Motorway Communications Services PFI Study

Transcription

1 Motorway Communications Services PFI Study Final Report Volume 3 - Financial Analysis Author Various Checker Name Redacted under Sec 40 of the FOIA Exemptions Personal Information Approver Name Redacted under Sec 40 of the FOIA Exemptions Personal Information This report has been prepared for in accordance with the terms and conditions of appointment for the Motorway Communications Services PFI Study dated 30/03/99. KHHS can not accept any responsibility for any use of or reliance on the contents of this report by any third party.

2 2/40

3 TABLE OF CONTENTS 1.0 FINANCIAL MODEL Introduction Wider socio-economic benefits Description of the financial model QA of the financial model FINANCIAL ANALYSIS AND RANKING HA risk adjusted NPV retained costs Commentary on results STRATEGY CAPITAL AND OPERATING COSTS Capital costs analysis by strategy Operating costs analysis by strategy THIRD PARTY REVENUES Third party revenue analysis by strategy MCS CO CHARGES MCS Co charges analysis by strategy Commentary on analysis HA MANAGEMENT AND ADMINISTRATION COSTS HA management and administration costs analysis by Commentary on analysis STRATEGY RISK ASSESSMENT Basis of risk assessment Risk modelling method Analysis of key risks for Motorway Communication Services Other risks considered by the Study Assessment of HA s retained risks under each strategy SENSITIVITIES Introduction Revenue sensitivities Break-even levels of third party revenues Discount rate sensitivity /40

4 THIS PAGE INTENTIONALY LEFT BLANK 4/40

5 GLOSSARY OF FINANCIAL ANALYSIS TERMS Capital allowances Corporation tax Debt Tax relief based upon the capital value of an asset, calculated each year as 25% of the tax written down value of that asset. The UK s corporate taxation system. A rate of 30% of taxable profits has been assumed. Borrowings from third party financing institutions, frequently banks, which have a priority over other investors in terms of their claim on the free cashflows of the SPV. Debt equity ratio The ratio of debt and equity in total funding. This is typically 90% in funding PFI projects, i.e. a project costing 100 in total is funded by 90 debt and 10 equity. Debt service cover ratio Equity Discounting PFI projects are generally funded on a project financing basis, where the lenders take security from a stream of cashflows. (eg payments from the HA to the PFI contractor) rather than from the more usual security of physical assets. Lenders monitor the cashflows available to repay their debt by means of banking cover ratios, the breach of which results in prescribed actions to protect the bank s security. The debt service cover ratio is one particular security, which is defined as cash before financing/capital repayment outstanding for the period. Funds invested in a project company by its owners through the subscription of share capital. The application of an interest rate to future cash flows to allow for the time value of money, e.g. if the interest rate is 10% per anum, a person should be indifferent between receiving 100 now or 110 in a year s time. Put differently, at an annual interest rate of 10%, 110 in a year s time has the same value as 100 now. This enables timing differences between cashflows to be reflected in project appraisal. Discount rate Latin hypercube sampling analysis The rate used to calculate present values. The Treasury discount rate of 6% is used in this study. This is a method of sampling which uses stratification to mitigate the effects of skewed distributions and enables results to be generated with smaller samples sizes than comparable methods such as Monte Carlo analysis. The method uses estimates of potential variations in selected items, to calculate probable risk values. These are used in the calculation of risk adjusted cashflows. 5/40

6 IRR NPV Present value PSC Rolled up interest Special purpose vehicle ( SPV ) Internal rate of return. The discount rate at which a project has a net present value of zero. Net present value. The present value of revenues less costs. The discounted value of future cash flows. At an annual interest rate of 10%, the present value of 110 received in a year s time is 100. Public sector comparator. A conventionally procured alternative to the PFI contract. The term given to interest on a loan (usually incurred during a construction period) which is not paid, but rather rolled into the principal loan amount for future repayment. A company set up by partners in a project, specifically for the purposes of that project. In a PFI context, SPVs are often used by bidding partners for the purposes of contracting to provide PFI services. The SPV has contractual relationships with its owners, financiers, equipment and service providers as well as the customer. From a financing perspective, the primary purpose of the SPV is often to limit the recourse of lenders to the project sponsors. There may also be taxation based or management/control motivation to use such a structure. In the context of this Study, it has been assumed that MCS Co would take the form of an SPV. Taxable profits Unitary charge Profit adjusted in accordance with UK tax legislation. The annual amount payable by the HA in respect of services provided by the SPV. 6/40

