Hacker News new | ask | show | jobs
How U.S. Cities’ Public Transit Stacks up (fivethirtyeight.com)
76 points by jgunaratne 4347 days ago
9 comments

What I don't like about this chart it it doesn't show how bad the USA's public transportation is relative to places like say Tokyo, Singapore, Amsterdam, Stockholm, Berlin, Helsinki, etc. They're night and day better than NYC which itself is night and day better than SF.
I think the subway in NYC is pretty great, though by no means perfect.

Amsterdam doesn't have great public transportation. It would drive me insane when I lived there (Indische Buurt / Science Park). Their system of public transportation relies a lot on connecting between bus/tram/subway/railway, but I never found the connections to work well. The frequency of most lines is also pretty darn bad.

Unfortunately a lot of people living in Amsterdam are fine with the sorry state of their public transportation because they feel the Amsterdam mindset is to ride a bike. (There's a lot of truth to that, of course, but the reality is that a lot of non-white Amsterdam-born people rely on public transportation.)

Stockholm's public transportation seemed fine last time I was there, but it's definitely not better than NYC. When I visited a university there, I had to take a subway + bus. Buses suck! Nevertheless, the connection was not bad overall.

The great thing about the subway in NYC is the how the stations are spread out on the map. There are a lot of stations and you will find them in sensible locations. The fact that the subway is operational 24/7 is also amazing. That's a lot more unusual than some people give credit for. You don't even find that in Singapore or Hong Kong or London or Paris. What's less than great is the reliability and the frequency of the trains to/from Brooklyn.

Going from one point in Brooklyn to another also sometimes requires going through Manhattan, which makes the trip take too long.

Another sore point is communication. The signs are confusing and there are virtually no CRTs/LCDs to guide you (if Paris and Brussels can handle the vandalism, so should NYC). The verbal announcements are often confusing or wrong (multiple times, late at night, I heard 'train X was canceled' just as it arrived). A lot of the time you can't even properly hear them. The train models are also incredibly outdated.

Unfortunately, in NY, buses are no good alternative because they are always late and absurdly slow. I always suspected they were intended foremost for old and disabled people since they accelerate very slowly after a stop and seem to stop every block.

I have never been to Helsinki, Tokyo, or Berlin.

One subway system I want to mention is the Beijing one. It's by no means perfect, but the frequency of the subway is just amazing. At many–perhaps most–stations a train stops every one or two minutes. My patience for waiting for trains never recovered after living there.

It's tricky comparing public transport in your city to public transport in cities you visit as a tourist. Tourists don't travel during peak hours. They aren't as bothered (so they don't notice) a 10 minutes delay because they're not in a rush. They don't frequent residential areas so they may not experience the trickier journeys. A lot of cities are great from getting from suburbs to residential areas, but getting from one residential area to another is hard.
Um how did you compare that? Amsterdam public transports sucks, so I question your other cities also.

I have been to Paris/London (multiple times) and their system sucks comparing to NYC. In NY I can get easily from point A to point B anywhere in the city aside from SI.

NYC has 24hours subway service, it's hard to beat that.

Exactly, if we're "stacking up" why not against international cities?
Other countries have legal, political, cultural, and social environments that differ significantly from the US. Finding out that something works better in London or Paris than it does in Seattle is interesting as trivia, but because of those environment difference usually doesn't tell much about how to improve that thing in Seattle.

Finding out that San Francisco does something better than Seattle does teach something about improving it in Seattle, because the legal, political, cultural, and social environments are much more similar.

Hence, domestic comparisons are more interesting, as they suggest more achievable improvements.

Indeed, no knowledge can be gained from the outside world. Those parts of the world run on magic and stinky cheese.
>Other countries have legal, political, cultural, and social environments that differ significantly

The same is true within the US. It is neither legally, politically, culturally, nor socially homogenous. Learning how to improve public transportation in Seattle would be more productive if you studied Amsterdam than if you studied Memphis or Baton Rouge.

However comparing internationally shows how far behind the US is on this front, and how much they need to improve.

No point giving pats on the back all round when nobody is even attempting to change the "legal, cultural and social" environment.

