Select Committee on Public Accounts Twenty-Second Report


TWENTY-SECOND REPORT


The Committee of Public Accounts has agreed to the following Report:

THE CHANNEL TUNNEL RAIL LINK

INTRODUCTION AND LIST OF CONCLUSIONS AND RECOMMENDATIONS

1. In 1996, following a competition, the contract to build the Channel Tunnel Rail Link and run the UK arm of the Eurostar international train service was awarded to London & Continental Railways Limited. The Department of the Environment, Transport and the Regions agreed to provide London & Continental with grants totalling £1,730 million for the construction of the Link and its use by domestic train services. Subject to London & Continental raising private finance from a stock market flotation and the issue of debt, it was expected that construction would start in 1998 and that the Link would open in 2003.

2. It was envisaged that London & Continental would draw on revenue from Eurostar to service the private debt raised and to provide a return for its shareholders. London and Continental's business plan therefore depended on whether revenue from passengers using the Eurostar service would meet or exceed forecasts made by the company during competitive bidding for the deal. Towards the end of 1997 it had become clear that the forecasts for the operating performance of Eurostar had been over optimistic, negating London & Continental's efforts to raise all the money it needed from private investors. In January 1998, the company asked for an additional £1,200 million in grants from the Department. This request was rejected and, following negotiations, the Deputy Prime Minister announced in June 1998 that the deal had been restructured in a way that avoided an increase in the public grants payable to the company.

3. On the basis of a Report by the Comptroller and Auditor General[1] our predecessor Committee took evidence from the then Department of the Environment, Transport and the Regions (now the Department for Transport, Local Government and the Regions), London & Continental and Railtrack plc on the reasons for restructuring the deal, on the implications of the restructuring for the taxpayer, and on the justification for the public grants the Link will require.

4. The key conclusions and recommendations to emerge from our examination are:

  (i)  There is now ample evidence that even expert forecasts of user numbers for new ventures are subject to very great uncertainty. When negotiating deals the eventual success of which is heavily dependent on the number of users and the income generated, departments should examine a wide range of sensitivities around bidders' forecasts before signing a deal.

  (ii)  One of the functions of risk capital in a project is to secure the commitment of those who subscribe it by giving them something to lose if the project fails. In this case, the shareholders in London & Continental were insulated from that risk in two ways: they stood to gain, and did gain, from substantial contracts placed with them by London & Continental and, when the project came close to collapse, the Department felt obliged to keep the shareholders on board, with undiminished capital. Although the proportion of risk capital in a project is a matter for negotiation with bidders who will balance the amount of risk borne against the higher cost of servicing such capital, it is a false economy to proceed with a PPP in which too little risk capital has been subscribed by the private sector.

  (iii)  In deciding to restructure the deal rather than pull the plug, the Department put in place complex arrangements that will expose the taxpayer to substantial risk for many years to come. For instance, some £4 billion of bonds were issued by London & Continental, subject to Government guarantees, and in addition it is likely that further substantial sums of taxpayer's money will have to be lent directly to the company to keep it afloat. The Department will need to ensure that the substantial risks to which the taxpayer remains exposed are monitored and managed carefully throughout the life of the project.

  (iv)  The Department's economic justification for the project depended heavily on the monetary value placed on economic regeneration which it expected to result from the Link. Without a monetary value for economic regeneration, the justification for proceeding relied on wider policy benefits, such as national prestige, which are inherently unquantifiable. Decisions to proceed with major projects should be based on a comprehensive assessment of expected benefits as well as costs. We look to the Department to monitor the realisation of these regeneration benefits after the Link has been completed.

5. Our detailed conclusions and recommendations are:

On forecasting passenger numbers

  (v)  The actual number of passengers using Eurostar, which had been in operation for less than a year when it was transferred to London & Continental, has been substantially less than the company had forecast when it bid for the project. Although London & Continental had based its forecast on numbers produced by British Rail and SNCF, the extent of independent validation carried out by the Department was inadequate. Rigorous testing of bids is all the more important in cases where a service is relatively new, as was the case here (paragraph 16).

  (vi)  There was an incentive, in the Department's assessment of bids for the Link, for bidders to be over-optimistic in forecasting passenger numbers. Whether or not bidders were in fact over-optimistic, it would have been prudent for the Department to have examined the robustness of the bidders' plans at much lower levels of passenger numbers than those used in the sensitivity tests that were carried out. Departments should ensure that they have a clear understanding of what bidders are telling them when submitting forecasts of future usage with their bids, in particular the sensitivity of these forecasts to future constraints on passenger numbers and pricing levels (paragraph 17).