7 1.0 FINANCIAL MODEL 1.1 Introduction The financial models in this report estimate the cost to the ( HA ) of each of the Strategies developed in Volume 2 of this Final Report so that they can be ranked where appropriate and assessed for potential value for money against their relevant Public Sector Comparator ( PSC ). The materiality and sophistication of the models, and the inputs and assumptions, reflect the feasibility status of the Study. Much of the information being modelled at this stage has been supplied without commitment on the part of interviewees and is substantially untested. Nevertheless, the financial model has been planned and developed to provide as accurately as possible at this stage in the project a high level indication of the potential benefits of one over another The financial models and supporting schedules are attached as Appendices Wider socio-economic benefits This financial analysis is concerned solely with the financial costs and benefits of each of the Strategies to the HA, i.e. those costs and benefits which ultimately result in cashflows to or from the HA. It does not attempt to place a value on the wider socio-economic benefits of each of the Strategies, that is those costs and benefits to the HA s customers resulting from increased safety, journey time savings, journey time certainty, vehicle operating costs and the like Thus, in this analysis, Strategies which are more limited in scope and thereby incur less cost, may compare favourably in terms of their NPV cost to the HA against more ambitious Strategies which on the face of it cost more, but which aim to provide a larger range of socio-economic benefits to the road user in furtherance of the HA s policy objectives An assessment of the socio-economic benefits was not required by the Specification and is inappropriate in anything other than qualitative terms at this feasibility stage of the Study. However, an assessment of these benefits, is likely to be appropriate when developing the chosen Strategies in more detail in Part B. An initial assessment of the Strategies against New Approach to Appraisal ( NATA ) criteria has, however, been undertaken and is presented in Volume 2 section Description of the financial model The Strategies Seven main iterations of the financial model have been prepared for the different Strategies and their PSC variants identified in the Final Report. These are as follows: a) 2, the Do Minimum; 7/40

8 b) 3 PSC, the PSC variant of the Private Finance Initiative ( PFI ) for a national network strategy; c) 3 PFI, the PFI for a national network; d) 4 PSC UMTS, the PSC variant of the PFI for an enhanced national network strategy based on a UMTS roadside to vehicle communications link; e) 4 PFI UMTS, the PFI for an enhanced national network strategy based on a UMTS roadside to vehicle communications link; f) 4 PSC DSRC, the PSC variant of the PFI for an enhanced national network strategy based on a DSRC roadside to vehicle communications link; and g) 4 PFI DSRC, the PFI for an enhanced national network strategy based on a DSRC roadside to vehicle communications link. Model outline The model is illustrated in Figure 1.1. Data Assumptions Calculations Outputs Triple package/ CSR Risk assumptions HA project management cash and admin assumptions Risk calculations Capex and opex calculations HA retained costs model UMTS Data DSCR Data 3rd P revenue assumptions Project management and admin assumptions 3rd P revenue calculations Project management and admin calculations MCS Co model HA Incremental costs model Mast Data Risk assumptions Risk calculations Financing and tax assumptions MCS Co financing/tax calculations Figure 1.1: Model outline Data on the triple package was generated using National Communications Programme ( NCP ) information provided by the HA. KHHS sought to identify cabling tasks within the triple package, which would be suitable for inclusion in a PFI project. 8/40

9 1.3.4 Information based upon industry knowledge and limited market research was used in respect of: GSM/UMTS systems; mobile phone masts; DSRC systems; and data management In the modelling process, data was separated between Project items and Non project items to illustrate the portion of the HA s communications services which are likely to be effected by a PFI from Year 4 onwards. Since the Strategies with a PFI Option, Strategies 3 and 4, concern the motorway communications network, Project costs broadly relate to these activities. Non project costs broadly relate to the HA s other communications activities included within the Study, namely provision and maintenance of the roadside equipment and provision of the business communications network Project costs and revenues relating to the activities which may be effected by a PFI are modelled separately from the HA s retained Non project costs and revenues in both the PFI and PSC scenarios, so that the costs, revenues and risks relating to these options are directly comparable, without distortion by the high levels of Non project expenditure retained by the HA The project data relating to the private sector PFI services provider, Motorway Communication Services Company ( MCS Co ) is used in the MCS Co model, together with financing and taxation assumptions to calculate a unitary charge to the HA which is set to enable MCS Co to attain a reasonable target rate of return. The unitary charge is reflected in the HA Project costs model in the PFI scenarios A fifteen year period of assessment was chosen to reflect a procurement period of approximately three to four years, commencing in April 1999, and an initial contract term of approximately ten years. 1.4 QA of the financial model The model has been subject to: a) review by KPMG s business modelling unit; b) walk-through checks; and c) review of output by the project Assistant Director and Partner. 9/40

