Rail usage

[Bald Rick]

Perhaps they are being too London Centric and not running trains from (example) Milton Keynes / West Midlands to Lake District which leisure travellers would like.

There’s trains from the West Mids to the Lake District every other hour in the morning and hourly in the afternoon…

 

[The exile]

Perhaps they are being too London Centric and not running trains from (example) Milton Keynes / West Midlands to Lake District which leisure travellers would like.

The real boom in leisure travel is surely medium distance days / evenings out, for which West Midlands - Oxenholme is really stretching it a bit (especially in the winter months).

 

[The Ham]

What part, if any, would split ticketing have on perceived passenger numbers?

Some, however if you look at the regional passenger numbers (i.e from London to another region) it's going to give you a reliable comparison.

As anyone splitting a ticket is likely to do so within a region (getting a train from one part of the South East or London to another to then get the direct train) or if between regions (say London to the West Midlands and then West Midlands to North West) it'll not show up in the growth numbers for the longer distance travel.

Generally, however, that intra regional growth is still happening.

For example, taking a long term view, from 2009 to 2025 passenger numbers between London and North West are up by comfortably over 2.4% per year (averaged over the whole timeframe)

That's including for the significant drop due to COVID and the fact that the 2025 numbers can still slightly lower than the 2019 numbers.

I don't know how that compares to how financially we'll the government wanted the railways to do, however 2.4% was the growth model used by HS2 and the 2025 numbers are broadly on track (West Midlands are a little lower than the others but still a year early, North West are 4 years early and Scotland up lots and it's about 20 years ahead!)

Whilst that's no argument for building more infrastructure faster, it's certainly one which shows that we shouldn't be abandoning building more infrastructure, as otherwise we're going to run out of railway capacity.

 

[Snow1964]

Dft have updated their post Covid transport usage spreadsheet

Daily domestic transport use by mode

Daily usage of selected domestic transport by mode for Great Britain.

www.gov.uk

There are two figures for rail (including Elizabeth line, and excluding Elizabeth line). Percentages of pre Covid usage for most recent 5 days are :

Wednesday 100% and 89%
Maunday Thurs 93% and 84%
Good Friday 101% and 91%
Saturday 112% and 101%
Easter Sunday 120% and 108%

So weekends are now busier than pre covid, and looking at these numbers, does rather reopen question of if less engineering work and more intensive service should happen on Sundays

 

[absolutelymilk]

So weekends are now busier than pre covid, and looking at these numbers, does rather reopen question of if less engineering work and more intensive service should happen on Sundays

I suspect daily passengers are still considerably lower on weekends than on weekdays, and revenue lower still due to the absence of peak tickets. Sadly the work has to be done sometime and better at the weekend than during the week!

 

[Bald Rick]

So weekends are now busier than pre covid, and looking at these numbers, does rather reopen question of if less engineering work and more intensive service should happen on Sundays

In absolute terms, Sundays have half the traffic of weekdays. This is across the whole network, there are of course some lines where Sundays are closer to a weekday, but then there always were.

 

[dk1]

I suspect daily passengers are still considerably lower on weekends than on weekdays, and revenue lower still due to the absence of peak tickets. Sadly the work has to be done sometime and better at the weekend than during the week!

That’s what I was thinking.

 

[Trainman40083]

In absolute terms, Sundays have half the traffic of weekdays. This is across the whole network, there are of course some lines where Sundays are closer to a weekday, but then there always were.

I guess with a lesser frequency on some lines on a Sunday, trains may be able to use two of four track, enabling engineering work on the other two or divert to other potential routes that have capacity.

 

[Stephen42]

Dft have updated their post Covid transport usage spreadsheet

Daily domestic transport use by mode

Daily usage of selected domestic transport by mode for Great Britain.

www.gov.uk

There are two figures for rail (including Elizabeth line, and excluding Elizabeth line). Percentages of pre Covid usage for most recent 5 days are :

Wednesday 100% and 89%
Maunday Thurs 93% and 84%
Good Friday 101% and 91%
Saturday 112% and 101%
Easter Sunday 120% and 108%

So weekends are now busier than pre covid, and looking at these numbers, does rather reopen question of if less engineering work and more intensive service should happen on Sundays

The daily figures for the National Rail timeseries are meaningless, the methodology remains a rolling 7-day average.

