What can governments in Asia learn from the UK’s various attempts to share with the private sector the costs and risks of running a national rail network?
The UK invented railways in 1825; nationalised them in 1948; denationalized them in 1994 -7; and last week published the Williams – Shapps Plan for Rail under which ownership of railways in England outside London will settle on a structure somewhere in between public and private. Keith Williams used to run BA then the Post Office and chaired the review; Grant Shapps is the Transport Secretary and quipped that “we won’t be going back to the days of [nationalised] British Rail with terrible sandwiches [pictured] and all the rest of it”.
No matter how it is owned, the financing of a railway system is invariably opaque. Revenues comprise a combination of visible fares from passengers and less visible support in the form of hefty subsidies from government; this support is justified because even less visible are the nonetheless real external benefits such as less congested traffic elsewhere, cleaner air, increased real estate values and more. This government support represents a transfer from the tax payer to passengers and external beneficiaries. Meanwhile, costs, whether capital or operating, can be difficult to assess so are prone to inefficiency.
Privatisation of the UK rail network has brought an impressive increase in utilisation – as of the eve of the pandemic, 21,000 services ran on an average day, up a third from before privatisation; passenger numbers had more than doubled; rail had achieved its highest share of all miles travelled in Great Britain since 1967; and a mile of track in Britain carried twice as much traffic (safely) as in, say, France.
However, whilst the UK may have pioneered the successful privatisation of many sectors in infrastructure, rail has not been one of them. A third of trains still ran late in 2019 / 20 and they were often crowded and uncomfortable. On the financing side, fares had risen 48% in real terms since 1997 despite the government contributing £150 billion, equivalent to a third of all revenues in that period. Worse, capex was being postponed – perhaps not surprising when the cost of electrifying the Great Western line, for example, had gone from £800 million to £2.8 billion.
Breaking the old British Rail monopoly into bite – sized chunks was intended to create some competition but this was easier said than done. Instead, came:
- A breakdown in coordination: 75 different types of train were in passenger service. Network Rail, which owned the track and stations, at one point lengthened platforms in Essex to cater for 12 – carriage trains only for the new operator to choose more frequent but shorter ones;
- Increased bureaucracy: 400 “train delay attributors” were employed to allocate responsibility for delays. Is a pheasant a small bird so the Train Operating Company was responsible for the delay caused by hitting one? Or is it a large bird in which case Network Rail was responsible? (It’s not even clear why there was a delay in the first place.) The Ticketing and Settlement Agreement between operators ran to 922 pages; and so on;
- Waning appetite from the private sector in that since 2012, two thirds of contracts had attracted only one bidder.
All in all, there was a complete loss of accountability.
The white paper envisages most roles once again being carried out by a public body to be named Great British Railways, a much larger organisation than Network Rail if not as huge as British Rail. GBR will take the risk on revenues, most costs and capex. With budgetary discipline swept away by Covid, the government will pitch in with £40 billion over the next four years for HS2 (Hmm …), the new line from Oxford to Cambridge and even some reopening of lines closed by the infamous Beeching report in 1963 (welcome). The private sector will still be involved (cue mention of public / private collaboration on the Astra Zeneca vaccine) but on a limited contractual basis via freight and Passenger Service Contracts. The detail is still to be sorted out but a more sensible balance of risk / reward has already been welcomed by the likes of FirstGroup.
If implemented, many of the white paper’s proposals look sensible. As ever, though, what is interesting is what isn’t in the report and here Logie Group offers some suggestions:
- If trains running late has been a perennial problem, did the study look to the likes of Japan? In fiscal 2019, the average delay on the Tokaido Shinkansen between Tokyo and Osaka was 12 seconds. And since 1987, the railways in Japan have been privately owned, by the regionally – organised Big Six in the JR Group but also by a further 100 or so smaller companies.
- Most infrastructure comprises a relatively inflexible supply built to meet a demand profile which varies enormously by time of day, week, or month. Rather than building capacity which will lie idle out of peak hours, how can the government flatten the demand curve (to borrow an image from Covid)? The power sector faces a similar imbalance and has toyed with the idea of paying for Negawatts (i.e. a reduction in demand) rather than Megawatts (i.e. yet more increase in supply). The rail study does foresee some immediate reduction in peak hour travel as two thirds of its passengers may change their behaviour as a result of Covid. However, more needs to be done, for example through more flexible pricing for peak / off – peak travel. Thus, a three – day season ticket is to be introduced; this should be attractive but it will cannibalise the traditional full season ticket.
- Infrastructure also features significant fixed costs. Track and rolling stock assets need to be sweated. Off – peak ought to mean all through the night.
- It is all very well to promise extra investment but how will this money be spent? Hopefully on signaling. Today’s technology can revolutionise signaling and, by running trains more efficiently, significantly increase the capacity of existing track.
- Where does this leave the ROSCOs? The three rolling stock leasing companies were hugely popular with investors.
- As examples of capturing external benefits, the MRT in Singapore and the MTR in Hong Kong are best thought of as property companies which happen to run a rail network since they fund themselves largely by developing real estate on top of their stations. The UK government may not own the land adjacent to its stations but it does own the air rights above them. The engineering would be challenging but it can be done.
- It is not clear how GBR will dovetail with the Scottish Government’s plans to nationalise train services north of the border by March 2022. This paper will add fuel to the devolution fire already simmering between Holyrood and Westminster.
Rail’s bicentenary is four years away. The Chinese, Japanese and French in particular may have stolen a march on the Brits but few other governments have attempted to share the costs and risks of a rail network with the private sector. As ever, they should not reinvent the wheel (or bogey) but learn from the UK all over again.
