MBTA Fare System (Charlie, AFC 2.0, Zone, Discounts)

Revenue Anticipation Bonds probably have covenants that would make it untenable to zero out fare box revenue.
 
Saw this on Reddit:
1685228399955.png


New fare readers are being installed on green line trains. Looks like we’ll be getting something like 10 readers per unit (2x at each main door and 1x by each cab). That feels like a lot?
 
That sounds good to me. SF has one per door, and sometimes it's an awkward reach on a crowded train.
 
Java King said:
I agree. This whole process of a transition system with both active seems to be a HUGE waste of money. Is fare evasion that widespread to be losing the millions of dollars spent on this system?
Click to expand...
Since they first announced AFC 2.0 what seems like 25 years ago, a common sticker-shock retort was "at this point, couldn't they save money by just eliminating fares?" - if that math didn't check out then, it's on track to by the time this goes live.
 
I've been seeing the new readers on my bus lately (62). So far the most I've seen is 2 near the front door. One right next to the existing fare box (on your right as you board) and one mounted to the pole as you get to the first seat (on the left as you pass). Haven't seen one at a rear door yet. I thought the plan was to have them at rear doors. Maybe the end state is 3 readers? Two at the front and one at the back. I don't remember ever hearing the specifics on that.
 
Have there been more details on how the PoP system will work? I don't see how a PoP system can work when fares aren't required when boarding SL1 at Logan and they frequently deactivate fare gates during construction to prevent people from having to pay two fares (right now the fare gates are open at Back Bay and Kenmore, Blue Line during tunnel work, etc.).
 
I'm mulling on a potential fare reform proposal and I've come to an interesting problem: I'm trying to figure out how many transit-using commuters conduct their entire commute on a 1-seat-ride within the "Core", which I'm defining roughly as Sullivan - Harvard - Kenmore - Ruggles - Andrew - Airport - Sullivan (excluding everything north of the Mystic), compared to the number of people who commute across that boundary.

I've looked at the BNRD's travel demand data but it unfortunately looks at commutes of all modes, and I don't think it really can give me the answers I'm looking for. I'm playing around with OnTheMap, but I don't love how many assumptions I'm having to make.

I think in particular this question boils down to how many make the following commutes:
  • Harvard <> Downtown
  • Central <> Downtown
  • Kendall <> Downtown
  • South End <> Downtown (SL4 + SL5)
  • 7
  • 8
  • 9
  • 10
  • 11
  • 43
  • 55
The bus routes are a little bit more tractable due to the Better Bus Profiles (and because the idea I'm toying with can potentially be less reliant on answering this question for buses). But the Red Line stations seem vexing -- the ridership figures the T provides are, AFAIK, unlinked trips, so don't distinguish between riders who walk to the Red Line vs transfer from buses.

Any ideas?
 
Riverside said:
I'm mulling on a potential fare reform proposal and I've come to an interesting problem: I'm trying to figure out how many transit-using commuters conduct their entire commute on a 1-seat-ride within the "Core", which I'm defining roughly as Sullivan - Harvard - Kenmore - Ruggles - Andrew - Airport - Sullivan (excluding everything north of the Mystic), compared to the number of people who commute across that boundary.

I've looked at the BNRD's travel demand data but it unfortunately looks at commutes of all modes, and I don't think it really can give me the answers I'm looking for. I'm playing around with OnTheMap, but I don't love how many assumptions I'm having to make.

I think in particular this question boils down to how many make the following commutes:
  • Harvard <> Downtown
  • Central <> Downtown
  • Kendall <> Downtown
  • South End <> Downtown (SL4 + SL5)
  • 7
  • 8
  • 9
  • 10
  • 11
  • 43
  • 55
The bus routes are a little bit more tractable due to the Better Bus Profiles (and because the idea I'm toying with can potentially be less reliant on answering this question for buses). But the Red Line stations seem vexing -- the ridership figures the T provides are, AFAIK, unlinked trips, so don't distinguish between riders who walk to the Red Line vs transfer from buses.

Any ideas?
Click to expand...
Buses are easier, there's a spreadsheet that has averaged data for every stop of every scheduled bus (And Silver Line) trip for a typical week between 2016-2022. It's 6 million rows which is too big for Excel, so here's a version I've already split into just 2022 and cleaned up. Rapid transit is harder, but you might be able to compare gated entries each half hour with boardings by time period to get some insights on how many people enter Park St and then ride to Harvard, for example. To deal with bus transfers you could potentially look at connecting the bus data with gated entries, but that might get tricky because you would either need to estimate bus travel times or look into this absolutely insane dataset of about 30 million bus arrival/departure times.
 
