Quick Note: Cars and Suburbs Commoditize Location
Trains and big cities are more efficient on a typical cost basis than cars. The operating costs of trains are such that even the unsubsidized costs of big-city metros and bus networks are a fraction of those of cars. For example, New York City Transit, despite its high operating costs, manages to serve a linked trip for around $6, which works out to an annual cost per user of around $3,500, and if it had the cost structure of London or Berlin this would be $2,000 or a bit less than that; American cars average $7,000/car per year in private spending.
And yet, cars have one singular advantage: they commoditize location. Public transit ideally works in large cities at specific locations, based on historic contingencies like national capitals, religious significance, or river crossings and harbors that may no longer be relevant with modern technology. It’s decommoditized, in that there is only one New York, one Philadelphia, one Chicago, etc., and the cost of moving is high. Public transit itself doesn’t lend itself to competition, because it requires extensive scale to ensure connectivity and high frequency. This is why public provision is almost universal, and the exceptions either involve a high degree of public coordination such as the Verkehrsverbünde in the German-speaking world even if elements are contracted out or are Japanese cities with such large systems that competition between a JR and a private operator still leaves each competitor with much scale; even generally privatization-happy states like Singapore keep the systems broadly public in planning.
What this means is that cities and public transit require a public sector that can keep up without the discipline of market competition. This means public-sector innovation, with competition taking place in the political sphere as in European cities or in the technocratic one as in Singapore. If this doesn’t happen, then the system suffers. If, for example, the New York MTA folds to a strike by the LIRR train drivers in which the union demands are so unreasonable that even the left-wing city mayor Zohran Mamdani doesn’t side with the union, and gives the drivers large increases in pay while still allowing them to collect double pay for driving both a diesel and an electric train, then there’s no easy way to move to a competitor.
Cars and suburbs instead commoditize location. If the city can’t provide adequate public services, people can just leave. It’s particularly easy if the municipality that falters in providing services is not a large city but a small suburb of one, as in the boroughitis of New Jersey. Cars facilitate that, in that they scale down better. There’s no way to squeeze anything the size of Midtown Manhattan or even the center of Paris into one auto-oriented place (Los Angeles has a weak central business district), but that’s fine, a region can take the hit on income and still function with worse scale; Dallas is not a poor region. There are real problems in this setup with higher transportation costs and with job centers with worse scale, but sometimes it’s worth it to take the hit if it means not having to deal with unaccountable government that one can’t leave. If there’s no mechanism to improve governance – say, if there is such democratic deficit at the local level that it’s not possible for voters to coerce the city into improving education or public transit or housing or any other devolved issue – then that usually equally affects city and suburbs, but one can move from one suburb to another at relatively low economic and social cost, and this has a disciplining effect to some extent.
Pedestrian Observations 
Excellent and insightful. I linked to your “Informed Voting and the Democratic Deficit” and found an idea that has struck me in recent weeks that American states should have state parties like in Canada to break up one party states, like Texas and California. Can’t run a economically liberal Republican in California or socially conservative Democrat in Texas because they won’t make it out of the partisan primary? Then form a state party that matches up with positions, policies, and ideology that a large number of citizens will vote for in a state election. Democracy without competitive elections is not democracy. It also might encourage more cooperation between a president and governor/mayor of different parties, like FDR and LaGuardia.
As a New York State voter we have our “fusion voting” on the ballet so I’ve voted under third parties for candidates, I have a friend who in the 1990s was chairman of the Conservative Party, he remember a meeting where they listen to a real estate developer deliver a pitch on why they should have him run for governor under their banner. They politely listen, wished him well, and after he left laughed out loud: “We’ll never put Donald Trump on the ballet!”.
On municipal consolidation, I agree bigger cities are better. In the last town and county election I only voted for the Town Highway Supervisor as I as I drive on the roads everyday they seem to be in good enough condition. For the rest of them, town and county tax collectors, town and county clerks, county coroner, county dog catcher, I had no clue of what type or how good a of job they were doing, so I did not vote. The county should have a proper legislature and the rest of these elected positions should be civil servants as opposed to elected positions — makes a lot more sense to me. Why are we electing the coroner?
