High Speed Rail (Preparation) Bill

Written evidence from BiblioFox Research (HSR 18)

1. BiblioFox Research is a research consultancy established in 2000. We specialise in document-based research and archival research and have recently been investigating the implications of the HS2 business case for existing intercity services on the West Coast, East Coast and Midland Main Lines.

2. Summary
The HS2 business case and cost-benefit ratio are based on an assumption that substantial savings to Government can be made through cuts to intercity services on the existing rail network once HS2 opens, and that these saved funds will be used to meet part of the cost of operating HS2. In offsetting such ‘classic line savings’ against the projected cost of operating HS2, and applying a contingency to account for optimism bias, HS2 Ltd appears to have made an elementary arithmetical error in their spreadsheet, as they have added the optimism bias to the classic line savings instead of subtracting it. The effect of this is to grossly overstate the savings that will be achieved and thereby to understate the cost of operating HS2 by some £126 million per annum. Given such an error, we would question whether HS2 Ltd has the competence to manage the preparation of such a vast and expensive project as HS2 and to keep costs under control. The Bill as drafted does not require the Secretary of State to prepare a first report on expenditure until 31 March 2015 so, while most of the preparatory work for HS2 Phase 1 is being undertaken, Parliament will not be in a position to scrutinise HS2 Ltd’s expenditure. Furthermore, the detrimental effect that the downgrading of existing intercity rail services will have on the economies of a number of towns and cities in the Midlands and North does not seem to have been taken into account in the business case. We believe that the negative as well as positive economic impacts of HS2 should be fully identified and made public before further public money is committed to the preparation of the scheme.


3. Recent reports on HS2 have focused on the likely cost of building the new high-speed line and acquiring the rolling stock. The predicted cost of operating HS2 once it opens has not received as much attention. In its August 2012 business case spreadsheets, HS2 Ltd calculated that the London-West Midlands phase of HS2 will cost £484 million per year to operate (at 2011 prices) when it opens in 2026, and that the operating cost will rise to £1.5 billion per year when the ‘Y’ network is complete in 2033 (again, at 2011 prices). [1]

4. HS2 Ltd has assumed that a substantial part of this cost can be met by diverting Government subsidy from existing rail services on the West Coast, East Coast and Midland Main Lines to HS2 services. Built into their business case are cuts in subsidy to services on the existing lines of £133 million per year in 2026, rising to £379 million per year in 2033. In its spreadsheet, HS2 Ltd describes these as ‘classic line savings’. [2]

5. The actual rail service changes that HS2 Ltd has modelled to obtain those savings can be identified from their Updated economic case for HS2 (August 2012): Explanation of the service patterns. [3] The changes comprise a reduction in the number of long-distance intercity services running on the existing lines and an increase in the number of commuter services, which together produce a net saving of £379 million per annum from 2033. In HS2 Ltd’s plan, from the year 2033 the West Coast intercity service between London and Manchester would be cut from three trains per hour to one, leaving intermediate stations with a poorer service than today. Stoke-on-Trent would receive half as many trains to London as today, Stockport’s service to London would be cut by two-thirds and Wilmslow left with no direct services to London at all. On the Midland Main Line, direct services from Nottingham and Sheffield’s city centre stations to London would be halved, as would Chesterfield’s and, on the East Coast Main Line, Wakefield would suffer the same fate. HS2 Ltd suggests that all ordinary intercity services from Liverpool to London (as opposed to HS2 ‘classic compatible’ services) could be re-routed via Birmingham, while passengers wishing to travel on traditional intercity services to London from Glasgow, the Lake District and Lancaster would see all of their trains diverted via Manchester. Both diversions would increase journey times for passengers by about an hour. Finally, Aberdeen, Inverness and all stations north of Edinburgh lose all direct trains to London in HS2 Ltd’s document. [4]

6. HS2 Ltd’s reasoning for these service changes is that, with most passengers from cities served by HS2 switching to high-speed services, there simply will not be the passenger numbers to warrant the same number of ‘classic’ intercity services as today. Indeed, demand modelling analysis undertaken for HS2 Ltd suggests that intercity services on the southern part of the West Coast Main Line would be only 27% full on average once HS2 takes the majority of passengers travelling to and from Manchester, Leeds, Sheffield and Birmingham. [5]

