The previous Government left HS2 curtailed, diminished in terms of its reach and capacity. True, a cut-back could reduce public sector capital outlays. But as this paper shows, high speed rail infrastructure, once complete, offers a great way – possibly a unique way – to raise private finance. It is a natural for consolidated public sector pension funds.
Here we look at the potential value of cash returns to HM Treasury from HS2. We have the model of HS1 to see what might be possible. This paper updates and replaces an earlier draft. We have now reflected some key differences between HS1 and HS2, and also the welcome decision taken in September to fund the previously held up Old Oak Common-Euston section of HS2.
The analysis remains preliminary. Financial values shown should be regarded as only indicative estimates of potential financial returns to HM Treasury.
1. High Speed One (HS1)
As the HS1 experience shows, once high-speed line construction is complete, and high-speed rail services become a day-to-day reality, the nation gains an asset which can attract large-scale private sector finance. Once built and handed over for day-to-day operation, most of the risks that make private sector finance more expensive than Treasury borrowing (legislative and planning risks, construction cost uncertainties, operational performance etc) have been resolved.
On 5th November 2010, after a two-stage construction programme and a joyous opening ceremony attended by HRH the Queen at St. Pancras International three years earlier, a concession to operate HS1 for thirty years was sold for £2.05bn. This amounted to a cash return to HM Treasury with a value equal to one third of the HS1 construction cost. The existence of a Government Guarantee through the ‘Domestic Underpinning Agreement’ was no doubt an important factor.
The concession was later sold on for an even higher price.
At the time of its initial sale, the high-speed line was complete and day-to-day operations had proven to be highly reliable, with average delays counted in seconds compared to minutes on the wider UK network. The period of high risk for private sector investors was over. The new railway could now be presented as a steady, long term earner, with a dependable income stream from track fees, and an upside potential of increased usage (more train services) in future. An ideal prospect: a steady, dependable and potentially growing return for Pension Fund investors.
New high-speed rail infrastructure has a unique and proven cash value to HM Treasury unmatched by other Government funded capital programmes. That’s because, unlike say schools or hospitals, for example, capital investment in rail yields a positive cash revenue stream, which in the case of a new high-speed rail line most readily takes the form of track fees.
Better still, in the UK there is an established tradition of independent regulation, through in this case ORR, which has proven successful in safeguarding a good investment climate for private investors. Through periodic reviews of the track charges this produces a predictable income stream for investors. On such occasions, the Regulator pays regard to and seeks to balance the real world needs of the several parties involved. This is what allows investors to take a long term view.
2. Careful rather than reckless curtailment of rail infrastructure investment
Given the constraints on public finances, the original wider national plan for high-speed rail infrastructure has been curtailed, at least for now.
But decisions on HS2 scope, we argue here, need to be taken in a way that has regard not just to Exchequer capital outlays on construction, but also to the subsequent cash inflows available.
The previous Government’s plan for HS2 – neither reaching central London at Euston nor re-joining the existing railway at Crewe and instead joining a bottleneck in Staffordshire – precludes the Exchequer from deriving full financial value from its outlay on the project.
It is one thing to curtail HS2, another to ensure that it still provides an asset of real financial value, as well as additional transport network capacity, so crucial for regional economic growth.
The integrity and quality of the infrastructure created determines the price of an HS2 concession. As with HS1, this could be payable to Government in short order following project completion, a unique and proven route to generate a cash return to the Exchequer.
3. Three scenarios for HS2
Here we contrast three scenarios for HS2:
1) The inherited position from the previous Government, with HS2 truncated to an Old Oak Common-Handsacre route – i.e. without either Euston or a new line onwards across Staffordshire to Crewe
2) The improved position, with government, following the October 2024 financial statement, now committed to funding the Old Oak Common-Euston section of HS2, which allows HS2 services to reach central London
3) The recommended position, delivering full value for tax-payers’ money, with HS2 completed to Euston in central London and with a new line built between Handsacre and Crewe to avoid the main bottleneck on the existing railway.
The value of an HS2 concession to tax-payers/Treasury is different between these three cases. Following the HS1 example, the value created for private sector investors in HS2 would be based on the number and quality of train paths that can be offered and taken up on the new line. Each train path, each day, generates the cash return to private sector investors, and is what determines the amount they would pay for a long term concession.
The first – ‘inherited’ – HS2 scenario (Old Oak Common-Handsacre) is seriously restricted in what services can be run, so the concession value we would expect to be low.
Following the Government’s welcome October 2024 commitment to provide the funding for the Old Oak Common-Euston section of line, the prospects for a better value concession for Treasury are improved (although a resolution of plans for Euston remains an absolute necessity too). This we term the ‘improved’ scenario.