Last edited:
Riverside said:
I'm mulling on a potential fare reform proposal and I've come to an interesting problem: I'm trying to figure out how many transit-using commuters conduct their entire commute on a 1-seat-ride within the "Core", which I'm defining roughly as Sullivan - Harvard - Kenmore - Ruggles - Andrew - Airport - Sullivan (excluding everything north of the Mystic), compared to the number of people who commute across that boundary.

I've looked at the BNRD's travel demand data but it unfortunately looks at commutes of all modes, and I don't think it really can give me the answers I'm looking for. I'm playing around with OnTheMap, but I don't love how many assumptions I'm having to make.

I think in particular this question boils down to how many make the following commutes:
  • Harvard <> Downtown
  • Central <> Downtown
  • Kendall <> Downtown
  • South End <> Downtown (SL4 + SL5)
  • 7
  • 8
  • 9
  • 10
  • 11
  • 43
  • 55
The bus routes are a little bit more tractable due to the Better Bus Profiles (and because the idea I'm toying with can potentially be less reliant on answering this question for buses). But the Red Line stations seem vexing -- the ridership figures the T provides are, AFAIK, unlinked trips, so don't distinguish between riders who walk to the Red Line vs transfer from buses.

Any ideas?
Click to expand...
This is a quicker response than I would have loved, but:

I initially wanted to say that we probably shouldn't focus entirely on commutes, which only account for 20% of all trips. (The BNRD travel demand data looks at all trips, not just commutes.) But then:
  • Many non-commute trips span much shorter distances, making transit, especially rapid transit, not very competitive. People would probably just walk or bike for those trips (e.g. grocery, restaurants), or do them by driving as they're often during less congested times.
    • This is illustrated very clearly in the BNRD data: regardless of which origin you pick, the biggest destination by far will almost always be the same region. That even applies to downtown.
  • "The Great Collapse of the 55":
1706459024304.png

The 55 had several changes during Covid. It was shortened to Copley at all times, whereas it used to run to Park St on weekdays; and rush hour trips are entirely removed. Here's the schedule summary for April 2018 (top) and August 2023 (bottom):
1706459456941.png

Weekday midday frequency slightly improved, I guess, which means techncially the schedule now is "less peak-focused". But the loss of all rush-hour trips means the route now has only 10% of pre-Covid ridership, compared to 60-70% on all buses.

While much of this could be due to the 40-min headways, it's still pretty striking for a very urban neighborhood like Fenway within the core. Indeed, most of the 55's demand came from rush hours pre-Covid:
1706460706140.png


Edit to add: Here's the 2022 US Census 5-year estimates of means of transportation to work for the three census tracks in Fenway:
1706462386002.png

1706462406866.png


I suspect you may be able to do an estimate like this for all downtown neighborhoods, combining this dataset and the "travel time to work" datasets.

Also, you may have missed bus routes 1, 92 and 93 depending on your definition of the core.
 
Last edited:
From an interview with Eng in the globe:

The other part you mentioned about the contactless payment: That is coming. I’ve been testing that. It’s not as far away as people think. It will be here this year. And the exact thing, as you said, you don’t need an account, you don’t need to buy a card. You could use your Apple Pay, your Google Pay, your Android, your iPhone, or a contactless credit card. . . .
 
sean99090900 said:
From an interview with Eng in the globe:

The other part you mentioned about the contactless payment: That is coming. I’ve been testing that. It’s not as far away as people think. It will be here this year. And the exact thing, as you said, you don’t need an account, you don’t need to buy a card. You could use your Apple Pay, your Google Pay, your Android, your iPhone, or a contactless credit card. . . .
Click to expand...
Does anyone know how the T is going to make up for all the credit card transaction fees that the contactless payment will cost them. Moving to a system where every ride is a separate CC transaction is a huge boon to the credit card processing companies (and a huge loss in revenue to the T).
 
JeffDowntown said:
Does anyone know how the T is going to make up for all the credit card transaction fees that the contactless payment will cost them. Moving to a system where every ride is a separate CC transaction is a huge boon to the credit card processing companies (and a huge loss in revenue to the T).
Click to expand...
I would assume there is some deal made in private. Honestly I wouldn't be worried about it that much, many of the people paying for single rides with contactless will have just been buying Charlietickets with one or two rides before, so it probably doesn't change a whole lot.
 
TheRatmeister said:
I would assume there is some deal made in private. Honestly I wouldn't be worried about it that much, many of the people paying for single rides with contactless will have just been buying Charlietickets with one or two rides before, so it probably doesn't change a whole lot.
Click to expand...
Contactless is apparently costing NYC a ton of fare revenue. Don't be surprised if the financial services folks pulled a fast one here. They are much better than the T at these scams.
 