And as for “local politics” — I know more about Mayor Zohran Mamdani and his positions and plans for the City of NY then that of my town and county (Saratoga) 180 miles to the north in Upstate.
In the age of the automobile and rapid transit, I think regional government works best in many places for most public needs. Does my village need its own government? Maybe like Tokyo we could just elected a local mayor to be a local voice. Buffalo and Erie County should be one large municipal government, not several cities and towns with duplicate public services and industrial development agencies still businesses not from Texas, but from each other… yeah, Town of Amherst NY I’m looking at you! Giving a tax break to an oil company to move from a downtown high-rise in Buffalo to a suburban office park in Amherst. How many jobs did that create in Western NY at what cost!?
California has unitary primaries. The top two candidates go onto the general election.
Probably not. That’s another weirdness of New York. That you can live in a village which is a separately incorporated municipality and in a town at the same time. And that the village can splay across town and county borders.
Does Saratoga Springs want to give up it’s control over zoning? Nah. Does Wilton want to give up all that lovely lovely sales tax revenue they get because Saratoga Springs doesn’t want chain stores? ( The “Saratoga Springs” chain stores are in Wilton. ) Nah. What happens to the school districts? Want to see the torches and pitchforks come out, think about consolidating school districts into one county wide one.
Someone has to sue to get them to change it. Until then they will have unequal representation.
On California’s primaries, let’s look concretely at Steyer vs. Becerra. If I’m a California voter, I can look at their platforms and endorsements – but it still won’t tell me what I need to know about how they’ll govern to the same detail I get if it’s a partisan federal election. If it’s a partisan federal election, I know the Republicans are the party of tax cuts, anti-abortion judicial nominations, political appointees at regulatory agencies who have lighter-touch regulations of big businesses, indifference to environmental protections, anti-union NLRB nominations, securitization of immigration, and less international cooperation with democratic allies. Individual Republicans may emphasize some of these elements more than others, but one votes for a package of appointees from the federal cabinet down and one knows what one gets. In contrast, I have no idea if there’s a difference between who Becerra and Steyer are going to appoint to agencies headed by politicals, what their legislative agendas are going to be, etc.
Of note, Mamdani, about the purest case of running as part of a slate against another slate, namely reform vs. machine, retained quite a lot of politicals and advisors from the thoroughly machine de Blasio and Adams administrations. Cid voted for Mamdani, not for Maria Torres-Springer. M. Torres-Springer isn’t even bad, but an election in which one can’t tell in advance who’s on the governing team is not meaningfully democratic. It’s not at all like that with federal elections, in which the exact list of cabinet members, judicial nominations, senior political appointees at agencies, etc. is not known in advance but their party affiliation and general agenda are.
Your love of authoritarians is showing. Again. The general disdain for democracy too.
Your upper middle class symbol manipulator is showing. Again.
They’re getting paid for 8 additional hours of work every day that they drive both a diesel and an electric train. They have to be trained on both types to qualify in the first place, but if they drive two different train types on one day, they get double.
Maybe as part of new labor agreement the LIRR engineers should be ewquired to drive steam locomotives instead of diesels to earn that extra eight hours? 😀
Sounds like a management problem to me. Ya know, the people who get paid a lot of money to …. manage… things could arrange it so there is enough staff scheduled so expensive things don’t happen.
It’s not a management issue, it’s a scam. The LIRR has dual mode locomotives so trains from diesel territory can reach Penn Station. If you operate one of those trains you get paid double for no reason. Driving from Greenlawn or Deer Park to Penn is the same distance over about 70% of the exact same track – but the driver making the trip from Greenlawn gets paid 16 hours for driving 8 hours while the one from Deer Park gets paid 8 hours for the same amount of work. Even though both stay in the same cab during their shift and use functionally identical controls.
It’s absurd to say it is management’s responsibility to “avoid expensive things” when the expensive thing makes no sense. The extra-full-day-pay rule is a holdover from the days of steam trains, either because running a steam locomotive was meaningfully different from driving an electric one, or because if you came into work in your clean uniform expecting to drive electric (not your already dirty coal uniform) and got switched to steam, the coal dust would soil your clothes and you then had to buy a new uniform. The LIRR hasn’t run a steam locomotive in 70 years, how can anyone justify the rule now?