10 THIS PAGE INTENTIONALY LEFT BLANK 10/40

11 2.0 FINANCIAL ANALYSIS AND RANKING 2.1 HA risk adjusted NPV retained costs The NPV cost of each to the HA and its risk adjusted cost to the HA is set out in Figure 2.1 below. Retained HA NPV costs Non project costs 2 3 PSC 3 PFI 4 PSC UMTS 4 PFI UMTS 4 PSC DSRC 4 PFI DSRC m m m m m m m HA capital costs (369.0) (346.4) (346.4) (346.4) (346.4) (346.4) (346.4) HA operating costs (105.5) (100.1) (100.1) (100.1) (100.1) (100.1) (100.1) HA mgt and admin (24.2) (24.3) (24.3) (24.3) (24.3) (24.3) (24.3) HA revenues Non project costs (498.7) (467.5) (467.5) (467.5) (467.5) (467.5) (467.5) Project costs HA capital costs - (42.8) - (42.8) - (42.8) - HA operating costs - (30.2) - (30.2) - (30.2) - MCS Co charges - - (32.7) - (27.3) HA mgt and admin - (2.1) (3.7) (2.1) (3.7) (2.1) (3.7) Consultants fees - - (2.8) - (2.8) - (2.8) HA revenues Project costs - (36.7) (39.2) (36.7) (33.8) (36.7) Total HA costs before risk Mean non project risk adjusted costs Mean project risk adjusted costs (498.7) (504.2) (506.7)) (504.2) (501.3) (504.2) (275.6) (520.6) (490.9) (490.9) (490.9) (490.9) (490.9) (490.9) - (56.4) (39.4) (56.4) (34.0) (56.4) th percentile - (40.4) (39.0) (40.4) (33.6) (39.5) th percentile - (75.4) (39.9) (75.9) (34.5 (76.0) Mean total HA risk adjusted costs (520.6) (547.3) (530.3) (547.3) (524.9) (547.3) (299.1) 95 th percentile (496.5) (504.7) (503.3) (504.7) (498.8) (503.8) (221.9) 5 th percentile (550.1) (591.0) (555.5) (591.5) (544.2) (591.6) (324.3) Figure 2.1: NPV of HA risk adjusted retained costs by The costs are shown over the 15 year evaluation period commencing 1 April 1999 and ending 31 March 2014, discounted at the Treasury s appraisal rate of 6%. 11/40

12 2.1.3 The risk adjustments made to the risk adjusted costs have been made in accordance with the risk assessments set out in section 7 using Latin hypercube sampling analysis. This provides a distribution of results for which we have provided the mean figure and the 90% confidence range bounded by the 95 th and 5 th percentile illustrated in figure 2.2 below: Probability 5 th Percentile Mean 95 th Percentile Risk adjusted NPV Cost Figure 2.2: Risk adjusted NPV cost Analysis of results As noted in section 1, Non project costs broadly relate to all the HA s activities up to the end of Year 3, which is considered to be the earliest date for the introduction of an MCS contract, and, from Year 4 onwards, to the provision and maintenance of the roadside equipment and provision of the business communications network Project costs and revenues relate to the activities which may be effected by an MCS contract from Year 4 onwards and, therefore, cover broadly the operation of the motorway communications network , the Do Minimum, relates to all of the activities included in the Study, and so includes both the Non project and Project costs. In comparing 2 with the other Strategies it is necessary to review the total HA risk adjusted costs, highlighted by the lower box in Figure Strategies 3 and 4 relate solely to Project activities, i.e. the operation of the motorway communications network and associated opportunities from Year 4 onwards. Therefore, when comparing Strategies 3 and 4 with each other and with their PSCs, it is necessary to review the mean Project risk adjusted costs to the HA, highlighted by the upper box in Figure Commentary on results Project and non project costs The striking observation for all of the strategies is that the non-project costs are so much greater than the project costs. The inference is that the majority of the cost in the HA s communications services resides in the provision and maintenance of its roadside devices. 12/40

13 2.2.2 However, KHHS believe that the wider opportunities for the HA, both in financial terms and in terms of achieving its broader policy objectives, rest in the development of its motorway communications network. That is why the two Strategies developed in most detail, Strategies 3 and 4, concern this part of the MCS and only 5 considers the procurement of the roadside equipment Furthermore, we believe that if the financial analysis were extended to include an assessment of the broader socio-economic benefits, we would expect to see a greater portion of the value being generated in the motorway communications network. This is because we would expect to see the emphasis move away from current dependency on roadside communications devices to the delivery of more detailed and tailored traffic information generally, to drivers via in-car and other journey planning and management services, and to the TCC, PCOs and other traffic management bodies. Public sector options has a lower NPV risk adjusted cost to the HA at million than each of the PSCs for the PFI Strategies 3 and 4 which each cost million. This is to be expected since 2 is a smaller scope option and no account is taken in this analysis of the broader socio-economic benefits which the HA would expect to achieve through Strategies 3 and 4. PFI options When reviewing the PFI options in Strategies 3 and 4, it is apparent that 3 appears to be marginally more expensive than its PSC alternative, and 4 UMTS appears to be marginally less expensive than its PSC alternative, before account is taken of risk (see Project costs line in Figure 2.1), but that each appears to demonstrate substantial value for money (i.e. is a lower cost to its PSC) when the HA s retained risk is also considered. Furthermore, 4 also provides better value for money than 2, the Do Minimum, even before taking account of broader socio-economic benefits Whilst PFI Strategies 3 and 4 UMTS appear marginal in comparison with 2, in terms of the total cost to the HA, they actually represent far greater value for money, since the HA will acquire far greater capability for a similar outlay. The overall cost of PFI is lower in each case when compared with the PSC The figure for the 4 DSRC option assumes that MCS Co would be the sole national operator (at least until such time as another operator started-up in competition) of the beacon-based roadside to in-vehicle communications services, because there is no existing DSRC network. Thus MCS Co would capture all subscription revenues from motorists using the service. With the UMTS option motorists are not assumed to subscribe to MCS Co, but to their usual mobile operator. In this scenario, MCS Co would only make a small margin on telematics communications traffic carried over the operators networks. 13/40