Data labelled ‘National Rail’ is calculated by summing the number of journeys over a seven-day period and expressing this as a percentage of the equivalent pre-COVID-19 week in 2019 or 2020.

The number of journeys is calculated from LENNON by taking earnings and estimating the likely number of journeys that would have been taken, using an industry-agreed methodology that is carried out by the Office of Rail and Road (ORR). For season products of more than a week’s duration, journeys are apportioned out across the duration of the product’s validity.

Historically, the ORR uses the ‘Date of Settlement’ for estimating the likely number of journeys that would have been taken. This approach has been applied consistently to maintain a stable time series. While the ‘Date of Settlement’ does not always match the actual date of travel, ORR considers any impact on quarterly or annual trends to be minimal.

The Date of Settlement aspect can cause timing differences so looking at one week in isolation will also be unreliable. The figure for the week before Easter Sunday are 97% and 87% which to me suggests timing may be a factor here.

The TfL figures are daily ridership figures which show the day of week pattern consistently. Again that won't necessarily mean more people are travelling on weekends than weekdays as the level of ridership will not have been the same pre-covid either.

 

[BorderCollie]

ScotRail say their busiest day of the week is Saturday.

 

[The exile]

Dft have updated their post Covid transport usage spreadsheet

Daily domestic transport use by mode

Daily usage of selected domestic transport by mode for Great Britain.

www.gov.uk

There are two figures for rail (including Elizabeth line, and excluding Elizabeth line). Percentages of pre Covid usage for most recent 5 days are :

Wednesday 100% and 89%
Maunday Thurs 93% and 84%
Good Friday 101% and 91%
Saturday 112% and 101%
Easter Sunday 120% and 108%

So weekends are now busier than pre covid, and looking at these numbers, does rather reopen question of if less engineering work and more intensive service should happen on Sundays

Is that comparing like with like though (due to Easter)?

 

[NCT]

I think fresh research into journey purposes by day of week and time of day is overdue and should yield some interesting results, though I suspect any results won't be quite as earth shattering as some might expect.

There is good recovery of peak time travel in London and the South East. If the theories around agglomeration and human interaction boosting productivity still holds, then commuting and work trips will continue to be viewed as being more economically essential than 'discretionary' trips, and it'll be difficult to argue for engineering works to be moved to weekdays. (Of course trips for work purposes happen on weekends as well but they are much less dominant and concentrated)

 

[Trainman40083]

I think fresh research into journey purposes by day of week and time of day is overdue and should yield some interesting results, though I suspect any results won't be quite as earth shattering as some might expect.

There is good recovery of peak time travel in London and the South East. If the theories around agglomeration and human interaction boosting productivity still holds, then commuting and work trips will continue to be viewed as being more economically essential than 'discretionary' trips, and it'll be difficult to argue for engineering works to be moved to weekdays. (Of course trips for work purposes happen on weekends as well but they are much less dominant and concentrated)

For many, weekend travel can be any weekend. For those attending concerts and sports events, for rail to be used they need travel that weekend

 

[Snow1964]

Dft has updated the usage compared to pre covid spreadsheet

Daily domestic transport use by mode

Daily usage of selected domestic transport by mode for Great Britain.

www.gov.uk

Column 1 is rail, column 2 rail excluding Elizabeth line

Most recent non bank holiday weekend
Sat 16th 105% 94%
Sun 17th 106% 95%

Weekdays
Mon 18th 106% 95%
Tue 19th 105% 94%
Wed 20th 109% 97%
Thu 21st 106% 96%
Fri 22nd 101% 91% (start of school half term)

Bank holiday weekend
Sat 23rd 101% 91%
Sun 24th 102% 92%
Mon 25th 104% 94%

Half term weekdays
Tue 26th 101% 92%
Wed 27th 102% 92%
Thur 28th 104% 93%
Fri 29th 102% 92%

Looks like on normal (non holiday) weeks Wednesday is slightly busier, but 97% of pre-Covid (excluding Elizabeth line) suggests nearly back to 100%.

Of course any Operator still running Covid reduced timetables with 97%of pre-covid is probably badly out of touch.

I realise that some lines, or specific passenger flows are still down, but if getting average of 97%, some must have grown to compensate. But if look at stock allocation today, it is little different to 2019 which suggests the railway (either franchise operators, or DTFO (shadow GBR) hasn't yet responded to growth in some areas and shrinkage in others by cascading stock to busier areas. Although maybe having lots of non standard designs means it is in too difficult bucket.

