Realistic outlook for new lines in London

[The Ham]

The lines out of Liverpool Street have had a massive uplift in capacity since 2013 with the introduction of new rolling stock and 240 metre trains, a much quicker and cheaper way of delivering new capacity than spending £billions on Crossrail 2.

In 2013 there was still an assumption that low economic growth was a temporary thing, not a rut that nobody would know how to get out of.

It is telling that the Stansted Airport expansion is going ahead on the assumption that existing capacity can cope with the growth.

Did you miss the bit where I said that the level of house building in the upper Lea Valley has been limited because of not having Crossrail 2.

First up, Airport flows are different to those generated by housing. For example a lot will be off peak or counter peak flow (i.e. adding more passengers to the emptier services out of London in the morning peak). As such not really adding significant numbers to the trains which are the busiest.

Of course the other factor is the scale of passengers, the airport is expected to add maybe 5 million passengers a year (the range is between 4 million and 5 million) which with an 18 hour a day, 4 trains an hour over 350 days a year means the average number of passengers per train will be about 100.

As such the impact isn't going tobe that big and so building Crossrail 2 just for the airport would be overkill.

Conversely building 150,000+ new homes (360,000 population increase) It's going to have a much bigger impact on rail use.

Rather than being 5 million, that could be 30 million (even a very low ball estimate using station usage at a station with high car ownership rates and a circa hour to a London terminus would put it comfortably over 10 million).

At 20 million (so between those estimates and potentially a little less than realistic) that's 4 times the numbers and mostly focused on peak (into London) services where the trains are already busy.

For example the Broxbourne Borough Local Plan 2018–2033 limited housing to circa 7,700, whilst under Crossrail 2 that district could deliver 17,000 new (additional) homes (range of 15,000 to 40,000, so 7,000 to 32,000 additional homes above those to 2033).

As I've said, there are also other areas (Central London passenger numbers, issues with connections at Clapham Junction for longer distance and outer suburban services, and the SWR routes) which also come into play so even if one or two are slightly off the original target that's less of a problem.

Especially given I've been comparing against the original 2033 target opening date, if that shifts back by between 7 and 17 years, then that means that passenger numbers need not to be so high. As such rather than Liverpool Street needing to be at 2030 numbers it can be at 2023 numbers.

From 2013, that's 10 years of growth (rather than 17), that means 137.2 million passengers for the London/East of England passenger flows rather than 150.0 million.

That's 7.2 growth between 2025 and 2030, which means my original point about getting to pre Covid levels at less than 2% growth per year (as 2% per year for 5 years is 10.4%) would mean that if we met the pre Covid numbers in or before 2030 then we'd be on track for the revised timeline of opening in 2040 and way ahead of 2050.

Politically, the chances of that are zero. This government isn't going to do it, and the government elected in 2028 or 2029 isn't going to do it either. Government's are going to struggle just to keep on spending on the projects that have already started.

Unpausing isn't likely to result in spending from Central Government any time soon, however would allow the taxes for Crossrail 2 to be started to be collected before construction starts with less political fall out.

For example, those current Crossrail levies if they just keep on being collected is less noticeable than if they stop and then restart.

Yes, there's going to be some who complain that they're paying for something which might not start for some time, but that's less noticeable than if they stop paying for (say) 5 years and then start again.

If by doing that it allows Central Government to shift a bit more of the cost to TfL then that's likely to be a positive thing.

Given this, that whilst we'renot that far off pre Covid passenger numbers and that by having a later opening date the target numbers are at least 7 years behind where they needed to be for the original opening date, I'm still convinced that it's not as dead in the water as some think it is.

Especially, given that there's a need for more housing and this could deliver over 100,000 additional homes over what could be delivered, just withinthe Hertfordshire and Upper Lea Valley areas.

Let alone (for example) helping to unlock the circa 35,000 new homes which are likely to be required over the next 15 years in the soon to be created North Hampshire unitary authority.

Let alone what it could unlock across the whole of the SWR area covered by the outer suburban and longer distance services where on average train frequencies could increase from 2tph to 3tph.

 

[Magdalia]

Did you miss the bit where I said that the level of house building in the upper Lea Valley has been limited because of not having Crossrail 2.

No I didn't.

The housebuilding in the upper Lea Valley hasn't been limited by lack of rail capacity, because that was provided by the introduction of 240 metre trains. There is plenty of rail capacity still unexploited.

What has limited the housebuilding in the upper Lea Valley is the absence of economic growth and the shift away from London commuting resulting from technological change. The latter of these is likely to continue with the application of artificial intelligence. Far fewer people want to live in the upper Lea Valley and commute to London, compared with what was forecast when Crossrail 2 was devised.

Housebuilders don't build if there is no demand. If there was demand they would be building and the new residents would be filling up the existing service.

Also, bear in mind that what little new demand that still exists for commuting into London from the north side is now likely to be met by Crews Hill New Town, something else that was unknown when Crossrail 2 was devised.

 

[DynamicSpirit]

No I didn't.

The housebuilding in the upper Lea Valley hasn't been limited by lack of rail capacity, because that was provided by the introduction of 240 metre trains. There is plenty of rail capacity still unexploited.

What has limited the housebuilding in the upper Lea Valley is the absence of economic growth and the shift away from London commuting resulting from technological change. The latter of these is likely to continue with the application of artificial intelligence. Far fewer people want to live in the upper Lea Valley and commute to London, compared with what was forecast when Crossrail 2 was devised.

I'm finding that rather hard to believe. I live in London and from personal experience of numerous people I know, London is full of people who are desperately looking for somewhere decent to live. People who are stuck in house-shares with far too little space for their need, or people who are forced to live with families or in boats, or even in often illegal overcrowded HMOs. And then there are all the people who would like to live in London but end up living in just about commutable places outside London because they can't find anywhere in London itself. It is not plausible that somewhere that would (with CR2) be so easily commutable into London wouldn't have massive demand for houses.

Housebuilders don't build if there is no demand. If there was demand they would be building and the new residents would be filling up the existing service.

Demand isn't the only thing that impacts house building. Building is constrained by the ability to get planning permission, the availability of suitable land, and the availability of skilled builders to build the houses. After all if you what said was true and housebuilders always built to meet demand, then there would not be a housing shortage in the UK. The fact that there is a massive housing shortage (and a particularly severe housing shortage around London) shows that demand by itself is not sufficient to cause houses to be built.

 

[JonathanH]

It is not plausible that somewhere that would (with CR2) be so easily commutable into London wouldn't have massive demand for houses.

With the best will in the world, Crossrail 2 is not being built in the commuting lifetime of people currently working in London.

People aren't going to move to the Upper Lea Valley now in the expectation of a railway existing in 30 or 40 years time, if ever, which is roughly the earliest it could be delivered.

As we have seen with the rise of populist politics, people want decisions that improve their life now, not in the distant future.

Given this, that whilst we'renot that far off pre Covid passenger numbers and that by having a later opening date the target numbers are at least 7 years behind where they needed to be for the original opening date, I'm still convinced that it's not as dead in the water as some think it is.

Where do you think the political impetus for Crossrail 2 is going to come from? It isn't something that the Mayor of London can progress without central government support. It isn't something Network Rail can deliver and it is way too big for the private sector.

It doesn't pass the populist agenda because it involves spending now for gain in the distant future.

 

[Magdalia]

It is not plausible that somewhere that would (with CR2) be so easily commutable into London wouldn't have massive demand for houses.

All of the Lea Valley is easily commutable now. People who want to move there don't need to wait for Crossrail 2.

Building is constrained by the ability to get planning permission, the availability of suitable land, and the availability of skilled builders to build the houses.

Crossrail 2 doesn't change any of these. For skilled construction workers and construction raw materials Crossrail actually makes matters worse, because it would be competing with housebuilding for some of the same resources.

London is full of people who are desperately looking for somewhere decent to live. People who are stuck in house-shares with far too little space for their need, or people who are forced to live with families or in boats, or even in often illegal overcrowded HMOs. And then there are all the people who would like to live in London but end up living in just about commutable places outside London because they can't find anywhere in London itself.

Here in the Fens, Cambridge has all of these issues too. But houses are being built, in Cambourne, Northstowe, Waterbeach and Eddington. None of those are waiting for some magic transport infrastructure to start building.

 

[Bald Rick]

All of the Lea Valley is easily commutable now. People who want to move there don't need to wait for Crossrail 2.

Indeed so, and in many places on the line, house prices are comparatively cheap compared to other locations a similar distance out of London.

 

[The Ham]

Here in the Fens, Cambridge has all of these issues too. But houses are being built, in Cambourne, Northstowe, Waterbeach and Eddington. None of those are waiting for some magic transport infrastructure to start building.