It’s the same for other rules. If you drive a special event train you get paid a full day, regardless of how many hours you actually work. Someone who drives 6 hours regular and 2 hours special event gets paid for 14 hours, while someone operating the same trains over the same track for 8 hours regular gets paid 8 hours.
You mention symbol manipulator bias as if you’re sticking up for the average working person, but the people benefiting are not the rank and file LIRR operators. The union uses seniority, so there is an aristocracy of long serving operators who ensure that they get the dual mode and special event slots. The rank and file go along hoping someday they will be senior enough to get in on the scam. The people who lose are the passengers that are not riding the service that could be run with all of the money being paid to a select few operators for doing nothing.
That is absolutely insane
Unions pull tricks like that and then have no clue why so many Americans are against them.
@henry, I don’t think unions are quite that bad these days over here. They were in the 1970s for sure.
It’s not a management issue, it’s a scam. The LIRR has dual mode locomotives so trains from diesel territory can reach Penn Station. If you operate one of those trains you get paid double for no reason. Driving from Greenlawn or Deer Park to Penn is the same distance over about 70% of the exact same track – but the driver making the trip from Greenlawn gets paid 16 hours for driving 8 hours while the one from Deer Park gets paid 8 hours for the same amount of work. Even though both stay in the same cab during their shift and use functionally identical controls.
It’s absurd to say it is management’s responsibility to “avoid expensive things” when the expensive thing makes no sense. The extra-full-day-pay rule is a holdover from the days of steam trains, either because running a steam locomotive was meaningfully different from driving an electric one, or because if you came into work in your clean uniform expecting to drive electric (not your already dirty coal uniform) and got switched to steam, the coal dust would soil your clothes and you then had to buy a new uniform. The LIRR hasn’t run a steam locomotive in 70 years, how can anyone justify the rule now?
It’s the same for other rules. If you drive a special event train you get paid a full day, regardless of how many hours you actually work. Someone who drives 6 hours regular and 2 hours special event gets paid for 14 hours, while someone operating the same trains over the same track for 8 hours regular gets paid 8 hours.
You mention symbol manipulator bias as if you’re sticking up for the average working person, but the people benefiting are not the rank and file LIRR operators. The union uses seniority, so there is an aristocracy of long serving operators who ensure that they get the dual mode and special event slots. The rank and file go along hoping someday they will be senior enough to get in on the scam. The people who lose are the passengers that are not riding the service that could be run with all of the money being paid to a select few operators for doing nothing.
Sounds like a management problem that they haven’t electrified. Or a management problem that they haven’t had the spine to explain to the riders in diesel territory that they are going to have to change trains. Ya know the people who get paid a lot to have someone else do their typing when they ….. manage… things.
All sorts of people get paid more because they have …additional qualifications…. Some of them also unionized.
If it’s not management’s responsibility to manage things whose job is it? If they aren’t managing things why are they getting paid? I know it’s rude to ask about symbol manipulator’s pay but I’ll go there.
That’s the way it works in most of the world. You have more experience, you can charge more. There’s probably something going on where …..management… doesn’t want to spend money training everyone for everything. And regulators look at how often those skills are used.
It’s a lovely sunny day here. The cat is off in the bedroom sunning herself. So I didn’t scare her when I laughed and laughed and laughed. I’m gonna have to pass that around to other people who have to wear uniforms at work. I’m sure they will think it’s hilarious too.
It’s not the worker’s fault the management allows antique work rules to continue. Or that they arrange to get paid more for work that needs… additional qualifications… even if it’s just seniority.
Bimodes are good.
Railfans get offended. Many ways. None of the offense is the fault of the staff on the train.
New York City Transit, despite its high operating costs, manages to serve a linked trip for around $6, which works out to an annual cost per user of around $3,500,
The average American takes 4 local trips per day:
https://www.bts.gov/statistical-products/surveys/national-household-travel-survey-daily-travel-quick-facts
At $6 per trip that would be $8,760 per year, or $9,000 if you use the figure of 1500 trips per year. That is more than the $7,000 average cost of a car.
