However, your second paragraph is not quite correct. If nobody steps up to the plate with funding, these cuts (or a similar set of cuts) will happen. If Albany comes up with most or all of the necessary funding, then most or all of them won’t happen. The proposed cuts do have a substantial impact on the bottom line – according to this document, $62 million in 2010 (assuming mid-year implementation) and $129 million in 2011, the bulk of that coming from NYCT.

None of this is meant to imply I support the cuts. I am very much against them. In fact, I believe that the cuts a political tool to get more money from Albany. The only proposed cut that has a real impact on the bottom line is the cutting of student MetroCards, which is so politically impossible that Paterson was on the microphone the next morning vowing that it won’t happen.

As for the Manhattan avenue buses which parallel subway lines, most of them get decent, if not spectacular ridership. I think they serve a useful function, though, for people who for whatever reason prefer the bus. Maybe it’s easier to board, or you want to sightsee, or whatever. I often take the bus over the subway for short trips and if punctuality is not a concern If the ridership is there, there’s no reason to cut the service.

There is parallel service less than half a mile away to the south, albeit parallel service making fewer stops, and one mile away to the north, making similar stops. Once again, there are entire neighborhoods elsewhere being left with no late night transit service at all.

The 2nd and 3rd Avenue els ran north-south. The M15 bus and Lexington Avenue subway run north-south. I thought your concern was with east-west service.

The large hospitals didn’t deliberately space themselves away from mass transit. Cornell built the medical school in 1898, when the mass transit in the area was the 2nd and 3rd Avenue els.

I doubt the 1 train feeds many riders to the crosstown buses at night. But why don’t you suggest running extra M72 service when the 1 runs express to the MTA? I suspect that the handful of late night crosstown bus riders know exactly when and where to get the bus.

I agree that not many people quit the lines for the bus to get cabs – most of the people who want cabs don’t walk to the bus stop first! And a lot of people going to the East Side use the other crosstown routes, just as not all of the cabs going to the East Side use the same transverse.

Operation between 12 and 1 is irrelevant. (Capacity for late-running events at Lincoln Center, perhaps?) Nobody’s proposing to cut the M66 between 12 ans 1. The question is after that. The ridership threshold for eliminating late night service was 15-20 passengers per hour – that’s incredibly low.

I haven’t objected to SAS; I’ve objected to its cost. I think even at $1.7 billion per route-km it would be about as cost-effective as the other subway projects in the US – Phase 1 would cost about $50,000 per rider, higher than for other recent US subways but not by much (e.g. the LA Red Line at $30,000). It would also have knocked off my commute this year and my girlfriend’s last year by about 15-20 minutes one-way. But I don’t know that there won’t be additional cost overruns, and I do know that the same project in Paris would cost $250 million per route-km.

You’re right that most lines operate differently at night, but NYCT is trying to minimize this as much as possible in order to simplify the system, or so I’ve read. But anyway, in the case of the D, I suspect another reason it runs express at night is that it skips so many stops it’s a real time saving for people in Harlem and the Bronx, and the Upper West Side stations are low-ridership. NYCT hasn’t tried to minimize or time transfers at night to avoid making people wait 20 minutes.

Most employment in Manhattan comes from large employers, not clusters of small employers. This is especially true at night – small employers close relatively early, and tend not to generate the same nighttime traffic as hospitals or even universities.

I don’t know how much demand there is for the nighttime M72. But it’s surely more than there is for a nighttime M66 or M79 when there’s no 1 train for them to connect to. At any rate, the cost of running the extra service isn’t high compared to capital construction cost…

In absolute terms the Lincoln Center demand is low, but relative to capacity, it’s huge. The buses are loaded after events, and, worse, the line of people boarding can take 10-15 minutes. Those people may be a minority of Lincoln Center attendees who come from the East Side, but I’m not so sure – I only see a few people quit the long lines to take cabs. (How would the cabs even deal with the capacity bottleneck across Central Park?). In either case, if it’s a minority, it’s going to be a larger minority if boarding times are halved.

Finally, what I said about the M66′s late night headways comes mainly from its operation between 12 and 1, before it switches to 40-minute frequency. It switches to 40-minute frequency earlier than the M86 but later than the M96 and M79, and runs late night buses at 10-minute frequency until it switches, on a par with the M86 and shorter than the M79 and M96.

The line along 60th St. is 90 years old. Blaming the line for being in the wrong place for being far away from the late night job centers seems kind of silly to me – especially coming from someone who has objected to SAS, which will have a stop much closer to Weill Cornell (with direct service to Times Square).

I do wonder, incidentally, why you are particularly concerned with individual large employers, which may well find it worthwhile to start operating shuttle buses when the convenient crosstown bus lines shut down, rather than with clusters of smaller employers, none of which individually have enough late night employees to make a shuttle bus worthwhile.

Plenty of subway lines have different service patterns at night than during the day – the D included (it runs local in Brooklyn at night; it could easily be modified to run local in Manhattan at night as well). The only lines that run exactly the same 24/7 are the 1, F, L, and Q (the G may soon be added to that list, with the Q removed).

Do you really think there’s substantial demand for the M72 when the 1 runs express at night – substantial enough to add the cost of running a bus to whatever capital project requires the 1 to run express (these days, probably the station rehab at 96th or 59th)?

The vast majority of people who take cabs after Lincoln Center events will take cabs regardless of the quality or quantity of bus service. And Lincoln Center generally has several simultaneous events that don’t necessarily end at the same time (but might happen to end at the same time on some nights) – how do you schedule the buses around that? Or do you just have the extra buses sitting around from the earliest time that any of the events might possibly end (with the drivers collecting time-and-a-half), pulling up when the crowds show up, kind of like what happens on the subway after Yankee games? I’m afraid we’re not looking at Yankee numbers here – sports events draw much larger crowds than Lincoln Center events, I’d be surprised if as many as 10% of Lincoln Center patrons ride the M66 home. Seems like you’re advocating a pretty major expense for little benefit.

All of the late night crosstown buses (M42, M50, M66, M79, M86, M96) run on a 40-minute headway – not because ridership levels require a bus every 40 minutes, but because it takes a bus 40 minutes to complete a round trip at night – in other words, increasing the headway would only marginally reduce operating costs (there wouldn’t be any labor savings). There are plenty of seats available on all six routes at night on the 40-minute headway. So I don’t see how you can conclude that the M66 has the second highest late night ridership judging by its late night frequency – they all have the same frequency, and even if they didn’t, frequencies at night are dictated by the cycle time, not by ridership.