  (vii)  The Department claimed to be reasonably satisfied that its monitoring of the early stages of the project had been effective, though it did not exercise its right to receive written monthly monitoring reports. The original deal left the Department exposed to very substantial risk in the event of failure by London and Continental to raise long-term finance. It cannot be right for substantial sums of taxpayers' money to be at once at risk and beyond effective monitoring. While avoiding excessive or premature involvement that might hinder the ability of a private sector contractor to get a PFI project off the ground, it is essential that Departments monitor and manage their exposure to risk if taxpayer funds are to be safeguarded (paragraph 18).

  (viii)  London & Continental's forecasts of passenger numbers were not subject to independent review until 1998, some two years after the contract had been awarded, and again in 2000. In both instances the forecasts had been subject to significant downward adjustments. If forecasts prove to be wrong, departments must ensure that realistic estimates are prepared as early as possible and are reviewed regularly (paragraph 19).

On the commercial basis of the original deal

  (ix)  The level of equity capital was insufficient to reflect the high level of commercial risk in this project, which depended on inherently risky forecasts of passenger numbers. If a project involves a high degree of commercial risk, then it needs to be financed with a commensurately high level of risk capital relative to bank debt (paragraph 25).

  (x)  Under the terms of the deal, London & Continental's shareholders would manage the construction of the Link and run the Eurostar UK train service. They had, prior to the restructuring of the deal, already awarded themselves contracts worth over £90 million. If it is considered necessary to proceed with a project in the absence of adequate levels of equity capital, it is not enough to rely on the reputational risk to shareholders, especially where the shareholders providing risk capital also stand to benefit from substantial contracts once the project is underway (paragraph 26).

  (xi)  The Department needed the co-operation of London & Continental's shareholders if the deal was to be renegotiated without further delay to the construction of the Link. As a consequence, the Department was not in a strong position to insist that the shareholders should bear full responsibility for the near collapse of the project. Under the PFI, the private sector is paid for taking risk. Responsibility should therefore remain with the private sector should these risks actually occur. Departments should ensure that equity risk in PFI deals is real and that over-optimism in bidding for contracts will lose money for the shareholders if things go wrong (paragraph 27).

On the restructured deal

  (xii)  The Department considered that direct public financing for the project would weaken the incentives for the private sector to manage construction to time and budget and to maximise the revenue yield from the Eurostar business. It accordingly preferred the option of Government guaranteed bonds, though these were rather more costly to service than conventional Government funding. The price of these private sector incentives amounted to some £80 million over the life of the project. For the future, departments need to consider the balance between the additional cost of retaining private investment in a project and the expected financial discipline and management expertise this will bring (paragraph 33).

  (xiii)  Under the terms of the restructuring, the taxpayer may have to lend significant sums of money to London & Continental if Eurostar UK continues to perform poorly against forecasts. The Department, however, was unable to give clear answers as to whether, and if so when, any of the guarantees might be called or when the company could be expected to start repaying any future borrowing from the public purse. Whenever a department enters into such open-ended commitments to support a private sector company it should have clear milestones for the eventual repayment of money borrowed, and should work toward these milestones in partnership with the company concerned (paragraph 34).

On the benefits expected from the Link

  (xiv)  The principal justification for the Link was that it would lead to significant regeneration benefits in the Thames Gateway area and in east London. Such benefits from the Link had been quantified in money terms, although guidance issued by the Department to others still advised against the inclusion of such estimates, which were considered too vague (paragraph 44).

  (xv)  In putting together value for money assessments for long-term projects such as the Link, the Department should use up-to-date assumptions, including the expected rate of future economic growth (paragraph 45).

  (xvi)  At our predecessors' evidence session, the Accounting Officer disagreed with factual material in the Comptroller and Auditor General's Report which was either not disagreed by the Department at the time or had not been noted as a point of difference between the National Audit Office and the Department. The Accounting Officer also presented the Committee with a revised cost estimate which had not been shared with the National Audit Office in advance of the session. This situation is unsatisfactory. We therefore reiterate that it is not acceptable for the Committee's inquiries to be frustrated by conflicts of evidence which could have been resolved in advance (paragraph 46).