7. The cuts to intercity services that HS2 Ltd has built into the business case for HS2 would, of course, badly affect those towns receiving no HS2 services in place of axed classic line services. [6] In HS2 Ltd’s plan, the line capacity that these cuts and diversions would free up would be used to provide more commuter services, largely into London from Milton Keynes, Peterborough and Bedford. [7]

8. It does not appear that HS2 Ltd has assessed the negative economic impact that these service cuts would have on Stoke, Stockport and other affected towns before incorporating these service reductions into the HS2 business case and cost-benefit ratio. Moreover, there is a significant error in the way HS2 Ltd has calculated these savings. We first noticed this error in an earlier iteration of HS2 Ltd’s costs spreadsheet. More importantly, the RAC Foundation publicly noted it in their review of the economic case for HS2, published in 2011. [8] Yet despite this published evidence, HS2 Ltd has done nothing to correct the error and it has persisted in all subsequent versions of the HS2 business case and cost-benefit ratio. The details of the erroneous calculation are given below.

9. After calculating the probable cost of running HS2 services (and before offsetting part of the cost with savings on the existing lines), HS2 Ltd has increased the headline cost of operating HS2 by 41% in their spreadsheet, to provide a contingency in case of cost increases (so-called ‘optimism bias’). When dealing with the projected savings from cuts to existing services, HS2 Ltd subdivides these into savings on existing rolling stock leases (£123 million per year from 2033) and other savings, namely a reduction in spending on drivers, guards and electricity on the West Coast, East Coast and Midland Main Lines (£256 million per year from 2033).

10. When it comes to assigning a figure for optimism bias to these savings, HS2 Ltd has opted for an 18% contingency on lease savings, as it believes savings from cancelling leasing arrangements on some rolling stock on the existing lines can be reasonably reliably predicted, and 41% on the rest of the savings, as these are less predictable. However, HS2 Ltd has made a basic arithmetical error in applying these 41% and 18% contingencies to the savings: they have added the sum for contingency to the savings instead of subtracting it - thus producing an even larger savings figure than before the contingency (risk) element was added.

11. In other words, having worked out which rail services could be cut, and calculated that this would bring in savings of £379 million per year from 2033, HS2 Ltd has then said that, actually, £506 million per year will be saved and used to help pay for the running of HS2. This is £126 million more, annually, than the proposed service cuts will deliver.

12. Any risk associated with making savings by cutting some existing intercity services must be that the required savings cannot be fully delivered, not, absurdly, that more services will be cut than expected. If a contingency is to be applied to the savings, we believe it ought to have been subtracted from, not added to, the projected savings.

13. The business case and cost-benefit ratio for the whole HS2 project are based on this erroneous figure of £506.1 million of savings per annum, but such savings would be impossible without further deep cuts to existing rail services – far beyond those set out by HS2 Ltd.

14. HS2 Ltd’s error appears to result from the fact that they have used the same spreadsheet formula to deal with a percentage of a negative number in the sheet (the classic line savings) as a percentage of a positive number, assigning a 41% optimism bias simply by multiplying the original number by 1.41 in both cases. But if a 41% risk factor is added to a saving, it should, we presume, be multiplied by 0.59 to reduce that saving - reflecting the fact that the risk is that the saving will not be fully made, not that it will be exceeded.

Annex 1

15. The problem can be illustrated most clearly using HS2 Ltd’s Profile of Costs sheet in the August 2012 spreadsheet Day 1 & Y costs.xls (Table 1). [9] Here, the total projected cost of HS2 over the 60-year appraisal period is represented in terms of present values, before and after optimism bias has been applied to operating costs and savings. Before optimism bias is added, the projected classic line savings are given as £5.746 billion. Once optimism bias is included, the required savings have been inflated to £7.704 billion.

