Actually just cutting a train also cuts the revenue that was partly paying for the fixed costs that do not go away by cutting the train and just reallocates them to other trains making them look worse. In the net, unless the trains had almost no riders and farebox recovery it turns out to be a net loss. No one has ever cut trains to prosperity in the annals of passenger railroad history. they have cut trains to eventually go out of business.
True if and only if the train is making enough revenue to cover its marginal costs of operation.
Agreed. In 2008, it started to become really hard to get the marginal costs information out of Amtrak, due to Congressional assaults on sanity (demanding meaningless "fully allocated numbers"). However, from the data we have, as far as we can tell:
-- most of the "state-supported" trains do not make enough onboard + ticket revenue to cover marginal costs of operation, hence the need for state subsidies
-- the NEC, the Lynchburger, the Norfolk & Newport News trains, Auto Train, Star, Meteor, Palmetto, and LSL do make enough to cover marginal costs of operation
-- Empire Builder, Coast Starlight, and Crescent are always pretty close to covering marginal costs and are probably doing so in some years (numbers have too much margin of error to tell for any given year)
-- the Cardinal would certainly cover marginal costs if it were daily
-- the Capitol Limited would probably cover marginal costs if the Cap/Pennsy through cars were operating
If you're trying to examine trains which are consistently not covering marginal costs, and where the reason isn't something totally bloody obvious which should have been fixed five years ago, really the only trains you can look at are:
-- California Zephyr (I continue to suspect that Chicago-Denver performs much better financially than Glenwood Springs-Reno)
-- Southwest Chief
-- City of New Orleans (the only "long-distance division" train whose financial performance seems to have definitely gotten worse from 2012 to 2015)
-- Texas Eagle
-- Sunset Limited
With the exception of the CONO, these have all had very substantially improved financial performance over the last three years. It is not outside the realm of possibility that the SWC could cover its marginal costs in 2016, i
If the financial improvements continued at the same rate (which they probably won't),
-- the SWC could cover its marginal costs in 2016
-- the CZ could cover its marginal costs by 2018
-- the SL and TE could cover their marginal costs by 2019
Railroads really aren't a marginal-cost business. Like telecoms they're a high-capital-cost, high-fixed-cost, low-marginal-cost business. That's why they thrive on high population density.