But as we show here, the potential ££ value to HM Treasury would still be diminished if the Handsacre-Crewe bottleneck is not addressed as well. HS2 completed both to Euston and Crewe we describe as the ‘recommended’ position.
It should be noted that there are other ways for HMT to repatriate some of its cash outlays following line completion. It could seek to realise values for instance through train service operating concessions, for example, rather than an infrastructure concession as used with HS1.
4. Timescales
For HM Treasury, the elapsed time to the start of a private sector finance cash return is likely to be crucial. A key determinant of the ‘how fast can we earn a payback? question’ is whether Parliamentary powers for construction have already been granted by Government.
There is little point in considering here options that envisage new line construction north of Crewe, for example, because parliamentary powers don’t (yet, at least) exist for any such schemes.
Only those parts of the original HS2 plan that have gained Parliamentary powers could be described as ‘ready to go’ into the construction phase (or are already in construction) as is the case for each of the three HS2 scope options considered here. The main section of HS2 between Old Oak Common and Birmingham/Handsacre is deep into the construction phase, and approximately half-way to completion.
Full Parliamentary Powers exist for HS2 to reach Euston and Crewe (but no further parts of the wider HS2 plan). With completion of HS2 to Euston and Crewe still possible in the early 2030s, a cash return of this new railway could be fully realised within the next ten years (two Parliamentary terms).
But this would need decisions on restoring these shorter (but crucial) sections of the line to be taken by (say) Spring 2025 for the Handsacre-Crewe line, quite possibly as a de-scoped version of Phase 2a. As we will see, an incomplete Phase 1 project, without Phase 2a which was always intended to be opened at about the same time, is much less valuable in cash terms.
Scope Changes
Any changes to the specification of the project, up to a point, may be permissible without needing to seek revised planning powers. The suggestion of the report from Sir David Higgins and a private sector group working on behalf of the Mayors of Greater Manchester and the West Midlands, is to seek to realise capital cost savings from changes to technical specification of the Phase 2a scheme (between Handsacre and Crewe). This would be achieved by dropping larger ‘European’ train gauge profiles and reducing the top design speed), and should not mean a need to re-visit Parliamentary consents.
More substantial changes, for instance adopting new alignments, or looking to incorporate new lines northwards from Crewe would generate a requirement for Parliamentary powers. Re-starting the planning process guarantees a fresh delay both to project completion and to the prospect of cash returns to HM Treasury. Getting planning consents is a lengthy process and is getting no easier.
Realistically, for this assessment, only the three options noted (inherited, improved and recommended) need to be considered.
5. Concession Value: Learning from the HS1 experience
We can see from the relevant National Audit Office report how the HS1 concessioning process was handled at the time:
“In restructuring LCR before the sale, the Department for Transport removed open-ended taxpayer support and made the line an attractive opportunity for investors. The Department handled the sale well and, at £2,048 million, the winning bid was higher than expected.”
[source: https://highspeed1.co.uk/media/vo1nqud3/structure-of-charges-review-initial-consultation-may-2021.pdf]
This was for a 30-year concession with the option to sell on.
“The UK government sold a concession to operate and maintain the stretch of railway and several stations between the St. Pancras station in central London and Ashford International in 2010 to two Canadian pension funds.
The stretch is the UK’s first high speed rail line and had cost the government £5.7 bn to build (or £6.2 bn undiscounted). The Canadian pension funds paid £2.1 bn to buy a 30-year concession from the UK government, with the consideration exceeding expectations by £600 million.”
Six years later the first concession was sold on to the highest bidder adding a premium to the initial sale price, and achieving an enterprise value of roughly £3.3 bn.
The recent Government policy decision to aggregate UK public pension funds could serve the country well in increasing the scope for UK-based pension funds to acquire an HS2 concession.
High Speed One in Kent
Photo: Partnerships Bulletin
HS1 Ltd’s charging framework was established in 2009 by the DfT in its Concession Agreement. Under this framework, track access charges and station usage charges are levied on train operators that use the HS1 infrastructure.
An Investment Recovery Charge (IRC) was set at a level designed to recover part of the long term capital costs of the HS1 project. The value of the IRC was capped and indexed to RPI. It was set by the DfT prior to the commencement of the HS1 concession and is fixed for the duration of the HS1 Concession Agreement.
The IRC was designed to recover the costs that HS1 Ltd directly incurs in operating, maintaining and renewing HS1, and long-term project costs incurred from the initial construction, and long-term operation of HS1. In practice it does not wholly recover the costs of HS1 construction, but – as we shall see – it is a very significant payback, nonetheless.
