On 3 January 2018, the London Assembly Budget and Performance Committee sat down with TfL to go through their current budget. Discussion largely focused on the various pressures on TfL’s finances – from the removal of TfL’s government grant, to fare freezes and falling advertising revenues.
About two and a half hours in, however, Assembly Member Caroline Pidgeon asked a question: “In here, there’s a line… £850m capital receipt… what exactly is this?”
The reply, from Simon Kilonback, TfL’s Chief Finance Officer, sent eyebrows at LR Towers skyward:
“So this capital receipt relates to how we finance our rolling stock… we have decided in this business plan to press ahead with the procurement of the new Piccadilly line rolling stock and we will be looking to award that contract in the first half of 2018.”
“We’ve therefore made a decision about taking some of our existing rolling stock and doing a sale and leaseback of that, to finance the Piccadilly line rolling stock.”
The outrage begins
Clearly, we were not the only people who had been watching the Committee. An hour later, an article appeared over at ITV:
Cash-strapped transport bosses plan to sell part of the London Underground fleet and lease it back to raise money for new tube trains.
That piece picked up on Pidgeon’s own in-Committee response the admission:
You’re going to be selling and leasing back rolling stock you’ve already got, that you wholly own, in order to give you the cash to buy new rolling stock? It sounds quite mad.
The Evening Standard (“TfL plans to sell part of its London Underground fleet of Tube trains”) and others were not far behind, nor was social media.
All these stories shared the same source – Kilonback’s statement – but they also had one other thing in common: they weren’t actually true.
Kilonback’s statement was very carefully worded – something that most of those watching seemed to have missed.
The reality is that TfL don’t plan to sell any London Underground trains. That would indeed be rather mad.
They plan to sell (and lease back) the Elizabeth line fleet.
Why the Piccadilly line matters
For a number of years we have reported on the problems facing the Deep Tube Upgrade – rather theatrically (and inaccurately) called ‘New Tube for London’ under the previous mayor. We concentrated on the technical issues the upgrade involved, why it was needed and the challenges it presented. It is a huge project, involving the upgrade of both signalling and trains across the Piccadilly, Bakerloo, Central and Waterloo & City lines at a cost that will likely equal the money spent on Crossrail by the time it is complete. As a result, we were not surprised when delivery dates started slipping.
When the Sub-surface Railway (SSR) signalling contract was cancelled, Deep Tube’s dates slipped further. The failed contract had been retendered and the winning bid (in fact the only bid) was more expensive, although at least this time the company providing it – Thales – had a good track record (pardon the pun) in this area. It made sense to assume that the substantial additional cost of this resignalling project would inevitably lead to unrelated follow-on schemes being delayed, not least to avoid a considerable cash flow problem.
Despite this, there seemed little reason to worry about Deep Tube’s financing. Once the SSR signalling upgrade is complete there are otherwise no really big (multi-billion pound) projects on the horizon that are explicitly signed off. There is still the dream that the Bakerloo Line Extension will happen and there is also the bigger issue of Crossrail 2. In both those cases though, separate funding-streams are being sought.
Indeed, one reassuring thing that came out of the cancellation of the short-term purchase of new Jubilee and Northern line trains was David Hughes, LU’s Director for Strategy and Network Development, telling the London Assembly Transport Committee that the cancellation helped ensure money was available to buy new trains for the Piccadilly line – trains, which he made clear, needed to be replaced as soon as possible.
Finding the money
Hughes’ comments were a relief because they seemed to address what remained one of Deep Tube’s biggest risks – that TfL wouldn’t be able to find the money to get things started. That cash is necessary because you can’t just buy Underground trains off the shelf. They tend to be bespoke. This isn’t a problem for most rolling stock suppliers in terms of construction, but it means that designing and building them isn’t without risk. As a result, the manufacturer tends to ask for an awful lot of money up front – or at least over a very short period of time.
As we’ve already mentioned, this lack of cash was one of the reasons behind Deep Tube’s earliest delays. Paying for the SSR upgrades depleted TfL’s petty cash jar. What was unanticipated, however, was that subsequent events, the general economic situation, grant changes and manifesto promises would prevent this from being topped up once again. Indeed TfL’s finances are still clearly under considerable pressure, something that the January Budget Committee laid out in stark detail.
In better times, this wouldn’t be a problem. In recent years, a favourite refrain of politicians (and amateur planners) with a transport project to push has been that TfL can leverage it’s size to borrow on favourable rates. To a certain extent this is true, but it’s also an oversimplification.