2nd paragraph: the data came from National Transit Database (NTD) and the American Community Survey (ACS). These data sets are USA-only.
Surely other countries have comparable datasets?
What are you talking about? NYC public transportation system is one of the best in the world and its 24 hours.
We had public transportation in Central Ohio, and then the automotive industry lobbied hard to have it dismantled. Ohio, Michigan, and the rest of the midwest were home to the automotive industry.

Now decades later, with the automotive lobby not being quite so strong, the asphalt industry has their guy as the head of the transportation department. Non highway projects of any sort are not just discouraged but actively squashed and ostracized.

I'm still astounded they are actually building the new US Bike Route 50 through Central Ohio. It must be forces outside our state government making that happen. Maybe it is a good contract for the asphalt contractors too. I would not be surprised, in spite of that, if our governor's administration figures out a way to stop it.

I have heard many, many times around here: "This is America. I have a RIGHT to drive the biggest car I can afford and I expect to be able to do so whenever and wherever I want. You are not going to force me to ride a bus or subway or train or bike, nor force any of my tax dollars to go to such un-American things."

(Apparently if you have any sort of public transportation system you are forcing people to use it.)

The motion that cars are a private, free market form of transport while trains are public, left wing sort of transportation is interesting.

Roads are publicly funded and owned just like tracks are. The cars themselves are privately owned, so I guess that's market points for cars. But, trains involve a direct transaction that (partially) funds the facility in a user-pays way. Roads are funded by taxes. They try to use registration and petrol taxes that are related to use of roads but it's still a tax, not a normal transaction. If you drive on a private road, you still pay petrol tax and the taxes can't be tied to use of a specific road. I'd say that's market points to the trains.

Getting to the more fuzzy trains have a social egality to them. Everyone sits in the same cars looking at each other, sharing space. You end up squished against people of races, creeds & classes that you mightn't associate with normally. They're seen as environmental which is left wing. They're inclusive of the marginal people who can't afford to drive, are to young or otherwise unable.

Cars have a personal liberty aspect to them. Your car. Your space. Your rules. Go where you want, when you want. Open roads. Wind in your hair. They don't have a schedule set by someplace else or stops decided on by some comitee. Cars are status symbols and and opportunity to show wealth. Car ownership is something to aspire to. They're symbols of the great capitalist/industrial age.

I have interesting associations with the late 19th & early 20th centuries, the formative years of trains and cars. I associate the industry of the late 19th more with socialist symbols. The plight of the working man, Marxism (pre-Lenin and the east-west associations), early labour movements, decaying empires. Coal soot. Europe. I associate the early 20th industry with American ascension. Iconic technicolour images of manufacturing wealth pampering American housewives with vacuum cleaners and weekly trips to a beauty parlors. All the values of that time and place. A certain type of clean shaven naivety.

I can't quite put a thorough argument together, but I think the symbolism is interesting.

But, trains involve a direct transaction that (partially) funds the facility in a user-pays way. Roads are funded by taxes. They try to use registration and petrol taxes that are related to use of roads but it's still a tax, not a normal transaction. If you drive on a private road, you still pay petrol tax and the taxes can't be tied to use of a specific road. I'd say that's market points to the trains.

Most U.S. public mass transit could not maintain itself based on "user pays", even forgetting about capital expenditures. Unsubsidized, fares would rise, ridership would fall, and prices would then need to be still higher to compensate.

On the other hand, the gas tax and other user fees (trucking, etc.) already fund the majority of U.S. highway spending both maintenance and capital (even with 1/6 of it redirected to mass transit!), and could easily fund all of it if the political will were there. Some people would drive less if we charged 2x the gas tax or had an odometer tax, but nobody thinks the system itself would be unsustainable in the same way that most public transit would be if it had to be funded entirely by its users.

For every mile of interstate in the US, there are over 100 miles of local roads that are poorly covered by the gas tax. Paying for a majority of the highway system shouldnt be much harder than graduating from kindergarten. Its pretty pathetic that the entire highway system can't be funded by the gas tax, given how little it contributes to car oriented infrstructure. 2x gas taxes still wouldn't come close to covering the costs of the infrastructure provisioned. Gas taxes in the UK are 8x what we pay, and even that doesnt cover the full cost of their infrastructure. Transit, in comparison, is magnificently cost efficient...only by deceptive means like yours can you appear to make the opposite case.

