Eliminating Barriers to Transit-Oriented Development
Transit-oriented housing developments are compact, mixed-use, pedestrian-friendly developments within walking distance of transit stations, typically defined as a half mile from a station. Advocates of transit-oriented development seek to direct population growth to locations where public transit and infrastructure already exist, with the expectation that the area’s residents, employees, and shoppers will increasingly walk or use transit rather than autos for many of their trips.
Our interviews with municipal officials and other knowledgeable individuals suggest that high-density housing development on infill and greenfield parcels near transit stations has been limited in the state of New Jersey for a number of reasons, including difficulty with land assembly, financial complexity, lack of developer knowledge, and public opposition. Current residents fear increased auto traffic, problems with parking, and an influx of number of school-age children straining public school budgets and leading to property tax increases. The main purpose of this study is to investigate the extent to which these perceptions are accurate.
In order to observe differences between households located close to transit and those located more distant, we conducted a survey of households located within two miles of ten NJTRANSIT rail stations. We compared households located within a half mile of the stations with households living up to two miles away. We also examined the behaviors and preferences of households living in newly constructed or extensively renovated housing, the kind of housing most relevant to new development proposals, located both within a half mile and up to two miles from a transit facility.
To control for the effect of supply factors, especially the availability of parking, we conducted field audits of parking availability and usage in the same ten areas. The audits examined both on-street and off-street parking. To control for school district quality we used data collected by the NJ Department of Education, primarily Scholastic Aptitude Test (SAT) scores and college matriculation rates for high school graduates, as well as New Jersey Assessment of Skills and Knowledge test scores for the third and fourth grade (NJ ASK).
We found that households choosing to live near rail stations have substantially fewer public school children than households living farther away, both in simple tabulations and when controlling for a number of other factors. But auto commuting and auto ownership are correlated more strongly with housing type and tenure, and larger geographical context, and not nearly as much by rail station access.
New homes near transit stations have about half the number of school children as new homes elsewhere, regardless of the type of housing, the quality of schools, the location within the state, or other factors. Specifically, the number of public school children present in new housing near rail stations is about 60 percent lower than in new housing farther away. Even when controlling for a number of other factors including school quality, the number is 50 percent lower.
In comparison to households in older housing farther away, households living in new housing near transit and households in new housing farther away from transit have substantially lower auto commuting and lower auto ownership. But lower auto ownership and usage for commuter purposes are not predicted primarily by rail station proximity. While households living in new housing near transit have about 30 percent fewer autos than those in new housing farther away, when controlling for other factors we find that variance in housing type, tenure, and area of the state accounts for most of the differences. Apartments and condominiums have much lower auto ownership, as do any rented units regardless of housing type. This likely reflects a self-selection process of more affordable housing for smaller families with fewer members.
Households living in new housing near transit are much less likely to use cars to commute to work, doing so 58 percent less than those living in new housing far from rail. But most of this difference has to do with the distribution of responses in the sample and the fact that most new housing near rail stops is in the most highly accessible station areas.
The policy implications of this study are in three areas:
- While auto commuting is not lower across the board within a half mile of stations, it is lower within a quarter mile of rail stations with low amounts of on-street parking. Parking policies should be reformed to maximize the potential of transit-oriented development. Lower on-street parking is highly correlated with less driving to work. Smaller amounts of on-street parking require managing on-street parking with permits and metering. This enables higher density development. The results strongly imply that parking availability should be taken into account when estimating the traffic impacts of new development near transit.
- Local land use policies for high density development should take into account substantially reduced auto use and ownership in high density housing and rental housing (whether in urbanized areas near or far away from rail stations). In this data set, auto ownership is a third lower in an apartment/condominium setting and 25 percent lower in a rowhouse/townhouse setting, compared to single family homes, when controlling for other factors. Auto ownership is also 22 percent lower in rental units regardless of housing type. These differences are roughly additive—in other words, our statistical model estimates a household living in a rented apartment will have about half the number of vehicles of a household living in an owner-occupied single family home. Development opportunities near transit facilities are often well-suited to high density and rental housing.
- Land use policies near rail stations should take into account lower school enrollment impacts of housing there. Although local context will vary, a reasonable starting point is to estimate the number of school children living in new development near transit stations at half that of new development elsewhere.
Single-family home development causes more driving, whether near rail stations or not. Dense new housing development reduces driving and auto ownership, as does lowering and managing the on-street parking supply. From a larger environmental and congestion management perspective, permitting such development should strongly be encouraged.
Permitting higher density development in transit-accessible areas also has clear benefits for the state of New Jersey, including lower congestion and pollution, and lower greenhouse gases. From the perspective of local municipalities, such development is assumed to be associated with higher fiscal and traffic burdens than lower-density development. However, the results of this study suggest those local burdens are significantly lower than has been conventionally assumed.
This research took place from 2007 through 2009, a period that includes an economic downturn in the United States. Interviews were conducted during the spring 2007, the field audit of parking availability and usage was conducted during the summer of 2008, and the survey of households living near stations was conducted during the summer of 2009.