I have for some little while been writing a book about how London Underground has expanded since it was first built. You will already realize its origins came in two bursts: the first was 1863-1868 when the first steam-operated sub-surface lines were opened and the second was in 1900-1907 when most of the central London deep tube network appeared. After each network got going, the original system was expanded to form what we have today. But the circumstances around which expansion took place, and even more expansion did not take place, is not covered at all well by the standard historical works. A huge amount of material has been left out and the nature of the difficulties that made any kind of expansion nearly impossible is quite overlooked. My book is still not ready for the world, so I thought it time to make known a few of my findings in the hope that they will be useful.
The thrust of my book will be that we have an incredible system in London, but not until the 1960s was one penny of public money contributed towards it. The problem was that the system was so marginal in making a return, and in truth probably didn’t make a real return at all, that private finance could not be freely obtained either. The outcome was a typical British fudge, the kind of thing that the British are apparently rather good at, and money was raised for extensions that were neither large enough or numerous enough, but fantastic when they happened. After WW2 things went from bad to worse and not only investment was very difficult but timely renewals (for which provision should have been available) suffered too. This has left us with a system that is in exceedingly variable condition and much harder to maintain than it should be (for which the ill-fated PPP was supposed to deal, but that is for another day).
The Core System
I will start off with a mere observation. In Victorian and early Edwardian times there was no shortage of schemes for improving railway transport in London. Railway Acts of Parliament tumbled through the parliamentary machine and probably more than half never got built. Companies were formed and prospectuses were issued for the most implausibly ambitious schemes. Given the difficulty that even the better amongst them had in raising money, it is curious that so much effort was wasted on concepts that had little chance of going anywhere.
Then there was an ‘event’. After that event there were virtually no further wild schemes for expansion of the Underground by the buccaneers who were so evident previously, and nor did the established companies rush to put schemes before Parliament. The ‘event’ was the near bankruptcy of the Underground Group during 1908 and its voluntary receivership to allow financial restructuring of a type which got it out of its immediate difficulty but made life awkward in later years. I think I am on fairly firm ground if I state that after 1908 there was no occasion when the Underground Group ever sponsored another extension of the system, anywhere, without some kind of third party support. At any rate, the rush to build Underground railways in London took on a very different complexion.
This, not entirely trivial, change of course seems to have avoided much prominence by those covering London Underground history, yet it is the very basis of why most of the Underground’s subsequent extensions were built where and when they were, or at all. Since most of the Underground’s mileage in fact constitutes extensions to the early lines then light should surely be shed?
The 1908 problem resulted from several factors, but the main ones are that the system of lines financed by Edgar Speyer and Charles Yerkes cost more to build or electrify than was expected and when opened or improved generated less traffic than was hoped for (these were not the only reasons but made dealing with any other issues virtually impossible). This was not only instantly discouraging for anyone thinking that more lines were needed but threatened bankruptcy of the Underground Group itself. Essentially what happened next was that repayment of a very large quantity of loan notes was only possible by issuing new bonds with a long repayment date and a near-exorbitant rate of interest, required to attract finance to a company that appeared a little unstable. It worked, but at a huge price for which Londoners should be grateful but of which they probably had no understanding at the time.
The Need to Expand
The situation now was that the Underground Group and its various subsidiaries wanted to expand, and needed to expand, to make better use of the new or revitalized lines, which had lots of capacity available and were proving expensive to run for the traffic actually being carried. The problem was that the companies had no spare money and no practical means of borrowing any. With the existing finances in disarray, the only way of obtaining money for an extension is if, as a direct result of that extension, the revenue increased sufficiently to pay the additional operating costs, the cost of financing the extension and included suitable provision for keeping it up to date. If the existing central London system couldn’t be certain to do that, any extension (by its nature less busy) would seem likely to make the financial position worse and not better given the traffic levels expected. Not only would this make raising money very difficult indeed but the directors had a legal responsibility to their shareholders not to behave recklessly by embarking on any scheme likely to disadvantage further its shareholders, who were already enduring some financial pain.
It is important to understand how all this worked. Take a modest, but hypothetical, extension of a railway line on a railway where the existing shareholders are receiving a dividend of (say) 2% against an expectation of 7% claimed in the original prospectus (the ordinary shareholders own the company and usually appoint the directors). A new extension is desired, but has to be paid for. Soundings amongst the shareholders indicate there is no chance whatever of their risking any more of their own money on buying new ordinary shares. If they had spare cash, they would be better off investing it elsewhere, such as nice safe government bonds. Furthermore, railways were usually forbidden by Parliament to borrow money in any proportion greater than a third of their total shareholding, and were often already mortgaged to the hilt. So, nothing happens.