The house building for Cambridgeshire and Peterborough as a whole is for an area of around 1,440 square miles, to fit in around 150,000 homes.

Compare this to around 15 square miles to fit in 89,000 homes.

Now consider why the former might not need a high density rail transit system to facilitate that nunnery of homes whist the later might.

Especially, given that quite large areas of Cambridgeshire and Peterborough are significantly more rural with lower traffic density for much of the area.

Where there's issues they can be delivered a much lower cost, due example the The Network Rail Cambridge south infrastructure enhancements which have a price tag of around £250 million and provide a lot of the capacity needed in the vicinity.

For example giving not only rail travel to London but also acting as a park and rail for local employment centres.

However, Crossrail 2 isn't only about the Lea Valley but the whole of London and large areas across the SWR area.

== Doublepost prevention - post automatically merged: 24 May 2026 ==

Indeed so, and in many places on the line, house prices are comparatively cheap compared to other locations a similar distance out of London.

Indeed, but with a further 89,000 homes (200,000+ people) that's potentially a lot of rail use.

Even for somewhere with 10,000 people and scaling up from that where that's got high car ownership and low housing density that's potentially 16 million rail passengers a year (and the Upper Lea Valley isn't likely to see that kind of low density development and high car ownership rates).

Original estimates (pre Covid) had numbers of around 40 million, accounting for changes to working habits might shift it to 30 million, so 16 million is a very low ball estimate.

As I said, by looking to be just 7 years later (and it could be 17 years later) the baseline numbers shift significantly.

Whist 7 years means a bit of growth to get to pre Covid levels, much beyond that and we're potentially above where we need to be to be on track to be needing to open around (say) 2046 with lowish (1.2% growth per year) for the eastern side.

 

[Recessio]

However, Crossrail 2 isn't only about the Lea Valley but the whole of London and large areas across the SWR area.

Indeed, I always thought Crossrail 2 main goal was mostly about the SWR Metro area (and freeing up capacity going into Waterloo), and that improvements to the Lea Valley side of things was just an added bonus?

 

[Silent]

Hi everyone, I know that this is a very vast topic but I just want to align on how bad things are as of today

I want to understand your outlook on the key 3 current projects in London: DLR to Thamesmead, West London Orbital and Bakerloo line extension, as well as a few more speculative ones. I propose the following categories depending on when you would expect the official go ahead: 2020s, 2030s, 2040s and “not in our lifetime”

The list of projects is as follows:
1. DLR to Thamesmead
2. WLO
3. BLE
4. Crossrail 2
5. Southern rail approach to Heathrow
6. Croxley link
7. Northern Line extension to Clapham Junction
8. Tramlink to Sutton

I would arrange them like this:
2020s: 1, 2, 3
2030s: 4, 5, 6, 7, 8

I appreciate that my view is quite optimistic. Please feel free to add more projects or just shoot everything down if you think that everything apart from the DLR extension is highly speculative. It would be great if you can reference specific political/economic justifications for the budget allocation.

DLR to Thamesmead. I don’t live in South east so don’t use it much but would expect it to be extended maybe early/mid 30s.

WLO, probably mid to late 30’s. Just making the new trains and stations seems like something that would take some time. Though I actually think this might still be speculative.

BLE. I think I would expect this to be done from the late 30’s to the 50s. The first section. I have read before that the earliest the Bakerloo line would receive new trains is actually the mid 30s because of all the prior work that would need to be done, looking at crossrail as well it seemed making the core section took a decade or longer to build. I imagine if the new Bakerloo line are similar to the 24’ ts they can pre plan the stations to fit them.

Crossrail 2 I imagine at earliest, maybe mid 40s, it’s still speculative I think, as it would probably take at least 10 years to build the core section. The northern line extension to Clapham junction wouldn’t happen without it but I would expect that to only take a few years to do.

== Doublepost prevention - post automatically merged: 24 May 2026 ==

Some extra things I’d like to see include:
-Victoria line extension southbound beyond Brixton;
-Splitting of the Northern line (TfL want this too, but it’s a long-term goal);
-If Crossrail 2 is ever built, then the taking over of Thameslink by TfL and bringing it under the Crossrail brand;
-DLR extension westbound from Bank (maybe to the old Jubilee line platforms at Charing Cross, or north towards Euston/St Pancras?).

Some of these were planned at one point or another but have since been shelved. It’s more of a wish list than anything else!

The only issue I have with Victoria line extension is that it has to be done really carefully to not affect its 36tph frequency and reliability.

== Doublepost prevention - post automatically merged: 24 May 2026 ==

Wouldn't Crossrail 2 provide capacity release between Victoria and Euston on the Victoria Line and so an extended Victoria Line could work in a post Crossrail 2 world.

I’m wondering if an issue with building relief line over relief line is that if the crossrail 2 has severe delays or part suspension the extended Victoria line might be overcrowded. So despite having relief lines maybe the lines should be built to operate higher capacity than expected.

 

[Magdalia]

The house building for Cambridgeshire and Peterborough as a whole is for an area of around 1,440 square miles, to fit in around 150,000 homes.

Compare this to around 15 square miles to fit in 89,000 homes.

Now consider why the former might not need a high density rail transit system to facilitate that nunnery of homes whist the later might.

Especially, given that quite large areas of Cambridgeshire and Peterborough are significantly more rural with lower traffic density for much of the area.

Where there's issues they can be delivered a much lower cost, due example the The Network Rail Cambridge south infrastructure enhancements which have a price tag of around £250 million and provide a lot of the capacity needed in the vicinity.

For example giving not only rail travel to London but also acting as a park and rail for local employment centres.

However, Crossrail 2 isn't only about the Lea Valley but the whole of London and large areas across the SWR area.

All of this is irrelevant to Crossrail 2. Area and targets do not create housing demand. The key point is that Cambridge has housing demand, and houses get built. If the Lea Valley had housing demand then housebuilders would be building there too, without the need for Crossrail 2.

Cambridge South is for employment, not housing. The nearby housing in Trumpington was built years ago, without any need for the new station. Cambridge South doesn't provide any capacity for current housebuilding in Eddington, Cambourne or Northstowe.

And you clearly don't know anything about traffic congestion in Cambridge, the city with less than 5 miles of dual carriageway. Traffic density is very unevenly distributed, with most of it wanting to be in the same places.

 

[SynthD]

It is telling that the Stansted Airport expansion is going ahead on the assumption that existing capacity can cope with the growth.

Is that the case, or do they just not want to pay for what is necessary?

I imagine if the new Bakerloo line are similar to the 24’ ts they can pre plan the stations to fit them.

Yes, same train but a different length. It should be possible to shorten a single one and send it to Queens Park for dynamic testing.

Area and targets do not create housing demand. The key point is that Cambridge has housing demand, and houses get built. If the Lea Valley had housing demand then housebuilders would be building there too, without the need for Crossrail 2.

Barking Riverside had permission to build 4000 homes before the DLR/Overground station opened, and another 16000 after. Lea Valley will have its own numbers, though I suspect the first is loosely defined for now.

 

[Magdalia]

Is that the case, or do they just not want to pay for what is necessary?

It is based on Greater Anglia data for 2024 quoted in the planning application.

The seated only loadings are:

up road AM peak 42.5%, PM peak 31.4%, daily 28.6%
down road AM peak 22.9%, PM peak 51.2%, daily 27.5%

 

[The Ham]

All of this is irrelevant to Crossrail 2. Area and targets do not create housing demand. The key point is that Cambridge has housing demand, and houses get built. If the Lea Valley had housing demand then housebuilders would be building there too, without the need for Crossrail 2.

House builers can not build of their development results in significant traffic congestion, it's why they almost every development has a Transport Statement (generally smaller developments) or a Transportation Assessment (larger developments).

Without the those reports (which depending on the complexity of the development can run to 10's of thousands of pounds) the development can't happen.

Building a small city (which is what you're taking about in the Upper Lea Valley) is very different to building a few thousand homes in each district across a vast area where each development may be quite a bit smaller.

One key difference is that if a development is less than 2% of traffic at a given junction that it's not all likely to do something to fix that junction, unless doing that releases capacity elsewhere.

Whist some developments will need infrastructure built in advance, others will have it retro fitted after so that by providing a modal shift from the existing car use to rail will allow the development to happen.

This is the case for the Cambridgeshire infrastructure project, the housing came first, but now the rises are full the rail is happening to allow further development.

As you say that's very different to Upper Lea Valley where the roads and rail are already significantly used with other development using up the additional capacity put in (i.e. Stansted filling up the longer trains, primarily off peak and/or counter peak).

Development which come forwards first and gets planning permission will get to use the available capacity and those which are follow then have to do more to mitigate against what they are delivering.

