Dumb question - who could pass such a rule? Can the federal government and if so which agency? Or a Biden executive order? Or just every local legislature has to enact this one by one? I agree with the idea but what’s the theory of change?
I am with Matt and the people at the Transit Cost Project when it comes to their critique of new transit infrastructure. I’m not completely following Matt’s examples of suboptimal use of existing infrastructure.
Matt seems to confusingly switch between the goals of ridership maximization and revenue maximization, particularly in his Amtrak example. Lowering the price of tickets on the Acela Train should increase ridership by some amount but might increase or decrease revenue depending on whether the price effect or quantity effect dominates. So, it’s important to clarify whether we’re truly talking about ridership maximization vs. revenue maximization.
It could be that, through a mix of inertia and lack of - or expense associated with - viable alternatives, price effect dominates because those riders who rely on Amtrak for their commute would pay a much higher price before switching to another commuting option. Similarly, based on how current infrastructure is situated, it may be the case that lowering the price hardly does anything to increase ridership and generate additional revenue. In which case Amtrak doesn’t have a financial incentive to maximize ridership for existing infrastructure, and a low ridership outcome is dictated more by incentives than lack of critical analysis.
Right as usual. I love RBT out here in Silver Spring because now we don't have to wait 30 minutes or more for a bus, and as a frequent Amtrak passenger in the NEC, I've never ever used the damned cafe car. I hate the very idea of the Purple Line; for what it costs, we could run free busses every five minutes till the second Brandon Trump presidency. Let's make public transit frequent on heavily traveled routes.
DCA, JFK, LAX, LGA, MSP, ATL, STC, DFW mainly for domestic (plus lots of small airports but those are fast through security for everyone). International is a different ballgame though even then priority lines make a huge difference (even at say TLV).
"... having the bus stopped every 600 (or even 800) meters rather than every 300 meters would generate higher ridership."
In San Francisco, the distance between stops is determined by the grade of the street. On the steepest streets, the stops are quite close together, as many people have trouble walking on the steepest hills. If a street is steeper than a 15% grade, then the stops are 300 to 400 feet apart. For a 10% to 15% grade, they are 500 to 600 feet apart. Only for grades under 10% can the stops be 800 to 1,000 feet apart. Other cities (Seattle?) may be similar.
1. Most urban light rail systems are spoke-and-hub and thus horribly inefficient and slow at moving people from where they are to most places they need to be. NYC is cross hatched, not spoke and hub.
2. In NYC one fare earns you transit to all destinations. Most other cities have you que up to slow kiosks to purchase tickets to specific stations (tickets which are never checked). And the name of the station can be obscure, different than that of the location, and know only to frequent users. Make the fare uniform, an average which will cover trips of all lengths. Install turnstiles and do away with tickets.
I have *never* heard of a US public transit system that made you "purchase tickets to specific stations" for their light rail system. All the ones I'm familiar with either use timed flat fares or scaled zone fares.
Denver, Seattle, you must purchase tickets specifically for your station. Of course you could purchase the most costly ticket for the final stop, and get off sooner.
Denver came up with corny station names. Try to find the correct one! And if you're wrong you start over, from the beginning.
In Denver I stood in line for one of the only 2 kiosks at Union Station after a Coors Field ballgame was over. The average time each person needed to purchase a ticket was roughly 5 minutes. Tenth in line meant waiting 50 minutes for a 20 min trip with trains running every 15. My wait for a ticket was far longer than my trip over the entire length of this route, from one terminus to the other. Pity those who were 20-30 places from the front! I never again used Denver light rail. Better to drive and pay for parking, 10 minute drive by car; no brainer to choose driving.
I can say with 100% confidence that you are completely full of it with regard to Denver as I live in Denver and regularly use light rail. The light rail system and the commuter rail line to the airport operate on a fare zone system -- under no circumstances do you "purchase tickets specifically for your station" and that would literally be impossible to do because tickets only come in "local," "regional," and "airport." It's all laid out very clearly, including with a map of the Denver rail fare zones ("A," "B," "C," and "Airport"), here: https://www.rtd-denver.com/fares-passes/fares You're also full of it to claim that there are only "2 kiosks" at Union Station unless there was some freak circumstance where the other half dozen or so kiosks were offline (there are two at the *end* of the track for the A line, there are several others along the sides of the track for the A line, I know this very well because I literally bought a ticket from one of the machines on the side just this past August).
