We’ve done this to death before, but the number of people giving up their cars and switching to rail because of reduced days in th office is going to be close to zero. Much more likely to be the other way round.
A 0.125% shift in the number of miles traveled by car (0.5% of all miles traveled) to rail would result in a 1% increase in miles traveled by rail. Such a shift, when the average distance per person driven is 7,000 miles would be a return trip of 4.5 miles each way a year.
Therefore one person using rail once a week for a 22 mile each way commute would cover 250 people's worth of modal shift, likewise a return trip of 110 numbers each way would be another 25 people's milage.
The other thing to bear in mind is that the number of people with a full driving license (as a percentage of that age group) is getting older. A decade ago it was those in their 40's with the highest percentage, now it's those in their 50's.
Now whilst there's (at least until a certain age) an ongoing increase within each cohort, this is typically fairly small after people reach 30.
Add in the inability to learn to drive and pass your driving test for many over the last 15 months and that's also going to have an impact on the numbers using rail (again fairly small overall).
As such whilst the numbers of each thing may be fairly small in terms of the number of miles traveled overall it could have a noticeable impact on rail use. Even so whilst it's unlikely to be huge, it's all likely to add to that rail use certainly isn't going to see ongoing use fall by 40%.
As such I'll still sick to my personal prediction of a fall in rail use in the 12 months after restrictions are lifted of 20% or less looking fairly likely and 15% or less being a reasonable probability with such a fall being fairly short lived (although still possibly down a little on 2019 figures, but probably still up on 2015 figures, for up to 5 years). Of course even such a prediction could be fairly pessimistic and it could get to within 5% of 2019 levels in that first year and overtake then soon after (bearing in mind that road use has already surpassed 2019 levels, it's just that much more of it is off peak).
Of course I could be overly optimistic and it could be that there's a 25% to 30% fall, however with significant numbers of city office workers still WFH full time (which defiantly isn't going to be the case) the current 55% rail use is most likely to only go one way and probably by quite some way. 1/3 more would get to to within the 25% to 30% fall range, whilst 40% more would take you back out the other side of that range.
Although the really difficult thing to predict is revenue, as if you only need to go in once a week would you bother living in Wimbledon if you could live in Winchester (as an example). As the time saving of living closer to work would be small but the house you could buy would be much bigger and you'd be much closer to countryside and the coast of you lived further out. As such your weekly rail costs could about the same or even actually a little more, but then that's offset by being mortgage free sooner/having a smaller mortgage.
Even if they stay put, by paying turn up and go prices then it doesn't save you 80% (probably closer to 2/3rds) of your session ticket costs if you go into the office 1 day a week (in the same way that driving in 1 day a week doesn't cut your overall car ownership costs by 80%, it's more likely to be less than 1/3, which is likely to alter the number of people getting their first car).
Even flexible season tickets, even if they were to save you some money on 1 day a week travel, wouldn't get you close to the same per trip cost as an annual season ticket.
As such there's the potential that revenue could reach 2019 figures before passenger numbers do.