Whilst a small city will have a lot of internalised movements (i.e. not driving on external roads to get to shops, schools, local facilities,.etc) that's only goingto get you so far.

Likewise, WFH has changed things, it's why rail isn't being used as much as it was. However, there's still a lot of employment which requires people to be in person, be that full time, or a few days a month.

Also, WFH also can change the economics of needing to have 2 cars in a home. For example, if there's two people working full time in a fixed location justifying two cars could be fairly easy.

However, if they each only need to be in 3 days a week, then most of the time one person WFH whist the other drives and it's then only once (or twice) a week when both need to be in a fixed location where then it maybe that they then look to use rail as its potentially cheaper than running a second car which isn't used much beyond that commuting.

Likewise, if you only have to be in in place a few times a year, or even a few days a month, then people may travel much further out and travel in by rail, but potentially off peak if that with an overnight stay is cheaper than paying peak rates.

Yes, that's still lower than it was, but if does mean that it's not as low as it could have been.

However, that changein working will also impact the developments in Cambridgeshire and Peterborough, and so what would have been required on 2018 might now not be required in 2028 as WFH has reduced road traffic volumes.

It's made doing transport planning a very different activity.

It's also means that the old 1 house equals 0.67 vehicle movement in the morning peak guide now a bit too high in some areas (mostly where WFH jobs are typical for those living there, but obviously not where WFH may not be offered for the typical jobs which residents of that area do) but crucially not everywhere.

Obviously, one thing which could reduce this rate of car use is by significantly increasing rail capacity along a corridor, especially if that gives an increase in frequency (going from 4tph to 6tph will be more attractive than going from 600 seats per train at 4tph to 1,200 seats per train even if the number of extra seats is higher for the lengthened services over the extra frequency.

 

[Magdalia]

House builers can not build of their development results in significant traffic congestion

That may be the case in the Lea Valley, but try telling that to the people of Cambridge. Each development makes an incremental contribution to congestion, but the effect is cumulative, and may not be nearby. Planning is a district council responsibility in two tier areas, Cambridgeshire is not a planning authority and is irrelevant. We do have Greater Cambridge planning for Cambridge City and South Cambridgeshire, the latter completely encircling the former, but anywhere beyond can give permission to build without worrying about how they are going to increase traffic congestion in Cambridge. There is a lot of commuter traffic on the arterial roads coming into Cambridge, generated by planning decisions made elsewhere in East Cambridgeshire, Huntingdonshire, North Hertfordshire, Uttlesford and West Suffolk.

Whilst a small city will have a lot of internalised movements (i.e. not driving on external roads to get to shops, schools, local facilities,.etc) that's only goingto get you so far.

These are also a major contribution to the traffic congestion in Cambridge, especially schools. But they are not going to make or break the case for Crossrail 2.

Building a small city (which is what you're taking about in the Upper Lea Valley) is very different to building a few thousand homes in each district across a vast area where each development may be quite a bit smaller.

That's what is happening with Cambourne, with significant further expansion in the draft local plan, and currently it has no significant transport infrastructure, apart from the A428. The Lea Valley has 2 railways going through it with spare capacity

However, that changein working will also impact the developments in Cambridgeshire and Peterborough, and so what would have been required on 2018 might now not be required in 2028 as WFH has reduced road traffic volumes.

The mix of employment in Cambridge is much less affected by WFH than London. Scientists doing research on high tech lab equipment on the biomedical campus don't do it at home. The healthcare staff in the major hospitals on the biomedical campus don't do it at home.

In Cambridge traffic volume is still increasing and congestion is getting worse.

Development which come forwards first and gets planning permission will get to use the available capacity and those which are follow then have to do more to mitigate against what they are delivering.

In the Lea Valley, stage one happening at pace needs to be happening before considering whether there is any prospect of ever reaching stage two. Now that the existing rail has more capacity, WFH has reduced commuting by up to 20%, and with the impact of AI still to come, I don't see stage two ever being reached.

 

[aron2smith]

All of this is irrelevant to Crossrail 2. Area and targets do not create housing demand. The key point is that Cambridge has housing demand, and houses get built. If the Lea Valley had housing demand then housebuilders would be building there too, without the need for Crossrail 2.

I think you miss the point, Lea Valley and all across North London and the immediate surrounding areas have significant demand for housing. Which is why towns like Cheshunt are growing and East Herts wants to build Gilston near Harlow. The existing infrastructure barely copes as it is, that and a lack of genuinely affordable housing are the 2 main concerns when people complain about development across the south east. The lack of a regional body coordinating the local plans of each council down here and protecting the London green belt as a whole is a significant block on progress. People want to live and work here, those of us from or near London want to stay where we are from. The fact that London boroughs dump those in need on towns like Stevenage, Harlow and Basildon and so on, pricing out people from those places shows there is a very serious problem.

I think what is telling is with Meridian Water, the station was upgraded and the land is there empty for years, building has been very slow despite a long waiting list and strong demand for housing. To me it stinks of corruption but also the increasing inability of councils to do things for their areas. The split between 3 counties and numerous councils also doesn't help.

I think Crossrail 2 is part of the same issue, it's structural in nature. Everyone knows it's needed but the ability to deliver is lost. Until that impasse is done with, things are stuck. In it's place, fragmented ribbon development, inadequate local services and increased pressure on what exists. Thing is Crossrail 2 is multiple projects in 1, it's 4 tracking up here (a major limit on growth on its own), it's increased capacity for 3 of London's busiest stations and it's about new connections. It allows for more local control across SW London's rail lines too. Direct connections to Clapham Junction and Kingston would be extremely useful. In Cheshunt, some of the new developments were based on this coming down the way, now there is grave uncertainty. While what exists is busier each year. There's only so many improvements to the timetables (eg Weaver Line or even Hertford Loop through nearby Cuffley) before things need an upgrade. The Victoria Line has no space to deal with its growth in demand either.

Also Crews Hill was a bit of a political issue in the borough election (only Labour publicly supported it), now there is grave uncertainty as Enfield Council is No Overall Control. Conservatives short of a majority with some unexpected Green gains. Strangely no ideas out there of who will run the council still.

Also with the Lea Valley, we can't simply build places like Northstowe without the ever essential infrastructure, geography also limits the available space. And other than the dual carriage A10, our main road is a 2 lane street with no alternatives, we need that great improvement in transport, buses included.

And I would say Crossrail 2 is pretty essential and the Elizabeth Line clearly shows this to be the case but I agree with OPs list to put it below DLR, WLO and the umpteenth attempt to extend the Bakerloo Line. For the Bakerloo, maybe this time. While Crossrail 2 is still very important and shows we really need some kind of regional body with the power to fund and build it itself, London is too small for this project and Westminster far too removed for it to ever happen from them. Same reason why trams and trains struggle to happen across the country.

 

[The Ham]

That may be the case in the Lea Valley, but try telling that to the people of Cambridge. Each development makes an incremental contribution to congestion, but the effect is cumulative, and may not be nearby. Planning is a district council responsibility in two tier areas, Cambridgeshire is not a planning authority and is irrelevant. We do have Greater Cambridge planning for Cambridge City and South Cambridgeshire, the latter completely encircling the former, but anywhere beyond can give permission to build without worrying about how they are going to increase traffic congestion in Cambridge. There is a lot of commuter traffic on the arterial roads coming into Cambridge, generated by planning decisions made elsewhere in East Cambridgeshire, Huntingdonshire, North Hertfordshire, Uttlesford and West Suffolk.

The issue you are experiencing is down to the fact that once traffic is less than 2% the planning system can't easily deal with it.

It's why (arguably) a single large site of 5,000 homes is actuality better for the existing communities than 25 sites of 100 homes, even though it had twice the number of homes.

These are also a major contribution to the traffic congestion in Cambridge, especially schools. But they are not going to make or break the case for Crossrail 2.

Indeed, the point I was making was that for a single large development there's a lot of movements which wouldn't use the existing infrastructure, butfor smaller sites would.

If you build 5,000 homes and 100% of them drive their kids to school before driving home, (as the schools would almost entirely be within the development) the impact on the existing roads is effectively zero.

However, those 25 sites of 100 homes, anyone who drives to schoolis then extra traffic on the existing network and so adds to the existing congestion.

Yes, it's gong to have zero impact on it Crossrail 2 is or isn't built. However, the point of talking about it, was to highlight how large developments are different to small, and how a single (small city) development would require different infrastructure than the same number of houses (or even more) but over a much larger area.

That's what is happening with Cambourne, with significant further expansion in the draft local plan, and currently it has no significant transport infrastructure, apart from the A428. The Lea Valley has 2 railways going through it with spare capacity

Cambourne currently has what planned? Circa 2,500 homes.

