That is a misunderstanding of how the railways are funded.
The money that is used to subsidise services in the North, comes from profitable commuter ToCs like SWR, GTR and C2C (the intercity and Thames Valley part of GWR cross subsidises the South West services to a degree also).
If the revenue from these ToCs is reduced then there is less money to go round for Northern rail etc. So service cuts will be widespread across the country.
No, the money paid back goes straight into treasury coffers and then the treasury divvies out subsidies to the local operators direct from the coffers. - There is no reason this money cannot be provided for a number of years to come - Furlough, Track and Trace, etc all cost significantly more than seeing the railways through just a few more months at the end of the pandemic will.
Using this argument to justify service cuts in places where passenger growth has been higher is absolutely ridiculous. In fact cuts are ridiculous right now when by the time they have been implemented (October) we can expect the landscape to be significantly different - my company wants 5 day a week office working again by then. Yet another short sighted move that will come back to bite us in a year. - Remember we're still recovering from the crisis, confidence is low, this is not the time to start making un-recoverable changes based on perceptions of what might happen.
Also worth pointing out that cutting services probably won't save much money. You're still paying ROSCOs for the same number of trains, infrastructure must still be maintained, stations must be staffed, trains must be cleaned - each service cut will not have a corresponding saving in costs, but will have an impact on the number of people choosing to travel by train. Cutting service during the pandemic was primarily there to reduce the number of staff at risk of infection and increase the resiliency with a reduced number of staff potentially available due to isolation and sickness - (most staff ended up on furlough anyway...)
This blog post has some interesting points on this - a 20% reduction in service is likely to equal approximately a 10% reduction in operating costs:
The whole thing really is an exercise in bean-counting, one that will really have a trivial impact on expenditure when all is said and done. Finding £2bn a year in operational savings is easy long term - invest in electrification and see your maintenance/fuel costs slashed, implement DOO more widely with guards focusing on revenue collection, and modernise legacy signalling systems that have large staff/maintenance requirements. All of these have the benefit of a better passenger experience too.
Capital projects aren't opex, but they could be managed significantly better, figures pointing to being able to reduce them by at least 30% with competent long term planning - likely more with good system specification and contractual negotiation methods (e.g. giving contractors bonuses VS punishing them for late work). Every time a project like HS2 goes for re-design to try and reduce costs, the price goes up by 10%... that's what supply chain uncertainty will do to you...
Let's also compare the costs of cutting now Vs waiting another year to be GENUINELY out of pandemic mode (remember many have received their last jab in only the past month, especially many of the younger office workers based in places like London). We cut services by 20%, which saves us 10% in opex, but conservatively reduces passenger numbers by at least around 5% (likely more). So now we've saved something like 5% in Opex. - Rail bosses haven't even agreed to cut jobs until the end of 2021, so no savings besides fuel until 2022 at the earliest!
Now in a year, say passengers with no other option are rammed on like sardines and the industry has to scramble to ramp up again. Now they have to hire new staff in the midst of a labour shortage, train them from the ground up and deal with reduced productivity while they learn, very likely on higher wages than the staff they replaced. This is really all starting to look like a bit of a farce, compared to just waiting until next year, seeing how the situation is then and making a long term judgement on service changes to be gradually implemented over a number of years to create a network better optimised for post Covid travel patterns...