 

[Horizon22]

Looks like on normal (non holiday) weeks Wednesday is slightly busier, but 97% of pre-Covid (excluding Elizabeth line) suggests nearly back to 100%.

Wednesdays and Thursdays are known to be the busiest day for rail commuting, particularly for London.

Of course any Operator still running Covid reduced timetables with 97%of pre-covid is probably badly out of touch.

But if look at stock allocation today, it is little different to 2019 which suggests the railway (either franchise operators, or DTFO (shadow GBR) hasn't yet responded to growth in some areas and shrinkage in others by cascading stock to busier areas. Although maybe having lots of non standard designs means it is in too difficult bucket.

I imagine there's a lot of variety as you say. The issue - as it has been for nearly 5 years now and has been posted many times here - is that the rebound in passenger numbers hasn't been met by an equivalent rise in revenue.

 

[Trainman40083]

Yes, in a perfect world, stock would be cascaded. But that would need them to be of the right type. Third rail electric units, wouldn't be much use to Manchester,Leeds or Newcastle. As Horizon22 points out, increased passenger numbers does not automatically translate into increased revenue. Less commuters, less business travel, a lot more advance tickets. I am sure many potential travellers are discouraged by overcrowding, due to short forming, maybe even increasing disruption. Yes, you are entitled to delay repay, but I'd rather get somewhere.

 

[Nicholas Lewis]

Dft has updated the usage compared to pre covid spreadsheet

Daily domestic transport use by mode

Daily usage of selected domestic transport by mode for Great Britain.

www.gov.uk

Column 1 is rail, column 2 rail excluding Elizabeth line

Most recent non bank holiday weekend
Sat 16th 105% 94%
Sun 17th 106% 95%

Weekdays
Mon 18th 106% 95%
Tue 19th 105% 94%
Wed 20th 109% 97%
Thu 21st 106% 96%
Fri 22nd 101% 91% (start of school half term)

Bank holiday weekend
Sat 23rd 101% 91%
Sun 24th 102% 92%
Mon 25th 104% 94%

Half term weekdays
Tue 26th 101% 92%
Wed 27th 102% 92%
Thur 28th 104% 93%
Fri 29th 102% 92%

Looks like on normal (non holiday) weeks Wednesday is slightly busier, but 97% of pre-Covid (excluding Elizabeth line) suggests nearly back to 100%.

Of course any Operator still running Covid reduced timetables with 97%of pre-covid is probably badly out of touch.

I realise that some lines, or specific passenger flows are still down, but if getting average of 97%, some must have grown to compensate. But if look at stock allocation today, it is little different to 2019 which suggests the railway (either franchise operators, or DTFO (shadow GBR) hasn't yet responded to growth in some areas and shrinkage in others by cascading stock to busier areas. Although maybe having lots of non standard designs means it is in too difficult bucket.

The SE heavy commuter operators generally report 70-80% peak use compared to 2020. Overall numbers across the day are higher as more usage has been moved into the shoulders of the peak. This is no bad thing has peak capacity is the most inefficient to operate and performance critical so overall it should make for a more reliable railway.

 

[Sonic1234]

This is no bad thing has peak capacity is the most inefficient to operate and performance critical so overall it should make for a more reclaimable railway.

It's more likely to be seen as not enough people are buying Anytimes.

 

[Horizon22]

The SE heavy commuter operators generally report 70-80% peak use compared to 2020. Overall numbers across the day are higher as more usage has been moved into the shoulders of the peak. This is no bad thing has peak capacity is the most inefficient to operate and performance critical so overall it should make for a more reliable railway.

But it was a big part of rail revenues which may now travel off-peak and also encompassed anytime business travellers, demand for which has dropped off a cliff.

So in some ways it is a challenge at the very least.

 

[The Ham]

But it was a big part of rail revenues which may now travel off-peak and also encompassed anytime business travellers, demand for which has dropped off a cliff.

So in some ways it is a challenge at the very least.

The issue with a very high peak flow is that there's the potential that some units run just 2 services a day (one inbound and one back again).

Assuming lease costs of £200,000 per coach and a 4 coach unit and lease costs being 1/3 of all TOC costs, that's £2.4 million a year.