  (xvii)  Regional Eurostar services have not yet been implemented, on the grounds that they are considered to be uneconomic. We note that the new Eurostar terminal at St. Pancras will allow passengers from some regions to connect more easily than at present with the international train service. It is disappointing that large sums of public money have been wasted on buying expensive trains in advance of need or evidence of a need (paragraph 47).

FORECASTING PASSENGER NUMBERS

The forecasts made by London & Continental

6. During competitive bidding for the project, London & Continental forecast that Eurostar would attract 9.5 million passengers in 1996-97, the second full year of operation of the train service. The actual number of passengers who used the service in that year was only 5.1 million.[2] The forecasts were a very important part of the deal in that revenue from passengers was expected to provide a significant element of the finance to fund construction of the Link.[3]

London & Continental forecasts compared with actual passenger numbers


Source: C&AG's Report

7. Asked why these forecasts had turned out to be so badly wrong, London & Continental said that it had had to rely on information available at the time. Although three major members of London & Continental had transport expertise (National Express, the Virgin Group and SNCF[4]) there had been no real experience of international rail travel in this country, as the Eurostar service had been in operation for less than a year when bidders were required to compile projections of passenger numbers. British Rail had estimated passenger volumes and SNCF had undertaken its own estimates, which had been subject to third party review. As SNCF was a member of the London & Continental group, it had been natural for executives and other shareholders in the group to have relied on the information that SNCF had brought to the discussions.[5]

8. In view of the importance of passenger forecasts in determining the amount of public grant required by a bidder and hence the likelihood of winning the deal, our predecessors asked whether there had been an in-built incentive for bidders to be over-optimistic. London & Continental accepted that there was a theoretical argument pointing to such an incentive but emphasised that it had been important, if the company was to stay in the competition and achieve the returns expected by shareholders, that it submitted a credible bid. As the bid that had been submitted contained passenger numbers that were consistent with those produced by British Rail and SNCF, London & Continental insisted that there had been no evidence of over-optimism at the time.[6]

9. Having previously stated that there was no element of over-optimism in its bid, London & Continental cited a number of additional reasons why the forecasts had proved to be so much higher than actual demand for the Eurostar service. The company had taken control of the Eurostar business much later in 1996 than had been envisaged when its bid had been made in 1995 and it had also taken time to introduce the marketing strategies that had been developed during the bidding period. This delay had been compounded by the fire in the Channel Tunnel in November 1996, which had affected passenger numbers very badly during the last five months of 1996-97. Although passenger numbers might have been higher if the company had been able to take control of the business earlier, the Eurostar market had developed in a way that nobody had envisaged in 1995 and 1996. For instance, the growth of competition from low-cost airlines was developing early on and had been a very significant factor over the course of the last two years.[7]

The Department's evaluation of passenger forecasts

10. There have been two recent examples of high profile start-up projects whose business plans depended on forecasts of usage by members of the public which have turned out to be highly optimistic.[8] In view of the failure to attract the forecast numbers of passengers for the Eurostar service, our predecessors asked what the Department had done to test the validity of the forecasts made by London & Continental before the deal was agreed. They were told that as the company had submitted a set of assumptions that were not very different from earlier estimates made by British Rail, the Department had been inclined to believe them. With the help of external advisers, however, the Department had conducted sensitivity tests on the forecasts to determine whether the bidders' plans were achievable and, in particular, whether they would be able to achieve their main project financing. The analyses noted that London & Continental had been more optimistic than its competitor for the contract, Eurorail, both in its estimate of construction cost and in the net revenues that could be generated by the Eurostar service.

11. The Department's advisers had estimated that there would be a one-in-two chance that construction costs would increase by 12 per cent but only a one-in-twenty chance that such costs would go up by more than 15 per cent. The advisers had not been able to be as precise when analysing the downside risks in London & Continental's passenger forecasts, as much depended on the ability of the consortium to initiate and realise its marketing plans. The analyses had noted, however, that London & Continental's own advisers believed that the business could support a 15 per cent reduction in Eurostar revenues and still raise the required finance to complete the project.