That's about 6,000 people once fully occupied.

Yes, there's talk of more (14,000), but only post East West Rail and C2C bus way - the latter will limit development (like Crossrail 2) if its not brought forwards.

The mix of employment in Cambridge is much less affected by WFH than London. Scientists doing research on high tech lab equipment on the biomedical campus don't do it at home. The healthcare staff in the major hospitals on the biomedical campus don't do it at home.

In Cambridge traffic volume is still increasing and congestion is getting worse.

Indeed, which is why local communities need to look at small scale improvements to fix barriers to people making their first choice of travel something other than getting in the car.

For some this could be obvious (like a controlled crossing), for others it's more subtle (like having seating so they can rest every 400m).

If the local communities had this, when developments happened the communities would more likely get these improvements, rather than something which wasn't actually helpful.

Locally to me a previous development put in traffic islands (no space for pedestrians), however they should have made them suitable for pedestrians, and with two short sections of footway would have provided a good link between two walking routes and the new development. As it is, it's a mess and difficult to use unless you are able bodied and are willing to walk a route which isn't obvious.

In the Lea Valley, stage one happening at pace needs to be happening before considering whether there is any prospect of ever reaching stage two. Now that the existing rail has more capacity, WFH has reduced commuting by up to 20%, and with the impact of AI still to come, I don't see stage two ever being reached.

Indeed, however if you recall I did say pre covid the talk was of 40 million, a 20% reduction would therefore be 32 million (again I mentioned use of 30 million, do did factor for that), however I'd gone even further and used rail use for somewhere with high car ownership rates, high WFH rates and fairly low traffic congestion to highlight that even with that it's still 20 million passengers.

That's significantly higher than the 6 million (which are mostly using up non peak capacity) linked to Stansted Airport's plans.

To put that into perspective, roughly 10% of all travel is in the morning peak hour. Given there's 250 weekdays (actually excluding bank holidays it's 253, but we'll just stick with 250) what does that look like in terms of peak demand?

10% of 20 million is 2 million.

Divide that by 250 and that's 8,000 movements in the morning peak.

Let's also say that it's 50:50, so half those moving in the morning peak are heading towards the new development and half are heading away.

That's 4,000 people, with empty trains with capacity for 1,250 people that's 3tph.

Is there that level of capacity in the Upper Lea Valley for rail passengers (and bearing in mind we've halved the numbers previously assumed, so AI would have to cut 37% from the current reduced rate due to WFH).

Happy to discuss where I may be being overly optimistic on rail use given the assumptions used.

However, as I've also said, a few more years of growth due to delaying the start by between 7 and 17 years means that some of those falls in rail use become irrelevant.

A 40% fall (due to WFH and AI) is still wiped out by 2% growth over 17 years, that compares with East of England to London growth from 2024 to 2025 of 5.7% (I'm not going to use other growth since 2020 as that's much higher due to Covid recovery) even if that rate of growth reduces that's still likely to be at least a few years of growth above 2% when the Crossrail 2 model was assuming less than 1.5%.

Again, happy to discuss where I might be putting forwards numbers which are too high, as I feel I've be robust in trying to not over state anything.

 

[Magdalia]

It's why (arguably) a single large site of 5,000 homes is actuality better for the existing communities than 25 sites of 100 homes, even though it had twice the number of homes.

This is why Greater Cambridge have Cambourne, Northstowe and Waterbeach, all single large sites with more than 5000 homes.

Cambourne currently has what planned? Circa 2,500 homes.

That's about 6,000 people once fully occupied.

You are a long way out here. Cambourne already had a population of 11000 in the 2021 census, and continues to grow rapidly. The latest plans for expansion of Cambourne are here:

https://consultations.greatercambridgeplanning.org/sites/gcp/files/2025-10/EBGCLPCPFSP1Oct25.pdf

Cambourne’s Spatial Framework Strategy 1 The Spatial Framework Strategy describes the opportunity to grow Cambourne to become a thriving regional town which is renowned for its natural environment.

Eventually Cambourne is likely to have 25-30k homes and a population up to 50k.

Lea Valley and all across North London and the immediate surrounding areas have significant demand for housing. Which is why towns like Cheshunt are growing and East Herts wants to build Gilston near Harlow.

I have to confess that I'd not heard of Gilston until last week, when it was mentioned in the context of the Hertfordshire Essex Rapid Transit proposal. So I looked it up, see here:

Gilston – Harlow and Gilston Garden Town

hggt.co.uk

Gilston​

Located north of Harlow and across the River Stort in East Hertfordshire, the Gilston area of the Garden Town will feature 10,000 homes and associated infrastructure for seven new villages.

I can see why I hadn't heard of Gilston: in the context of Crossrail 2 it is not significant. Gilston, with 10000 new homes, is smaller than Waterbeach New Town

Waterbeach New Town

Waterbeach New Town planning development in South Cambs

www.greatercambridgeplanning.org

Planning progress​

Western Parcel (Urban&Civic)

• Outline planning permission for up to 6,500 homes, employment space, schools, community facilities, and infrastructure was granted in September 2019
• The development is being delivered in phases, in line with approved design codes
• The Design Code for Phase 1 (up to 1,600 homes) was approved in June 2020
• The first phase of infrastructure works is underway. The first housing scheme was approved in July 2021.

Eastern Parcel (RLW Estates)

• Outline Planning Permission for up to 4,500 homes, plus other uses and infrastructure was approved December 2024.

Waterbeach is getting a new station. Gilston justifies a similar level of investment, which appears to be a "sustainable travel corridor":

Work Begins on Garden Town Travel Route – Harlow and Gilston Garden Town

hggt.co.uk

Work is set to get underway this month on new transport infrastructure to support Harlow & Gilston Garden Town.

I think what is telling is with Meridian Water, the station was upgraded and the land is there empty for years, building has been very slow despite a long waiting list and strong demand for housing. To me it stinks of corruption but also the increasing inability of councils to do things for their areas. The split between 3 counties and numerous councils also doesn't help.

I pass through Meridian Water on the train, and I'm always interested in what I can see there. Two key points impact on how much development takes place, irrespective of the demand that you assert exists. One is supply of building companies and skilled construction workers. This creates a capacity constraint that limits the speed of development. The other is house prices, which in the London area are falling. This makes houses a bit more affordable, but less profitable to build, so the available building capacity goes elsewhere. Furthermore, flooding a falling market with new houses could lead to a much bigger fall in house prices, which, while good for first time buyers, can plunge others with big mortgages into negative equity.

Also Crews Hill was a bit of a political issue in the borough election (only Labour publicly supported it), now there is grave uncertainty as Enfield Council is No Overall Control. Conservatives short of a majority with some unexpected Green gains. Strangely no ideas out there of who will run the council still.

Also with the Lea Valley, we can't simply build places like Northstowe without the ever essential infrastructure, geography also limits the available space. And other than the dual carriage A10, our main road is a 2 lane street with no alternatives, we need that great improvement in transport, buses included.

Crews Hill is much more like Cambourne, Northstowe and Waterbeach than the Lea Valley. Furthermore, as a New Town, the government will be able to designate it as national infrastructure. For these reasons I think that Crews Hill is more likely to develop quickly than Lea Valley, with an impact on where rail investment will be needed.

Happy to discuss where I may be being overly optimistic on rail use given the assumptions used.

I haven't the time or inclination to mark your homework but a few things to highlight are:

Divide that by 250 and that's 8,000 movements in the morning peak.

But that is not a big number. Each 10 car class 720 has more than 1000 seats. The Lea Valley has 10tph if you count Stansted, Cambridge and Hertford trains. Over a 3 hour peak that's more than 30000 seats. 8000 more movements is an increase of about 25%, well within what can be absorbed by the existing service.

A 40% fall (due to WFH and AI) is still wiped out by 2% growth over 17 years, that compares with East of England to London growth from 2024 to 2025 of 5.7%

These growth assumptions are heroic, especially if you remember that the 2024/25 on 2023/24 comparison is hugely affected by the level of strike action in 2023/24.

And it still only gets back to 2018/19 traffic levels, more evenly spread through the day, and with a significant increase in capacity in the meantime.

 

[Farnborough]

To put that into perspective, roughly 10% of all travel is in the morning peak hour.

Depending on how you define the morning peak, that may not be the wow! statistic you thought it might be!

10% of a day is 2.4 hours (2 hours and 24 minutes) so unless your morning peak is significantly shorter that this, you're into statistical noise from the average...

For example, suppose two hours for the morning peak (10% over two hours, 5% per hour), and two hours for the evening peak (10% over two hours, 5% per hour) that leaves 80% across the other 20 hours, or 4% per hour!

