Crossrail: A Hole New World


At the beginning of August 2018, seven Crossrail core stations were due to be handed over to TfL. Instead, all remain in some state of construction. For Crossrail, this is a problem. For the Elizabeth line it is an even bigger one.

The problems currently affecting Crossrail are not limited to one thing. Broadly, sources suggest they fall into two categories: issues with the signalling and issues with the stations themselves.

The first is a subject we have touched on before here on LR. The Elizabeth line will require its trains to interact with a number of different signalling systems, all of which have their own quirks. For some time now, there have been issues with the signalling on the Heathrow section of the line, but sources suggest that it is not these that are the primary cause of the current delay. Instead, it is issues in the core and at the interface points, at least in part to do with the CBTC signalling.

As with the issues at Heathrow, this is a solvable problem, with sufficient application of time and testing. Yet it seems that the second problem – that with the stations themselves – has limited opportunities to solve the first.

This is because a number of stations remain unfit for handover. Indeed sources suggest that TfL’s internal assessment was that only three major stations in (or around) the central tunnel section (Paddington to Abbey Wood, which had been due to open this December) were likely to be ready for full use in service by December 2018: Bond Street, Canary Wharf and Abbey Wood. All of the major stations outside of this grouping are progressing, but are not yet finished, with Whitechapel being the furthest behind. Here, the problem isn’t just fit-out but construction – sources suggest that the station is still not structurally complete.

We will dig into the full details of what has gone wrong in a future article, not least because more information is only just beginning to emerge. But the above is enough to paint a clear picture as to why a decision was made to postpone. From a technical perspective, you might plausibly open a line lacking several stations, but operating only three out of ten on the Elizabeth line core would have been an extreme failure. From a service delivery level it would have made no sense, and the concentration of passengers at Bond Street would have been a major risk.

The thermocline of truth

One of the questions that will need to be asked of TfL is the timeline that led up to the decision to delay. We will tackle the validity of that decision shortly, but its timing – and suddenness – suggests that they may have been some element of the ‘thermocline of truth’ in play. This is something to which large rail projects – most notably GTR’s new timetable rollout – have repeatedly shown they are susceptible to. It is the principle that bad news tends to accrue at a lower management level, because no one wants to be the person who moves a project risk marker from ‘yellow’ to ‘red’ on a RAG chart. As a result, pessimism and a belief that the project will overrun ‘bubbles up’ to a certain decision-making level but never beyond, as if hitting the thermal layer that exists in the ocean. Eventually, the issues reach critical mass and force their way through, leaving senior management wonder why everything ‘suddenly’ went wrong, when in fact the signs that the project was troubled existed at a lower level for some time.

Given the various senior staff changes that have happened at TfL in recent years, along with their own restructure which will inevitably have robbed the organisation of some experience and talent, it is valid to question how quickly TfL spotted that these issues were occurring. On the other hand, this decision does suggest that they did at least catch the problems in time to make a valid decision to delay.

Ultimately, what has begun to emerge is a picture of a line that is progressing, but has simply run out of time. Beyond the details then, what is critical to ask now is what the impact of this decision will be on TfL, on London and on the railways at large.

The financial elephant in the room

Much has been made of the role that the Elizabeth line is expected to play in TfL’s future finances. It has become axiomatic that the Elizabeth line’s revenues are critical to balancing TfL’s budget line. On a broad level, this is true. Indeed one of the reasons this has become such an accepted truth is because it is something TfL have repeatedly acknowledged themselves. The reality though is far more complex and nuanced.

A delay is bad news for TfL’s finances, but not quite yet.

Letting Lizzie grow

Contrary to reports, TfL never expected to make quick money from the Elizabeth line this December. Indeed weirdly, today’s announcement actually means that on one level (net operating costs) their 2018/19 financial outlook has now actually improved.

This is because metro systems don’t burst into being fully utilised (although we suspect Crossrail will fill up faster than most). Despite this, the operator has no real choice other than to run a heavy service pattern from day one. New metro systems are loss-leaders, at least to begin with.

The Elizabeth line is no different. As LR’s own Nicole Badstuber pointed out on Twitter, TfL’s own forecasts don’t show the Elizabeth line turning a profit until 2020. So, contrary to a lot of the initial reporting, 2019 TfL is currently not out of pocket, but 2020 TfL most definitely is.

Still with us? Good.


The nuance behind the profitability impact highlights the real problem with this delay announcement – Crossrail being delivered ‘on time and on budget’ in December 2018 has become such an anticipated milestone within TfL’s – and indeed National Rail’s – calendar that many other future plans (and budgets) have been built upon the assumption that it will be in place. The removal of this milestone means that a significant number of TfL’s, Network Rail’s and even the DfT’s plans for the next five years will need to be redrawn.