I lived there 2011-18. This was my experience. I used the line from Golden to Union Station. Maybe it's different now. I am reporting facts, personally experienced, not BS.
I am curious about whether transit agencies even try to figure out what would make people more likely to use transit. Specifically people who have stopped or reduced their use of transit. But it’s weird how many transit agencies just don’t see their riders as ther primary stakeholdera
"Why do non-NYC American cities have such a hard time attracting mass transit ridership?"
[Ctrl+F "density"]
[0 hits]
From the book "Green Metropolis" by David Owen: "[t]he most significant factor in determining the viability of any transit system... is population density." A 1980 study found that the "threshold density" for getting good public transit is seven dwellings per acre. One of the authors of that study, Jeffrey Zupan, said: "People often said they want public transit but then aren't willing to zone their area to achieve the density needed to support it."
This whole column has lots of interesting information but it also feels beside the point. If you don't have the density to support more than one sad little bus that comes by every 45 minutes on a weekday and not at all on weekends, then *of course* you aren't going to have high ridership!
Anecdote time: I live in the sprawling-ass suburb of Orange County, California, which makes me sad, because I miss my former city, Boston (I moved to OC for a very good job). Sometimes, for kicks and giggles, I look at Google Maps and compare how long it would take me to get to a local destination by car vs. public transit. A typical result:
Car = 20 min, mostly spent going at 65 MPH down a freeway.
Transit = 1 h 30 min, including a 20 min bus ride, a 5 min walk to another bus stop, waiting 25 min for the next bus, followed by another 40 min bus ride.
Then I let a single tear roll down my cheek as I remember Boston, and I drive to my destination.
I once tried starting a project to see how long it would take to get to each NFL stadium from the airport by mass transit. I abandoned it right at the beginning when I got a big fat "LOL nope" from Google Maps on going from DFW to JerryWorld.
Why is that very good job located in OC and not Boston? Is there something about a “sprawling ass suburb” that is more attractive to employers? And what’s wrong with being able to hop in your car and be somewhere in 20 minutes?
i completely agree with this but it's funny that you keep focusing on Ohio b/c I can't really drive anymore and was in Athens -> Columbus -> Cleveland over New Year's and oh boy was the bus ride from Columbus to Cleveland extremely unpleasant and weirdly expensive compared to East Coast, European or even Chicagoland/southern Wisconsin travel.
I spent a solid 20 mins on the bus ride theorising about why this was the case (Greyhound is the worst, almost everyone in Ohio drives except for students and very low-income folks and the students were away because of New Year's, etc.) and while I completely agree that putting a train from Athens to Columbus to Cleveland is so not the best use of government money, I was actively thinking on the bus ride just how much I would've liked one then.
I agree with the narrow argument "ridership should be the key performance indicator for public transit spend". However, if you take this as the only KPI you quickly run into the obvious problem with US public transportation spend - outside of a few areas in the Northeast, there are very few areas where public transportation spending makes sense on a dollar per passenger basis.
I'm in Los Angeles, so I can see this dynamic pretty clearly both in local transit spend and with the CA HSR. The new Purple Line extension (running down Wilshire, our busiest transit corridor) is estimated to cost $435 in upfront capital costs per rider in the first year. At $1.50 a ticket, you're estimating 300 years before you make back your upfront costs, without even taking into account NPV of the investment.
So if you're a booster for public transportation, that's out - you don't want to look at that, you don't want to talk about that. But you DO want public transportation built. What do you do instead? You start adding more variables. Yes, maybe this $4.5B project will only have 10 million riders the first year. Ah, but each rider takes a car off the road - isn't that worth something? Or maybe ultrapoor riders can get to better jobs farther away - that's got to be worth something too! And maybe those jobs are working on the subway itself, that's an advantage!
