As an industry the railway is 'bankrupt'. For the individual operating businesses they are not trading insolvently because their revenue is being guaranteed by Government.
But the level of reliability (and coverage) before the pandemic was nowhere near good enough to encourage modal shift. I am questioning whether the cost of changing that in order to encourage modal shift is worth it. It is a little sad that the current state of affairs is meaning that you would consider the situation in 2019 as acceptable. It wasn't - it was **!"*.
Unfortunately, so many of the population (and so many voters) do not wish to take part in the pleasure of public transport, so the remaining use is such that does not economically justify utility service. So it has become a niche activity (albeit a fair sized niche) which has to be increasingly propped up with public money. This money needs to be spent wisely and efficiently to get the best coverage with the most appropriate mode.
You are right. If we do not want to/cannot afford the current service/network size at an acceptable level of reliability then we reduce their size to the level that can be run reliably. One of the ways to do this is to improve efficiency and another is to identify the worst revenue/cost ratio parts and lop them.
We all accept a certain level of unreliability in our lives (the car sometime does not start/puncture/etc) and a certain amount of inconvenience (the shop we like is further away than an alternative we don't like etc) . However, reliability, and convenience costs money and there is always a trade off. There is such a large gap between private car and public transport (and I accept that this changes depending on the journey being undertaken - but those instances where public transport wins is niche) that it would be an impossibly large sum of money to be poured down to make a much of a difference.
The issue is that lopping off one part of the network doesn't always bring the cost savings expected (for example National Highways has been given the costs of maintaining historic rail routes, i.e. those from the Beeching report, and the infilling of bridges which has caused complaints) and even if it does it may not result in the passenger numbers being retained on longer distance routes.
That's not too say that there can't be cuts, just that there should be a viable business case which considers these factors (effectively the opposite of a reopening business case).
As I highlighted previously we can't just focus on reducing the £25bn of spending on rail by the public sector if we ignore the £18bn of income. As we could end up cutting the £25bn to £15bn but also end up cutting the £18bn of income to £6bn and so the level of support would rise from £7bn to £9bn.
For example the cheapest way (at least in staffing costs)to provide for a longer with 2 coach trains every 2 hours would be to run a single 10 coach train. The problem is that no one would use such a service and so the income would fall much further than the savings made.
Of that £7bn (pre Covid) an easy short term cut could be to enhancement spending (£3bn), however it maybe better to keep that spending if it results in lower costs (for example through electrification). Likewise some of that spending would also be on HS2, which in their could reduce the costs now.
However with more passengers and services per driver and a reduction in coaches needed (even though the trains are longer) to run the same frequency of service actually if you want to maximise the income while reducing costs HS2 actually could be sensible to do.
11 coach 390 taking 5 hours to do the round trip London to Manchester at 3tph requires 165 coaches, do the same for HS2, but it's 16 coaches and the round trip is 3 hours and the number of coaches needed falls to 144.
Likewise 1 driver per 589 seats on a 390 vs ~1,100 on a HS2 train. However given that over a 10 hour window two return trips can be run with a 390 vs three by HS2 this further reduces the cost of staff on a per seat basis.
As an example of currently it costs £25 million a year to run 3tph the cost per seat would be £12, this could fall to £20 million (due to the better use of rolling stock and staff) however even if it didn't (let's for argument say we had extra maintainance and energy bills which offset those savings) the cost per seat would already be at about £7 due to the extra seats per train. That could allow slightly lower ticket prices and still generate the same income if more people used the servicess. (Numbers are for illustrative purposes, as the proportions stay the same, factor up by double and it goes to £24 vs £14, factor up by 10 and it's £120 vs £70)
The other thing to bear in mind is that rail use hasn't fallen to 1980 levels, or even 1990 levels but rather circa 2010 levels (at least this year). At the time rail was having significant levels of investment due to carrying near record numbers of passengers. Yes costs have increased over the time between then and now, as well as a shift in who is traveling and when and how much they pay, however few were suggesting cuts at the time. Yes there's a need for change, but it's useful to put passenger numbers in context.