Fare Collection and Cash Use: Effects of the Current Cash/Ticket Surcharge

The MBTA is currently implementing the next generation of its fare collection system. “AFC 2.0” (as we’re currently calling it) will provide many benefits to passengers, including fully integrated fare payment across all modes of T travel, greatly improved account management options, and broader payment options (such as using smartphones to pay). As part of this process, we are conducting extensive outreach and analysis in order to craft the best policies to take advantage of the new technology.

One of the most exciting, and most challenging, changes will be moving cash “off of” buses, surface light rail, and commuter rail. Passengers will still be able to use cash to pay for their MBTA travel, but rather than paying cash on a vehicle after boarding, passengers will need to pay cash at a fare vending machine (FVM) or a retail location before boarding a vehicle. We are writing several Data Blog posts on different aspects of this topic and posting some relevant data to the AFC 2.0 web site. We have written in the past about some of the exciting benefits of this change—how it will improve dwell times, overall speeds, and off-peak bus reliability of these vehicles. We will also be exploring how cash is used on the MBTA in order to help identify and minimize any inconveniences of this change for our passengers who currently rely on cash to pay their fares onboard.

In this blog post, however, we are looking at a different effect of taking cash off of vehicles: eliminating the current price differential between a cash and CharlieTicket fare. Currently, the cash fare is $2.00 for Local Bus travel and $2.75 for Rapid Transit while the CharlieCard fare is $1.70 for Local Bus and $2.25 for Rapid Transit. This differential exists in order to promote CharlieCard usage over CharlieTickets and direct cash payments onboard vehicles. In the future, cash payment onboard vehicles will not be allowed and there will be no fare difference for accepted fare cards, so that promotion will no longer be needed. In addition, CharlieCards will be available at all fare vending machines.

In order to understand the future effects of removing this price differential, we analyzed where our passengers are currently paying the premium for using cash and CharlieTickets. We have mapped what price people are currently paying on average for stored value trips whether using cash, ticket or card. For a rapid transit station or vehicle, if everyone was using stored value on a CharlieCard to board, we would see an average fare equal to the card price of $2.25. If everyone was using tickets or cash to board, the average fare would be the cash/ticket fare: $2.75. For a combination of the different types of media, we would see an average fare somewhere between the two—a lower average fare meaning more people are using cards, and a higher average fare meaning a higher percentage of passengers are using tickets or cash. (For buses, the range would be $1.70 to $2.00.)

To conduct this analysis, we queried our AFC 1.0 database to select all transactions during October 2017, limiting to adult stored value fares. Passes were excluded since all passes cost the same regardless of which type of media they are on, and reduced fares were excluded since they are by definition card-only. We also ignored short fares and express bus fares, since we wanted to see the average fare produced by the card versus ticket differential instead of other factors. For the purpose of this analysis, transfers on CharlieCards were included and the transfer fare was replaced with the full CharlieCard fare. This allowed us to focus on the price differential for different media instead of differences in transfer rates and policies. (By paying a step-up transfer fare, an MBTA passenger effectively pays the full CharlieCard fare for their trip, and when we initially excluded CharlieCard transfers it dramatically skewed the apparent mix of media and average fare at major bus-rail transfer locations.) To make better sense of the data, we grouped the fares by “cluster”, which are groups of stops and/or stations that OPMI has developed based on a clustering algorithm, weighted by service frequency at each stop and/or station. Generally, these clusters group our many stops into understandable areas, weighted around subway stations. We then calculated the average fare for each cluster, separately for bus boardings and for rapid transit boardings.

The following maps display the data grouped by cluster (Click to enlarge).

Average Fare on Local Bus Boardings
Average Fare on Rapid Transit Boardings

Both maps show that the average fare tends to be higher in places you might expect to see visitors and tourists, as well as places that are far from fare vending machines. In the bus map, we see the highest average fares in locations far from fare vending machines—North Shore communities like Lynn, Salem and Revere, and some of the routes in Quincy and places north of Cambridge/Somerville. People in areas near the rapid transit system are likely to have a CharlieCard even if they use the MBTA infrequently.

In the Rapid Transit map, the clusters with the highest average fares are the ones containing Riverside, Quincy Adams and Braintree, and Logan Airport. All of these stations are places you might expect infrequent visitors (who don’t have CharlieCards) to enter the system, whether they are from another state or country, or driving in from suburban locations. We also see high average fares at North Station, Kenmore, Back Bay and Aquarium for likely similar reasons. Places in the city that tend to have mostly local residents boarding see a low average fare—even if people in these areas are riding the T infrequently, they tend to have and use CharlieCards (remember that pass users, including 7-day passes, are excluded from these analyses).


These maps show how the differential between card and cash/ticket fares plays out geographically. As expected, access to fare vending machines and frequency of travel seem to have the largest impacts on whether passengers use card or tickets to pay their stored value fares, and therefore the average fare that passengers in each area are paying. This analysis is informing the MBTA’s prioritization of fare vending machines in areas not near rapid transit stations. Future posts will examine some other aspects of the current fare collection system and analyze how things may change as new policies are developed and implemented. For more about AFC 2.0 read the AFC 2.0 project page.