Eirik Melaa Skjelsvik (p), John Östh

João Pedro Saldanha Corrêa

In recent years there has been an increasing interest in developing methodologies for evaluating the equity impacts of public transport projects, such as the impact of transit investments on accessibility to employment opportunities. Most studies have tended to focus on either the spatial, horizontal, or vertical equity impact of public transport projects. Given limited public resources there will in most cases be important trade-offs between different transit network design principles and the different equity standards. To understand the distributional impacts of restructurings in public transport services it is therefore important to have methodologies that can evaluate the impacts on potentially diverging equity targets.

In this study a holistic methodology is proposed for evaluating the spatial, horizontal, and vertical equity impact of transit network restructurings on job accessibility using a combination of GINI coefficients/Lorenz curves and the Theil Index. Of the three outlined equity principles, we will operationalize spatial equity as the equal distribution of job accessibility across space and horizontal equity as the equal distribution of job accessibility across the population. Finally, vertical equity will be operationalized as a distributional principle where socially disadvantaged groups should not be worse off than more advantaged population groups.

The study will use the restructuring of the bus system in the Trondheim Metropolitan Area in 2019 to test the evaluation methodology. In the restructuring in the Trondheim Region a bus system primarily operating direct lines through the city center was replaced with a feeder-based bus rapid transit (BRT) system operating a reduced number of lines through the city-center. Using a GTFS-based network model, public transport accessibility to jobs before and after the BRT-implementation will be calculated. Combined with geocoded information from the Norwegian Entity registry, detailed residential grid data, and an individual-level micro dataset with socioeconomic, workplace and residential information, the data allows for developing a spatially disaggregated and comprehensive methodology for assessing equity impacts.

Stefano Magrini (p), Margherita Gerolimetto, Marco Di Cataldo, Alessandro Spiganti

Eirik Melaa Skjelsvik

This paper analyses institutional changes in local governance structures as determinants of noise and air pollutants reductions in urban areas. We exploit the designation of Transport Management Areas (TMAs) as a quasi-experimental framework. TMAs are designated by the US Secretary of Transportation for urbanized areas that overcome the population threshold of 200,000 as defined by the Bureau of Census, in recognition of the complexity of transportation issues. When an urban area is designated as a TMA, the Metropolitan Planning Organization (MPO) responsible for that urban area (an MPO is mandatory for urban areas with population over 50,000) is subject to several transportation planning requirements among which a Congestion Management Process and an Air Quality Plan (for non-attainment areas). This transformation provides certain benefits, but also implies additional requirements. In particular, the planning processes in MPOs in TMAs also must be certified by the Secretary of the Department of Transportation as being in compliance with Federal requirements. In addition, the MPO serving the TMA has a more formal governance structure as its board must include local elected officials, officials of public agencies that administer or operate major modes of transportation, and appropriate State officials. We rely on regression discontinuity design techniques, a long-standing way to obtain credible causal estimates, when specific conditions are met, that is gaining increasing popularity in recent times. In particular, among other sources, we employ data on noise and air pollution in US Transport Management Areas provided by the US Environmental Protection Agency (EPA).

Marina Toger (p), Umut Türk, John Östh

Stefano Magrini

Sustainability of mobility for leisure and tourism in exurban areas gains increasing attention following growing popularity of outdoors leisure and avoidance of people-dense activities during and after the COVID-19 pandemic. Scarce public transit connections and distance friction exacerbate inequalities in transport accessibility to non-urban areas. Understanding motivations and mechanisms of mobility flows for leisure and tourism towards the non-urban areas is critical for addressing the difficulties in sustainable development of non-urban destinations.
Using spatial interaction models, this paper examines spatial mobility between non-urban and urban locations around the Stockholm-Uppsala greater area. Mobile phone data records were analysed, and OpenStreetMaps data was complemented to provide information on land cover and amenities such as schools, markets, hospitals, and cultural amenities. The data from OpenStreetMaps was utilised as proxies for push and pull factors. The study's hypotheses are as follows: i) natural amenities are pull factors to non-urban areas, and urban amenities are push factors from non-urban and towards urban areas; ii) push factors from non-urban areas will have a weaker effect during holidays and leisure times; iii) push factors will have an even weaker effect during the COVID-19 pandemic, but this effect will vary according to individuals' socioeconomic characteristics and ability to travel; and iv) distance decay will follow a similar pattern, with less decay from urban to non-urban during the pandemic but high decay from non-urban to urban. The findings of this study will reveal inequalities in mobility by socioeconomic characteristics and between urban and non-urban areas.

João Pedro Saldanha Corrêa (p), Eduardo Amaral Haddad

Marina Toger

We explore data from the São Paulo Metropolitan Region (SPMR) to determine the effect of accessibility to job opportunities on house prices. Large datasets on housing, transportation, and production are used to map price surfaces for housing in SPMR. Simulations on the distribution of projected population growth throughout the region were run based on the price surfaces. SPMR is divided into 517 traffic zones (TZ) and data is available for the employment level in each TZ as of 2017. We use the r5r package for R to estimate travel times by private transportation and public transit between each pair of TZs. An accessibility index for each transport mode was calculated for each TZ using the data on employment and the travel time estimates. This index measures how many job opportunities are plausibly within reach of an individual dwelling in a given TZ. A conventional hedonic prices model was employed to determine the marginal price of accessibility to job opportunities in the offer price of new housing units in SPMR. The dataset we use contains price and structural details of more than 1,000 types of apartments and houses built between 2014 and 2017. The marginal price allows us to map the housing price surface in SPMR as of 2017. Several macroeconomic scenarios for the trajectories of final demand components are projected up to 2040 employing a computable general equilibrium model (CGE). Based on these trajectories, an input-output model (IO) calibrated for the SPMR provided estimates for changes in production and employment levels for each municipality in the region, as well as for the rest of the São Paulo State, and for the rest of Brazil. Both CGE and IO models were developed in previous works. The municipal results were then downscaled to the level of the TZ. Changes in TZ’s employment levels result in new price surfaces for housing, which are used to distribute projected population growth among TZs for each scenario, based on demographic projections from São Paulo State’s census bureau.