Assuming 70 seats per coach plus 30 people standing, that train basically needs to be rammed with people paying £6,000 a year in season tickets.

Conversely, if it's only doing off peak travel and runs 4 return trips a day (we'll assume 330 days a year to allow for maintenance) running at 29% of seating capacity (80 seats per unit). The ticket price per person would need to average £23 return.

In a real world example Woking to Southampton evening out return* is £27 and the season ticket is £6,528 (so close numbers so we can reasonably scale them.

If we scale the season ticket down that would be £5,560, which is lower than the required £6,000. On that basis the off peak trains would break even whilst the rammed peak trains would make a loss.

In reality there would be a mix of tickets being used across trains across the day.

However, the reality is that there were likely fleets in the southeast where, if there wasn't the need to cater for the very high peak flow that the number of units could have been reduced.

Typical peak travel is 1 in 10 people traveling between 8am and 9am, yet there's typically 13 hours of train services (excluding the reduced frequency often seen after around 9pm). For rail that's more typically 15% and sometimes rising to 20%.

If we have a baseline of 1,000 heading towards "City A" that's 150 in the morning peak and an average of 71 each of the remaining hours.

If Covid means that 150 has reduced by 20%, so that the morning peak is now 12% of daily flows, if the total passenger count is still 1,000 the numbers alter to the following.

Morning peak hour is 120 and the off peak average is now 74.

The issue the railway's really have is the distances traveled, whilst peak hour travellers may well travel for 30 minutes, off peak they are likely to travel shorter distances.

This means that to earn £100, the railway's need to have more people traveling off peak than peak. However, as these are shorter trips it could be that the peak capacity needed isn't any higher (for example on a train between Southampton and Woking a seat maybe used from Southampton to Eastleigh, Winchester to Basingstoke and Fleet to Woking and generate a similar income as one person traveling Southampton to Woking (yes there's gaps, but due to higher costs per mile for short station hops this can result in a similar total).

The complexity is that WFH can mean that peak hour travel may have only fallen 10% on the busiest days but 20% on the quietest.

I do wonder if the railway's are missing a trick in that they could offer a flexible season ticket. What it could offer is a monthly ticket based on allowing unlimited peak travel, but with a tap in uplift to "upgrade" a given day to permit peak hour travel.

In doing so it could allow people to use it for other travel (visiting friends, going shopping, going to work off peak for a meeting on a day they don't normally go to the office but the meeting time means they don't have to use peak time trains, etc.).

Pricing might be tricky to pitch, but based on Southampton to Woking if it were 16 days at £27 (£432 for the month) with a £6.50 daily upgrade for peak travel (compared to the £7.50 price difference). For anyone traveling less than 175 days of peak travel that's cheaper than the annual season ticket.

For within the month, chances are it'll only be worth doing if you make some extra trips (even if they are one station hops) and are working in the office 3 days a week (and even then the advantage of just paying a monthly amount, plus a tap in charge which is taken automatically for the "anytime" days, and not having to buy tickets each day could be of value to some).

The point being, that if they then wanted to go somewhere else (London, Reading, Guildford, etc) they'll almost certainly pay the extra to extend their ticket rather than drive. Overall, whilst some may get it to work in their favour, it's likely to act as a boost to the railway's income (in the same way Railcards do).

*yes I know that an evening out return is only available for a part of the day, but wished to use the cheapest off peak ticket available as a comparison to highlight that season tickets, whilst a lot of money on one go may not make the railway the profits people assume.

 

[crablab]

I do wonder if the railway's are missing a trick in that they could offer a flexible season ticket. What it could offer is a monthly ticket based on allowing unlimited peak travel, but with a tap in uplift to "upgrade" a given day to permit peak hour travel.

UK rail pricing penalises discretionary travel outside of forced purchases for commuting.

If someone is paying £6k (or £70 a day) to commute to work there is no incentive to spend even more money on the railway for leisure, unless you have to.
As has been mentioned numerous times, most households (and certainly those in the SE regularly commuting) have access to a car. That's a sunk cost with a small incremental expense for usage.

A season ticket is just a sunk cost and, given the changes in working patterns, it represents pretty poor value for money these days.

If, for a nominal fee (or nothing), you could use it to travel anywhere on the network then maybe we'd actually encourage some uptake & modal shift.