14 2.2.8 It is apparent from the results that the 4 DSRC PFI appears prima facie to be the best option for the HA. Indeed, such are the assumed potential revenues under the DSRC option (even when reduced to 40% of expected total revenues for the purposes of calculating the cost to the HA), MCS Co would appear to be paying the HA to provide the services rather than vice-versa. This results in an NPV risk adjusted cost to the HA of million against million for the DSRC PSC, and million and million for each of the other PFI options However, it must be pointed out that the revenue numbers, particularly, are very speculative, and would need thorough testing in the market. The HA would not seek, in requesting tenders for a 4 bid, to specify either a UMTS or DSRC solution, and it would be for the bidders, based on their knowledge of the market conditions and the technologies at the time to decide on the solutions and to carry the risk of achieving forecast revenues. The results of the financial analysis are useful though in giving an idea of the range of potential savings the HA may seek to achieve through PFI over a conventional procurement. 14/40

15 3.0 STRATEGY CAPITAL AND OPERATING COSTS 3.1 Capital costs analysis by strategy The NPV capital costs incurred by the HA in each, and those incurred by MCS Co in the PFI Strategies, are set out in Figure 3.1 below. Retained HA NPV capital costs Non project costs 2 3 PSC 3 PFI 4 PSC UMTS 4 PFI UMTS 4 PSC DSRC 4 PFI DSRC m m m m m m m Infrastructure (367.4) (338.5) (338.5) (338.5) (338.5) (338.5) (338.5) Transmission (1.6) (1.6) (1.6) (1.6) (1.6) (1.6) (1.6) Mast costs - (6.3) (6.3) (6.3) (6.3) (6.3) (6.3) Non project costs (369.0) (346.4) (346.4) (346.4) (346.4) (346.4) (346.4) Project costs Infrastructure - (28.9) - (28.9) - (28.9) - Transmission - (2.0) - (2.0) - (2.0) - Mast costs - (11.9) - (11.9) - (11.9) - Project costs - (42.8) - (42.8) - (42.8) - Total capital costs (369.0) (389.2) (346.4) (389.2) (346.4) (389.2) (346.4) MCS Co Project costs Infrastructure - - (23.6) - (23.6) - (23.6) Transmission - - (3.2) - (3.2) - (3.2) Mast costs - - (12.0) - (12.0) - (12.0) GSM/UMTS (5.0) - - DSRC Beacon costs (8.5) MCS Co project capital costs - - (38.8) - (43.8) - (47.3) Figure 3.1: NPV of capital costs by The costs are shown over the 15 year evaluation period commencing 1 April 1999 and ending 31 March 2014, discounted at the Treasury s appraisal rate of 6%. 15/40

16 3.1.3 In the above analysis, Non project capital costs broadly relate to all the HA s activities up to the end of Year 3, and, from Year 4 onwards, to the provision and maintenance of the roadside equipment and provision of the business communications network. Non project expenditure is in effect expenditure which is made regardless of whether a PFI or PSC option is taken. Project capital expenditure in the PSC Strategies relates to the national motorway communications network activities from Year 4 onwards Non project expenditure includes masts cost which the HA is assumed to incur prior to the commencement of operation of MCS Co. Project expenditure represents continued expenditure on masts, infrastructure and transmission systems in the absence of MCS co The expenditure apportioned to MCS Co from the HA s continuing projects relates to cabling work. Expenditure retained by the HA is assumed to relate to roadside devices and business communications Infrastructure costs are based upon a review of HA plans. Other costs are KHHS best estimates based upon industry knowledge. The PFI scenarios assume that a 20% cost saving is possible on the HA s infrastructure costs through more effective procurement practices It is assumed in the 4 PSC scenarios that the HA would not undertake the development of the roadside to vehicle links due to the risks involved and the capabilities required. Consequently, the HA incurs no expenditure on GSM/UMTS or DSRC equipment in the 4 PSCs. 16/40