Regardless, the effect of even a tiny subsidy for cars affects the cost recovery of transit. By subsidizing cars, you actually force larger subsidies for transit. The operating and maintenance costs of car use scale linearly with vehicle miles travelled, whereas the costs of transit scale as a step function (you don't need a new bus for every user, you need a new bus for every 60 users...riders 2 through 60 ride for free) on the scale of an individual transit vehicle, and roughly log linear in aggregate. Therefore, every transit user you subsidize into a car increases the per user cost of all the rest of the transit users. If car users paid even a tiny fraction more than the pittance they currently do, it could push enough users onto transit to make it sustainable. And if they acctually paid their true costs, almost nobody would drive at all.

"Nationwide in 2011, highway user fees and user taxes made up just 50.4 percent of state and local expenses on roads. State and local governments spent $153.0 billion on highway, road, and street expenses but raised only $77.1 billion in user fees and user taxes ($12.7 billion in tolls and user fees, $41.2 billion in fuel taxes, and $23.2 billion in vehicle license taxes).[3] The rest was funded by $30 billion in general state and local revenues and $46 billion in federal aid (approximately $28 billion derived from the federal gasoline tax and $18 billion from general federal revenues or deficit financed)."

That a direct quote from here: http://taxfoundation.org/article/gasoline-taxes-and-user-fee...

Which cites its source as the, "U.S. Census Bureau, State and Local Government Finance 2011."

I was careful to say "highway system", by which I meant the IHS. Yes, local governments fund local roads out of property and other local taxes, but a) Everyone depends on their own local roads, even if they never drive, walk or bike on them, not so mass transit b) Local gas taxes are hard to make work because it's possible and tempting for gas stations to locate in the lowest taxed locality.

Even so, your own link says 50% of state and local roads are funded with user fees. The IHS number is around 70%, even with 1/6 of the gas tax redirected to public transit. If you doubled the user fees and eliminated all other sources of funding, the IHS and the state/local road system would still thrive. Not so most public transit.

You convince car people to support transit with arguments like: Adding another lane to the freeway encourages people to take more cars. It actually doesn't reduce congestion on the freeway much; the primary effect of adding lanes to the freeway is to increase congestion on secondary roads. The best way to reduce congestion so that you can drive to work faster is transit. Every person riding a bus is somebody not driving a car, clogging up your lane.
Similar thing happened in the south. We used to have reasonably good, popular and well used bus services in most of or cities. But then the civil rights movement happened and state and local governments cut funding and dismantled bus systems lest they have to desegregate.
Seemed like he stopped at a visualization of the data, without any real interpretation.

I'd like to see some information on the density of the areas. For example, Chicago is listed as the metro region (8 million people) when the city itself only has about 3 million, and public transport hardly exists outside the city.

If the total rides had been divided by the city population, it would equal much closer to 200 rides per capita.

Good point - I found this a confusing analysis without digging much into what it was saying - Urbana and Athens were high on the list...why? Do they have public bus systems that students use extensively? If so, that would tend to give a false impression of how much public transit is used there. I liked that he later pulled apart the small and large cities, but it would have been nice to understand what the data was saying in the context of the cities it examined.

Your point on Chicago is well taken - I was surprised to see SF/Oak ahead of it, as they "feel" harder to navigate with transit only, but I was forgetting that the suburbs are included in Chicago, which is basically impossible once you leave the city limits.

Yes, college towns often have great public transit usage because students ride for free, the school pays the local agency or runs their own vehicles, and you have a lot of young people without cars.

Ann Arbor, MI (#20) is a great example where University of Michigan runs a fleet of buses that puts many small cities' systems to shame. Lafayette, IN (#30) is partially supported by serving Purdue University. Bloomington, IN (#40) has two bus systems - the city's, and Indiana University's. (Disclaimer: these are all customers of DoubleMap.)

Not sure why you think it would be a false impression - public transit is used a lot in college towns.

