Association for Public Policy Analysis & Management

For the past century, mobility in the United States has been dictated by cars. Furthermore, cars—and even more so, the storage of cars—have dictated urban form. With cities dedicating more space to parking than even streets and roads, parking has become baked into city land use, regulations, codes, ordinances, master plans, and even finances. This paper explores the idea of what happens when a car trip no longer ends in a parking space? Both transportation network companies (TNCs), such as Uber and Lyft, and, eventually, autonomous vehicles (AVs), enable personal mobility without nearly as much parking. As such, these new mobility services necessitate the rethinking of parking policy for a future in which demand for parking will likely be greatly reduced or eliminated—and along with it a potential vital source of revenue. However, very little research has investigated the relationship between TNCs and parking or examined how TNCs may help us better prepare for the future of AV mobility.

In this project, we use Seattle as a case city to develop data, models, and tools to aid in understanding how TNCs are already affecting the demand for parking and the revenue it generates for the city. Specifically, we ask three questions. First, what is the relationship between TNC use and parking demand and revenue over a five-year period (2012 – 2017)? Second, what localized factors—such as transit ridership, land use, TNC use, and car ownership—explain weekly temporal and spatial patterns of parking demand and revenue? And third, how can TNCs inform planning for the future introduction of AVs? In this paper we specifically investigate all areas of the City of Seattle with paid on-street parking and introduce tract-level data from Uber and Lyft to assess the relationship between parking and TNC use between 2012 and 2017. The long timeframe combined with detailed TNC data allows for an assessment of both the short-term patterns and long-term trends of TNC use and parking demand and revenue. Findings of weekly spatial and temporal patterns will inform parking policy today, while the multi-year analysis will yield insights into how cities should manage parking infrastructure to prepare for a new age of AV mobility.

To do the analysis we combine data analysis in ArcGIS and statistical models. Regression models assess how the urban form and transportation choices people make influence the demand for parking and the flow of revenue to the city during the study period.

Initial results demonstrate that parking demand and revenues are currently positively related to TNC use, but that as use of this new form of transportation grows more common demand and revenues will drop if no changes to current parking policies are made.