It is true that TfL has a good ‘credit rating’ – AA (Standard & Poor’s), Aa3 (Moody’s) and AA- (Finch) – but they are also subject to strict borrowing limits imposed by the Treasury. And as they laid out plainly in their recent refusal to build the Metropolitan Line Extension without further funding – they are very close to maxing out that limit.
Although this may seem like a perilous situation, this lack of ready cash probably isn’t causing too many sleepless nights at TfL. Indeed one is tempted to suspect that any jittery young TfL accountants will have been gently reminded that things used to be considerably worse. When the Underground was renationalised, its cash reserves were so low that it rarely had enough in its current account to make it through the day. For some time, the solvency of the world’s oldest metro system rested entirely on daily, semi-formal lunchtime chats with the financial people over at the Greater London Council (GLC). After each lunch, the GLC would remit enough cash over to cover the day’s bills.
Today’s situation is far less seat-of-the-pants. At the end of the last financial year TfL actually had just shy of £2bn in the bank. It’s just that this is already earmarked for use elsewhere (£500m of it is Crossrail money, for example) and – without the ability to borrow either – this is preventing TfL from doing the one thing they like to do – buy their Underground trains outright.
Owning your own trainset
If you are TfL (or, more specifically, London Underground) then there are considerable advantages to owning your own fleet. The biggest of these is that – for what is practically a bespoke product – it is considerably more financially attractive than the usual alternative – leasing.
If you want to lease something – be it a car, plane or something else – you quickly learn that the lessor will normally be very careful not to put himself in a position where he loses out if you stop paying.
First of all, there will be a clause that the goods remain the property of the lessor until they are paid for in full. Secondly, there will be a requirement to lease the goods for a minimum period of time – basically to cover the early period where the value of the goods drops faster than the payments received. Thirdly, once the lessee is free to terminate the lease, the lessor will try to make sure that there is some residual value in the product throughout its life and that this is greater than the sum of payments still due.
This is fine if the product is a car or standard production aeroplane. It doesn’t really work though if the item is bespoke – such as a Tube train. Put simply, if your only post-London option as the Lessor is the Isle of Wight, then you don’t really have another option.
It should quickly be apparent that a train lessor is going to be very wary about leasing Tube stock unless very strict terms are applied. And, if they naively think that nothing will go wrong leasing vehicles run by TfL, they only have to remind themselves what happened when a different mayor decided he didn’t like bendy-buses.
This is not to say trains on the Underground haven’t been leased. Most notably TfL actually lease the current Northern line trains from a Special Purpose Vehicle controlled by Alstom, a legacy of the disastrous PFI period. This lease is a little bit different though as, unusually, Alstom also maintain the trains as part of the lease. Indeed we suspect that in this instance that arrangement must be working quite well and the terms, set up a long time ago, are actually quite favourable to London Underground. Certainly they didn’t exercise the contractual break clause when they had the chance.
Despite the apparent success of the Alstom leasing arrangements, London Underground mostly looks after its own stock without outside help. Caroline Pidgeon suggested that not doing so would be ‘quite mad’ and TfL would, in most cases, be the first to agree.
What makes little sense underground, however, can make a lot more sense over it. With far more commonality in fleet design on the surface, the options for re-leasing there are much greater and thus there is less risk (and cost) to be passed on to the lessee. This makes it a much more economically viable model and so there, like any other Train Operating Company (TOC), TfL have not been afraid to lease.
This still isn’t to say that they do things the same way as everyone else. On the national rail network, the leasing market is dominated by a relatively small number of companies known as the ROSCOs (Rolling Stock leasing Companies). When TfL found themselves in need of a fleet of trains to run the newly-acquired Overground, however, they couldn’t just turn to a standard deal with one of the traditional big three. This was because TfL (stop us if you’ve heard this before) wanted something a bit different: in this case high-capacity, walkthrough trains with longitudinal seating – something practically unheard of on the national network at the time. They also didn’t want to be tied into an excessively long contract. The huge amount of infrastructure work (both in terms of stations and electrification) meant they could easily see a future where they wanted more cars or even different power supplies. They needed a fleet arrangement flexible enough to support that.
The traditional ROSCOs weren’t entirely enamoured with this proposition. They looked at the high capacity setup TfL wanted from Bombardier in the Class 378s and they saw something that was thoroughly untransferable or (just as bad) unconvertable. The door arrangement would ensure that a large vestibule would still remain if they wanted to put 2+2 seating in and all the electrical control equipment would have to be under the seats – something that would normally be at the carriage ends. That would need relocating. Today, that setup is considerably more attractive (more on this later) but back then there seemed very little certainty of being able to re-lease without major expenditure.