Table 1

Discounted Market Prices

Total (£2011, PV), No Optimism Bias



Track Renewals


Rolling Stock Purchase


Rolling Stock Maintenance






Track Maintenance


Station OpEx


Station Maintenance






Capacity Charge


Electrification Asset Usage Charge




Variable Overheads and Admin Costs


Other On-Train Staff


Classic Line Savings

Total (exc. Driver, Diesel & Elec)








Lease Costs


Total With OPEX Optimism Bias



Track Renewals


Rolling Stock Purchase


Rolling Stock Maintenance






Track Maintenance


Station OpEx


Station Maintenance


Classic Line Savings (Excl. Lease)


Classic Line Savings (Lease)






Capacity Charge


Electrification Asset Usage Charge




Variable Overheads and Admin Costs


Other On-Train Staff


16. The impact that erroneously multiplying the various classic line savings figures by 1.18 and 1.41 respectively has on the amount of savings that it is claimed can be achieved annually is shown in Table 2.

Table 2

Predicted annual savings from changes to existing services set out in HS2 Ltd’s Service Patterns document

Optimism bias (contingency)

HS2 Ltd’s spreadsheet calculation

Predicted annual savings including wrongly-applied optimism bias

Savings on rolling stock leases: £123 million


£123 million x 1.18

£145.14 million

Other savings (drivers, guards, electricity etc): £256 million


£256 million x 1.41

£360.96 million

Total savings per year: £379 million

Total savings per year with optimism bias included:
£506.1 million

[1] http://www.hs2.org.uk/news-resources/publications/economic-documents (August 2012 z ipped a ppraisal spreadsheets for Phase 2 - Day 1 & Y Costs.xls - Y Profile of Costs sheet )

[2] Ibid.

[3] https://www.gov.uk/government/publications/updated-economic-case-for-hs2-august-2012-explanation-of-the-service-patterns

[4] https://www.gov.uk/government/publications/updated-economic-case-for-hs2-august-2012-explanation-of-the-service-patterns . All of these service reductions can be identified by comparing today’s service patterns on the West Coast, East Coast and Midland Main Lines with the service pattern set out by HS2 Ltd in their ‘Do Something’ scenario in Appendix A (‘Do Something’ being the scenario that HS2 goes ahead as planned).

[5] Demand and Appraisal Report: HS2 London-West Midlands. Report for HS2 Ltd by MVA Consultancy, in association w ith Mott MacDonald and Atkins (April 2012). http://www.hs2.org.uk/sites/default/files/inserts/Demand%20and%20Appraisal%20Report%20London-West%20Midlands.pdf . The map in Figure 6.2 shows the forecast daily load factors on classic line long distance services in 2037 – i.e. after the HS2 ‘Y’ opens . Long-distance services on the southern section of the West Coast M ain L ine would be on average 27% full ; those on t he southern Midland Main Line would be 38% full on average and , on the East Coast Main Line , they would 35% full in the south and only 26% full on the approach to Leeds. Furthermore, these are high estimates, according to the legend beneath the diagram.

[6] The full lis t of places that, in HS2 Ltd’s plan, would receive fewer or slower intercity services to London and no HS2 service in recompence is : Aberdeen, Arbroath, Aviemore, Carlisle, Chesterfield, Coventry, Derby, Doncaster, Dundee, Falkirk Grahamston, Gleneagles, Inverkeithing, Inverness, Kingussie, Kircaldy, Lancaster, Leicester, Leuchars, Montrose, Nottingham (city centre station ), Oxenholme, Penrith, Perth, Pitlochry, Sheffield (city centre station), Sterling, Stockport, Stoke-on-Trent, Stonehaven, Wakefield, Wilmslow .

[7] Explanation of the s ervice p atterns , pp. 7 and 9. https://www.gov.uk/government/publications/updated-economic-case-for-hs2-august-2012-explanation-of-the-service-patterns

[8] Chris Castles & David Parish: Review of the Economic Case for HS2 (RAC Foundation, November 2011), p. vii. They wrote that "optimism bias has been incorrectly applied to the cost savings on the existing network after the opening of HS2, thus inflating this item of benefits attributed to HS2 by 41%".

[9] http://www.hs2.org.uk/news-resources/publications/economic-documents [August 2012 zipped appraisal sheets]


[9] July 2013

Prepared 16th July 2013