12. The banks that were to lend money to London & Continental in the early stages of the project had taken into account a downside scenario of a 22.5 per cent increase in construction costs, a delay of a year in building the Link and a 20 per cent decrease in forecast Eurostar revenues. Even at these levels there were still available bank facilities of over £400 million. In the light of this analysis, the Department's advisers concluded that the downside scenarios had appeared to be at the limits of financeability. Nevertheless, the Department accepted in its evidence that the downside scenarios for Eurostar revenues should have been much harsher about the likely prospects for the project.[9]

The Department's monitoring of the early stages of the project

13. Formal and regular progress meetings between the Department and London & Continental did not commence until April 1997, just over a year after the award of the contract. At the inaugural meeting the Department agreed that, rather than receive detailed monthly reports on progress towards the achievement of the main financing for the project, it would receive outline reports which would be expanded upon at subsequent meetings.[10] Our predecessors asked why the Department had not insisted that the company should comply fully with its obligations to supply financial information.

14. The Department said that it had been aware of the state of the project but did not want to be seen to be getting involved in its "micro-management". Instead, it had decided to enter a dialogue with London & Continental to try to build confidence in the project. The Department had been satisfied with the nature of that dialogue and with the answers that had been given by the company. The Department had not insisted on written reports because it would have been required to pass such documents to the banks that had lent money to the project early on. If the reports had led to the banks calling the project into question, there could have been a loss of confidence. Although the lending banks would have recovered their loans from the Department, this loss of confidence would also have had an adverse effect on others involved in the project. The Department stated that a more effective form of monitoring would not have avoided the problems which surfaced later on. The project had run into difficulty because Eurostar passenger and revenue forecasts had been over-estimated, and because of the Channel Tunnel fire, which the Department and London & Continental could not have foreseen.[11]

Revised passenger forecasts

15. London & Continental had revised its forecasts downward for the restructured deal in 1998. On this occasion, however, the Department employed a transport consultant to provide an independent review of the forecasts and to produce its own estimates of Eurostar patronage and revenues. This independent analysis produced a central case scenario that had been lower than the revised forecasts prepared by London & Continental, and a downside case reflecting a more pessimistic scenario. In April 2000, the Department revised its forecasts downward for a second time, based on the actual performance of Eurostar,[12] bringing its forecasts broadly into line with the downside case figures produced in 1998. The Department expected about 11 million passengers to be using Eurostar by 2009-10 and about 16 million by 2020.[13]

Conclusions

16. The actual number of passengers using Eurostar, which had been in operation for less than a year when it was transferred to London & Continental, has been substantially less than the company had forecast when it bid for the project. Although London & Continental had based its forecast on numbers produced by British Rail and SNCF, the extent of independent validation carried out by the Department was inadequate. Rigorous testing of bids is all the more important in cases where a service is relatively new, as was the case here.

17. There was an incentive, in the Department's assessment of bids for the Link, for bidders to be over-optimistic in forecasting passenger numbers. Whether or not bidders were in fact over-optimistic, it would have been prudent for the Department to have examined the robustness of the bidders' plans at much lower levels of passenger numbers than those used in the sensitivity tests that were carried out. Departments should ensure that they have a clear understanding of what bidders are telling them when submitting forecasts of future usage with their bids, in particular the sensitivity of these forecasts to future constraints on passenger numbers and pricing levels.

18. The Department claimed to be reasonably satisfied that its monitoring of the early stages of the project had been effective, though it did not exercise its right to receive written monthly monitoring reports. The original deal left the Department exposed to very substantial risk in the event of failure by London and Continental to raise long-term finance. It cannot be right for substantial sums of taxpayers' money to be at once at risk and beyond effective monitoring. While avoiding excessive or premature involvement that might hinder the ability of a private sector contractor to get a PFI project off the ground, it is essential that Departments monitor and manage their exposure to risk if taxpayer funds are to be safeguarded.

19. London & Continental's forecasts of passenger numbers were not subject to independent review until 1998, some two years after the contract had been awarded, and again in 2000. In both instances the forecasts had been subject to significant downward adjustments. If forecasts prove to be wrong, departments must ensure that realistic estimates are prepared as early as possible and are reviewed regularly.

THE COMMERCIAL BASIS OF THE ORIGINAL DEAL

The capital structure of the deal

20. There were two stages to London & Continental's financing of the project. The first stage would fund the first two years of the project while the bulk of the funds needed would be secured through a second stage financing. The first stage was financed by £430 million of short—to medium-term bank debt and £60 million of equity from London & Continental shareholders.[14] In view of the significant transfer of passenger volume and revenue risk to London & Continental, our predecessor Committee asked the Department whether the proportion of equity to debt in this deal fully reflected the risks involved.

