Multimodal travel and the poor: evidence from the 2009 National Household Travel Survey

Abstract

Most travel behavior studies focus on discrete mode-choice outcomes. They predict the likelihood of traveling by a single mode (e.g. solo driving, carpooling, taking public transit, walking, and biking). Yet qualitative studies focusing on low-income households suggest that their mode choice does not fit neatly into a single category; they regularly “transportation package,” use multiple modes of travel in a single day. The authors use data from the 2009 National Household Travel Summary to examine the extent to which individuals’ engage in multimodal travel and to determine whether low-income individuals transportation package more than higher-income individuals, controlling for other factors.

The authors find that multimodal travel is less – not more – prevalent among low-income adults than higher-income adults. However, there are important differences in the number and mix of modes that appear to be influenced by income. Moreover, low-income multimodal travelers took far more trips than even higher-income unimodal travelers. This finding suggests that providing viable avenues for multimodal travel may enhance low-income individuals’ mobility, particularly if they face barriers to automobile access.

Keywords:

• Multimodal travel
• Travel behavior
• Low-income travel
• National travel survey
• Modal mix

Notes

1 There is some debate in the literature on the exact distance that is “reasonable.” The distance may depend on the access mode, either bus or rail. As Zhao et al. (2003) find, however, it is commonly accepted that people are willing to walk up to ¼ mile to a transit stop or station, beyond which the likelihood of their using transit declines.

2 “Introduction to the 2009 NHTS” (nd) http://nhts.ornl.gov/introduction.shtml (accessed August 11, 2011).

3 van Nes (2002) and van Nes and Bovy (2004) analyze trip data; Nobis (2007) and Kuhnimhof et al. (2006) use travel data for a 1-week period. Krygsman and Dijst (2001) define multimodal travel as a trip (from origin to destination) in which the primary mode is public transit.

4 The authors also experimented with a 15-min threshold for walk trips, but this change did not alter their results substantively.

5 Automobile ownership is strongly associated with income. Therefore, the authors tested their model by excluding the automobile ownership variables. The relationship between income and multimodal travel remains unchanged.