17 3.2 Operating costs analysis by strategy The NPV operating costs incurred by the HA in each, and those incurred by MCS Co in the PFI Strategies, are set out in Figure 3.2 below. Retained HA NPV operating costs Non project costs 2 3 PSC 3 PFI 4 PSC UMTS 4 PFI UMTS 4 PSC DSRC 4 PFI DSRC m m m m m m m Infrastructure (84.8) (86.3) (86.3) (86.3) (86.3) (86.3) (86.3) Transmission (20.7) (12.2) (12.2) (12.2) (12.2) (12.2) (12.2) Mast costs - (1.6) (1.6) (1.6) (1.6) (1.6) (1.6) Non project costs (105.5) (100.1) (100.1) (100.1) (100.1) (100.1) (100.1) Project costs Infrastructure - (1.3) - (1.3) - (1.3) - Transmission - (9.7) - (9.7) - (9.7) - Mast costs - (19.2) - (19.2) - (19.2) - Project costs - (30.2) - (30.2) - (30.2) - Total operating costs (105.5) (130.3) (100.1) (130.3) (100.1) (130.3) (100.1) MCS Co Project costs Operating expenditure - - (1.3) - (1.3) - (1.3) Transmission maintenance expenditure - - (5.6) - (5.6) - (5.6) Mast maintenance - - (20.5) - (20.5) - (20.5) Data Management Systems Microcell/beacon maintenance - - (1.0) - (1.0) - (1.0) (7.5) - (17.7) MCS Co project operating costs - - (28.4) - (35.9) - (46.1) Figure 3.2: NPV of operating costs by The costs are shown over the 15 year evaluation period commencing 1 April 1999 and ending 31 March 2014, discounted at the Treasury s appraisal rate of 6%. 17/40

18 3.2.3 Operating expenditure for non project activities in all scenarios includes maintenance costs and maintenance contract arrangements consistent with those already in place. This includes RMC and RMCMC costs in respect of maintenance of the roadside equipment. Project costs include NTMC arrangements in respect of the cable and transmission infrastructure in the PSC scenarios and MCS Co s own maintenance in the PFI scenarios Transmission expenditure in MCS Co is estimated to cover an upgraded transmission network In the PFI scenarios, the cost streams and associated revenues relating to mobile phone masts pass to the MCS Co on commencement of its operations. In PSC scenarios these cost streams and associated revenues continue in the HA and are shown as project costs In the MCS Co scenarios, operating costs are incurred on DSRC and UMTS activities. As mentioned above, it is assumed that the HA would not develop this additional functionality Infrastructure costs are based on a review of HA plans. Other costs are based upon KHHS best estimates derived from industry knowledge. 18/40

19 4.0 THIRD PARTY REVENUES 4.1 Third party revenue analysis by strategy The NPV revenues received by the HA in each, and those received by MCS Co in the PFI Strategies, are set out in Figure 4.1 below. HA NPV revenues Non project revenues 2 3 PSC 3 PFI 4 PSC UMTS 4 PFI UMTS 4 PSC DSRC 4 PFI DSRC m m m m m m m Mast revenues Non project rev s Project revenues Mast revenues Project revenues Total HA revenues MCS Co revenues Mast revenues Data management Sale of traffic data Sale of comms/airtime Microcell rentals/dsrc Beacon rentals MCS Co revenues Figure 4.1: NPV of third party revenues by The revenues are shown over the 15 year evaluation period commencing 1 April 1999 and ending 31 March 2014, discounted at the Treasury s appraisal rate of 6% In 2, the Do Minimum, no revenues accrue to the HA as it is assumed that the HA would not undertake any revenue generating opportunities. 19/40

20 4.1.4 In Strategies 3 and 4, it is assumed that the HA will initiate an interim strategy in respect of the masts opportunity. This is reflected in the 3.3 million Non Project mast revenues. In the PFI scenarios, these activities are transferred during year 4, to MCS Co and are reflected in mast revenues. In the PSC scenarios, the HA continues these activities itself and the revenues generated from year 4 on are reflected in Project revenues The PFI revenues in Strategies 3 and 4 also reflect additional revenue generating activities which would be undertaken by MCS Co, including data management in 3 and the UMTS/DSRC activities in The revenues shown in the PFI scenarios represent 40% of the gross potential revenues which may be expected to accrue. This is a conservative estimate of the proportion of expected revenues which the private sector may offer to offset against charges to the HA. In reality this figure would be derived in the competitive tendering process and negotiated with the preferred bidder. Revenues are based upon KHHS best estimates and are stated net of direct costs Revenues accruing to HA directly in the interim arrangements and PSC scenarios have been reduced to 30% of potential income on the assumption that the private sector will possess more finely honed commercial and selling skills To demonstrate the effect of these revenue assumptions on the total costs to the HA in each scenario, sensitivities assuming 0% and 100% achievement of gross potential revenue are provided in section 8. In addition, we have run a sensitivity for each strategy illustrating the level of revenue at which the matches the PSC in NPV value and at which the NPV cost to the HA is zero. 20/40

