Toward Accessibility-Based Planning: Fid-Bau Portal

Toward Accessibility-Based Planning

Yan, Xiang

A growing consensus has formed among planners, especially planning scholars, that promoting accessibility is a major policy goal. However, efforts to promote accessibility-based planning face a conceptual impediment: a common assumption that equates the benefits of accessibility to travel cost savings (TCS). Starting from this assumption, many researchers have interpreted the absence of TCS (e.g., savings in commuting costs and reductions in vehicle miles traveled [VMT]) as evidence undermining the rationale for accessibility-promoting strategies such as jobs–housing balance and transit-oriented development (TOD). In this study I challenge these interpretations by suggesting that accessibility improvements can result in not only TCS but also destination utility gains, which means the individual satisfaction from interacting with or choosing desirable destinations. The absence of TCS from accessibility-promoting policies can be explained by accessibility gains manifesting as destination utility gains. To analyze the importance of destination utility gains, I engage with literatures in economic geography and travel behavior and examine some recent urban trends (e.g., the rise of city-to-suburb commuting). I further estimate residential location choice models to test whether households value accessibility beyond the benefit of TCS. Results from the Puget Sound (WA) and Southeast Michigan regions support the hypothesis, demonstrating that destination utility gains shape residential location decisions.

Takeaway for practice

Planners should not automatically interpret the absence of TCS as evidence that accessibility-promoting strategies are not working. To advance accessibility-based planning, planners should start to explore and measure the various forms of destination utility gains. The evaluation framework for land use and transportation policies should shift from being centered on TCS alone to being accessibility based.