 

[The Ham]

Depending on how you define the morning peak, that may not be the wow! statistic you thought it might be!

10% of a day is 2.4 hours (2 hours and 24 minutes) so unless your morning peak is significantly shorter that this, you're into statistical noise from the average...

For example, suppose two hours for the morning peak (10% over two hours, 5% per hour), and two hours for the evening peak (10% over two hours, 5% per hour) that leaves 80% across the other 20 hours, or 4% per hour!

Specifically I said

10% of all travel is in the morning peak hour.

I.e. exactly 60 minutes, and generally 8am to 9am (although the time period can change, and with rail commuting the percentage can be higher).

The thing is yes you're right the flows over the rest of the day can be tiny - but the rest of the day almost doesn't matter.

To give a practical example; very few people care if there's congestion at 2am. Unless that traffic is going past their houseand potentially disturbing their sleep adding 500 cars to any road at 2am isn't going to impact anyone.

Why, well two reasons if there's congestion it doesn't impact them, and secondly capacity is such that it's unlikely that those 500 cars would create any congestion.

Conversely, adding 250 cars to the roads in the morning peak HOUR is going to cause significant congestion.

It's why generally developments only really need to look at their impact during the morning and evening peaks, with some types of development then testing other periods (like retail will also look at Friday evening and Saturday day times) when the traffic generation for the site is highest, but the traffic use on the wider network may not be as busy as a weekday morning.

It's why assessments are on the busiest days (even pre Covid typically it may have excluded Monday evenings and Friday mornings) and the busiest time period as well as potentially the busiest train or bus service.

This is why Greater Cambridge have Cambourne, Northstowe and Waterbeach, all single large sites with more than 5000 homes.

You are a long way out here. Cambourne already had a population of 11000 in the 2021 census, and continues to grow rapidly. The latest plans for expansion of Cambourne are here:

https://consultations.greatercambridgeplanning.org/sites/gcp/files/2025-10/EBGCLPCPFSP1Oct25.pdf

Sorry, I wasn't clear, 2,500 new homes in the current local plan period (now to 2041).

The point being that what's existing is coping with the existing infrastructure (there's a separate debate as to if there should have been more infrastructure in the past) and it's the extra.

If we're gong to compare existing populations, the Upper Lea Valley is circa 750,000.

Obviously, existing population will have an impact on the need for infrastructure, it also has an impact on future use of new infrastructure.

Eventually Cambourne is likely to have 25-30k homes and a population up to 50k.

Indeed.

Existing 4,600 homes
Plus 2,500 through to 2041
Plus the 14,000 beyond that which are being talked about (and may well be higher)

Total of at least 21,000 homes and even at that level of housing 50,000 people could be likely (2.4 people per dwelling isn't uncommon, depending on demographics, although it's likely to reduce due to falling numbers of children).

Although, again, the beyond 2041 numbers are reliant on new infrastructure in the form of East West Rail and the C2C bus infrastructure project.

However, compared to the Upper Lea Valley where the existing population is circa 750,000 and the extra homes (89,000) are adding a further circa 200,000 people to that, the level of scale is different.

If one were to charge developments £2,500 for each new home for infrastructure improvements Cambourne would get £41 minion (note this isn't far off the circa £37 million from developer funding expected forthe C2C bus project).

Upper Lea Valley would get £220 million.

Now whilst that's peanuts compared to the cost of Crossrail 2, it's only one small element of that project.

However, does further highlight the differences in scale.

I haven't the time or inclination to mark your homework but a few things to highlight are:


But that is not a big number. Each 10 car class 720 has more than 1000 seats. The Lea Valley has 10tph if you count Stansted, Cambridge and Hertford trains. Over a 3 hour peak that's more than 30000 seats. 8000 more movements is an increase of about 25%, well within what can be absorbed by the existing service.

Whilst later on (where breaking down the numbers) I wasn't clear (as I did only use the term "morning peak") when I initially talked about the 10% of flows it was in the context of peak hour (I.e. 60 minutes).

Likewise, when i talked about the impact on the train services (and for clarification this was still in the context of one hour) the difference in passenger numbers we are both using is small:

That's 4,000 people, with empty trains with capacity for 1,250 people that's 3tph.

Note that the difference between the 8,000 in the extract quoted and this extract were I use 4,000 for train capacity is allowing for 50% of movements to be arriving into the Upper Lea Valley and 50% for departures.

Also, that's already assuming a 50% reduction vs what was being assumed pre Covid (pre Covid was 40 million passenger movements and that's on 20 million, when WFH's 20% fall would be 32 million).

Also, whilst a 720 could probably manage to carry more than 1,250 people, the "official capacity" when full and standing is 1,252 (C2C configuration) and up to 1,435 (Greater Angela configuration for the highest layout) 4,000 isn't actually 3×1,250, but rather 3×1,333.

3tph (all empty) is therfore about 30% of the existing train capacity of 10tph.

Whilst that's smaller than the circa 50% increase in capacity from delivery of the 720's, there's a few things to consider:
- new rolling stock is generally only delivered when trains are already full, so at least some of that 50% was used up on day one
- rail use on Greater Angela services in 2025 is the same as 2019 (using quarterly data, see below)
- whilst (say) 20% reductions in commuting have been seen due to WFH, that's not equal across the week and so someone needing to travel on a Tuesday could be left behind due to trains being full, but gets a choice of seats on a Monday morning (again the busiest services are the ones which need to be assessed)
- if the fall in rail use for the new homes isn't as much as 50% when comparing to pre Covid assumptions, let's say it's 33%, then instead of 3tph of empty trains capacity it's 4tph of empty trains capacity
- we've assumed zero impact from increasing frequencies for a post Crossrail 2 world on the travel habits of the existing ppopulation (which is unlikely)

These growth assumptions are heroic, especially if you remember that the 2024/25 on 2023/24 comparison is hugely affected by the level of strike action in 2023/24.

And it still only gets back to 2018/19 traffic levels, more evenly spread through the day, and with a significant increase in capacity in the meantime.

First off Greater Angela (rolling 4 quatre) had a peak annual flow of 86.7 minion passengers (period to the end of December 2019), for the latest data (to the end of December 2025) it's 86.1 million.

Whilst there's going to be changes in flows, if rail commuting has seen the following for peak hour travel:

Monday was 100, now 50
Tuesday was 100, now 100
Wednesday was 100, now 100
Thursday was 100, now 100
Friday was 100, now 50

That was 500, and is now 400, in percentage terms that's a 20% fall, however the busiest trains are still at 100 and so if you're looking to add more people there's not the extra space that 20% (overall) fall due to WFH would imply.

Yes it's an overly simplified example, but then is saying that WFH means 80 on every day, the reality is somewhere between the two. Again, the point being that the busiest services are the ones which are those which need to be assessed.

On the annual growth for the calendar year 2024 to 2025 (using the quarterly data) seeing growth of 6.1% is high. Yes, the strikes will have had a significant impact on that rate (as will, still, Covid recovery) and it's why I've always referencered the growth rate and then talked about potential growth rates which are much smaller.

Yes, 2% could be too high (and certainly higher than the assumed circa 1.3%), however that was also in the context of a combined WFH and AI fall in rail use of 40% from the 2019 peak AND assuming that it was evenly spread over the working week.

As such, it's worth looking at where we currently are at, based on the latest numbers, basically back to pre Covid, whilst there's scope for there to be impacts within that total for falls in commuting (including the 20% referenced), and whilst AI may well have an impact, it's likely that over the next 5 years (which is the timeline I've mentioned about unpausing Crossrail 2 with then the potential for services starting between 2040 and 2050) we should be able understand more clearly the impact and the growth we see during 2026-2028.

Yes Crossrail 2 may still be unlikely to happen.

However, I've never said that isn't the case, I've been saying that when looking at opening at a time beyond the original 2033 opening date, the falls we've seen doesn't necessarily mean that Crossrail 2 is never gong to happen and may still be something which is planned to be delivered in the first half of the 21st century.

Especially if the falls seen don't necessarily have a significant capacity reduction on the busiest services.

 

[Instinct04]

But that is not a big number. Each 10 car class 720 has more than 1000 seats. The Lea Valley has 10tph if you count Stansted, Cambridge and Hertford trains. Over a 3 hour peak that's more than 30000 seats. 8000 more movements is an increase of about 25%, well within what can be absorbed by the existing service.

The Stansted and Cambridge trains do not stop anywhere in London along the Lea Valley (other than Tottenham Hale), rendering them useless for all commuters in Enfield, including ones forcast as part of the Meridian Water development.

All the ones that currently do stop are AT CAPACITY. Hence why many morning bound Hertford East trains no longer call at Cheshunt, they do this to spread passenger loadings easier.