21 5.0 MCS CO CHARGES 5.1 MCS Co charges analysis by strategy The NPV of MCS Co (charges)/royalty to the HA for each of the PFI Strategies is set out in Figure 5.1 below. NPV MCS Co (charges)/royalty 3 PFI 4 UMTS PFI 4 DSRC PFI m m m (32.7) (27.3) Equivalent annual (charge)/royalty (5.9) (4.9) 36.0 NPV MCS Co charges (no third party revenue) (106.1) (116.2) (135.4) Equivalent annual charge (18.4) (21.1) (24.6) Figure 5.1: NPV of MCS Co charges by The charges are shown over the 15 year evaluation period commencing 1 April 1999 and ending 31 March 2014, discounted at the Treasury s appraisal rate of 6%, together with the equivalent annual (charge)/royalty MCS Co s (charges)/royalties are calculated so as to recover the MCS Co capital and operating expenditure in Figures 3.1 and 3.2, allowing for MCS Co revenues in Figure 4.1, plus sufficient profit to attain a post tax project IRR of 15% for Strategies 3 and 4 UMTS. A return of 25% is assumed under 4 DSRC to reflect the risk arising to the MCS Co from heavy dependence on DSRC revenues The charges have been modelled based on the following assumptions: a) MCS Co will be funded by 90% debt and 10% equity; b) interest is charged on balances outstanding at 7.5%; c) interest is rolled up for the first two years of MCS Co s operation; d) a debt service cover ratio of 1.3 is required; e) loan repayments commence in MCS Co s third year of operation and the capital balance is repaid in equal annual instalments at each period end commencing in MCS Co s third year of operation; f) corporation tax is assumed to be charged at 30%; and 21/40

22 g) capital allowances are available on capital expenditure at 25%. 5.2 Commentary on analysis The benefit of third party revenues on the level of charge to the HA is clearly demonstrated in the Figure above as the charge is expected to be lower for the higher functionality 4 than for 3. Indeed in 4 DSRC, the MCS Co would pay the HA for the right to provide the services due to the third party revenue opportunities this would provide. 4 DSRC is heavily dependent upon growth in new technologies and markets, whereas the UMTS option could build on existing networks. Detailed research will be necessary to establish more firmly the market opportunities for the MCS Co. This will be a key issue for a private sector investor assessing the viability of a potential PFI project and for the HA in assessing likely levels of charge. 22/40

23 6.0 HA MANAGEMENT AND ADMINISTRATION COSTS 6.1 HA management and administration costs analysis by The NPV cost to the HA for management and administration costs in each is set out in Figure 6.1 below. HA NPV retained costs Non project costs HA management and admin costs 2 3 PSC 3 PFI 4 PSC UMTS 4 PFI UMTS 4 PSC DSRC 4 PFI DSRC m m m m m m m (24.2) (24.3) (24.3) (24.3) (24.3) (24.3) (24.3) Project costs HA management and admin costs - (2.1) (3.7) (2.1) (3.7) (2.1) (3.7) Total management and admin costs (24.2) (26.4) (28.0) (26.4) (28.0) (26.4) (28.0) Figure 6.1: NPV of HA management and administration costs by The costs are shown over the 15 year evaluation period commencing 1 April 1999 and ending 31 March 2014, discounted at the Treasury s appraisal rate of 6% In order to quantify costs resulting directly from the HA s anticipated capital expenditure programme a staff model has been developed. This staff model reflects proposed expenditure under the NCP and relates this to numbers of HA staff against different grades and pay bands. Averaged HA pay bands for London and outer London based staff have been used along with a fixed rate for overhead costs, again based on an HA standard figure. The ratio of capital expenditure managed per employee has been used as a means to gauge the output of the staff model. Output figures have been compared with typical HA figures for various types of project work. This exercise would suggest that the staff model is reasonably accurate. An analysis of staff numbers has been provided in Figures 6.2 to 6.4 below Where it is proposed that the cabling works in schemes are transferred to MCS Co under Strategies 3 and 4, then a similar model has been used to determine MCS Co s staff costs. Efficiency savings within the private sector are catered for by an anticipated increase in the ratio of capital expenditure managed per staff employed. Capital costs transferred out of the HA domain require an adjustment to the basic HA staff cost model, firstly to reflect the reduced amount of capital expenditure and secondly to allow for the HA to manage effectively its interests in respect of MCS Co s operations. This actually results in a small increase in HA staff costs. 23/40

24 Project staff numbers The Project staff costs in the PFI Strategies reflect the costs of some 19 staff of varying grades to manage the contract and two staff to oversee the technical aspects of the procurement. It is assumed that financial assistance is obtained from FS and from external advisers. These may be analysed as follows; Grade Years 1 3 Years Figure 6.2: HA Project staff to manage PFI contracts Project staff numbers in each PSC are as follows: Years Grade Figure 6.3: HA Project staff numbers PSC 24/40

25 Non project staff numbers The non project staff in Strategies 2, 3 and 4 are as follows: Years Grade Figure 6.4: HA staff - 2, 3 and Commentary on analysis The PFI scenarios are consistently more expensive than the PSC conventional procurement, reflecting the need set out above for relatively senior staff to manage the contract. 25/40