This makes it genuinely difficult to assess just how big, or wide, the impact of this announcement is likely to be. Right now, on a practical level, London and the South East’s rail and transport strategies are perhaps best represented by a giant shrug emoji ( ¯\_(ツ)_/¯ ).

Some things, however, are clearer than others. Particularly as the Crossrail press release provided a timescale for the anticipated delay – ‘Autumn 2019’. Here, it is always worth following the ‘Roger Ford rule’ for transport press releases: if something is announced for a season, that means it risks being delivered on the last possible date to which that seasonal definition can be applied. In this case, that’s the last day of November, or even later than that should it be felt useful to tie it in to the December 2019 timetable change. This isn’t necessary, as the first operational phase of the Elizabeth line only interacts with the mainline between Portobello and Old Oak Common, but it may be felt to be a convenient marker to hit.

Our prediction is thus that the Elizabeth line will likely open exactly one year late, though a more limited preview service may arise sooner, for operational setting-in as much as anything else.

Spreading the load

On a basic level, this means that we can at least predict the operational financial impact on TfL of this delay (excluding other potential costs, to which we will return later).

Regardless of when it opens, the adoption curve for the Elizabeth line isn’t likely to be much different from what it would have been had it opened on time. This means that if TfL expected the Elizabeth line to be profitable in 2020 at the earliest, then they can now expect it to be profitable in 2021 instead. That’s a £300m hole in their future, based on their own budgeting estimates, that they will need to address.

What is less obvious is the indirect effects. These occur because Elizabeth line passengers are not going to spring from the earth fully formed. In most cases, they are expected to come from services elsewhere.

From TfL’s perspective, two of the most important sources for those passengers are their own services – the Tube (Central and Sub-Surface lines) and London bus services.

It may seem counterintuitive for TfL to be looking forward to drawing passengers away from their own services, but there are good reasons for them to be pushing for (and reliant on) this shift.

As any regular Central line passenger will tell you, the service there, in rush hour, is heavily oversubscribed. This means that there is an awful lot of suppressed demand (i.e. potential extra passengers) on London’s red corridor. The Elizabeth line will also relieve the north side of the Circle line.

Drawing passengers away from the Tube onto Crossrail will be a major win for TfL. Not only will it increase the comfort and reliability of those corridors, but it will also release capacity there for passengers who would be using the Central and Circle line if they thought they could actually get on a train. Essentially, there is much suppressed demand, particularly on the Central line, to compensate for any major shift of passengers now better served by the Elizabeth line to those services instead. So for TfL, opening the Elizabeth line offers an opportunity to syphon off a percentage of Central line passengers onto a new service (where they will still be paying TfL) and make extra money from those new passengers who emerge and fill the created gap. Some Circle and Hammersmith passengers may also prefer a more frequent ‘express Tube’, in the style of Paris’ Reasau Express Regional (RER).

The bus benefit

When it comes to buses, the benefits of the Elizabeth line to TfL are slightly more abstract, but no less important. TfL’s bus services are under enormous financial pressure, due to the massive cuts in government subsidy introduced by the previous Chancellor, George Osborne. To put things in context, just a few short years ago TfL’s government subsidy was £700m. Now it is nothing, which makes the former Chancellor’s selective amnesia about the causes of TfL’s current financial woes as the current Editor of the Evening Standard somewhat difficult to understand.

Even without this, TfL’s bus services have been long overdue for an extensive service pattern review. The financial pressures though make that even more critical and leave TfL in a position where it has little choice but to make some kind of service cuts.

Plans are already underway for a full bus consultation (it is a topic we will be covering at length in future), and the assumption within TfL was that the Elizabeth line’s opening would open up major opportunities to refactor (or, more bluntly, consolidate) services along various parts of the new line’s route. Handily for TfL, the Elizabeth line route covers one of the few areas of London where they both genuinely believe changes are required and (usefully) know the Mayor would be unlikely to intervene – Tottenham Court Road and Oxford Street.

The decision to delay the Elizabeth line’s opening throws a huge spanner in the works for TfL’s bus review. No Elizabeth line means no indirect relief in the areas where they were keenest to cut services, and that means TfL’s options are now far more limited – go forward with their review and make fewer cuts now, or delay it until opening and cut more in future.

Similarly, TfL’s ability to redirect some bus mileage to become feeder services to Crossrail (such as from new suburban housing areas at the Elizabeth line railheads) is made more difficult, although even that could benefit TfL in two ways: by improving the Elizabeth line’s accessibility from growing catchments, and simultaneously improving bus operational efficiency by removing some slow-moving and costly-per-hour bus operation from central London.