But there is one thing a railway wanting to extend can do, but it isn’t cheap: it can issue shares that guarantee a worthwhile return. These shares come in a wide variety of forms but typically take the form of ‘preference’ shares to which is attached a stated rate of income. This income is not guaranteed, but because preference shares are paid out of profits before any distribution is made to ordinary shareholders (hence the name) a railway would have to be doing disastrously badly for a preference shareholder to get nothing. [Some Underground railways did do disastrously badly, the Metropolitan District, for example, only rarely paid a dividend to ordinary shareholders and preference shareholders rarely received their full amount either (and most tube lines struggled to pay more than 1% on ordinary shares)]. For this reason preference shares had to make the ‘promised’ return quite high or nobody would be interested. Amounts between 4% and 6% were quite usual. Preference shareholders rarely had the same rights as ordinary voting shareholders. Of course, because the voting shareholders were now shunted to the back of the queue for receiving any return on their money, they, too, had to be happy that they would actually be better off and that their railway would be a better business as a result of the proposed extension being built.
So we now have a position where an extension is financed by means of a mixture of preference shares and further borrowing, which was also expensive (probably also 5 or 6%). In addition, borrowing required, at least in theory, some provision for paying it back, which was not necessary in the case of shares. Because extensions had to be financed this way, it made them even more expensive than they otherwise would have been, and therefore even more difficult to justify.
Let us look at an example of a hypothetical extension perhaps in the 1920s
An extension will cost £2 million to build, and therefore £2 million must be borrowed. The financing cost even at 5% is £100,000 a year. The additional costs incurred as a result of operating the extension (staff, electricity, maintenance and so on) will be (say) £225,000. The total annual costs to be met are therefore £325,000.
Traffic, once it has settled down, is expected to reach an average level of 15 million trips a year and it is estimated that the average fare paid will be 6d, generating £375,000. Costs appear to be £50,000 less than income, so we do it? No we don’t. Here are some reasons.
- The 15 million a year figure takes ten years to build up to that level and does not arrive overnight when the extension opens. A loss is inevitable initially and this needs factoring into the costs.
- We need to ask where the people come from. Many will be new people filling up empty suburbs, but some, say a quarter, will transfer from bus services also run by the Group. If we considered that transfer of passengers (albeit a good thing) will reduce bus profits by (say) £15,000 a year, then that cannot be counted as new money from the perspective of the group as a whole, and we must discount it too. This can equally work the other way around if bus feeder services are possible, but we will assume not in this particular example.
- The shareholders and lenders expect a return on their money from the moment they part with it, while the extension might take three years to build before it becomes revenue earning. Provision is therefore made for paying interest immediately out of capital, so more capital needs to be raised than is actually needed for the railway. That has a cost.
- Raising the money isn’t free either. Commissions have to be paid to the brokers and possibly discounts have to be offered if people don’t rush to subscribe to the shares.
When all these factors are taken into account, we find we are actually borrowing £2.1 million and perceived financing costs are more like £375,000, when all the factors I have mentioned are considered (and there are other’s I haven’t mentioned, like parliamentary costs). Meanwhile it seems that actual revenue (that means ‘extra’ money after allowing for passengers transferring from other modes) looks to be nearer £355,000 a year, and the extension, in reality, appears likely to be loss making. We therefore do not do it.
The Need for Another Way
The point is that many railway extensions are fairly marginal because of the high costs of building and operating them. What has put the above example into the ‘unbuildable’ category is the method of financing. We are adrift by a mere £20,000. If the financing costs were 3% and not 5% then we would be looking at a £20,000 profit and not a £20,000 loss. The way the money is raised is therefore absolutely crucial, and has hugely affected the way the Underground has expanded.
The only extensions proceeding after 1908 on the basis of raising finance this way were Bank to Liverpool Street, Edgware Road to Paddington and Paddington to Queens Park, but these received some financial support (or its equivalent) from the Great Eastern, Great Western and London & North Western Railways which assisted schemes that were not all that expensive and quite likely to be busy anyway. There were certainly no more extensions built on this basis (it will be recalled that services to Watford and Ealing ran over main line railway tracks and the financial arrangements were completely different and involved little Underground capital).
So, how did the extensions and reconstructions come about? First, both the Underground Group and the Metropolitan Railway (which had exactly the same problem) did a lot of planning. The Underground planned on an extraordinary scale under the leadership of the good Mr Pick, whilst the good Mr (shortly to be Lord) Ashfield worried about how to pay for it. Pick never intended his official’s ideas to represent any kind of plan for virtually none of it could be funded, but the process flushed out where the Underground considered new lines would be needed. Nearly all of the proposals in mind were in the open air or involved inter-running with main line trains. By 1920, some discussions had already taken place with main line railways about the possibilities. This was sensible as (with some exceptions) the main line railways found it very expensive and disruptive to handle large quantities of local traffic, which wanted to travel, all at once, for just four hours a day. The Underground was better equipped to do this and would also divert the traffic away from main line terminals which in those days were quite unsuited to such crowds.
I will set out what actually happened in my next blog, and why Pick’s grand ideas were initially sidelined for reasons of expediency. That we have the suburban extensions we do is at least in part down to Ashfield’s remarkable circles of friends and his great fondness for lunch.