26 THIS PAGE INTENTIONALY LEFT BLANK 26/40

27 7.0 STRATEGY RISK ASSESSMENT 7.1 Basis of risk assessment The aim of the risk assessment process is to arrive at a valuation of the risks retained by the HA under each, distinguishing between the PFI options and the PSC options, to enable a like with like comparison of the value for money provided by one strategy over another and by a PFI strategy against its PSC In assessing cost of retained risk by the HA, KHHS have focused only on those risks which are likely to have a material impact on the likely benefit of each, and are likely to differ materially between Strategies and between the PFI and PSC options of each The identification of the risks to be assessed is based both on KHHS s experience in this and related projects and on the basis of a risk workshop held with HA management and staff For the purposes of this risk assessment, we have focused on the risks associated with the upgrade and enhancement of the motorway communication services, since it is the Strategies 2, 3 and 4 affecting the MCS for which the risk adjusted financial analysis is being prepared. 7.2 Risk modelling method The risks are modelled using triangular distribution with three points - a minimum, most likely and maximum. As shown in Figure 7.1 below, the direction of the skew of the triangular distribution is set by the size of the most likely value relative to the minimum and the maximum. The probability of the occurrence of the minimum and maximum values are zero. Probability Minimum Most Likely Maximum Value of risk Figure 7.1: Triangular distribution The most likely values for each of the risks modelled are the annual input data for the cost or revenue input provided in respect of each. The minimum and maximum values are estimated as variances in percentage terms from the most likely value. 27/40

28 7.2.3 The effects of the risks on the total cost of each are combined using Latin hypercube sampling analysis. This provides a distribution of results from which we can obtain the mean risk adjusted cost and the 90% confidence range bounded by the 95 th and 5 th percentile Latin hypercube is a relatively new stratified sampling technique used in simulation modelling. Stratified sampling techniques, as opposed to Monte Carlo techniques, tend to force convergence of a sampled distribution in fewer samples. 7.3 Analysis of key risks for Motorway Communication Services In PFI projects, the basket of risks that are generally considered in the risk assessment are: a) design and construction risk: including fitness for purpose, failure to deliver against requirements, time and cost overrun; b) commissioning and operating risks: including the availability of the services; c) demand risks: risk of variability of usage; d) residual value risk: the risk concerning asset value at the termination of the contract; e) technology and obsolescence risk: risk that the assets will cease to be the most technically effective means of delivering the services; and f) regulatory and legislation risk: risk of changes in law and regulations affecting the services Against these typical risk headings KHHS have identified the key risks particular to the MCS. These are described below in general terms and assessed against the particular strategies below. Time and cost overrun during construction A key design and construction risk in most PFI projects is time and cost overrun. Experience in the public sector suggests that historically it has a poor record of managing cost overruns in large projects and the aim in PFI is to transfer much, if not all, of this risk to the private sector. In the case of this project, the particular issues concerning land ownership, the right access to the roadside verges and the scheduling thereof and ground conditions experienced in civil works all contribute to this risk The risks of Capital costs risk - overrun has been modelled in this feasibility study by reference to percentage increases and decreases on the annual input amounts to give maximum, most likely, and minimum inputs. The percentage increases and decreases are based on assessments described in section 7.5 which vary between Strategies and between the PSC and PFI inputs for the same. 28/40

29 Transition/migration - physical and contractual Among the commissioning and operating risks, transition risk is seen as a key risk to the HA in terms of underestimating the HA resources required to manage the transition from its existing contractual arrangements to new PSC or PFI arrangements proposed in the Strategies. Shortfalls in resources or skills will either need to be met through additional recruitment or result in project delays HA management and administration cost risk has been modelled in this feasibility study by reference to percentage increases and decreases on the annual input amounts to give maximum, most likely, and minimum inputs. The percentage increases and decreases are based on assessments described in section 7.5 which vary between Strategies and between the PSC and PFI inputs for the same There is also a technical risk relating to the transition and migration from the HA s existing largely bespoke, non-standard and multiple standard equipment to a new, modern, commercial equivalent. The HA is concerned that it or a PFI supplier may underestimate the work required in moving its systems and devices across to a new platform and the reconfiguration this is likely to entail This technical transition risk has been included for the purposes of the feasibility study in the assessment of Capital costs risk - transition which is again modelled by reference to percentage increases and decreases on the annual input amounts to give maximum, most likely, and minimum inputs. The percentage increases and decreases are based on assessments described in section 7.5 which vary between Strategies and between the PSC and PFI inputs for the same. Latent defect risk Another commissioning and operating risk which has been highlighted by potential private sector suppliers to the HA concerns latent defects in the HA s roadside equipment. Concerns focus particularly on some of the HA s fibre optic cable which is direct buried and more than five years old. It is feared that this may have deteriorated and not be of sufficient quality to be able to convey modern high data rates This is noted as a risk in this report but is not included in the risk assessment as we understand that the HA has issued instructions for tests to be carried out on the fibre concerned to establish the extent of this risk. The results of these tests are expected well before any contract involving third party exploitation of the cable is to be negotiated. However, in Part B of the project, we will have to examine in more detail a range of risks concerning the condition of the existing equipment, if it is to be transferred to the private sector, and the risks posed by the limited number of suppliers with the skills and knowledge to service it. 29/40