 

[Magdalia]

First off Greater Angela (rolling 4 quatre) had a peak annual flow of 86.7 minion passengers (period to the end of December 2019), for the latest data (to the end of December 2025) it's 86.1 million.

You have quoted these numbers before. I have already explained that the aggregate position for Greater Anglia hides some significant differences. The growth is mainly on the rural routes, Norwich-London and the Stansted Express. For commuting traffic in and out of Liverpool Street, from places like Bishops Stortford and Harlow Town, traffic is still about 20% down, and that's what matters for the case for Crossrail 2.

Whilst there's going to be changes in flows, if rail commuting has seen the following for peak hour travel:

Monday was 100, now 50
Tuesday was 100, now 100
Wednesday was 100, now 100
Thursday was 100, now 100
Friday was 100, now 50

That was 500, and is now 400, in percentage terms that's a 20% fall, however the busiest trains are still at 100 and so if you're looking to add more people there's not the extra space that 20% (overall) fall due to WFH would imply.

But that isn't what happened. Working from home is used for many reasons, and lengthening the weekend is only one of them. In particular, your hypothetical distribution does not take account of parents with childcare responsibilities. Here is an alternative that is probably much nearer to what is actually happening:

Monday was 100, now 75
Tuesday was 100, now 85
Wednesday was 100, now 85
Thursday was 100, now 85
Friday was 100, now 70

Yes, 2% could be too high (and certainly higher than the assumed circa 1.3%)

1.3% would be a heroic assumption now, given that economic growth has been less than that for many years.

5 years (which is the timeline I've mentioned about unpausing Crossrail 2

As I've already explained, politically the prospects of this are zero. This government isn't going to do it and neither will the government elected in 2028 or 2029.

 

[The Ham]

You have quoted these numbers before. I have already explained that the aggregate position for Greater Anglia hides some significant differences. The growth is mainly on the rural routes, Norwich-London and the Stansted Express. For commuting traffic in and out of Liverpool Street, from places like Bishops Stortford and Harlow Town, traffic is still about 20% down, and that's what matters for the case for Crossrail 2.

OK, so let's look at what has happened specifically at Bishop's Stortford.

Whist it's seen a 20% fall since 2019, how did that compare to the 2013 baseline (I.e. when Crossrail 2 was put forwards).

Because if it had seen almost no growth and then a 20% fall you'd have a point, however if growth was higher than 1.3% and the 20% fall brought the numbers back to an average of 1.3% then maybe you don't.

Passenger numbers to/ from Liverpool Street
2013 - 2.67 million
2019 - 3.26 million (@1.3% from 2013 would be 2.88 million)
2024 - 2.80 million
2025 - 3.00 million (@1.3% from 2013 would be 3.12 million)

As such, based on pre Covid yes 20% fall, although pre Covid had seen a lot more growth than the 1.3% growth model used for Crossrail 2 had assumed.

That means that the current numbers are a bit behind where the Crossrail 2 model had assumed so rightly so we shouldn't be looking to open Crossrail 2 in 2033.

Now if we were to see 1.3% growth for the next 3 years we'd reach the 3.12 million expected for 2025 and therefore basically be on track for an adjusted start date of 2036 if we could then assume 1.3% growth going forwards.

Only, last year (and obviously strikes will have a bearing on this) growth was 7.3% (at Bishop's Stortford), obviously that's very high and is likely to be unsustainable.

Likewise, I'm not talking about 3 years of delay, but at least 7 and maybe 17 years.

If the target was 3.80 million in 2040 (I.e. 7 years after opening) that target is now 3.8 million in 2047 to 2057 (opening in 2040 to 2050).

That's 27% growth over 22 to 32 years which is at most 1.09% growth.

That's not significant and over the next 3 years we'll see just how likely any sort of growth is.

But that isn't what happened. Working from home is used for many reasons, and lengthening the weekend is only one of them. In particular, your hypothetical distribution does not take account of parents with childcare responsibilities. Here is an alternative that is probably much nearer to what is actually happening:

Monday was 100, now 75
Tuesday was 100, now 85
Wednesday was 100, now 85
Thursday was 100, now 85
Friday was 100, now 70

I understand that, which is why I said:

Yes it's an overly simplified example, but then is saying that WFH means 80 on every day, the reality is somewhere between the two. Again, the point being that the busiest services are the ones which are those which need to be assessed.

Based on the 85 value means that:

2013 - 74
2019 - 100 (@1.3% from 2013 would be 80)
2025 - 85 (@1.3% from 2013 would be 86)

Based on that, the busiest services would be close to being on target to what was expected to be the case in 2025.

Arguably (if it hadn't already been paused and growth wasn't such a roller coaster ride over the last decade) few would have been saying that we should cancel Crossrail 2 based on being out by about 1 year's worth of growth.

Yes, the Monday and Friday trains are less busy and so if there's growth which resulted in the Tuesday to Thursday trains being overly busy it could be that people opt to work in the office more on Mondays and Fridays.

However, even on the Friday numbers (70) those services aren't even quite 20% (18.6%) lower than where they "should" have been on the original modeling.

Yes, the lower rates of income will be having an impact, but if TfL can get a few years of business rates by continuing the current taxes from 2031 before needing to actually start building then that's likely to more than cancel out the difference.

1.3% would be a heroic assumption now, given that economic growth has been less than that for many years.

Rail usage growth hasn't followed economic growth for some time, the classic extreme example is 2008 where GDP FELL by 5.5% yet rail use ROSE by nearly 5%.

Likewise last year rail growth was circa 7% vs GDP of 1.4% and whilst that's got a lot of other factors (like strikes) it'll be interesting to see what rail growth does over the next few years.

As I've already explained, politically the prospects of this are zero. This government isn't going to do it and neither will the government elected in 2028 or 2029.

The government wants lots of house building, not building Crossrail 2 is already restricting house building and will further impact it the longer it is paused for.

If there was really zero appetite for it then it would have been cancelled rather than just paused.

Again, I'm not saying that it'll definitely happen, I'm saying it's bit as dead as it is made out to be by some, especially given that we're taking about a 7 to 17 year delay in starting services.

The next 3 years worth of data will show if 1.3% growth even has the slightest chance of happening or not, given that the project is currently paused there's no risk in seeing what happens.

 

[Farnborough]

The Ham presents excellent evidence...

Furthermore, with the spectre of Crossrail 2 hovering, there are lots of other plans and proposals that haven't been put forward, because Crossrail 2

eg SWML capacity into Waterloo is already max-ed out - notwithstanding covid service cuts that haven't returned... Clapham remodelling... Woking junction... etc, etc

 

[Magdalia]

Rail usage growth hasn't followed economic growth for some time, the classic extreme example is 2008 where GDP FELL by 5.5% yet rail use ROSE by nearly 5%.

Likewise last year rail growth was circa 7% vs GDP of 1.4% and whilst that's got a lot of other factors (like strikes) it'll be interesting to see what rail growth does over the next few years.

Both of these are unusual examples and they don't invalidate the underlying economics.

2008 was the year of the global financial crash, as we have already established last year's high growth is due to the high levels of strike activity in previous years.

That underlying economics is:

• travel to and from work grows when employment grows, and employment grows when the economy is growing
• leisure travel grows when consumers are spending, and consumers expenditure grows when the economy is growing

Both of those relationships operate with variable time lags. In particular, big economic shocks, such as the 2008 financial crash and Covid, can cast a very long shadow, with the impact felt for years afterwards.

For a project like Crossrail 2,what matters is the prospects for economic growth in the future, the past is only relevant to the extent that it helps to form a view of future prospects. In 2013 most economists still thought that the UK economy would recover to 2-3% annual growth, rates that happened in most years leading up to 2008, and Crossrail 2 would have been assessed on that basis. But it would be very hard to find an economist in 2026 who still thinks that, and any case for Crossrail 2 now has to operate on the assumption of 0-1% annual economic growth. You being able to find unusual single years where rail travel grew much more rapidly than the economy does not change this, and a decades long project is not going to be decided on cherry picked economic comparisons.

Furthermore, what matters for the case for Crossrail 2 is pressure on peak time capacity, not overall usage spread through the day. Two significant things have happened here since 2013:

• fleet replacement has delivered am increase in peak time capacity that I estimate to be more than 50%, from about 18000 seats in a 3 hour peak with classes 317/379 to about 30000 seats with classes 720/745
• changes in working patterns since Covid have delivered a permanent cut of about 15% on maximum peak hour flows

If we assume that those 18000 seats were at 100% capacity, then a 15% cut in passengers from changes in working patterns brings that down to just over 15000, yet capacity is now up to about 30000. Peak time usage can almost double before peak time capacity constraints are going to be felt. This is borne out by the numbers quoted in the Stansted Airport planning application, where peak time loadings from Greater Anglia in 2024 are 42.5% in the morning and 51.2% in the evening.