With the bus network continuing to haemorrhage money, it’s entirely possible that, for TfL, this might actually be the largest short-term, negative financial impact caused by the delay. Certainly, it will be interesting to see how their plans for a bus consultation change.

Counting the pennies

Reading between the lines in the TfL Draft Budget, the anticipated benefit in the directly affected Tube and bus operations and incomes is relatively small (tens of millions at most) but every penny helps right now. These numbers worse with the postponement of some ex-main line revenues and the hiatus on growth. The government subsidy cut isn’t the only pressure on TfL’s finances. Estimated passenger growth figures for the Underground have, in recent years, habitually been over-optimistic but last year saw growth drop to fractional levels. Statistically, it may not have represented (in most cases) an actual drop – despite the way most of the mainstream media reported it – but this doesn’t change the fact that TfL’s own profitability forecasts were based on seeing a continuing increase. In truth, nobody knows the reason for the collapse in these growth figures yet – or indeed whether it may turn out simply to be a temporary blip – but the financial consequence is obviously unpleasant.

This unpleasantness is compounded by three other factors. Firstly, that TfL have, for some time, been skirting their borrowing limits. They are certainly not over them, but they have very little room left on the corporate credit card. This was one of the drivers for the decision to leaseback the Elizabeth line’s train fleet in order to free up capital to purchase the next generation of rolling stock for the Deep Tube lines. That was a perfectly reasonable and sensible thing to do, but there is no doubt that TfL would not have done it had they been able to raise the necessary up-front cash required any other way.

The other two factors TfL face come from decisions outside their control. Firstly, a DfT ‘rethink’ on the way A road maintenance is funded has placed a large, unexpected financial burden on TfL. Again, it is a subject worthy of an article in itself, but London contains a disproportionately high percentage of the country’s A road network, most of which fall under the governance of TfL as transport authority rather than the local boroughs. The new rules leave TfL footing the bill for all of these, a grossly unfair situation but one that the government seem to show little motivation to resolve.

The second factor is one with which most followers of London transport news (or general London politics) will be familiar: the Mayor’s manifesto promise of a fare freeze. Contrary to the somewhat hyperbolic press on the subject, this is not in imminent danger of bankrupting TfL. Indeed its impact is thoroughly dwarfed by the subsidy cut. Nonetheless, it was a reckless promise by the Mayor as a candidate – the London transport equivalent of starting a land war in Asia – and its impact has the potential to increase exponentially the longer it remains in force.

The path to profitability

All of the above pressures highlight why it is important that the Elizabeth line open as quickly as possible for TfL. Again, not because it will immediately create a substantial new revenue stream via its farebox, but because the sooner it is open, the sooner it can begin to move closer to that goal. As we’ve looked at above, TfL’s broad figures suggest this will leave them £300m out of pocket with a year’s delay, but we suspect the reality is probably a bit more nuanced.

For while the residents of LR Towers have long suspected that TfL has a habit of being a bit too optimistic with regards to passenger growth figures in general, we think that they may actually be too pessimistic with regards to the Elizabeth line. Nor are we alone in that view – it’s something that Sir Peter Hendy, Chairman of Network Rail, has commented on too. The betting money here has long been that the Elizabeth line will actually mature in closer to nine months, if not as little as six months. If that’s the case, then the long-term impact of this delay may turn out to be less than anticipated too.

This makes it relatively important to think of the losses this delay is causing TfL less in terms of full years, and more in terms of month-by-month cost. Handily, TfL provide sufficient depth of information with regards to their budgeting and passenger numbers for us to do that.

Broadly, taking TfL’s forecast budget figures for existing TfL Rail services and the Elizabeth line and adding some conservative estimates for Heathrow passengers, it’s possible to put a rough value to TfL of a single Elizabeth line journey, and set that against the cost of delivering the service. If the maths here in LR Towers is correct, then once it starts running, and until the line fructifies, every Elizabeth line journey taken will cost TfL roughly £1, or a total of £15m a month.

Multiply that figure by the passenger number estimates for the first few months of service and this is why not opening in December actually has a positive impact on TfL’s finances to the potential tune of £60m (ignoring any contractual penalties that they may be triggering elsewhere). TfL’s 2018/19 financial situation arguably just got better, not worse.

The hidden (and not-so-hidden) costs

All the above means we can place an operational cost of this delay for TfL to somewhere between £200 – £300m per year, depending on how quickly the line matures. There are no doubt further costs though, the value of which is yet to be revealed, which may accrue to TfL too.

