The Distribution of Access: Evaluating Justice in Transport

University of Minnesota Ph.D. dissertation. December 2018. Major: Civil Engineering. Advisor: David Levinson. 1 computer file (PDF); xii, 134 pages.In academics, transport planning, and transport project implementation there is growing interest in better understanding the justice associated with transport systems and projects. Considerable effort has been expended on understanding the negative unintended side-effects of transport systems and their disparate impacts on minority and low income populations since the establishment of the Civil Rights Act of 1964 and implementation of Executive Order (EO) 12898. The Civil Rights Act of 1964 established discrimination based on race, sex, color, religion and/or national origin as illegal in public accommodations. This bill was intended to provide against discrimination in public settings such as shops, schools and theaters, as well as to protect the right to vote. In addition Title VI of the bill outlawed discrimination by government agencies receiving federal funds, indicating that agencies found in violation could lose said funding. Despite the advances of the Civil Rights Act of 1964, instances of widespread and institutional discrimination continue to be observed. In the 1980s a study performed by the United States Government Accountability Office found that minority and low-income neighborhoods were the sites of a disproportionately high number of toxic waste facilities throughout the South. As a result of the study, President Clinton signed EO 12898 which requires all federally funded agencies to identify and mitigate adverse health and environmental impacts of their programs, particularly as those impacts pertain to target protected groups. In practice the following groups are targeted in investigations of equity: African Americans, Hispanics, Asian Americans, Native Americans, Alaskan Natives, Native Hawaiians, Other Pacific Islanders, low-income populations, the elderly, the disabled, and children. In the 1990s the Los Angeles MTA began to prioritize the implementation of rail projects to the benefit of mostly white suburban commuters at the expense of the bus system which serviced the mostly black inner-city poor. This prioritization led to significant cuts to funding for bus programs resulting in proposals for a decrease in both the frequency and quality of service available via the bus network. In addition the MTA sought to increase fares to cover the gap in bus funding and as part of the fare increase to eliminate the bus pass, which provided for use of the buses at a reduced rate for regular bus riders. As part of a federally mandated process, the MTA held a public hearing on its proposals for making up for the shortfall in bus funds. The bus rider community came to the meeting en masse and provided insight into the hardship that the service cuts and fare increases would generate within the community. Despite the turnout at the meeting, the MTA board voted to raise fares, eliminate passes, and cut service as planned in order to meet budget shortfalls, and a week later approved a plan which provided funds nearly matching the bus system funding deficit to a single light rail project. In the end the bus riders unionized and successfully settled a class action lawsuit against the MTA on the grounds of intentional racism, but the enforcement of the settlement continues into this century. The experience of the Los Angeles Bus Riders Union demonstrates the prevalence of justice issues in a modern context and a need for a better understanding of justice and a means by which to systematically evaluate justice in the accrual of negative impacts, costs, and benefits of government projects among the populous. However the effort focusing on the distribution of the benefits of transport systems has not been as extensively addressed as the focus on the negative impacts of transport systems. Further the research that has been performed to date is varied in nature with little agreement about what the theoretical justice foundation is or much attempt to define it outside of the mention of a few leading economic theories, primarily Rawls' Theory of Justice. In addition, there is little agreement on what the operationalization of these concepts of distributive justice should be. Further, of the measures currently in use few are capable of addressing non-segregated populations, with much existing analysis being dependent on arbitrarily chosen concentrations of target populations as representative of the experience of all members of those populations. These issues make it difficult to understand the implications of transport policies on the distributive justice of transport networks. This thesis systematically reviews justice as a standalone concept separate from law that guides the actions of individuals and states in the division of goods and labor as well as towards individuals and populations. In this dissertation philosophical literature dating back to the times of Confucius and Plato is discussed and reviewed and followed through history to further the understanding of justice throughout human history and how it relates to various government structures. Within the context of a western government structure justice is then discussed as an economic concept specifically relating to the distribution of resources. Finally the chapter concludes with a review of the literature relating to justice in transport noting the tendency to utilize accessibility, which accounts for both land uses and transport networks, as a measure of the transport good. This thesis further utilizes accessibility and four theoretical concepts of justice from economics to operationalize measures of justice which are based on the distribution of access. The four concepts of justice utilized are absolute need, equality of opportunity, maxi-min justice, and relative need. There is also a discussion of the meaning and measurement of accessibility as applied in this dissertation. The result is a series of multiple potential operationalized measures of justice. This thesis applies the measures of distributive justice developed to the Sioux Falls Network, a commonly used toy network in transport research, as a demonstration of the measures. This thesis also indicates other information, namely levels of segregation, that is useful to consider in conjunction with the accessibility data and distributive justice measures. The measures are then applied to a case study. The case study is laid out as a before and after analysis of the Harris County Re-Imagined Bus Network, which was fundamentally a rescheduling of the entire bus network for the Houston metropolitan area which occurred shortly after the opening of two new light rail services. The thesis details the phases of the project itself, discusses the measurement of accessibility for the various phases of the project, applies the accessibility based distributive justice measures, and discusses the implications of each. The distributive justice measures are used to compare 48 of the 50 largest metropolitan areas in the United States by population. This provides a description of the metropolitan areas and the accessibility in those areas. The distributive justice measures are calculated based on the access data, and predictive curves are developed for the measures to account for the impacts of some land use factors on the measurements. The results provide a basis for comparing the strengths and weaknesses of each of the measures as well as providing information on which metropolitan areas under/over perform relative to prediction in distributive justice. It is interesting to note that this varies depending on the distributive justice measurement applied. Finally regression analysis is performed using network characteristics and land uses on distributive justice at the sub-county level. The thesis reviews graph theory and its implications for network analysis, particularly in its ability to simplify the inherently complex nature of transport networks. Further the thesis includes preliminary analysis of the various network characteristics, land use measures, and distributive justice measures developing a series of hypothesis for how the network characteristics might theoretically impact distributive justice. Applying these measures in a regression analysis with land use and network characteristic measures led to the preliminary conclusion that in order to improve justice without decreasing access and opportunities across the board policy should focus on increasing worker residential density, decreasing average circuity, reducing average edge length (the distance along a road between nodes), increase the number of edges in a network, decreasing average closeness centrality, and reducing the radius of networks. However in order to validate that the trends noted here are causal rather than merely correlated it will be necessary to perform a time series regression, which will require more data than is currently available