That’s not going to work out in your favor for the simple fact that roads already exist and maintenance is extremely cheap. It’s paid for with gasoline taxes.

Whereas rail is unbelievably expensive. The MTA budget for the NYC metro area is $18.6 billion. Compare that to the $11.8 billion budget for the entire New York State Department of Transportation.

…and that’s for an existing rail system.

>...roads already exist and maintenance is extremely cheap. It’s paid for with gasoline taxes.

According to [0] no state derives more than 71% of its "State & Local Road Spending" from gas taxes. Unclear to me what proportion of the spending is maintenance though.

The best number I could find for maintenance cost was [1], which says in 2015, the average per-mile maintenance cause of US roads is $28,020.

Can anyone help find a comparable figure for rail maintenance?

[0] https://taxfoundation.org/states-road-funding-2019/ [1] https://azmag.gov/Portals/0/Documents-Ext/FLCP/Roadway-Maint...

That combines maintenance with new projects though, the GP was only refering to maintaining existing roads.

No, in "expected to break even" I was referring to total costs.

Including maintenance, externalities and construction.

(though while I would apply the same standards - I would not expect such infrastructure to directly pay for itself. But expecting it only from rail while not expecting it from roads is extremely weird.)

I've seen it estimated elsewhere that a km of heavy rail track is slightly cheaper to build than a single freeway lane. To what degree all the surrounding infrastructure, maintenance and externalities change that equation I'm not sure, but a single track is surely capable of moving a lot more passengers per hour than one freeway lane. Of course there may be relatively few situations it actually does, but if your goal is to get as many people from A to B as quickly and cheaply as possible, rail is surely cheaper.

> maintenance is extremely cheap

200+ Billion each year & 5th largest category of state and local public spending behind education, general welfare, and healthcare

> It’s paid for with gasoline taxes.

Generated 52 billion so ~25%

The NYC subways alone delivered over a billion rides a year, even with Covid.

Highway maintenance is anything but cheap, unless you only do the barest minimum as has been done in the US for decades.

The MTA is also known for being riddled with mismanagement and waste. If they worked on optimizing their use of funds/directed funds towards worthwhile improvements, credible vendors, and rooted out waste/corruption, that number would likely be a bit (maybe more than a bit) lower.

Not to mention, the 2nd ave line seemed like a big waste.

How are you pricing in car externalities?

> roads already exist

The implication in this is that railroads don't already exist, which is simply false. In fact, nearly every town in the Midwest established before 1950 were built on a railroad line. A few notable examples are Las Vegas, Los Angeles, Sacramento, Birmingham, Chicago, St. Louis, and Portland Oregon. Moreover, many suburbs were formed around a streetcar line, which were later torn up to let cars drive there instead.

> maintenance is extremely cheap. It’s paid for with gasoline taxes

That is demonstrably false. Gas taxes as of 2011 do not cover road maintenance (at most it's RI with almost 80% and median of about 45% [1]). Furthermore, the federal gas tax (which is supposed to pay for the interstate system) hasn't been increased since 1993, and isn't indexed to inflation. The United States Highway Trust Fund has been relying on non-fuel tax revenue since 2008, and that doesn't look to be changing any time soon [2]. Of course, regardless of the actual costs of effective maintenance, you can make any public works project much cheaper by simply not maintaining it. If maintenance was actually "extremely cheap," then it stands to reason that the US would be getting more than a 'D' on roads [3]. Even after all of that, it still doesn't factor in the legal requirements for private expenditure on roads, such as minimum parking requirements and extensive R1 zoning. It also doesn't factor in the vast negative externalities of cars (yes, even electric cars).

> Whereas rail is unbelievably expensive

I'm going to ignore that you chose the MTA, a famously corrupt organization that is responsible for tunneling under the densest metro area in the country, as a baseline for the cost of rail. Instead I'm going to point out that car roads have nearly every bureaucratic advantage in the US, for a multitude of reasons. There is a lot of nuance to this, but for anything short of HSR, the expense in railroads is not engineering. If you're interested in a jumping off point for learning more here, check out [4].

[1]: https://taxfoundation.org/gasoline-taxes-and-user-fees-pay-o...

[2]: https://www.enotrans.org/article/ten-years-of-highway-trust-...

[3]: https://infrastructurereportcard.org/cat-item/roads-infrastr... (The Infrastructure Report Card is published by the ASCE, which does have an incentive to create more work for civil engineers, so it's worth taking it with a grain of salt)

[4]: https://palladiummag.com/2022/06/09/why-america-cant-build/

> I'm going to ignore that you chose the MTA, a famously corrupt organization that is responsible for tunneling under the densest metro area in the country, as a baseline for the cost of rail.

Sidenote, that's called apophasis - https://en.m.wikipedia.org/wiki/Apophasis

> Whereas rail is unbelievably expensive.

Note that rail is uniquely more expensive in USA than elsewhere for various reasons.