JeffDowntown said:
Contactless is apparently costing NYC a ton of fare revenue. Don't be surprised if the financial services folks pulled a fast one here. They are much better than the T at these scams.
Click to expand...
I wonder how contactless fees compare against increased usage & fare evasion. I think just tapping your phone (not even having to activate Apple Pay/Google whatever) makes it so easy to pay that it's a no brainer to take the subway. There's no longer the hurdle of tourists/first time users needing to understand how the charlie card system works, and it gives less reasons to hop a turnstile.

I will say having used OMNI for 2 years, sometimes it didn't feel like paying a fare, more like "oh i need to tap my phone to unlock the turnstile." Maybe not the best situation for all riders, but definitely something that impacts usage.
 
TheRatmeister said:
I would assume there is some deal made in private. Honestly I wouldn't be worried about it that much, many of the people paying for single rides with contactless will have just been buying Charlietickets with one or two rides before, so it probably doesn't change a whole lot.
Click to expand...
Not a private deal, the MBTA is explicitly part of the MA statewide contract for Electronic Payment Processing Services. If I've read the cost submittal correctly, every credit card transaction costs the T $0.0738 in processing fees, not including what the networks (Visa/Master Discover) charge. Using Mastercard, their published public sector fee is 1.55% + 0.10. So each $2.40 MBTA Subway fare would cost the T $0.211 to collect.

But, lets compare that to AFC 1.0, the traditional Charliecard. If every passenger pays for each trip separately on Charlietickets, this would likely remain unchanged under AFC 2.0. But lets take the occasional rider who stores value. For simplicity's sake lets say I put $24, 10 rides, on it. That cost the T 54.5 cents, 2.27% of their fare revenue. a weekly pass is $22.5, costing the T 52.5 cents, or 2.33%. The same 10 rides run as individual transactions would cost the T $2.11 - approximately fourfold increase, now 8.79% of revenue.

What we don't know is how Cubic and the MBTA will run things, or individuals will behave. Are individuals going to get new stored value cards, or still buy passes? or will the convenience of tapping an ApplePay win? - will Cubic aggregate all taps from a single card and run it at the end of each day, so that your round trip is a single charge? Or will the T do what San Diego, and OMNY have done, and introduce a "best value" system, where your charges are capped to a daily/weekly pass? Both of those methodologies could reduce the total number of transactions going through the system. Keep in mind that people moving from weekly passes to 10x individual tickets would still represent an increase in processed charges.
 
Last edited:
Stlin said:
Not a private deal, the MBTA is explicitly part of the MA statewide contract for Electronic Payment Processing Services. If I've read the cost submittal correctly, every credit card transaction costs the T $0.0738 in processing fees, not including what the networks (Visa/Master Discover) charge. Using Mastercard, their published public sector fee is 1.55% + 0.10. So each $2.40 MBTA Subway fare would cost the T $0.211 to collect.

But, lets compare that to AFC 1.0, the traditional Charliecard. If every passenger pays for each trip separately on Charlietickets, this would likely remain unchanged under AFC 2.0. But lets take the occasional rider who stores value. For simplicity's sake lets say I put $24, 10 rides, on it. That cost the T 54.5 cents, 2.27% of their fare revenue. a weekly pass is $22.5, costing the T 52.5 cents, or 2.33%. The same 10 rides run as individual transactions would cost the T $2.11 - approximately fourfold increase, now 8.79% of revenue.

What we don't know is how Cubic and the MBTA will run things, or individuals will behave. Are individuals going to get new stored value cards, or still buy passes? or will the convenience of tapping an ApplePay win? - will Cubic aggregate all taps from a single card and run it at the end of each day, so that your round trip is a single charge? Or will the T do what San Diego, and OMNY have done, and introduce a "best value" system, where your charges are capped to a daily/weekly pass? Both of those methodologies could reduce the total number of transactions going through the system. Keep in mind that people moving from weekly passes to 10x individual tickets would still represent an increase in processed charges.
Click to expand...
My understanding about OMNY, though, it the cap system actually increases the number of discrete credit card transactions. What used to be a once per month pass transaction becomes a series of discrete ride transactions up to the cap. That is a ton more discrete CC transactions being processed given the large percentage of riders on former passes.
 
JeffDowntown said:
My understanding about OMNY, though, it the cap system actually increases the number of discrete credit card transactions. What used to be a once per month pass transaction becomes a series of discrete ride transactions up to the cap. That is a ton more discrete CC transactions being processed given the large percentage of riders on former passes.
Click to expand...
It also means that if you're visiting NYC from a Wednesday to next Tuesday you'll cap twice, which is annoying. Last time I went to NYC I got a Metrocard because it lets you use a 7-day pass starting at any point.
 
You must log in or register to reply here.

Back
Top