Checking my compound interest tables, 3% compound growth will double in 24 years, and 2% compound growth will double in 35 years. 1% compound growth only delivers 42% growth in 35 years.

So clearly there is no prospect of peak time capacity being constrained any time within the next 25 years, and therefore no case for Crossrail 2 expanding capacity in the Lea Valley.

not building Crossrail 2 is already restricting house building

We have already established that this is not the case. Houses can be built in the Lea Valley and there is capacity on the existing railway for people in those houses to commute to and from London. This will continue to be the case for at least 25 years. Whatever is restricting housebuilding in the Lea Valley, it is not Crossrail 2.

 

[Recessio]

We have already established that this is not the case. Houses can be built in the Lea Valley and there is capacity on the existing railway for people in those houses to commute to and from London. This will continue to be the case for at least 25 years. Whatever is restricting housebuilding in the Lea Valley, it is not Crossrail 2.

I can't comment on Lee Valley as I'm not familiar with the area at all. But it seems TfL are pushing the Bakerloo and DLR extensions because in those cases, there are now planned housing along the Old Kent Road and in East London respectively that are contingent on the extensions. It's probably why they're pushing these as a priority over other projects such as Crossrail 2 (not to mention they'll also be quicker and cost less than CR2).

 

[The Ham]

Both of these are unusual examples and they don't invalidate the underlying economics.

2008 was the year of the global financial crash, as we have already established last year's high growth is due to the high levels of strike activity in previous years.

I did specifically say that 2008 was an extreme example, but other years haven't shown a link between the two:
- 2012 - GDP 1.5% rail passengers 3.0%
- 2013 - GDP 2.2% rail passengers 5.7%
- 2014 - GDP 2.9% rail passengers 4.4%
- 2015 - GDP 2.3% rail passengers 5.1%
- 2016 - GDP 1.9% rail passengers 1.3%
- 2017 - GDP 1.8% rail passengers (fall) 1.4%
- 2018 - GDP 1.3% rail passengers 3.0%

That underlying economics is:

• travel to and from work grows when employment grows, and employment grows when the economy is growing
• leisure travel grows when consumers are spending, and consumers expenditure grows when the economy is growing

Both of those relationships operate with variable time lags. In particular, big economic shocks, such as the 2008 financial crash and Covid, can cast a very long shadow, with the impact felt for years afterwards.

Indeed, and whist there's been the odd year where GDP grew at broadly the same as passengers, and the odd year where there was a fall in passengers, the general trend is that rail growth isn't closely linked to GDP growth.

For a project like Crossrail 2,what matters is the prospects for economic growth in the future, the past is only relevant to the extent that it helps to form a view of future prospects. In 2013 most economists still thought that the UK economy would recover to 2-3% annual growth, rates that happened in most years leading up to 2008, and Crossrail 2 would have been assessed on that basis. But it would be very hard to find an economist in 2026 who still thinks that, and any case for Crossrail 2 now has to operate on the assumption of 0-1% annual economic growth. You being able to find unusual single years where rail travel grew much more rapidly than the economy does not change this, and a decades long project is not going to be decided on cherry picked economic comparisons.

4/7 years where rail growth significantly out performed GDP, hardly odd years.

However, the point I'm making is that at the very least we should look at the next 3 years of data before we write Crossrail 2 off.

Furthermore, what matters for the case for Crossrail 2 is pressure on peak time capacity, not overall usage spread through the day. Two significant things have happened here since 2013:

• fleet replacement has delivered am increase in peak time capacity that I estimate to be more than 50%, from about 18000 seats in a 3 hour peak with classes 317/379 to about 30000 seats with classes 720/745
• changes in working patterns since Covid have delivered a permanent cut of about 15% on maximum peak hour flows

If we assume that those 18000 seats were at 100% capacity, then a 15% cut in passengers from changes in working patterns brings that down to just over 15000, yet capacity is now up to about 30000. Peak time usage can almost double before peak time capacity constraints are going to be felt. This is borne out by the numbers quoted in the Stansted Airport planning application, where peak time loadings from Greater Anglia in 2024 are 42.5% in the morning and 51.2% in the evening.

30,000 seats appears to be (based on 1,000 seats per train) 10tph.

Although, others have pointed out not all those services call in the Upper Lea Valley and some services are at capacity:

The Stansted and Cambridge trains do not stop anywhere in London along the Lea Valley (other than Tottenham Hale), rendering them useless for all commuters in Enfield, including ones forcast as part of the Meridian Water development.

All the ones that currently do stop are AT CAPACITY. Hence why many morning bound Hertford East trains no longer call at Cheshunt, they do this to spread passenger loadings easier.

Also, whist I've got no complaint about the 15% fall, it appears you've assumed that there was zero growth from 2013 to 2019 and then a 15% fall from 2019 to 2025.

In other words:
2013 : 100
2025 : 85

However what happened was (using Bishop's Stortford to Liverpool Street passenger numbers):
2013 : 100
2019 : 122
2025 : 112

To put that into the assumed seating numbers used above:
2013 : 18,000
2019 : 21,960 (73% of 30,000)
2025 : 20,200 (67% of 30,000)

As such better than the 50% assumed above.

Also, because it's roughly 2/3 full to get to full that's a requirement to add 50% of existing numbers to get to full.

Also, something like 1 in 5 commuters were standing in 2014, (2013 wasn't this bad, but was still carrying a noticeable number of passengers in excess of capacity). As such, in the morning peak hour (maybe morning 3 hour peak) it would likely have been more than 100% of seats were occupied, even if that was only 5% more than the number of seats (about 4 people standing per coach), that'll increase the current loading numbers to around 70% whilst 12% standing (about 9 per coach) would increase seating capacity to around 80%.

On percentage loadings cited, these look very much like the two way loading factors, so include some fairly empty services out of Liverpool Street and some very busy services into it.

Also, if you are right and all 89,000 homes can be built without Crossrail 2, that's increasing the population of the Upper Lea Valley by roughly 1/3 and yet the above assumption is that rail growth wouldn't even meet that (1.3% over 25 years is roughly 1/3).

Even though rail use is likely to take a larger proportion of travel from that growth than is the case for the existing housing stock - due to limitations on the existing roads.

Checking my compound interest tables, 3% compound growth will double in 24 years, and 2% compound growth will double in 35 years. 1% compound growth only delivers 42% growth in 35 years.

So clearly there is no prospect of peak time capacity being constrained any time within the next 25 years, and therefore no case for Crossrail 2 expanding capacity in the Lea Valley.

Which may be the case, however, the point is that we'll have a better idea of rail growth by looking at the next 3 years of growth.

That will be clear of the Covid rebound and the impacts of the strikes.

If there's still growth of over 2% per year between 2025 and 2028 then (and again remembering the original modeling was based on 1.3% per year) then Crossrail 2 isn't as dead as implied.

We have already established that this is not the case. Houses can be built in the Lea Valley and there is capacity on the existing railway for people in those houses to commute to and from London. This will continue to be the case for at least 25 years. Whatever is restricting housebuilding in the Lea Valley, it is not Crossrail 2.

There's a difference between some houses being delivered and all 89,000 homes.

Especially given that future development would likely result in more rail use than the existing housing stock.

Also, the Upper Lea Valley is only one small part of the case for Crossrail 2, if it's not quite there but other areas justify it, then it's likely to still happen.

Especially given that rather than needing to meet a given number of passengers in 2033 that target is moved back by 7 to 17 years, which either allows for a growth rate lower than 1.3% per year or for the average to 1.3% for 20 years (2013 to 2033) but then an average of zero growth for between 7 and 17 years (the thing with averages is that could be a 20% fall in one year but 1.4% growth for 16 years).

On time to double passenger numbers the original 1.3% growth rate from 2013 would be 2067 (54 years of 1.3% growth gets you to double). Even without a fall to get from 18,000 to 30,000 (+67%) would still be 40 years. As such, getting no where close to the numbers cited above.

== Doublepost prevention - post automatically merged: 30 May 2026 ==

I can't comment on Lee Valley as I'm not familiar with the area at all. But it seems TfL are pushing the Bakerloo and DLR extensions because in those cases, there are now planned housing along the Old Kent Road and in East London respectively that are contingent on the extensions. It's probably why they're pushing these as a priority over other projects such as Crossrail 2 (not to mention they'll also be quicker and cost less than CR2).

The DLR is currently planned to open in 2033 whist the Bakerloo is currently planned to be opened in 2040.

Given that I'm taking about 2040 at the earliest but potentially 2050 (and even then it could be a bit further off), neither are likely to be a significant barrier to it's business case.