The upshot is that because your primary KPI doesn't make sense, you want to intentionally obfuscate that by promoting other indicators that will always look good. Of course CA HSR wants to call out small business spend - we're spending $1.7M PER DAY on the project, so shoveling some of that to small businesses creates a "good" number. Or if your new transit connector takes 3000 cars off the road a day (for the low low price of $1.8B), you get to stand in front of the metro community outreach meeting and say that this is traffic by ONE MILLION TRIPS annual. That's a pretty good number too! And because they're second-order and arbitrary, no one's going to argue against them, or say it's not worth $1800 to remove one car from the road, or ask why we don't just give the HSR money directly to small businesses without the hassle of a massive, time-consuming infrastructure project.
This is a bit of a straw man. American transit agencies absolutely prioritize ridership. The problem is that there isn't operationally much a transit agency on its own can do to increase ridership. They can't put a tax on cars. They can't remove road lanes or parking. They can't build TOD. They can't run more frequently unless they have more money for operators.
Jarrett Walker likes to pretend that transit agencies don't know about the frequency/coverage issue because, like many consultants, he makes money by telling management that their workers are idiots and the only way to improve things is by hiring him.
As a tax and spend liberal I almost always vote for bond measures on the California ballot -- I think we need to improve our infrastructure and I like the idea of creating new jobs to do that work, but I did not vote for the HSR bond authorization because I read a fair amount about the projected ridership and it was clear that it would never be able to be self-sustaining and that it just was not a good fit for the geography and population of California. My basic understanding of the economics of HSR is that you need to be able to stop the trains every 100 kilometers or so and have a fair number of people get on and off -- that you need these intermediate distance passengers as well as folks travelling the distance that would previously have taken a flight. But in California, once a train departs San Jose, there is nowhere with a large population until you get to LA and yet the cost of running the rail over the mountains is huge. I concluded that while it would in fact be cool to have a HSR option for getting from SF to LA, that it was never going to pencil out, and there would be a taxpayer revolt at some point when the subsidy cost became apparent. But most people I know voted for it because they liked the idea of doing the right thing environmentally and being more European and thinking big and being bold. I feel sad that all of the money that's been spent was not instead directed at improving transit in the many California cities with horrible commutes and traffic and the need for better transit, although with the new work from home world that we are now in that might have been a waste as well. I really hate driving for more than 30 minutes, and when Amtrak is running on time it's a slow but pleasant way to commute from Sacramento to the Bay Area-- I would love to see those trains run at Acela speed but I am not sure that the ridership would justify it (there is a cadre of health care workers who commute from Sac to Oakland because of cost of living but probably not enough to justify the upgrade).
I have a hard time seeing why, especially in a blue state, a transit agency would follow thru on prioritizing riders per cost. Wouldn’t that just get thrown in the many factors? How would that factor have any teeth? Many people here feel righteous and good about paying high taxes (or so they say). And, I don’t know if I’m the only one who’s noticed this, transit reformers seem loathe to ally with the kind of people who would lead a tax revolt. I’m sure they have their reasons but until the connection happens *shrug*
Wanted to hear more about the tradeoffs (or lack thereof) between ridership and access. I think the most relevant argument to defeat here is that focus on ridership will come at the expense of marginalized communities. I can absolutely see accusations of building transit only for rich gentrifier neighborhoods, and since the political environments for transit are going to skew heavily left, these arguments are the ones that matter.
The way I would counter that argument is that people in marginalized communities are either riding the bus or they aren't. If they are, then a ridership-maximizing system will serve them. If they aren't, then it doesn't really matter whether they are served or not.
Sometimes, I think people have a tendency to way overthink this.
Dumb question - who could pass such a rule? Can the federal government and if so which agency? Or a Biden executive order? Or just every local legislature has to enact this one by one? I agree with the idea but what’s the theory of change?
I am with Matt and the people at the Transit Cost Project when it comes to their critique of new transit infrastructure. I’m not completely following Matt’s examples of suboptimal use of existing infrastructure.
