the lack of information on the flows of goods and people. This is most common in urban studies, though data on region-wide freight movements is also commonly deficient. New research is needed to develop ways of using existing resources to improve the quality and availability of data needed to demonstrate the economic consequences of transportation investments in a way that enhances public understanding and steers clear of black box models.
Illustrating this problem is the fact that most analyses of the relation between transportation investment and economic productivity do not take into account the intensity of use of transportation systems. Productivity studies, particularly those using production and cost function frameworks, typically treat all transportation systems as if traffic flows are the same. This simplified assumption potentially leads to biases in the estimates of the productivity of transportation infrastructure. Moreover, this approach ignores the very activity—shipment of goods—that generates the productivity gains.
Another deficiency in data collection is the lack of information that links the location of businesses and households to the location of the transportation systems that provide them with services. Most productivity studies are conducted using data that are aggregated by some level of government jurisdiction. Studies performed at the state or national level fall far short of establishing a spatial link. It is likely that state-level analyses could attribute the efficiency gains experienced by a business in one part of the state to an Interstate highway located in another part of the state. The problem is even more acute for national-level studies. Furthermore, estimates from state-level analyses do not address the questions that policy makers or planners have to consider concerning the location and type of future projects.
Among regional and corridor studies, the links between highway benefits and patterns of use are typically developed, but the net productivity effect is often not measured. Such studies tend to pay more attention to gross levels of job and income attraction than to the measurement of net productivity effects.
One important direction in coming years is to explore better ways to generate and collect the data necessary to conduct useful studies. Because important effects go beyond the outcomes of the transportation system itself, many types of data are needed. These data include not only transportation system characteristics, but also firm-level characteristics, transportation financing information, commodity flows, and accompanying characteristics of the regions included in the economic analysis. Furthermore, the data should be both cross section and time series to improve the reliability of the estimates.