Measuring the Causal Impact of Plans on Development

While creating plans is central to the field of urban planning, there is little evidence that plans have any impact on development.

Impact of TOD Plans Analyzed

In his new article for the Journal of the American Planning Association, "Planning as Bargaining: The Causal Impacts of Plans in Seattle and San Francisco," Adam Millard-Ball begins to fill this gap by examining the impacts of transit-oriented development (TOD) plans in Seattle and San Francisco.

To measure this impact, Millard-Ball uses a mixed methods approach. First, he uses a regression discontinuity design, which assumes that data points on either side of an arbitrary threshold are the same on average other than for a key variable and which side of the threshold they are on. He applies this method spatially to compare the quantity and character of development for parcels immediately inside and outside the bounds of the TOD plans. Figure 3 from the article gives a sense of the similarity of parcels on either side of the threshold by showing photos of the divide in both San Francisco and Seattle. Millard-Ball then supplements this quantitative study design with 20 qualitative interviews with planners, developers, and neighborhood activists to understand the mechanisms that led to changes in development within the plan areas.

Figure 3. Urban fabric at the plan boundaries. The plan boundary runs through the middle of these photos in San Francisco (left) and Seattle (right). In the SF photo, the buildings in the foreground are in the plan area; those in the background (the green building and beyond) are outside but the light industrial character is the same. In the Seattle photo, the house on the left is in the overlay zone, but the house on the right is not. Photos by Adam Millard-Ball.

TOD Plan Impact in San Francisco

Ultimately, the author finds that San Francisco's TOD plan did indeed cause the character of development to change — namely lower parking ratios and greater heights and densities for new buildings, though it is difficult to tell whether it led to a greater quantity of development. In Seattle, however, the study showed no statistically significant differences in parking ratios or densities, suggesting that any differences that did arise could not be attributed to the station area plans themselves, but rather to the city's broader policy reforms.

These findings and the accompanying interview data lead Millard-Ball to identify three ways in which plans affect development. The first and most obvious way is through zoning changes that are outlined by and adopted in conjunction with the plan. The second is that plans serve as anchors for negotiation for developers, city staff, and community activists, meaning they set a standard for "what is fair and reasonable in subsequent negotiations" and thus influence what is feasible in those negotiations. However, plans can only effectively serve as anchors for negotiation when there are opportunities for future negotiations between planning staff, developers, and community groups in the development review process. And third, plan boundaries create what the author terms laboratories of planning, in which planners can pilot progressive strategies that when successful, can be adopted by other neighborhoods within the city.

The article will be particularly helpful for planners who are looking to either better understand or explain how plans can meaningfully impact development, as well as researchers interested in how quantitative research methods (and particularly regression discontinuity design) can be applied spatially for planning research.

Top image: JMWScout/iStock/gettyiimagesplus