In the end, two of the big three ROSCOs submitted bids, but presumably their price wasn’t seen as competitive enough. In the end, TfL looked to two helpful banks, Sumitomi Mitsui from Japan (one of the world’s biggest asset leasing companies) and National Australia Bank, who formed a hands-off leasing company – QW Rail. Each contributed roughly 25% to its formation whilst the European Investment Bank (EIB) loaned the rest. The final loan repayment to the EIB is due in 2027 with the funding from the two banks repayable in 2044 (by which point the original stock will be 35 years old).
The resulting arrangement was one that, in like-for-like terms, seemed less beneficial than a standard lease to many in the industry. It was on the terms that TfL wanted though. Since then, further leasing deals have followed on the Overground – including more standard leasing arrangements for the Class 172s used on the GOBLIN, which will be the first Overground stock to be cascaded elsewhere.
All of which brings us back to the Elizabeth line (nee Crossrail).
To a certain extent, one of the things that made the Elizabeth line’s Class 345 trains (now mid-rollout with TfL Rail) unique was that they weren’t leased. This was largely because TfL were in the (relatively) luxurious position of not having to do so. The money was there in the project and (at least at the time they began shopping) the Overground-esque layout they desired was something which still made the traditional ROSCOs suspicious. Taking all that into account, owning the rolling stock wasn’t just playing to TfL’s preference, but also taking the path of least resistance.
Kilonback’s statement at the Budget and Performance Committee, however, shows just how much that situation has changed.
The simple truth is that TfL now find themselves needing to build two new train fleets – one of which they are currently deploying, the other which they are about to tender for – whilst only being in a financial position to outright own one of them. Owning the Deep Tube fleet (or at least being able to pay for it up front) is a “must”, whilst owning the 345s has only ever been a “nice to have”. So it seems they plan to give up the chance to do the latter to do the former.
Scoping the contract
Of course with the Elizabeth line fleet already specified, detailed design work completed well over two years ago and almost 40% of the fleet already built, this means that the leasing deal will be far from a standard ROSCO arrangement once again. With a maintenance deal for the Class 345s already signed with Bombardier for up to 32 years, there is also far less opportunity for an ongoing technical and management role from an interested ROSCO too. Essentially, many of the elements of a conventional leasing deal have already been specified and will be managed and undertaken by Bombardier as part of their maintenance role (for example corrosion repairs and repainting).
All this means we are likely to see TfL continue in their role as the hipsters of train leasing (“your favourite ROSCO sucks”) and are unlikely to see one of the traditional big three ROSCOs take part in the financing – unless it is as part of a syndicate, providing rail expertise which the other members might otherwise lack.
Indeed all of the traditional ROSCOs will be feeling some financial pressures over the next few years, as the leases on various existing stock they own begin to come to an end with limited signs of renewal, for a variety of reasons. These include both the ageing out of some of the UK’s most familiar rolling stock but also – somewhat ironically – the fact that TOCs across Britain are increasingly asking for the same style of high-capacity trains that can now be found on the Overground. Unfortunately for TfL, being able to say “we told you so” doesn’t get you a ROSCO discount.
My first ROSCO
With big ROSCO budgets tight, the purchase and lease back of the Elizabeth line stock is, in many ways, ideally suited to a more hands-off investor or, more practically, a group of them, given the £875m that TfL are looking to raise. This is more than most potential investors would be willing or able to consider on their own.
Indeed the Elizabeth line rolling stock would be the ideal candidate for one or more new (or relatively new) players, and one suspects TfL might be pitching it as such to the market. With many trains already completed, in service or involved in testing and the new signalling and other systems included in the Crossrail project, the risk involved is far lower than for an early stage investment in the first order of a new rolling stock class elsewhere.
This means there is no need for the complex project or structured finance (equity and debt) type deal seen on Thameslink – which ultimately involved three private equity investors (one of whom was Siemens’ own private equity arm), a whopping 19 lending banks and two years of negotiations post-contract-award. As a side note, the Thameslink lenders included TfL’s own reliable standby lender – the European Investment Bank. Indeed the fact that the EIB will shortly no longer be available to TfL (or any UK body) as a source of relatively cheap finance may well be another factor playing into TfL’s decision to sell and lease the 345s. As with rail devolution, Brexit is perhaps once again proving to be a reason why London can’t have nice things.
The good news for TfL is that in the last decade a number of organisations outside the traditional ROSCOs have entered the great leasing game, often outlining their grand plans in press releases that start with the phrase “The first of many…” In most cases, it actually proves to be their last deal as well as their first and they often exit the market when a suitable refinancing or portfolio sale opportunity arises. So far only three new organisations have actually done multiple deals in a significant capacity. Two of those are likely to prove the potential models for the investors in Elizabeth – or may actually invest themselves.
