Effects of Urban Density on Rail Transit

Judy S. Davis and Samuel Seskin, May 1, 1996

Despite the long-term and continuing trend away from central business districts and toward suburban development, a number of factors are motivating recent attention to rail transit. These factors include:

concerns about the negative impact of auto-oriented sprawl desires to reduce air pollution and energy consumption interest in rebuilding urban communities need to provide access and mobility to those without autos desires to save the costs and avoid the impacts of new or widened roadways

Many metropolitan areas in the United States are considering the addition or expansion of light rail and commuter rail systems to link employees with business centers. The land use characteristics of the corridors where transit lines operate have been shown to influence transit ridership, but much of the previous work is more than 20 years old and based on data from a limited number of regions.

Our national research project, conducted for the Transit Cooperative Research Program with Jeffrey Zupan, expands and updates earlier research. We analyzed information on 261 stations on 19 light rail lines in 11 cities, including Baltimore, Cleveland and St. Louis, and 550 stations on 47 commuter rail lines in the six city regions of Boston, Chicago, Los Angeles, San Francisco, Philadelphia and Washington, DC.

The study shows that light rail and commuter rail serve distinctly different markets and land use patterns. Light rail with its closely spaced stations attracts more riders per station when it is located in denser residential areas. Feeder bus service helps to boost ridership. Light rail can function in regions with a wide range of CBD sizes and employment densities. Commuter rail depends more on park-and-ride lots at stations in low-density, high-income suburban areas farther from the CBDs, which tend to be larger and more dense than those in light rail areas.

Light rail, with its more frequent service, averages about twice as many daily boarders per station as commuter rail, even though light rail is more often found in smaller metropolitan areas. Figure 1 shows that light rail is most effective in attracting passengers close to the CBD. Figure 2 shows that commuter rail attracts the largest number of its riders about 35 miles out from the CBD. In both figures, other factors affecting ridership, except CBD employment density, are held constant.

Because most transit systems emanate from and focus on a region’s CBD, the amount of employment concentrated downtown clearly affects the demand for transit. Figures 1 and 2 also show that ridership increases with CBD density for both light rail and commuter rail. For light rail, the effects of CBD density on ridership are most pronounced for stations within 10 miles of the core, while for commuter rail the larger impacts occur at stations 20 to 50 miles outside the city.

Changes in Employment and Residential Density

CBD employment density (as measured by employment per gross CBD acre) is nearly twice as important for commuter rail ridership as for light rail. Our study shows that a 10 percent increase in CBD employment density produces 7.1 percent more commuter rail riders, but only 4.0 percent more light rail riders. Commuter rail boardings are more strongly influenced by CBD employment density because these systems usually have a single downtown terminal. Higher-density CBDs assure that more jobs are within walking distance of the commuter rail station. Employment density in city centers is less important in light rail regions since they have more stations distributed throughout the CBD.

On the other hand, a 10 percent increase in station area residential density (as measured by number of persons per gross acre within two miles of a station) boosts light rail boardings by 5.9 percent and commuter rail boardings by only 2.5 percent. Throughout the study these effects are measured holding constant transit system characteristics such as parking availability, station distance to the CBD and station area income levels.

Light rail, with its relatively short lines, is most effective in attracting passengers when stations are in higher-density residential areas close to the CBD. Commuter rail ridership rises more slowly with residential density because commuter rail is a high-fare mode, and its higher-income riders tend to live in more expensive, lower-density places. Moreover, the higher speeds and longer distances on commuter rail tend to increase ridership to the CBD from precisely those places outside the city where residential densities tend to be low.

Cost-efficiency and Effectiveness

In this study, cost-efficiency is measured by annual operating costs plus depreciation per vehicle mile. Effectiveness is measured by daily passenger miles per line mile. For light rail, these measures indicate a strong positive relationship with CBD employment size and residential density. A weaker but still significant relationship occurs for CBD employment density and for the line distance from the CBD. This suggests that medium to large cities with higher density corridors work best for light rail. For commuter rail, larger, denser CBDs attract more riders per line mile, but add to the cost per vehicle mile, creating a trade-off between effectiveness and cost-efficiency.

The length of the rail line is important for both light rail and commuter rail. Longer light rail lines are both slightly more cost-efficient and effective, but ridership diminishes beyond 10 miles. Commuter rail lines are much more cost efficient when they are longer, but their effectiveness declines beyond 50 miles.

This summary does not address many other significant factors in rail transit usage and land use patterns, including operating, capital and environmental costs saved as a result of not using other modes of transportation, notably automobiles and buses. Cities considering investment in new or expanded rail systems need to examine carefully all transportation alternatives in a corridor, including site-specific conditions and local preferences. Further, our study makes clear the need to integrate transit planning with land use planning at the earliest possible stage.


Judy S. Davis is an urban planner and Samuel Seskin is a senior professional associate with Parsons Brinckerhoff Quade and Douglas in Portland, Oregon. As a faculty associate of the Lincoln Institute, Seskin also develops and teaches courses linking land use and transportation. This article is derived from a report titled Commuter and Light Rail Transit Corridors: The Land Use Connection. It will be published by the Transit Cooperative Research Program in the summer of 1996 as part of Volume 1 of An Examination of the Relationship Between Transit and Urban Form, TCRP Project H-1.