Matt seems to confusingly switch between the goals of ridership maximization and revenue maximization, particularly in his Amtrak example. Lowering the price of tickets on the Acela Train should increase ridership by some amount but might increase or decrease revenue depending on whether the price effect or quantity effect dominates. So, it’s important to clarify whether we’re truly talking about ridership maximization vs. revenue maximization.
It could be that, through a mix of inertia and lack of - or expense associated with - viable alternatives, price effect dominates because those riders who rely on Amtrak for their commute would pay a much higher price before switching to another commuting option. Similarly, based on how current infrastructure is situated, it may be the case that lowering the price hardly does anything to increase ridership and generate additional revenue. In which case Amtrak doesn’t have a financial incentive to maximize ridership for existing infrastructure, and a low ridership outcome is dictated more by incentives than lack of critical analysis.
Right as usual. I love RBT out here in Silver Spring because now we don't have to wait 30 minutes or more for a bus, and as a frequent Amtrak passenger in the NEC, I've never ever used the damned cafe car. I hate the very idea of the Purple Line; for what it costs, we could run free busses every five minutes till the second Brandon Trump presidency. Let's make public transit frequent on heavily traveled routes.
Clear is huge at peak times.
DCA, JFK, LAX, LGA, MSP, ATL, STC, DFW mainly for domestic (plus lots of small airports but those are fast through security for everyone). International is a different ballgame though even then priority lines make a huge difference (even at say TLV).
"... having the bus stopped every 600 (or even 800) meters rather than every 300 meters would generate higher ridership."
In San Francisco, the distance between stops is determined by the grade of the street. On the steepest streets, the stops are quite close together, as many people have trouble walking on the steepest hills. If a street is steeper than a 15% grade, then the stops are 300 to 400 feet apart. For a 10% to 15% grade, they are 500 to 600 feet apart. Only for grades under 10% can the stops be 800 to 1,000 feet apart. Other cities (Seattle?) may be similar.
Not sure if anyone else has posted this piece from the Philadelphia Inquirer. It directly describes the problem Matt addresses
https://www.inquirer.com/transportation/septa-bus-revolution-route-changes-complaints-20230105.html?utm_medium=referral&utm_source=ios&utm_campaign=app_ios_article_share&utm_content=WDGEH7PKJVALLMWAT27ZRTEVUU
Two very simple facts:
1. Most urban light rail systems are spoke-and-hub and thus horribly inefficient and slow at moving people from where they are to most places they need to be. NYC is cross hatched, not spoke and hub.
2. In NYC one fare earns you transit to all destinations. Most other cities have you que up to slow kiosks to purchase tickets to specific stations (tickets which are never checked). And the name of the station can be obscure, different than that of the location, and know only to frequent users. Make the fare uniform, an average which will cover trips of all lengths. Install turnstiles and do away with tickets.
I have *never* heard of a US public transit system that made you "purchase tickets to specific stations" for their light rail system. All the ones I'm familiar with either use timed flat fares or scaled zone fares.
Denver, Seattle, you must purchase tickets specifically for your station. Of course you could purchase the most costly ticket for the final stop, and get off sooner.
Denver came up with corny station names. Try to find the correct one! And if you're wrong you start over, from the beginning.
In Denver I stood in line for one of the only 2 kiosks at Union Station after a Coors Field ballgame was over. The average time each person needed to purchase a ticket was roughly 5 minutes. Tenth in line meant waiting 50 minutes for a 20 min trip with trains running every 15. My wait for a ticket was far longer than my trip over the entire length of this route, from one terminus to the other. Pity those who were 20-30 places from the front! I never again used Denver light rail. Better to drive and pay for parking, 10 minute drive by car; no brainer to choose driving.
