How Fares Can Impact Access to Opportunities

by Willem Klumpenhouwer

Sunday, May 22, 2022

Despite free-fare transit (sometimes misleadingly referred to as "free" transit) recieving continually growing attention, I don't tend to talk much about fares and fare policy. This is mainly because the evidence suggests that while the fare price certainly matters to people, there are more impactful ways to improve people's mobility and access, and to induce ridership: Frequent and reliabile service.

But fares do have an impact on people's lives, especially if they make up a significant portion of someone's income. This is exacerbated when there are different types of service with different levels of quality at different prices. A typical example is regional or "commuter" rail or bus service, where you can pay a premium to take a faster train or bus; a service usually designed for and marketed to commuters.

This creates a two-tiered system of service. The very nature of premium service necessitates that there will be gaps between what the two types of service have to offer and raises some important transportation justice questions:

  • How big is the gap between premium and non-premium service?
  • How does this gap vary accross different demographic groups and income levels?

Developing quantified answers to these questions was a big part of what the TransitCenter Equity Dashboard was aiming to achieve, and taking a deeper look at the resulting data was the focus of a recently published paper in the Journal of Transport Geography entitled Living on a Fare: Modeling and Quantifying the Effects of Fare Budgets on Transit Access and Equity (if you don't have access, you can download a preprint version). In this article I will provide a high-level summary of what we did, and what we found.

Access: A Measure of Potential

In order to quantify how fares can limit transit for people, we need a metric. While all metrics have flaws, some are better than others and for public transit service, the measure of access to opportunities is gaining growing use due to its ability to capture aspects of the both transit service design and land use patterns within a city.

The basic premise is this: From any given starting point, we count up all of the destinations of a certain type that can be reaached with a certain amount of effort. Destinations can be anything we feel is important, from more abstract concepts like "total number of jobs" or "acres of park space" to more concrete thigns like "grocery stores". You can read a more detailed description and discussion of access as a measure of transit service on Jarrett Walker's blog.

In the study, in large part because we were interested in the gaps created by premium transit service with higher fares, we focused on low-income jobs as our destination. For a measure of "effort" and to quantify the gap between these premium and non-premium services, we looked at how many low-income jobs are reachable from a given origin in 45 minutes of transit travel time both with a limit on the money that a rider could spend on fares, and without.

Defining the Fare Gap

Consider the two possible methods of getting from an origin to destination by transit in the diagram below. Option A uses regular non-premium transit and costs $2.25. In this case, you reach your destination in 52 minutes, and so the jobs available at that destination are considered out of reach for our metric and are not counted. If you can afford the higher fare of $5.80, however, you can reach that destination in under 45 minutes and so the jobs at that location are within reach. This is how the gap between premium and non-premium transit is measured.

Two paths to a destination, one more expensive than the other.

Two paths to a destination: One is slower and more expensive, the other faster and cheaper.

The bulk of the work for the project went into finding ways to relatively accurately quantify the fare between two points within the seven U.S. urban regions we studied. We looked at two networks for every city: One with the whole transit network, and one with premium services removed to find the best paths between two points on the non-premium network. We also assumed a "fare budget" for a single trip when quantifying the gap between premium and non-premium transit of about double the standard fare, adjusted slightly for cost of living in a given city.

With these methods in place, we were able to calculate two numbers for every origin within a study area: The total number of low-income jobs reachable in 45 minutes when no restriction was put on the monetary cost of the trip, and the total number of low-income jobs reachable in 45 minutes when a limit was put on the monetary cost of the trip. The discrepancy between these two numbers is what we were interested in. To capture the difference between these two numbers we created a measure of "fare disparity", which depends on the ratio of the total number of jobs reachable with a fare budget to the total number of jobs reachable without a fare budget: \[ \mbox{Disparity} = 1 - {\mbox{Jobs reachable with a fare budget} \over \mbox{Jobs reachable without a fare budget}} \] In this measure higher is worse: A value of 100% disparity means no jobs are reachable without exceeding the fare budget limitation, and a value of 0% means that no difference exists between what you can reach with or without a budget. A city with completely flat fares and no premium service would score 100% in all zones.

To answer the second question about how these gaps are distributed among different population groups, we used data from the U.S. Census to look at who lives in each zone, and calculated the average disparties experienced throughout a study region by different groups. In theory, a city without any historic segregation or with a transit system that provides service to all population groups evenly, the gaps between these groups would be zero, even if fare disparities in the city were high due to an abundance of premium transit.

Here's what we found

Across the seven major U.S. urban areas we studied, New York City had the highest level of average fare disparity (see the diagrams below) with the average person losing access to 31% of low-income jobs when a fare budget was imposed. This gap was different accross population groups: the average Black person saw a fare disparity of over 39% while the average white person saw a disparity of 25%. We also found that 20% of the population experiences a fare disparity of 50% or greater, meaning that they lose access to over half of the low-income jobs when a fare budget is imposed. Again, gaps between groups exist: Only 13% of white people fall into that category, while 27% of AAPI/Hawaiian people experience this dramatic drop in access.

Fare disparity matrix for seven U.S. urban areas and eight population groups.

Fare disparity averages for each region and demogragrphic group (left), and the percentage of the population living in areas with a fare disparity above 50% (right).

To understand how these differences happen, let's have a look at a map showing how the fare disparity varies throughout the New York City metropolitain area. In map below, each origin zone is shaded based on the fare disparity calculated for that origin. Darker purple means higher disparity. The map shows how the different boroughs of New York and portion of New Jersey across the river are very different: Staten Island has low disparity likely due to the relatively long trip into the urban core, while Queens has generally high fare disparity due to the regional railways that serve the borough. Combined with the high levels of demographic segregation in New York and other U.S. cities and you end up with the gap between demographic groups.

Fare disparity map showing New York City.

Fare disparity for each origin zone in New York City's Metropolitan Statistical Area. Darker purple means higher fare disparity.

This pattern exists to different extents accross all of the study areas, but is the least prevelant in Los Angeles, where transit fares throughout most of the city are a single flat price, and very little premium transit exists in general. This shows an important caveat to the way we measured things: If transit service is relatively low everywhere, then disparity can also be low. Nevertheless, Los Ageles' flat-fare policy means that differences in the opportunities created by differences in fares is very small. Take a look at a map of Los Angeles

Fare disparity map showing Los Angeles.

Fare disparity for each origin zone in Los Angeles' Metropolitan Statistical Area. Darker blue means higher fare disparity.


While charging a higher price for a faster or higher quality transit service may seem like a logical approach, it can be problematic, especially when it creates a two-tiered level of service between the same origin and destination. The gaps between what these services provide to individuals can be substantial, and the makeup of our cities means that large differences can exist between demographic groups. Given that public transit is intended to be available and open to all, it should not be contributing to the existing systemic gap between marginalized and non-marginalized groups of people.

With these and other quantifications, we can start to adjust our fare policies to avoid these pitfalls. Integrated fare plans where agencies coordinate their fares based on the principle of "one origin-destination trip, one fare" are a good place to start. In large regions with multiple agencies, this requires negotiation and understanding of a common goal. But first, the scale and systemic nature of the problem needs to be pointed out. Starting the conversation is the goal of this study, and of the open data and visualizations available on the TransitCenter Equity Dashboard.