40 The comparison of the Public Sector Comparators and the PPP bids involves a series of financing and economic assumptions. These cover:
■ the source of funding for the public sector operations - essentially conventional funding or bond funding;
■ the cost of capital relevant to each source of funding (for example, the interest costs on bonds); and
■ the government's time discount rate, which is used to discount future costs and present them as net present costs.
41 London Underground's conventional funding scenarios assume that the infrastructure operations are funded through a mixture of fare revenues and general government revenues. London Underground has supplemented this with bond funding scenarios, which envisage bonds issued by Transport for London in three tranches and with a credit rating based on an indicative view provided by a major credit rating agency. Including a bond scenario reflects London Underground's view that conventional funding can impose constraints on operations because, with only at best a three-year horizon of secure funding, it can hinder long-term planning of capital works. Raising bonds directly from capital markets could provide London Underground with more secure long-term funding and enhance long-term planning. London Underground did not prepare a scenario with a government-backed bond.
42 In the bond scenario, London Underground assumes additional efficiency savings arising from secure finance and the end of annual funding negotiations, amounting to 7 per cent per annum off investment costs and 1.5 per cent off maintenance costs. The bond scenarios also assume some improved performance by the public sector. There is inevitably little by way of quantified evidence available on the level of possible efficiencies with bond finance, and London Underground recognises that the 7 per cent figure is a matter of subjective judgement based on its consideration of the limited evidence.
43 Some of London Underground's scenarios assume a government cost of capital and discount rate both at 6 per cent real. This is in line with standard investment appraisal practice in government, although a non-standard approach has also been examined in this case. The non-standard approach arises from concerns about the appropriateness of the 6 per cent real rate. In 1997, the Committee of Public Accounts noted7 that the Treasury's assumed discount rate of 6 per cent was at the top end of the range of values that could be justified. The Committee was concerned that the use of too high a discount rate would introduce a bias in favour of PPP-style deals over conventional funding. The Committee observed that using more recent figures for interest rates on government debt would lead to a significantly lower figure for the discount rate. The Committee recommended that it would be good practice for authorities to examine the sensitivity of financial analysis of this sort to the choice of discount rate.
44 In accordance with this recommendation, and given that the standard 6 per cent rate is significantly higher than the prevailing rate on government debt (between 2 and 3 per cent) and professional economists also question it as a time discount rate, London Underground is examining the impact on the financial analysis of using a discount rate of 3.5 per cent and has developed a matrix of assumptions using various discount rate and interest rate assumptions. At the time of writing, London Underground had only just completed initial figures for some of these assumptions, so we are unable to reach any conclusion on whether the figures require further refinement, their impact on the financial analysis and whether this impact has been fairly taken into account by London Underground.
45 Where appropriate, some of these scenarios include an adjustment to capture the impact on the government's reputation for prudence of extra borrowing, known as reputational externality8. There is some theoretical basis for making such an adjustment, but also, in our view, considerable scope for argument over its size. We note that the issue largely disappears if the discount rate is set at a more realistic level of 3.5 per cent.
46 The various financing and discounting scenarios are technically complex, and would not be necessary for most PPP contracts. They have sensibly been chosen as scenarios for this transaction because it involves unusually large sums of money, and because alternative funding sources are conceivable for London Underground, with secure fare revenues that could be used to finance direct borrowing.
_____________________________________________________________________________
7 47th report, 1997-8
8 The standard Treasury approach uses a cost of capital and discount rate at 6 per cent. Scenarios which use a lower cost of capital than the discount rate create a bias in favour of public borrowing by lowering the implied interest costs of such borrowing. But Treasury considers that increasing public borrowing has an external cost on the government's reputation for prudence, and the reputational externality is a calculation to reflect this assumed external cost.