In its January 2020 report entitled High Speed Two: A progress update, the UK Auditor General concluded that the government underestimated the cost and complexity of its new high-speed rail initiative. By way of background, the report stated that the “High Speed Two programme (the programme) aims to construct a new high‑speed, high-capacity railway between London, Leeds and Manchester, via the West Midlands.” The new railway will “join with the existing rail network to enable journeys to Liverpool, Newcastle, Edinburgh and Glasgow.” As the Auditor General observed, the project “is the government’s largest infrastructure programme by value and many of its component parts, such as works at Euston station, construction of the railway infrastructure and purchase of land and property for the route, are large and complex projects in their own right.” The Auditor General found that the government has underestimated the complexity of the program, and had already overshot its original cost estimates and budget allocations:
The Department and HS2 Ltd underestimated the complexity of the programme, and Phase One is now forecast to cost between £31 billion and £40 billion, £3.9 billion to £12.9 billion (14% to 47%) more than its available funding. High Speed Two is a highly ambitious programme intended to be world-leading in its specification, including features such as a service frequency of 18 trains an hour to and from London compared with between a typical two to six trains an hour on European high-speed railways. The programme is constructing a new railway through city centres where site logistics are more challenging, and the UK’s high population density means a large number of roads, utility pipes and cables must be moved. The Department set the available funding in 2013 when the programme was at an early stage, based on a basic design, and uplifted this for inflation in the Spending Review 2015. Since then, forecast costs have increased on all elements of Phase One except the purchase of new trains (paragraphs 2.2, 2.10 to 2.12, and Figures 5, 7, 8, 11 and 12).
The Auditor General also found that the programme’s initial costing assumptions were based on inappropriate comparisons to other high-speed rail projects, and that these flawed assumptions led to the inaccurate cost forecasts:
HS2 Ltd did not account for the level of uncertainty and risk in the programme when previously estimating the costs of Phase One. HS2 Ltd’s previous cost estimate was agreed in April 2017, when Phase One was still at a relatively early stage of design maturity. At this stage, it was not possible to know what the specific requirements of the programme meant for the design or the costs of the railway. HS2 Ltd used existing international high-speed rail programmes as a basis for its costs, but these programmes have significant differences to High Speed Two. HS2 Ltd also had to make assumptions about elements such as ground conditions, as it did not yet have the legal powers to undertake detailed surveys. Despite these uncertainties, HS2 Ltd adopted a detailed method for calculating contingency appropriate for a programme at a much greater stage of development and certainty. This method resulted in £7 billion of contingency being set, 37% of forecast future Phase One costs. The amount of contingency was not enough to address the significant increases in cost that emerged as the design became more detailed, and issues such as poor ground conditions came to light (paragraphs 2.14 to 2.16, 2.19, 2.22, and Figures 11 to 13).
The Auditor General also found that the programme underestimated the costs that would arise from design changes:
The Department and HS2 Ltd underestimated the impact on costs of changes made to the design and construction of the railway by the hybrid bill. To gain the legal powers to build the railway, in 2013 the Department and HS2 Ltd deposited a hybrid bill in Parliament. As part of this process, the Department and HS2 Ltd committed to meeting certain requirements during the construction and operation of the railway. Some of these commitments were requested by people affected by the programme and were added through additional provisions to the bill, or through commitments known as ‘undertakings and assurances’. Some of these requirements made High Speed Two different from the international comparators on which HS2 Ltd based its April 2017 cost estimate, although it is difficult to isolate the impact of these commitments from other design choices and legal requirements. HS2 Ltd commissioned analysis which suggested that environmental and visual impact measures, such as lowering the railway beneath ground level, have made High Speed Two’s main civil construction around £1 billion more expensive than international comparators. Although there is some overlap with the measures above, indicative analysis by HS2 Ltd also suggests that its previous estimate of £245 million for meeting undertakings and assurances may have also been underestimated, and that the impact on costs may be £1.2 billion. Not all commitments have an impact on cost, although some of the commitments made can restrict contractors’ ability to avoid costs and create interdependencies that add to the challenge of delivering the programme (paragraphs 2.16 to 2.18).
Furthermore, the Auditor General found that the failure to meet the cost targets could largely be attributed to the government’s over-optimism regarding the estimated savings it could achieve when compared to the actual costs incurred during the first phase of the project:
HS2 Ltd included ambitious savings targets in its forecast costs to get within the available funding on Phase One but did not change its approach in order to deliver them. To make Phase One affordable, HS2 Ltd incorporated £4.9 billion of efficiency savings, changes to the design and scope of the programme and price estimate reductions within its April 2017 estimate. These were based on internal scrutiny of the estimate, the benchmarking approach it used to cost the programme and reviews by the Cabinet Secretary. For example, HS2 Ltd assumed it could make savings by encouraging UK industry to work closer together, as is common in European countries, to reduce administrative costs. When we last reported, HS2 Ltd had plans identifying where the savings might be found but did not develop these further into a programme of activity to achieve them. The elements that had the most assumed savings have seen significant cost increases. Had HS2 Ltd assumed no savings and applied a more appropriate amount of contingency, it would have forecast a cost comparable with the current estimate for Phase One (paragraphs 2.20 to 2.22, and Figure 13).
The Auditor General concluded that this optimism bias led to the government’s failure to adequately manage value-for-money risks in the long-term project:
Conclusion on value for money
High Speed Two is an ambitious national programme, the construction of which will take decades. The Department, HS2 Ltd and government more widely underestimated the task, leading to optimistic estimates being used to set budgets and delivery dates. In not fully and openly recognising the programme’s risks from the outset, the Department and HS2 Ltd have not adequately managed the risks to value for money. If these risks had been recognised and managed earlier, then the significant activity in a pressured environment over the past year trying to understand and contain cost increases may not have been necessary. There are lessons to be learned from the experience of High Speed Two for other major infrastructure programmes.
We welcome the increased realism on the estimated cost and schedule for the programme. However, significant risks remain. While the estimated cost and schedule for Phase One are now on a stronger footing, the challenge of getting Phase One into construction, and of monitoring and managing the programme as it progresses, is considerable. Phase Two is at a far earlier stage of development with many important decisions to be made before HS2 Ltd and the Department can improve cost and schedule estimates. Completing High Speed Two will require sustained focus and support from the Department and across government to ensure the programme is re‑established on a sound basis, balancing cost, time and benefits, and delivered in a way that achieves long-term value for money.
The report also found that the underestimation of project complexity resulted in a “significant, unanticipated, volume of work required in a compressed amount of time” and that this “stretched HS2 Ltd’s capacity and put staff under considerable pressure.”
The Auditor General recommended that in the future, the government should take “appropriate account of the limitations of cost benchmarking information” when compiling its cost estimates. With respect to anticipated savings, the Auditor General warned that the government should be more “realistic about the cost of complex and ambitious programmes and not assume arbitrary efficiency savings to make them affordable within available funding.” As this case study illustrates, underestimating the complexity of a major project can result in unrealistic cost estimates. While in the short term, lowballing project cost estimates may facilitate upfront project funding approvals, in the long term, these flawed assumptions set projects up for failure by imposing unrealistic downstream expectations.