The first of those is that of the overrun works themselves. Here, the decision to go for such an extreme delay makes a certain amount of sense. Even if the stations are nearer to point of delivery than sources suggest, one of the major issues (and indeed causes of the existing cost issues) has been the clear attempt to try and do more work in less time. That’s meant overtime, additional kit hire and the cost of needing to both fit-out and test the line at the same time. The current delay means these costs will inevitably continue, although how much will fall on TfL rather than penalties to the various contractors remains to be seen. A long delay, however, mitigates those costs somewhat, as it allows fit-out and testing to proceed at a slower pace.

The second cost is far less predictable, and something that TfL have so far been unable to provide us information on, but which we strongly suspect exists – penalty clauses payable to Canary Wharf.

That these exist is a reasonable piece of conjecture, simply because Canary Wharf have form for it. When the Jubilee was extended at the end of the last millenium, Canary Wharf ensured that their contribution to the project was closely linked to the planned opening date, and thus the point at which they had projected they would see an increase in revenue themselves. In the end, this ended up saving them an awful lot of money, lowering their final contribution by well over 50% to roughly £90m.

Luckily for TfL, the Jubilee line and the Elizabeth line are not directly comparable. Completion of Crossrail is not subject to the Millenium Year drop-dead date imposed (with costly budget consequences) on the Jubilee Line Extension project. Indeed the Jubilee line extension may actually have limited the liability with Canary Wharf this time around. Critically, Wood Wharf, the area’s big expansion (bringing an estimated 15,000 jobs) is not contingent on the Elizabeth line opening, as the planning decision deemed that there were sufficient transport links in the area already (Jubilee and DLR) to support the new development. This is something over which TfL are likely breathing a sigh of relief, but that doesn’t mean that Canary Wharf won’t have put in place any kind of penalty clause at all. And if they have, then one suspects it will scale over time.

May day?

Given all the above, it might seem surprising that TfL have decided to announce such a long delay in the delivery schedule. As we’ve already mentioned, the opportunity to spread out fit-out and testing again (that is, not trying to ‘cram’ the build in) likely yields some small financial benefit, but seemingly not enough to outweigh the various financial risks of a longer delay.

There are, however, other factors that come into play too.

One of those factors is likely human – The Elizabeth line’s Director of Operations, Howard Smith. Throughout his career Smith has demonstrated a firm belief in (and commitment to) implementing service patterns correctly and reliably. As the experience with Heathrow signalling has shown, he is a person who will happily make – and implement – a backup plan, but sources suggest that he was simply unable to find an acceptable path to one here. As a result, running nothing was deemed better than running a fragmented or unreliable service on day one.

There is also a more practical reason. Once rumours began to surface within the industry that Crossrail might be delayed, thoughts began to turn in LR towers as to how long it would be possible to do so for. Broadly, the general consensus was ‘not long’. Not, as you might think, for financial reasons but for logistical ones.

This is because although the timetable changes required to make the first phase of the Elizabeth line’s service pattern (i.e. the core service) work are limited in impact, these scale up rather quickly once the wider reaches begin to open. Indeed a glance at the calendar showed that it was the May and December 2019 timetables, not the December 2018 one, that would see the largest set of changes to other services that were reliant on the Elizabeth line being open. Most particularly, because of the changes to Anglia services in May 2019 with through running to Shenfield, and throughout Great Western in December 2019 with the Elizabeth line to Heathrow and Reading.

May 2019 thus presented a rather glaring first opportunity for TfL to ‘Thameslink’ themselves. If the Elizabeth line’s opening was delayed much beyond the end of January, then there would be insufficient time to train additional drivers on the various required routes, and that might mean implementing a service pattern in May which they would be unable to meet. Alternatively, it might have meant delaying that first phase until May and trying to do that, and the second phase, at the same time.

This would have been a huge ask, but politically, such a move might still have been attractive. One wonders how much pressure TfL found themselves under to do exactly that.

Ultimately, it seems our ‘not long’ prediction may have been based on the right reasoning, but underestimated how brave TfL might be prepared to be – because delaying the opening by an entire year removes that timetable risk almost entirely from the table. Indeed one wonders whether quietly whispering the word ‘Thameslink’ in the ear of the Secretary of State for Transport may have been enough on its own to ‘sell’ him on the current delay.

Eddies in the transport continuum

Such a long delay does, however, bring consequences elsewhere. It throws an awful lot of other rail plans into question, because just as the Elizabeth line will no longer be extracting passengers from other TfL services (in the short term), it will no longer be doing so elsewhere either.

