A Justice-theoretic Exploration of Accessibility Measures
K Martens, Radboud University of Nijmegen, NL; A Golub, Arizona State University, US
The recent shift towards accessibility measures directs the attention to the distributive question in transport. Based on equity argument we dismiss a number of widely used accessibility measures and propose others instead.
The recent shift towards accessibility measures directs the attention to the distributive question in transport: who reaps the accessibility benefits from investments in the transport system? As the distributive question is highly political, so is the use of accessibility measures. This does not imply, however, that the choice of an accessibility measure is merely a political issue. Rather, we argue that the choice for a certain accessibility measure should be based on a clear understanding of the responsibility of both government and the transport system user for the level of accessibility of that user.
We start our argument with the notion of motility, which we define as a person's ability to participate in desired activities, i.e. the access of a person in a particular time and place. A person's motility is determined by four factors: the transport system, the land use pattern, the extent to which a person appropriates certain parts of the transport system, and a person's plans and projects. Since motility includes a person's particular space-time situation, we need a person-based or utility-based accessibility indicator to measure motility.
We differentiate motility from person-based access. Person-based access is defined as the set of activities that can be reached given the components of the transport system that are appropriated by that person. Like motility, it thus indicates a potential and is determined by the transport system, the land use pattern, and the extent to which a person has appropriated certain parts of the transport system. Person-based access can vary widely, even for persons in the same location, as people may differ in terms of the transport modes that are appropriated. For instance, a person may have a permanent access to her own car, be able to use the transit system and able to cycle, while another person may have only irregular access to a (borrowed) car and not be able to cycle. As a result, person-based access may differ substantially. However, we argue that, from a distributive perspective, it is not this wide variety that counts, but the extreme values. In terms of accessibility measure, an improved location-based measure that accounts for various combinations of available travel modes would be sufficient to assess the accessibility gaps between the extremes.
We distinguish a third "leve" termed system-based access. This is defined as the set of activities that can be reached using a certain transport mode. Like person-based access, it indicates a potential. In contrast to person-based access, it only indicates the potential for one particular mode of transport, not for the particular set available to a person or group of persons. System-based access can be measured with standard location-based accessibility indicators.
We subsequently argue that, while ultimately the motility of people is what counts, the transportation planning effort could actually start the other way around, from the level of system-based access. We have two arguments for this. First, differences in system-based access affect much more people than differences in personal mode appropriation or in personal projects. It is a systemic phenomenon. Thus, the analysis that informs the transportation planning effort should start at the system level, providing a first insight into the distribution of access.
Second, we argue that differences in motility are only relevant if they can be considered the responsibility of the government. For instance, a poor level of motility or person-based access due to a person's failure to appropriate a certain transport mode cannot be considered a responsibility of the government, unless this failure can be attributed to systemic features (e.g. lack of bicycle capabilities among ethnic minorities in the Netherlands) rather than to a person's personal responsibility (e.g., refusal to learn to cycle or to understand the transit system). Likewise, a person's space-time constraints are only relevant from a policy perspective if they are systemic, i.e. a characteristic of a particular group. In that case, no insight into the activity schedules and limitations of real persons is needed, but merely a translation of general limitations (e.g. people working night shifts, single parent households) to access levels for the particular group in a particular spatial setting.
Taken together, we argue that while highly sophisticated accessibility measures have been developed in recent years, the application of simpler measures is sufficient to inform accessibility planning at the macro level. High gaps in access levels between transport modes or neighborhoods call for additional analyses, which may require more detailed accessibility measurement, although here, too, the focus should be on systemic factors shaping access levels rather than individual characteristics and activity patterns.
Association for European Transport