 

[The exile]

UK rail pricing penalises discretionary travel outside of forced purchases for commuting.

If someone is paying £6k (or £70 a day) to commute to work there is no incentive to spend even more money on the railway for leisure, unless you have to.
As has been mentioned numerous times, most households (and certainly those in the SE regularly commuting) have access to a car. That's a sunk cost with a small incremental expense for usage.

A season ticket is just a sunk cost and, given the changes in working patterns, it represents pretty poor value for money these days.

If, for a nominal fee (or nothing), you could use it to travel anywhere on the network then maybe we'd actually encourage some uptake & modal shift.

That’s where, for the main centres of population at least, a multi-mode zonal fares structure is an absolute must. That enables precisely that kind of flexibility at least in local areas, which would go a long way to satisfying most of the people most of the time. You can then introduce things like the ticket permitting free travel for accompanied kids on weekdays in the school holidays (which is where the capacity is but is going to be self limiting because the holder probably cant take the entire time off work).

 

[The Ham]

UK rail pricing penalises discretionary travel outside of forced purchases for commuting.

If someone is paying £6k (or £70 a day) to commute to work there is no incentive to spend even more money on the railway for leisure, unless you have to.
As has been mentioned numerous times, most households (and certainly those in the SE regularly commuting) have access to a car. That's a sunk cost with a small incremental expense for usage.

A season ticket is just a sunk cost and, given the changes in working patterns, it represents pretty poor value for money these days.

If, for a nominal fee (or nothing), you could use it to travel anywhere on the network then maybe we'd actually encourage some uptake & modal shift.

People might own a car, but not everyone does and quite a lot (if there were the right tools) could probably reduce the number of cars that they own.

In the example given, for someone doing another trip from Southampton to Reading rather that Woking (point of work) the return ticket price is £9.70 from Basingstoke to Reading.

Even without parking, given the round trip would be about 95 miles (and be around 20 minutes faster, even without any traffic delays) just fuel costs (unless you've got an EV) are likely to be comparable.

As such, it's very much the sort of travel which the railways should be making it easy to be the first choice of travel.

Yes that's not much money, however, the point is that by doing this it encourages a few percent more people to use rail for their non WFH travel and boosts other railway spending too, that's got to be good in reducing the cost of the subsidy for running the railways.

Likewise, if you could get students to use it (where they may not have to travel during peak hours every day) and it helps keep their costs of rail travel down, they are less likely to drive to college (yes they may still learn to drive and drive a patents car at other times, but the "need" to own a car is reduced). By reducing the cost advantage of driving they are likely to use the railways for longer. Again, increasing rail use and reducing the cost to the tax payer.

It's no magic bullet, however it's likely to be more useful than not.

 

[crablab]

People might own a car, but not everyone does and quite a lot (if there were the right tools) could probably reduce the number of cars that they own.

Most households do have access to a car.
If we (society) wanted to achieve that shift then we have to treat public transport as a public good and invest accordingly.

That’s where, for the main centres of population at least, a multi-mode zonal fares structure is an absolute must. That enables precisely that kind of flexibility at least in local areas, which would go a long way to satisfying most of the people most of the time

Indeed. And if you look at the map I linked above, it's not unrelated that car ownership is lower in areas with such public transport networks (there are other significant factors too, such as population density).

Yes that's not much money, however, the point is that by doing this it encourages a few percent more people to use rail for their non WFH travel and boosts other railway spending too, that's got to be good in reducing the cost of the subsidy for running the railways.

The majority of journeys are not from outside one station to the other and rely on multi-modal transport options. The quality, availablity and awareness of local connections (eg. bus, e-scooter, walking) affect whether an overall journey is perceived as desirable, or even possible, via public transport.

A journey which connects to a bus with a 2h frequency and takes 30% longer (assuming you make the connection) is not attractive, and people who can drive just will.

But I think we're generally agreeing. My point more broadly is that I don't think the current season ticket offering is compelling, and the market has moved on from being fleeced. We can see that reflected in the sales statistics.

There's many more significant structural problems with how we deliver & fund public transport in the UK, which affect how attractive it is as the first choice of transport, rather than last resort.

You could turn every season ticket into a Bahncard 100 equivalent and it still wouldn't move the needle on overall journeys, because convenience and connectivity are really important considerations. But it's pretty galling if you're paying Bahncard prices and only getting a ticket to get you 60 miles to London and back; that's definitely not going to incentivise people to choose the train over car for leisure travel.