That’s potentially good news for the DfT, in some small measure at least. This is because it means another rethink of the floundering bid process for South Eastern. Sources suggest that bids here have been far from positive when it comes to suggested profit or passenger increases. With a year’s less Elizabeth line interference on the cards, the DfT may finally be able to coax some face-saving figures from the successful bidder.

Overall, the arrival of Crossrail will add to passenger volumes and revenues on main line services, because of the extra rail distribution capacity in central London and at its three strategic Satellite Activity Zones (SAZs) at Canary Wharf, Stratford and in future at Old Oak Common. Rail as a whole around London and the Home Counties will become a more attractive mode to use.

Due to this, several train operating companies and the DfT will have been relying on Crossrail opening on time if not to budget, in order to see some positive growth in passenger numbers and income. Now this could be shifted to the right by a year. The Crossrail investment directly benefits the Great Eastern and Great Western corridors, with through Elizabeth line services, but this too will be deferred.

In the particular cases of the South Eastern North Kent lines (NKL) and Thameslink, the impacts are more moot. Crossrail represents a TfL opportunity to attract passengers and revenues from the South Eastern network, and provide a basis for refocusing the services operated there – both a loss in the inner section of SE NKL and a gain on the outer section, and also an opportunity for refilling of seats on SE trains inwards from Woolwich.

For Thameslink, TfL has been refusing to market Thameslink on the tube map even though it provides a new strategic north-south rail corridor and a more comprehensive combined rail and Tube network in the London urban area. This is because TfL doesn’t want to risk losing revenue at the margin from the parallel Tube lines (even though those are under huge capacity pressure with the delays to NtFL). It is tempting to wonder whether this delay might lead London politicians to pressure TfL to put more of Thameslink on the tube map.

The delay represents more neutral news for Network Rail. It means an inevitable postponement of the much-needed improvement works at Liverpool Street that were scheduled to start in 2019. That this is neutral, rather than bad, news is mostly for blunt, pragmatic reasons: there is significant reason to doubt whether Network Rail were really in a position to complete this work anyway. The organisation faces its own management and financial pressures, so in that regard the ability to quietly triage away the Liverpool Street work and blame it on circumstances outside of their control may actually prove to be a spot of luck for Network Rail.

Boldly going where no transport authority has gone before

All of the above highlights just how wide-ranging the impact of this delay is. That, in itself, brings us back to the main question: why do it?

Overall, the picture seems to be one in which TfL have had the sense to realise that it was better to ‘go big or go home’. If a delay was needed, then make it a proper one which genuinely allows enough time for the mistakes to be corrected. This still leaves plenty of questions to be answered – not least why it took so long to spot that all these issues still remained – but it’s a bold move and one that deserves a certain amount of praise.

One of the biggest criticisms to emerge from the GTR and Northern debacles has been the unwillingness of both the rail industry and its political masters to make hard, painful decisions. It is impossible to deny that in announcing this delay TfL, the Mayor and even the Secretary of State, Chris Grayling, have just done exactly that.

If this delay means that the Elizabeth line eventually opens in December 2019 as a finished, reliable railway then it will have been a brave – but correct – decision to make. Indeed even delayed, and with the current projected increase in costs, the Elizabeth line will still represent both an astonishing engineering success and very good value for money. As megaprojects go, Crossrail remains on course to be a major success.

By delaying the launch as much as they have though, TfL have effectively drawn a line in the sand. December 2019 may offer the best, most realistic chance of delivery but it also arguably represents the longest TfL could have delayed the project whilst keeping the financial impact of such a decision just about manageable. ‘YOLO’ as, they say.

In effect, TfL now have three ‘big ticket’ financial priorities – the delivery of Crossrail, the completion of the Sub-Surface Signalling upgrades (4LM) and the purchase of new rolling stock for the Deep Tube lines. Anything not on that list – or anything not yet underway – will have to bear the financial weight of any further cost issues with those three things. Those hoping for a Bakerloo line upgrade in the near future may need to manage their expectations, as even the cost of the preliminary work is something that TfL may not be able to bear right now.

There remains, however, one elephant-shaped Oyster card left in the room. For although it is not the primary cause of TfL’s current financial woes, there is one thing that would now make TfL’s immediate book-balancing considerably easier:

TfL Farebox * (RPI + 1%) = £250m (approx.)

LR is funded largely by its community. Like what we do? Then help us to do more!

Donate £1Visit our shop

Written by John Bull
John Bull is the Editor of London Reconnections. A transport journalist and historian, his writing often focuses on the political or strategic challenges facing London's transport network and beyond.