I can say with 100% confidence that you are completely full of it with regard to Denver as I live in Denver and regularly use light rail. The light rail system and the commuter rail line to the airport operate on a fare zone system -- under no circumstances do you "purchase tickets specifically for your station" and that would literally be impossible to do because tickets only come in "local," "regional," and "airport." It's all laid out very clearly, including with a map of the Denver rail fare zones ("A," "B," "C," and "Airport"), here: https://www.rtd-denver.com/fares-passes/fares You're also full of it to claim that there are only "2 kiosks" at Union Station unless there was some freak circumstance where the other half dozen or so kiosks were offline (there are two at the *end* of the track for the A line, there are several others along the sides of the track for the A line, I know this very well because I literally bought a ticket from one of the machines on the side just this past August).
I lived there 2011-18. This was my experience. I used the line from Golden to Union Station. Maybe it's different now. I am reporting facts, personally experienced, not BS.
Maybe the fare base is zoned, but the tickets needed to specify the destination.
I am curious about whether transit agencies even try to figure out what would make people more likely to use transit. Specifically people who have stopped or reduced their use of transit. But it’s weird how many transit agencies just don’t see their riders as ther primary stakeholdera
"Why do non-NYC American cities have such a hard time attracting mass transit ridership?"
[Ctrl+F "density"]
[0 hits]
From the book "Green Metropolis" by David Owen: "[t]he most significant factor in determining the viability of any transit system... is population density." A 1980 study found that the "threshold density" for getting good public transit is seven dwellings per acre. One of the authors of that study, Jeffrey Zupan, said: "People often said they want public transit but then aren't willing to zone their area to achieve the density needed to support it."
This whole column has lots of interesting information but it also feels beside the point. If you don't have the density to support more than one sad little bus that comes by every 45 minutes on a weekday and not at all on weekends, then *of course* you aren't going to have high ridership!
Anecdote time: I live in the sprawling-ass suburb of Orange County, California, which makes me sad, because I miss my former city, Boston (I moved to OC for a very good job). Sometimes, for kicks and giggles, I look at Google Maps and compare how long it would take me to get to a local destination by car vs. public transit. A typical result:
Car = 20 min, mostly spent going at 65 MPH down a freeway.
Transit = 1 h 30 min, including a 20 min bus ride, a 5 min walk to another bus stop, waiting 25 min for the next bus, followed by another 40 min bus ride.
Then I let a single tear roll down my cheek as I remember Boston, and I drive to my destination.
I once tried starting a project to see how long it would take to get to each NFL stadium from the airport by mass transit. I abandoned it right at the beginning when I got a big fat "LOL nope" from Google Maps on going from DFW to JerryWorld.
Why is that very good job located in OC and not Boston? Is there something about a “sprawling ass suburb” that is more attractive to employers? And what’s wrong with being able to hop in your car and be somewhere in 20 minutes?
>>Why is that very good job located in OC and not Boston?<<
Any number of reasons. You can find lots of good jobs in Orange County. And lots of good jobs in Boston, MA. Both locations are affluent.
i completely agree with this but it's funny that you keep focusing on Ohio b/c I can't really drive anymore and was in Athens -> Columbus -> Cleveland over New Year's and oh boy was the bus ride from Columbus to Cleveland extremely unpleasant and weirdly expensive compared to East Coast, European or even Chicagoland/southern Wisconsin travel.
I spent a solid 20 mins on the bus ride theorising about why this was the case (Greyhound is the worst, almost everyone in Ohio drives except for students and very low-income folks and the students were away because of New Year's, etc.) and while I completely agree that putting a train from Athens to Columbus to Cleveland is so not the best use of government money, I was actively thinking on the bus ride just how much I would've liked one then.
I agree with the narrow argument "ridership should be the key performance indicator for public transit spend". However, if you take this as the only KPI you quickly run into the obvious problem with US public transportation spend - outside of a few areas in the Northeast, there are very few areas where public transportation spending makes sense on a dollar per passenger basis.
I'm in Los Angeles, so I can see this dynamic pretty clearly both in local transit spend and with the CA HSR. The new Purple Line extension (running down Wilshire, our busiest transit corridor) is estimated to cost $435 in upfront capital costs per rider in the first year. At $1.50 a ticket, you're estimating 300 years before you make back your upfront costs, without even taking into account NPV of the investment.