To make things worse for transit, the cost of a car can serve more than one persons trips per day (a couple going out together, a parent taking children to the doctor, etc.). But transit costs are per person, so a family of four would cost ~$35,000 per year in transit at NYC costs, which is far more than most spend on two cars.
Of course, at the London/Berlin costs the value is $5,000 per year on transit, which is less than a car per individual and about even for a family of three with three vehicles.
Just like automobile owners don’t pay, directly, for roads and parking, New York City transit users have a fare cap of $35 for each seven day period. 35 times 52 is 1820. 1820 times 4 is $7280. If the household member is eligible New York City public school student they get a student Omny card with 4 rides per day all year long.
Transit users don’t pay directly for tunnels and track (equivalent to roads) those are capital expenses funded through sources other than fares. Parking it depends. Some is free, some is not, some is a cost captured elsewhere (your rent/mortgage if your home comes with parking). Alon’s $6/trip value has nothing to do with what riders pay or Omny, it has to do with how much MTA spends to actually provide service to each riders. Equivalent to what people pay yearly for their car in gas, maintenance, insurance, etc.
Question for Alon: does the $6/trip and $7k/year include cost of rolling stock (for transit) and car payments respectively (non infrastructure capital costs)?
Difficult to believe a car costs $7k/year excluding the car payment.
But the honest thing is I am not sure this per-trip costing is particularly helpful given the variable costs of both cars, buses and especially trains are pretty small.
Comparing the MTA’s total costs to what a household spends annually is cooking the books. Hard.
The page Alon linked to says “These figures represent “out-of-pocket” expenditures, which are the costs households pay directly from their own funds” Which is the equivalent to transit fare. I have no idea what Alon marinated which way to come up with $7,000 dollars. IRS reimbursement rates for 2025 were 70 cents a mile. It’s 72.5 for 2026. The AAA estimates that the mythical average new car costs $11,577 per year if you drive 15,000 miles a year and keep it for five years. Things are more complicated than drive times because a an automobile trip into Manhattan includes a lot of time arranging parking. So does a trip into most big cities. And tolls, all the lovely tolls. …. Driving to 57th and Park costs $9 more than driving to 62nd and Park.
It doesn’t change that at $35 per seven day period it’s $1820 a year. Which compares to how Alon marinated things to come up with $7,000.
All of you can continue to cook the books which ever way you want to. Including believing Google’s hallucinations about New York City drive times. Rich people in New York City suburbs, who own cars, can afford the tolls and paid parking, will continue to take the train into Manhattan. And a few other places for special events.
In a transit oriented city, most local trips aren’t actually made with transit, they are made on foot (approximately $0 per trip), or in some cities, on bike (almost $0 per trip).
Even in Tokyo or Hong Kong, where very few trips are done by car, people average under 1 round trip on transit per day. $6/trip -> $3500/user-year is high, representative more of a transit commuter than a a regular transit user.
This is why I bang the cars-are-in-the-way thing so hard. Higher capacity transit enables a density of urbanization that means that most things are so close you can just walk to them, taking transit perhaps to reach special amenities that aren’t ubiquitous (museums, parks, concert venues, geographical features). The comparison shouldn’t necessarily be between cars and transit, but between cars and walking, as rail transit enables the existence of more things walkably close together; simple as.
Of course, a lot of people will use transit to commute to work–much as they do cars–but as Alon has pointed out repeatedly, job concentration is what can really drive transit usage, whereas the spatial inefficiency of cars disperses jobs. For all other purposes, there’s zero reason that people can’t just use their feet to meet 75% of their transportation purposes, and there’s nothing cheaper than walking (except, arguably, on a caloric level, where cycling’s efficiency wins out).
And yet, cars have one singular advantage: they commoditize location.