(I am not, and have never been, an annual season ticket holder)

 

[Porty]

ScotRail say their busiest day of the week is Saturday.

...and at the May timetable change the North Berwick to Edinburgh Waverley service was doubled to twice hourly during the daytime on Saturday compared to only being at peak hours on Monday to Friday.

 

[The exile]

...and at the May timetable change the North Berwick to Edinburgh Waverley service was doubled to twice hourly during the daytime on Saturday compared to only being at peak hours on Monday to Friday.

A line where the importance of day-trip traffic has been recognised for years. Back in the 1970s when it was hanging on for dear life the service was, IIRC, peak hours only on weekdays but probably one DMU working back and forth all day on Saturdays.

 

[sh24]

Our household doesn't own a car, and haven't since 1999. Outside of London this revelation is treated with real shock and disbelief, often with a side order of have you been disqualified from driving. I manage a lot of UK travel by train, a bit of plane and a lot of taxis and Uber. It works for me but a lot of people would far prefer to be in their car, not least as they are big markers of status and social class.

 

[The Ham]

Most households do have access to a car.
If we (society) wanted to achieve that shift then we have to treat public transport as a public good and invest accordingly.

Indeed, however over 1 in 5 don't and that's not an insignificant number.

Indeed. And if you look at the map I linked above, it's not unrelated that car ownership is lower in areas with such public transport networks (there are other significant factors too, such as population density).

Indeed, although around 4 in 5 people live in an urban location.

Also, it's worth being aware of the UK definition of Urban, which is a settlement with a population of over 10,000. As such (and bearing in mind that for every small town of 9,000 people you need 3 villages of 3,000 or 18 hamlets of 500) a significant percentage of those who are rural are likely to live somewhere where public transport is viable.

The majority of journeys are not from outside one station to the other and rely on multi-modal transport options. The quality, availablity and awareness of local connections (eg. bus, e-scooter, walking) affect whether an overall journey is perceived as desirable, or even possible, via public transport.

Indeed, and that is something which government policy is trying to deliver.

A journey which connects to a bus with a 2h frequency and takes 30% longer (assuming you make the connection) is not attractive, and people who can drive just will.

Indeed, however 55% of people live within a 30 minute walk of a train station.

Also, a 0.5bph bus frequency is fairly rare, and (bearing in mind the other things I've said) is likely to be something which only a fairly small number of people would encounter.

But I think we're generally agreeing. My point more broadly is that I don't think the current season ticket offering is compelling, and the market has moved on from being fleeced. We can see that reflected in the sales statistics.

There's many more significant structural problems with how we deliver & fund public transport in the UK, which affect how attractive it is as the first choice of transport, rather than last resort.

You could turn every season ticket into a Bahncard 100 equivalent and it still wouldn't move the needle on overall journeys, because convenience and connectivity are really important considerations. But it's pretty galling if you're paying Bahncard prices and only getting a ticket to get you 60 miles to London and back; that's definitely not going to incentivise people to choose the train over car for leisure travel.

(I am not, and have never been, an annual season ticket holder)

Overall it may not move the needle by a significant amount, that wasn't the point of the suggestion.

The point was that even a tiny shift in behaviour could reduce the cost of subsidy of running the railways.

Even a 10% uplift in ticket spending nationally would be an extra £1bn (which if that was also a 1% shift in miles would mean nationally rail accouted for an extra 0.1% - as you say not shifting the needle).

Although that's a reduction on subsidy from £12bn to £11bn (circa 9% less).

Even if DfT only got to retain 10% of that, it could support an uplift in bus subsidy of around 5% and creating better multi modal options for more people.

Or put it another way, if we could reduce the rail subsidy to £0, that would be enough to increase defence spending without having to make any other cuts.

 

[absolutelymilk]

Even a 1% uplift in ticket spending nationally would be an extra £1bn (which if that was also a 1% shift in miles would mean nationally rail accouted for an extra 0.1% - as you say not shifting the needle).

Not sure how you've calculated this - passenger income is only a bit over £10bn so 1% of this would be £0.1bn

 

[Sonic1234]

could reduce the cost of subsidy of running the railways

And/or the cost of running the railways could be reduced. But no one wants that fight with staff, especially a Labour government.