Arguably they (by feeding more passengers from outer London into the Central core of the TfL network) could actually make the case for Crossrail 2 stronger.

Also, whilst Crossrail 2 is a lot more money, it also (like the original) adds something like 10% of the existing tube network capacity, and so the benefits are significant.

The Elizabeth Line has smashed its planning estimates, with the observed passenger numbers in 2024 matching those expected in 2030 and then has seen 10.25% growth from 2024 to 2025.

During planning growth rates of around 1.5% were assumed with some modeling looking at 2%, as such, even with significantly higher than expected passengers in 2024 there was still at least 5 years of growth in just one year (putting those who actually traveled inthe Elizabeth Line in 2025 on a par with 2035 in terms of expected passenger numbers).

However, taking the 12 month rolling total the amount of growth between March and December 2025 (so just 3/4 of a year) and the amount of growth is 4.89% (this compares with GDP of 1.4% for the whole of 2025, and rail passenger growth of 4.82% for the rolling 12 months but comparing same March to December 2025 data points - excluding Elizabeth Line and OAO's).

(For clarification March 2025 is the standard annual data set for the 12 months ending in March, December 2025 is the total of the calendar year of 2025 using the quarterly data).

 

[Magdalia]

30,000 seats appears to be (based on 1,000 seats per train) 10tph.

I'm not looking backwards to what growth there has been since 2013, I'm looking at what capacity there is now, and looking forwards to how long will it take to fill that capacity, if at all. The key point is that there is about 50% more peak time capacity than there was in 2013, and a one-off drop in peak time passenger numbers of about 15%. This means that existing capacity can cope with an almost doubling of traffic from current levels before peak time capacity is fully utilised. What the growth numbers were between 2013 and 2025 are irrelevant to this, except to the extent that they inform the prospects for future growth. I'm not assuming anything about growth in 2013 to 2025 because I don't need to, I'm starting from where we are now.

Also, something like 1 in 5 commuters were standing in 2014, (2013 wasn't this bad, but was still carrying a noticeable number of passengers in excess of capacity).

And that could apply again in the future. Greater Anglia assume that a 10 car class 720 has standing capacity for 290, in addition to the 1088 seats. With people standing there is a peak capacity of about 40000. That makes reaching full capacity even more unlikely.

Also, because it's roughly 2/3 full to get to full that's a requirement to add 50% of existing numbers to get to full.

The Stansted Express numbers don't support that, they support my 50% assumption, and they are 40% of the peak time service. But even if it was the case, 3% growth takes 14 years to increase by half, 2% growth takes 21 years, and 1% growth takes 35 years.

I did specifically say that 2008 was an extreme example, but other years haven't shown a link between the two:
- 2012 - GDP 1.5% rail passengers 3.0%
- 2013 - GDP 2.2% rail passengers 5.7%
- 2014 - GDP 2.9% rail passengers 4.4%
- 2015 - GDP 2.3% rail passengers 5.1%
- 2016 - GDP 1.9% rail passengers 1.3%
- 2017 - GDP 1.8% rail passengers (fall) 1.4%
- 2018 - GDP 1.3% rail passengers 3.0%

A key factor here is that GDP is for the whole of the UK, but rail travel is heavily skewed towards London. When the London economy is growing more rapidly than the rest of the UK, then rail travel will grow more rapidly than UK GDP. And it is important to remember that it is what is happening at peak times which determines the need for new capacity, not the overall picture. Increases in off peak travel are absorbed by existing capacity.

But the London economy growing more rapidly than the rest of the UK can't go on forever, either economically or politically. Just because rail passengers grew more rapidly in a period where London was the engine of UK economic growth, doesn't mean that will happen forever into the future. Looking forwards I expect the fastest growing parts of the UK to be outside London, especially the Liverpool-Manchester-Leeds and Oxford-Cambridge corridors. Consequentially, rail travel is unlikely to continue to outpace GDP growth, and London isn't going to be at the front of the queue for transport infrastructure.

 

[Mgameing123]

I’m not the poster addressed, but I do think there were many benefits:
- Access to Watford Junction from many more places: WCML access for many as well as St Albans
- Watford as a shopping and employment centre for outer NW London (plenty of company hqs, retail, civic/local employers too)
- Support a major health campus and a stably top ish flight team (at the time)
- connect Harrow and Watford town centres
- build more counter and off peak demand for what are very empty daytime Met trains
- not a crazy amount of station work (Watford Junction itself has the platforms ready) and close a random, very quiet station to add more purpose to the line.

When HS2 is under construction for Euston (closures) - it would be another route to Central London. And after that, Watford Junction will likely have many more calls.

Also it has become a lot more well used for the Harry Potter studios tours - again, demand and purpose in the off hours and counter direction.

That all said, what it does really need is either:
A single double ended station instead of two - or -
A f-ton of new home building

It won’t be another diversion route to Central London unless if WCML trains are equipped with tripcocks.

 

[The Ham]

The Stansted Express numbers don't support that, they support my 50% assumption, and they are 40% of the peak time service. But even if it was the case, 3% growth takes 14 years to increase by half, 2% growth takes 21 years, and 1% growth takes 35 years.

As I pointed out the Stansted Express numbers appear to be the two way flow. As such that 50% they talk about isn't the busiest services.

It's no good having 10,000 seats for travel heading out of London when it's heading inwards where the trains are the busiest and people are wishing to travel towards London.

The reason that Stansted Express are interested in the teo wat total is that for serving the airport, relatively few people are traveling into London during the morning peak from the airport.

This is because the first fights arrive just before 7am and even allowing for a fairly rapid exit few passengers will be arriving before 8:30, maybe passengers from around 12 fights could make it before 9:00.

I'm not looking backwards to what growth there has been since 2013, I'm looking at what capacity there is now, and looking forwards to how long will it take to fill that capacity, if at all. The key point is that there is about 50% more peak time capacity than there was in 2013, and a one-off drop in peak time passenger numbers of about 15%. This means that existing capacity can cope with an almost doubling of traffic from current levels before peak time capacity is fully utilised. What the growth numbers were between 2013 and 2025 are irrelevant to this, except to the extent that they inform the prospects for future growth. I'm not assuming anything about growth in 2013 to 2025 because I don't need to, I'm starting from where we are now.

The thing is, passenger numbers aren't something which happens in isolation.

However, in 2024 the passenger capacity data for the 3 hour period shows Service provision (Standard class capacity, including an allowance for standing passengers) for Liverpool Street of 212,000.

This then compares with the Standard class critical load (I.e. the number of passengers which needs to be accommodated) of 112,000.

As such over the 3 hour period is suggesting that average loadings into Liverpool Street are 64.6% (or roughly 2/3).

That's 108 services with passengers who are standing out of 175.

It gets more congested when you look at peak hour, with 81,500 being the capacity and 61,500 being the passenger numbers (75.5% of the capacity, or roughly 3/4).

With then 40 out of 64 services having people who are standing.

Now it may well be that most or all of those people are on Elizabeth Line trains.

As such, why build Crossrail 2 as the Anglia services aren't where there's the capacity issues?

Whilst that's true, it's also true that Crossrail 2 would provide extra capacity within the core by giving an alternative route between Liverpool Street and Tottenham Court Road. Whilst the capacity might not be needed for standing capacity issues on Anglia services, other services would then benefit from the relief that Crossrail 2 provides.

A key factor here is that GDP is for the whole of the UK, but rail travel is heavily skewed towards London. When the London economy is growing more rapidly than the rest of the UK, then rail travel will grow more rapidly than UK GDP. And it is important to remember that it is what is happening at peak times which determines the need for new capacity, not the overall picture. Increases in off peak travel are absorbed by existing capacity.

But the London economy growing more rapidly than the rest of the UK can't go on forever, either economically or politically. Just because rail passengers grew more rapidly in a period where London was the engine of UK economic growth, doesn't mean that will happen forever into the future. Looking forwards I expect the fastest growing parts of the UK to be outside London, especially the Liverpool-Manchester-Leeds and Oxford-Cambridge corridors. Consequentially, rail travel is unlikely to continue to outpace GDP growth, and London isn't going to be at the front of the queue for transport infrastructure.

Some fairpoints, however, I'm not asking for is to make a decision on Crossrail 2 now, rather based on what happens over the next 3 years.

Also, London would be funding 40% (and if the levies roll on from the current levies, it might be that more could be funded from London) of the costs of Crossrail 2, as such that means that only part of the costs needs to nationally funded.

 

[BritishGuy54]

So, while this idea isn’t necessarily in London, it’s just outside of it, and I wonder if there would be any benefits to it…

Extending the c2c’s Rainham branch from Grays to Gravesend via a rail tunnel under the Thames.