I think you are missing another major advantage of cars: from a transit perspective they offer zero-headway express service over infinite span. That is the car leaves as soon as you get to it, you generally don’t have to wait at your parking spot for your car to arrive like you have to wait at the bus stop or on the platform (with some exceptions like valet or parking stackers), you don’t have to make stops except at your destination(s), and your car runs anytime day or night, weekend or not. As a result, cars are almost always faster than transit, even in major dense metros like NYC, which a quick check of Google maps directions almost always shows. Only in certain times/locations (rush hour, and/or to/from downtowns or other major destinations) is transit faster than automobiles.
In the United States. It is much less true in Europe and rich Asia – especially when you include parking time which is far from instant.
The car hs overwhelming mode share in Europe, even when including air travel:
https://www.eea.europa.eu/en/analysis/publications/sustainability-of-europes-mobility-systems/passenger-transport-activity/modal-shares-in-eu-27-passenger-transport-activity
Cars are almost always faster than transit for trips in Europe too. There are places where it is not, but again places and times where it is not in the US as well. But the vast majority of the time transit simply takes longer.
This includes parking time. Parking time (and then the time to get from there to destination or vice versa) is an access time that contributes to trip time despite not being a component of travel time. But transit has access time as well, the time to walk to a bus stop or station or time to get from the entrance to a given platform. In some cases the station is right at the destination and transit access beats parking. Overall though, parking, like roads, is widely distributed, while stops/stations, like lines, are discrete and parking access takes less time. Even when it doesn’t, the trip time savings from no waiting (zero headway) and no intermediate stops (express service) makes cars faster than transit. Again, trip time calculators like Google maps that show alternatives and account for walking to parking/a station and real time arrival info show this again and again.
Yet people take transit anyway. If you want to avoid it you are more than welcome to continue to ensconce yourself in a expensive tiny little cage.
There are a lot of considerations. If time is your only factor than a car is almost always better. However most people consider lots of other factors. Cars are usually more expensive. Transit lets you do something other than stare at the road during that time. Transit can be used safely when you are drunk. There are a number of other considerations that haven’t been brought out. Each person gets to figure out what is the best optimization for their particular situation. Every time transit comes out worse transit advocates should consider if they can fix the problem – but sometimes the problem is not fixable, and sometimes it could but shouldn’t (don’t make transit more expensive.
OK so in Britain doing e.g. Manchester Piccadilly to Cardiff Central by train takes 3h20 vs 3h40 by car, plus parking time as @adiron points out. So some trips not including London are competitive by public transport.
Also cars are only used for 60% of trips here, with 30% of trips using active travel. And yes by distance travelled cars are 3/4s – https://assets.publishing.service.gov.uk/media/68e8f5af57038b5739b98656/NTS_Factsheet_2024.pdf
“No intermediate stops”.
Perhaps you are unfamiliar with these things called traffic lights. Or stop signs. Or traffic, because the capacity of cars as a mode is so very very low. Or accidents, because driving a car is inherently complicated–even if we’re not cognizant of that while we do it–and people mostly don’t care about driving, they just want to get where they’re going.
Hypothetically, my transportation costs–if I don’t own a car–can be functionally zero, if I have enough things close to me that I can walk to them and live a pretty full life. Not common, for sure, but certainly not impossible–with sufficiently dense transit-oriented development. However, if I live in an area built to accommodate cars, I must own a car, practically speaking, and it will spend the vast majority of its time parked, depreciating happily away.
This is kind of the point of “mobility freedom”: I am free to pay for as little or as much for my own transportation as I want. This is strictly impossible with suburban, car-centric development, if I don’t want to end up unemployed and homeless.
A correct comparison would be for modal share in the geographic regions where rail transport is available. “In Europe” is meaningless.
The cat looked at me when I giggled. Google is …. optimistic… It doesn’t include the time it takes to park.
Bicycles also are zero-headway express vehicles, and are much easier to park than cars. In my experience, bicycles are roughly equivalent in speed to metro service if cycling infrastructure is good; and judging by the speed of speed pedelec users, those are up to approx. 25-50% faster than pedal bicycles, which is similar or better speed vs autos in an urban context factoring parking.
Of course, parking can be disregarded if one is taking a taxi; but then the trip is not zero-headway.