Travel time costs in the near- (circa 2020) and long-term (2030–2035) for automated, electrified, and shared mobility in the United States. (May 2021)

• Record Type: Journal Article
• Authors: Compostella, Junia; Fulton, Lewis M.; De Kleine, Robert; Kim, Hyung Chul; Wallington, Timothy J.; Brown, Austin L.
• Is Part Of: Transport policy. Volume 105(2021)
• Page Start: 153
• Page End: 165
• Publication Date: 2021-05
• Subjects: Travel time costs -- Value of travel time -- Generalized cost -- Automated vehicles -- Electric vehicles -- Shared mobility -- Road user fee -- Time incentive; Transportation and state -- Periodicals; Transportation -- Rates -- Periodicals; 388
• DOI: 10.1016/j.tranpol.2020.12.014
• Languages: English
• ISSNs: 0967-070X

Abstract

Abstract: The introduction of automated, electrified, and ridesourcing mobility options will likely revolutionize mobility over the coming decades. Adoption of these new mobility technologies will be shaped by their economic competitiveness. Economic assessments typically focus on fixed and variable out-of-pocket monetary costs such as vehicle purchase, fuel, maintenance, parking, and insurance cost. Non-monetary costs, while difficult to quantify, are also important in determining the relative attractiveness of competing mobility options. Travel time cost (TTC) is an important non-monetary hedonic cost. Vehicle automation and shared mobility enable more productive use of travel time and have a large impact on TTC. We build on our recent monetary cost (MC) analysis by adding TTC for driving a personal vehicle, riding as a passenger, waiting for pickup, searching for parking, and walking to and from vehicles and destination points. We show that TTCs can be of major importance and are sometimes greater in value than the MCs for competing choices. The sum of MC and TTC which we define as "generalized cost" varies with trip type, travel time, and the traveler's value of time which is a function of their income. In our circa 2020 scenarios driving a personal car is the lowest cost mode for all income levels and trip lengths. Pooled ridesourcing is typically lower cost than solo ridesourcing. In marked contrast to the situation in circa 2020, in our hypothetical future automated scenario in 2030–2035 ridesourcing becomes cost competitive with private vehicles; pooled ridesourcing is generally the least attractive option for short trips. Finally, we review road pricing programs around the world. We find that even those systems with the highest fees per mile do not change the relative costs of private and ridesourced trip options significantly, with the exception of congestion area fees (such as London's). These can substantially disfavor the use of private vehicles especially for short trips. Highlights: The sum of the monetary cost (MC) and travel time cost (TTC), which we define as "generalized cost", varies with trip type, travel time, and the traveler's value of time that is a function of their income. TTC can be of major importance for travel choices, in some situations of greater value than the MC for many travelers. Vehicle automation and ridesourcing enable more productive use of travel time than driving, and can result in significant reductions in TTC. In our circa 2020 scenarios, driving a personal car is the lowest cost mode for all income levels and trip lengths. Pooled ridesourcing is typically lower cost than solo ridesourcing. In marked contrast to the situation in circa 2020, in our hypothetical future automated scenario in 2030-2035 ridesourcing in automated vehicles becomes cost competitive with private vehicles; pooled ridesourcing is generally the least attractive option for short trips. As of today, road fees in existing programs do not change the relative costs of private and ridesourced trip options significantly, with the exception of a few congestion area fees (such as London's) that can substantially disfavor the use of private vehicles for short trips. Future pricing systems would need to overcome generalized cost advantages of solo driving, especially once this is automated.