No asset sale besides the precedent of HS1 has generated such a substantial financial return for Government in the last 15 years.
So what would an equivalent arrangement for HS2 look like? We think there are three scenarios to consider.
6. Three scenarios for HS2
These are:
Scenario 1: Truncated to an Old Oak Common – Birmingham/Handsacre route – i.e. without either Euston or a new line onwards to Crewe (this we call the inherited position)
and
Scenario 2: Extended from Old Oak Common to a new station at Euston but not northwards from Handsacre to Crewe(this is the improved position)
and
Scenario 3: An HS2 completed to Euston in central London and with its northern limit set at Crewe rather than at Handsacre (the recommended position).
As with HS1, in letting a concession for HS2, HM Government will be looking to recoup as much of its capital outlay as possible. Bid prices for an HS1-style concession will depend on the quantity and quality of the train paths that are sold for the daily operation of the line. These vary between the three scenarios.
There will be a charge for each path payable by the train operating companies using the line, and this will likely form the primary income stream for the concession holder. The number of, and quality of, train paths is what differentiates the three scenarios and the price that can be expected from any HS2 concession.
Two relevant factors that impact on concession price are line capacity (the number of train paths) and the quality of these train paths. In this preliminary analysis, neither attribute can be quantified with any certainty. The following assessment needs to be treated as a preliminary indication accordingly.
1) Capacity
In terms of line capacity, in Scenario 1, trains on HS2 will need to terminate at Old Oak Common. But this station is not designed as a terminus. This means that arriving and departing trains need to cross paths with each other approaching/leaving the station. It is estimated that at most 6 trains/hour could be operated in this way, well under 50% of HS2’s line capacity.
Indeed, it might well be more realistic to assume that services would operate only between Old Oak Common and Birmingham over HS2 in Scenario 1 which means only 3 trains/hour. With HS2 services running to Old Oak Common, Birmingham will also continue to have direct train routes to central London terminus stations (Euston and Marylebone).
So for Birmingham, the initial absence of a central London terminus on HS2 is mitigated by having other routes available. But this option doesn’t arise for other HS2 services (to Manchester, Liverpool and Glasgow).
The implication is that Scenario 1 could well be best represented as a 3 trains/hour case.
In Scenario 2 (Euston-Handsacre), it might be realistic to operate 6 trains/hour over HS2 to/from Euston Three of these trains each hour would operate between Euston and Birmingham, and three would join the West Coast Main Line (WCML) at Handsacre (serving Liverpool, Glasgow and Manchester). Each of these services would likely require the withdrawal of an existing WCML service, given the network capacity constraints between Handsacre and Crewe.
The original business case for HS2 suggested 10 trains/hour would be operated once Phase 1 is open to Euston (which is Scenario 2 here). But while there is certainly scope to remove some of today’s west coast ‘Pendolino’ services that will in effect be replaced by HS2 trains, other fast services WCML services will need to remain in order to continue appropriate service provision to locations not served by HS2 – places that include Milton Keynes, Rugby, Nuneaton, Tamworth, Chester, North Wales, Warrington, Blackpool, and others.
In Scenario 2 there would be no Birmingham-Manchester HS2 services (as per the original business case assumptions).
In Scenario 3 (Euston-Crewe), there would be 10 high-speed London trains per hour, three serving Birmingham and seven serving Liverpool, Glasgow and Manchester. There would also be two trains/hour operating on HS2 between Birmingham and Manchester via Crewe, so twelve trains/hour in total.
This is four times the assumed Scenario 1 throughput and twice the Scenario 2 capability in terms of train paths.
Scenario 1 has just 3 trains hour on HS2; Scenario 2 has 6 trains per hour on HS2 and Scenario 3 has12 trains an hour (two of which would be Birmingham-Manchester services).
2) Quality of train paths
Under Scenario 1 (Old Oak Common-Handsacre), travellers to/from central London will need to take the already busy Elizabeth line to complete their journeys. No other Underground lines are available at Old Oak and the surrounding road network is poor. So the quality of HS2 train paths and the perceived value they bring for travellers in a case where HS2 only reaches Old Oak Common, is low. Fewer people would choose to travel by HS2 than if the same services were extended to Euston in central London, and so the commercial value of HS2 train service provision would be diminished.
In general, shorter high-speed train paths are less valuable, diminishing concession value. In the HS1 case, the infrastructure cost recovery charge available under the concession was set by the time each train spends on HS1, rather than distance travelled along it (apparently to provide an incentive for service providers other than Eurostar to operate trains at or close to Eurostar’s line speed).