Indeed one possible candidate is TfL’s old friend from the Overground – Sumitomi Mitsui (SMBC). They entered the UK rolling stock market in an attempt to diversify their asset holdings and have – by most accounts – been happy with the result. SMBC have been lending to the Japanese railway sector for decades, typically acting as the lead investor (and lender) in a leasing syndicate. What marks them out from the traditional ROSCOs is their ability to re-partner with the most suitable other investors at any given point in time. ln the UK this allowed them to quickly move beyond the Overground, becoming one of the many parties involved in Thameslink as well as various Scotrail orders. Crucially for TfL, SMBC seem to have got a real taste for oranges – they recently returned to London to fund the Overground stock that will soon enter service on the recently electrified GOBLIN, Watford and West Anglia Overground routes.
Another potential investor model is that of newcomer Rock Rail, which has been able to raise significant funding from pension funds and insurers (as well as a small number of banks who wish to invest though remain hands-off). Some of this is the result of pension fund mergers, which make the size of rail leasing deals a better investment for larger funds as well as changes in financial regulation.
The finer details
Any sale and lease back deal may also have to including financing for the options on additional Elizabeth line stock that Tfl have yet to take up (they have until 2023 to do so). This will inevitably make any deal more complex, as additional provisional funding will need to be arranged, despite the fact it may never be used.
This flexibility will add extra cost on any leasing deal, so the suspicion here at LR Towers is that TfL are currently mulling over a number of choices:
- Include the options as part of any sale and lease back
- Finance the options separately if (or more likely when) they decide to take them up
- Decide right now to take some or all of the options up, so they can factor them into the current deal.
The flexibility of the first choice will carry additional cost and the later requires TfL to make choices about increasing service levels before the Elizabeth line has officially opened. That said, the likelihood of TfL taking up some (or all) of the unused options is high, with the Elizabeth line’s western turnback station moving from Paddington to Old Oak Common when HS2 opens in 2027 and the possibility that the Elizabeth line might end up being the main operator of services on the new Western Rail Link to Heathrow (WRLtH) – that rare thing on which both Chris Grayling and Sadiq Khan agree.
Counting the readies
The key points of interest, of course, are what this raises for TfL up front, and what it’ll cost them over time. Kilonback confirmed that they are looking to get £875m from the deal in their 2018-19 financial year (which runs April – March). That should cover the cost of the Piccadilly Line stock at least. Whether it leaves much for additional work required as well though remains to be seen. If it does, one suspects it will not cover the full cost and the bulk of the money for the infrastructure on the Piccadilly line – including resignalling costs – will have to be found from elsewhere.
£875m is close enough to the price TfL paid for the Elizabeth line rolling stock that it suggests TfL are valuing the stock ‘as new’, even if the order was placed four years ago. This is not as weird as it sounds – the fleet is still mid-rollout and the devaluation of Sterling against the Euro and other currencies, as well as other inflationary costs since the original deal was signed, mean that the cost of purchasing identical ‘new’ rolling stock today would be up to 12% higher anyway, based on other recent deals.
Depending on the length of term TfL is looking for (presumably 20 – 25 years – which suits many current investors), where ownership would reside after the end of that term, and given that the associated risk is relatively low, the betting here at LR Towers is that TfL will probably end up paying out £55-80m a year in leasing costs, as a result of this deal. That will put total leasing repayments at 50-75% more than the sale amount.
On the plus side, that puts the repayment costs well within the projected Elizabeth line operating surplus, but it is tempting to wonder what TfL had that slice of money earmarked for originally.
Whatever it was, it won’t be happening now. But it is hard to see a world in which kicking off the Piccadilly line upgrades wasn’t likely more important. Certainly, TfL seem to believe so or they would not be doing this at all. Selling and leasing back the 345s is a pragmatic solution to a problem that must be solved. It is not ideal, certainly, but it hardly warranted the burst of outrage Kilonback’s statement seemed to provoke.
Indeed much of that outrage seemed to rest on a fundamental misunderstanding of how rail leasing actually works, or at least how prevalent it is (currently over 90% of UK rolling stock is managed by one of the big three ROSCOs).
In recent months, it has increasingly seemed that a new class of railway purist has begun to emerge alongside the traditional crayonista. Rather than drawing lines on maps, however, the renationalistas seem determined to reduce all rail decisions to questions of ideology. As with the debate over nationalisation, how rail leasing might work here is a far more complex one than those on both sides of the political spectrum would like most Londoners to believe. Ideology is a wonderful thing, but pragmatism builds railways or – in this case – buys trains.