State College, PA is where Penn State University is located. Even if you drive to the university as a student you still need to take a bus to get to class. That makes around 45,000 students using public transportation everyday.
Yeah, I was very surprised to see Athens, GA in the top 5. I was not even aware that they had a substantial public transit system and I lived in GA for ~15 years.
You are probably mentally discounting bus infrastructure. Busses are much more prevalent than people think but their demographic skews much lower on the economic scale than other transit options (often because they are abysmally slow).
Well, it's not clear if the ride count includes metra. If it does, then you need to include the entire metro area. If it's CTA only, then it should be just the city.
CTA does actually service parts of the metro area outside of the city such as Evanston & Oak Park.
In general, I do wish more people would explicitly say whether they're listing city population or metro population. In this case the numbers are all clearly metro area. It would be interesting to see how the rankings differed for just the cities. I imagine they'd be very similar, but a few cities would move around.
In many cases I think you'd want a third number, something like the core urbanized area. Metro area is too big, but in some cases city limits are either too big or too small for historical/political reasons. For example Houston includes many of its suburbs, because Texas law is very annexation-friendly. Whereas parts of the urban core of Boston are legally in Brookline and Cambridge, and part of the legal core of Copenhagen is legally in Frederiksberg, but for transit purposes those distinctions aren't that important (Brookline/Cambridge are still part of the "T", and Frederiksberg has a line of the Copenhagen Metro).
Public transportation outside the city of Chicago (PACE + METRA + northern and western L extensions) is part of a unified system with Chicago, and is huge, heavily used, and critical (as opposed to hardly existing.)
However, most people out there own cars. Metra is very infrequent outside of commuting hours, and seems to be mostly used as a way for car owners who live in the suburbs to avoid freeway congestion and parking costs downtown, rather than a complete car replacement.

In urban neighborhoods on the red and blue lines, for example, car ownership is basically superfluous.

There is a lot more interesting data in the NTD that I wish someone with more programming skills than me could put together in a graphical way. For instance, you can calculate the amount of CO2 emitted per passenger and per passenger mile. You can calculate the subsidy per rider and per passenger mile. You can even determine the average utilization of the system by checking the average number of riders on each bus.

From doing some of that work manually, I found that many (if not most) public transit systems are emitting more CO2 than if all of the passengers had ridden in a car, with the national average of 1.7 riders per car.

I live in the 50th largest city in the US and we have zero public transportation. Yet we are the sports and entertainment center of the DFW metroplex including the Dallas Cowboys spaceship stadium.
I am surprised that Chicago is not stacked higher. I was under impression we here have one of the better transit systems, compared to others in the US.
it's quite good within the city limits, but it drops off very quickly when the suburbs are included. These numbers included the suburbs.
That being the case, you'd assume that SFBA would plummet down the list, since public transportation on the peninsula is atrocious.
If you remove buses from the equation, I believe the US transit system would like considerably worse.
DC's subway system is awesome, for a city of its size.
The only thing I learned from the article is that NY has a really popular subway. Duh.

My question is: Why is SF's public transit so popular? It can't be because of the quality. LA is a much bigger city, why isn't public transit more popular there?

Geography's a factor. It's not just SF, it's Oakland too, and there's a ton of commuting between them on BART on a daily basis due to the cost of living difference. They're separated by water so with BART (an underwater subway) you avoid bridge traffic. Also parking in SF and real estate in general is at a premium due to it being on the head of a peninsula.

SF alone might rank lower. Muni is certainly not "popular" with the population.. it's much less reliable than BART. My terribly anecdotal impression is a lot of people bike or take cabs within SF if at all possible.

San Francisco numbers are misleading. They likely counted both BART and Caltrain. Since Oakland is included, Amtrak trains (Capitol Corridor, ACE, San Joaquin) might also be included. All the systems I mentioned are heavy rail and carry hundreds of thousands of passengers daily.

Muni has an always packed light rail subway system as well. Unfortunately, unlike San Jose's VTA, that subway has no bike carrying capacity. This makes it difficult to ride the bike to a light rail station and then use it again to reach your final destination in downtown or elsewhere in the city.

Regional heavy rail systems like BART have enormous ridership.

San Francisco buses need dedicated lanes in residential areas. Based on my personal experience, the root cause of traffic slowdowns appears to be double-parked cars. Buses run quickly at night when they are not slowed down by cars.

http://sf.streetsblog.org/2013/10/02/the-sorry-state-of-doub...

But this is changing. Dedicated lanes are being put in place. http://sf.streetsblog.org/wp-content/uploads/2014/07/Picture...

Maybe because SF is denser?
Yes, and part of that is the area used. The SF-Oakland area used for aggregation contains 3.3M people. The LA area contains 12.3M.