7. Comparative concession values for HS2
With the October 2024 commitment to fund the Old Oak Common – Euston section of HS2, and assuming that the HS2 station at Euston is provided, perhaps the most important comparison becomes between Scenarios 2 (HS2 from Euston reaching Handsacre) and 3 (HS2 from Euston reaching Crewe).
Assuming Government elects to adopt an infrastructure concession model approach as per HS1, there would be a significant difference in the realisable concession value for these two scenarios. This is because a combination of factors – fewer paths and less value/path in Scenario 2 – it offers only half as many train paths to sell through any concession, and (London-Birmingham aside) high-speed train paths are shorter in Scenario 2. These two factors compound and mean that the Scenario 2 concession (‘improved’) value derived from path charges might be only around 35-40% of Scenario 3’s value (the ‘recommended’ Euston-Crewe version).
We conclude based on this preliminary assessment that the concession value of the Scenario 3 railway is between 2 and 3 times higher than the Scenario 2 railway. Payment to HM Treasury could be realisable in the 2030-2034 period, if HS2 is progressed without further delay.
8. How much would the concession raise for the Exchequer?
If the same ratio of concession value to capital outlay was realisable as achieved with HS1 (a third of capital outlay), the HS2 concession when complete from Euston to Birmingham and Crewe would be worth around £15-£20bn. But we must acknowledge that the ‘if’ here is a ‘big if’. In fact such a price is almost certainly unachievable.
There are some key differences between HS1 and HS2. While HS1 carries the international Eurostar trains, actually most of the train paths are taken by Southeastern’s high-speed services across Kent. Interestingly, these were made subject to a +30% fares premium when the go-ahead was given in 2003 to fund the Class 395 ‘Javelin’ trains that operate this service. With a strong and growing commuter market, the effect was substantial revenue growth, which ultimately flows through to DfT through the franchise (now contract) agreements.
Southeastern high speed Class 395 ‘Javelin’ train on HS1 at Ebbsfleet, Kent
Photo: Greengauge 21
No decisions have been taken on fare levels on HS2, and so it cannot be assumed that a similar commercial premium would be applicable for HS2 train operating companies. This matters, because ultimately the value of the HS2 infrastructure realisable through a concession sale depends on the overall commercial value in HS2 train services. It is likely that HS1 with its fare premium allows a higher infrastructure concession value than would be realisable on HS2.
There is a second factor to consider when making the HS1 vs HS2 comparison. In estimating an HS2 concession value based on pro-rating the commercial return from a given capital spend, it needs to be recognised that the capital cost of creating high-speed rail infrastructure has increased substantially in the 25 years or so since HS1 was built. Part of this is due to higher design standards applied to HS2, including long term weather resilience specifications to deal with the challenge of climate change.
For these two reasons, we believe it is necessary to reduce the estimated value of an HS2 concession given above.
Both of the factors noted (premium fares payable on HS1 and higher capital costs per mile constructing HS2) suggest that the estimates of HS2 concession values for HS2 shown above would be too high. We therefore halve the values estimated above, to account for the two key factors noted in making estimates of HS1 vs HS2 concession values.
9. Conclusion
On this basis, our preliminary estimate of the value of an HS2 concession lies in the range £7.5bn – £10bn. This applies to our Scenario 3, the ‘recommended’ case, with HS2 fully open from Euston to Birmingham and Crewe.
If HS2 is concessioned without the Handsacre-Crewe section its value would be much less, because of the inability to provide HS2 train paths through the existing WCML bottleneck in Staffordshire. With fewer train paths available and HS2 restricted to a Euston to Birmingham/Handsacre facility (our Scenario 2), we estimate the concession value to be between £2.75bn and £3.75bn.
The differential in concession value is striking and reflects in particular the increased line utilisation that becomes possible once HS2 is able to reach Crewe as well as Euston.
Reaching a fully-functioning terminus at Euston is important, because the most likely outcome if HS2 trains go no further than Old Oak Common would be just an Old Oak Common-Birmingham Curzon Street high-speed service of 3 trains/hour.
The ’no Euston case’ was our Scenario 1 and we found its likely concession value to be lower still, possibly returning around only £1.5bn to the Exchequer. As with HS1, the HS2 concession holder would benefit from a charge in using the London terminus as well as track user charges, provided the HS2 station at Euston is built.
So this analysis makes clear that there is a much higher ££ return to Treasury if the HS2 project is completed to both Crewe and Euston.
And it remains the case that the concession value that HS2 can generate for HM Treasury is most unlikely to be matched by any other comparable public sector asset sale.
Greengauge 21
We actively encourage people to use our work, and simply request that the use of any of our material is credited to Greengauge 21 in the following way: Greengauge 21, Title, Date.