So if you're a booster for public transportation, that's out - you don't want to look at that, you don't want to talk about that. But you DO want public transportation built. What do you do instead? You start adding more variables. Yes, maybe this $4.5B project will only have 10 million riders the first year. Ah, but each rider takes a car off the road - isn't that worth something? Or maybe ultrapoor riders can get to better jobs farther away - that's got to be worth something too! And maybe those jobs are working on the subway itself, that's an advantage!
The upshot is that because your primary KPI doesn't make sense, you want to intentionally obfuscate that by promoting other indicators that will always look good. Of course CA HSR wants to call out small business spend - we're spending $1.7M PER DAY on the project, so shoveling some of that to small businesses creates a "good" number. Or if your new transit connector takes 3000 cars off the road a day (for the low low price of $1.8B), you get to stand in front of the metro community outreach meeting and say that this is traffic by ONE MILLION TRIPS annual. That's a pretty good number too! And because they're second-order and arbitrary, no one's going to argue against them, or say it's not worth $1800 to remove one car from the road, or ask why we don't just give the HSR money directly to small businesses without the hassle of a massive, time-consuming infrastructure project.
This is a bit of a straw man. American transit agencies absolutely prioritize ridership. The problem is that there isn't operationally much a transit agency on its own can do to increase ridership. They can't put a tax on cars. They can't remove road lanes or parking. They can't build TOD. They can't run more frequently unless they have more money for operators.
Jarrett Walker likes to pretend that transit agencies don't know about the frequency/coverage issue because, like many consultants, he makes money by telling management that their workers are idiots and the only way to improve things is by hiring him.
As a tax and spend liberal I almost always vote for bond measures on the California ballot -- I think we need to improve our infrastructure and I like the idea of creating new jobs to do that work, but I did not vote for the HSR bond authorization because I read a fair amount about the projected ridership and it was clear that it would never be able to be self-sustaining and that it just was not a good fit for the geography and population of California. My basic understanding of the economics of HSR is that you need to be able to stop the trains every 100 kilometers or so and have a fair number of people get on and off -- that you need these intermediate distance passengers as well as folks travelling the distance that would previously have taken a flight. But in California, once a train departs San Jose, there is nowhere with a large population until you get to LA and yet the cost of running the rail over the mountains is huge. I concluded that while it would in fact be cool to have a HSR option for getting from SF to LA, that it was never going to pencil out, and there would be a taxpayer revolt at some point when the subsidy cost became apparent. But most people I know voted for it because they liked the idea of doing the right thing environmentally and being more European and thinking big and being bold. I feel sad that all of the money that's been spent was not instead directed at improving transit in the many California cities with horrible commutes and traffic and the need for better transit, although with the new work from home world that we are now in that might have been a waste as well. I really hate driving for more than 30 minutes, and when Amtrak is running on time it's a slow but pleasant way to commute from Sacramento to the Bay Area-- I would love to see those trains run at Acela speed but I am not sure that the ridership would justify it (there is a cadre of health care workers who commute from Sac to Oakland because of cost of living but probably not enough to justify the upgrade).
I have a hard time seeing why, especially in a blue state, a transit agency would follow thru on prioritizing riders per cost. Wouldn’t that just get thrown in the many factors? How would that factor have any teeth? Many people here feel righteous and good about paying high taxes (or so they say). And, I don’t know if I’m the only one who’s noticed this, transit reformers seem loathe to ally with the kind of people who would lead a tax revolt. I’m sure they have their reasons but until the connection happens *shrug*
Wanted to hear more about the tradeoffs (or lack thereof) between ridership and access. I think the most relevant argument to defeat here is that focus on ridership will come at the expense of marginalized communities. I can absolutely see accusations of building transit only for rich gentrifier neighborhoods, and since the political environments for transit are going to skew heavily left, these arguments are the ones that matter.
The way I would counter that argument is that people in marginalized communities are either riding the bus or they aren't. If they are, then a ridership-maximizing system will serve them. If they aren't, then it doesn't really matter whether they are served or not.
Sometimes, I think people have a tendency to way overthink this.