21. The Committee was told that there had been a small amount of equity relative to the total amount of money being put in because the equity element was more expensive to service than bank borrowing. Eurorail, the losing bidder, had proposed an even smaller proportion of equity to debt, so it had not been a choice between one bidder that had a very large equity component to its structure and one that had a much smaller equity element. The shareholders in London & Continental were all highly reputable groups which would have been concerned about the reputational risk they were taking as well as the financial risk. The Department accepted that there was an argument that the proportion of equity in this deal should have been somewhat more, but not a lot more.[15]

The interests of shareholders

22. During the expected gap of two years between the two finance raising stages, project development costs amounting to £92 million were paid to London & Continental's shareholders.[16] Asked whether the amount of equity was sufficient to balance shareholders' interests in becoming major contractors to the project, the Department agreed that there was some doubt whether it was sufficient and that, in such circumstances, the shareholders might have been asked to supply more equity for the project.[17]

23. One of the Department's objectives in the 1998 restructuring of the deal was to remove the existing shareholders from management control but require them to maintain an economic interest in the success of the project. Under the terms of the restructuring, the original shareholders converted 95 per cent of their equity into preference shares, accruing interest at seven per cent a year, and Railtrack was brought in to take on the construction of Section 1 of the Link.[18]

24. Asked why the shareholders did not lose their money since the deal had come close to collapsing, the Department said that it could not have agreed a way forward with London & Continental's shareholders that would have meant losing their original investment. If this had looked likely, then there was a risk that the shareholders would not have agreed to the restructuring and the project would have had to be restarted. What the Department had done in the restructuring was to apportion risks to those best placed to manage them. For instance, in relation to construction risk for Section 1, Railtrack's money was now exposed and a new management company had taken on part of the revenue risk inherent in the Eurostar business.[19]

Conclusions

25. The level of equity capital was insufficient to reflect the high level of commercial risk in this project, which depended on inherently risky forecasts of passenger numbers. If a project involves a high degree of commercial risk, then it needs to be financed with a commensurately high level of risk capital relative to bank debt.

26. Under the terms of the deal, London & Continental's shareholders would manage the construction of the Link and run the Eurostar UK train service. They had, prior to the restructuring of the deal, already awarded themselves contracts worth over £90 million. If it is considered necessary to proceed with a project in the absence of adequate levels of equity capital, it is not enough to rely on the reputational risk to shareholders, especially where the shareholders providing risk capital also stand to benefit from substantial contracts once the project is underway.

27. The Department needed the co-operation of London & Continental's shareholders if the deal was to be renegotiated without further delay to the construction of the Link. As a consequence, the Department was not in a strong position to insist that the shareholders should bear full responsibility for the near collapse of the project. Under the PFI, the private sector is paid for taking risk. Responsibility should therefore remain with the private sector should these risks actually occur. Departments should ensure that equity risk in PFI deals is real and that over-optimism in bidding for contracts will lose money for the shareholders if things go wrong.


1   C&AG's Report, The Channel Tunnel Rail Link (HC 302, Session 2000-01) Back

2   C&AG's Report, para 1.11 Back

3   Q157 Back

4   Societe Nationale de Chemins de Fer Francais (the state-owned rail network in France) Back

5   Qs 2, 158, 250-251 Back

6   Qs 168-169 Back

7   Qs 204-206, 208-210 Back

8   C&AG's Reports, The Millennium Dome (HC 936, Session 1999-2000) and The Re-negotiation of the PFI-type Deal for the Royal Armouries Museum in Leeds (HC 103, Session 2000-2001) Back

9   Qs 139, 256; Ev, Appendix 2, p32 (Q277) Back

10   C&AG's Report, para 1.20 Back

11   Qs 65-69, 75, 136, 179-180 Back

12   C&AG's Report, paras 3.14, 3.40 Back

13   Q216 Back

14   C&AG's Report, para 1.7 Back

15   Qs 171, 174, 259 Back

16   C&AG's Report, para 1.12 Back

17   Qs 258-261 Back

18   C&AG's Report, para 2.13 Back

19   Qs 150-152, 221 Back


 
previous page contents next page

House of Commons home page Parliament home page House of Lords home page search page enquiries index

© Parliamentary copyright 2002
Prepared 21 March 2002