30 Reconfiguration risks The HA has identified a risk concerning the work currently required when new items of roadside equipment or new schemes are added to the existing communications network. These additions are often time critical to ensure the continuing safety of the highways. Given the hierarchical structure of much of the existing network, the addition of new equipment to it requires substantial, specialised reconfiguration. To complicate matters, the TSS team at the HA is often not given notice of the new requirements until very late in a project, requiring an immediate and considerable mobilisation of effort. This was demonstrated most recently by the bus lane project between Heathrow and Heston on the M The risk has several parts to it. First, there is the risk of perpetuating a hierarchical system design which renders additions to the network difficult and costly. Second, there is the risk that a new system or contract structure would not be flexible enough to allow a rapid response to safety requirements. Third, there is the risk that HA work practices and highways conditions generally will continue to generate unplanned and immediate requirements not covered by contractual obligations The first risk of unforeseen reconfiguration costs has been included for the purposes of the feasibility study in the assessment of Operating cost risk - reconfiguration which is modelled by reference to percentage increases and decreases on the annual input amounts to give maximum, most likely, and minimum inputs. The percentage increases and decreases are based on assessments described in section 7.5 which vary between Strategies and between the PSC and PFI inputs for the same The second risk in respect of the threat to improvements in road safety has not been assessed since a consideration of the benefits from improvements in road safety resulting from the project is not within the scope of the Specification The third risk relating to additional operating costs from unforeseen additional works has been included for the purposes of the feasibility study in the assessment of Operating cost risk - overrun. This also has been modelled by reference to percentage increases and decreases on the annual input amounts to give maximum, most likely, and minimum inputs. The percentage increases and decreases are based on assessments described in section 7.5 which vary between Strategies and between the PSC and PFI inputs for the same. Demand risk Demand risk relates to the HA s use of its communications network, and whether its actual usage justifies the investment in infrastructure. This risk will require careful consideration in Part B of the project as the scope of the requirement and potential demand for communications services is developed in detail. 30/40

31 Perhaps more significant for this project is the risk attaching to the demand from third parties for commercial services. As noted in section 8.2 of the Volume 2, MCS Co s assessment of this risk will determine the amount by which it is prepared to subsidise its charges to the HA. This has been treated as a sensitivity rather than a risk adjustment in the financial model as described in section 8. Additionally, revenues accruing to the HA have been subject to a risk assessment using a maximum value of +5% and a minimum of -100%. 7.4 Other risks considered by the Study During the course of the Study, other risks specific to the motorway communications network have been identified by KHHS. These risks are described here but have not been included in the financial assessment for the reasons noted. Capacity/availability With ever increasing demands for capacity, the HA is faced with the risk that whatever the arrangements it makes now for its motorway communications network, be they public sector procured or PFI, its capacity requirements in the future are misjudged and it finds itself constrained by its existing provision and is forced to make additional arrangements at additional expense. This risk is not considered material by the Study as modern technologies are providing ever increasing capacities over existing infrastructure. The HA s capacity requirements for the foreseeable future are thought unlikely to exceed the capacities which its existing infrastructure, and any new infrastructure, should be capable of providing with the necessary upgrades and modifications to its transmission The HA also faces a specific risk regarding the availability of its services given the current prevalence of civil contractors on the highways damaging the HA s buried equipment during the course of civil works. Whilst this practice causes considerable irritation and disruption to the HA, it is unlikely to result in a material financial risk. However, it does highlight the consideration which will need to be given to all of the interfaces an MCS Co will have with the HA and its existing contractors during Part B of the Project. Technology risk/residual value Given the speed of developments in telecommunications and the time taken in the public sector to formulate and receive authorisation for projects, be they public sector funded or private sector funded, there is a risk to the HA that it adopts or is provided with a communications network which over the course of the project ceases to be the most technically effective means of delivering the service. The result would be that the HA is locked into a higher operational cost than is necessary or has to reinvest in order to secure a more cost effective service or one that meets new requirements enabled by new technology. 31/40

32 7.4.5 This risk is common across all telecommunications solutions for communications traffic, but it is greater for more specific applications such as roadside to vehicle links. It is not possible to quantify this risk, but it will require a qualitative consideration when determining between bidders solutions in 4 and between Strategies 3 and Assessment of HA s retained risks under each strategy Following on from the narrative in section 7.3 above, KHHS have identified the following risks for inclusion in the financial assessment of risks retained by the HA under each : a) Capital costs risk - overrun; b) Capital costs risk - transition; c) HA management and administration cost risk; d) Operating costs risk - overrun; e) Operating costs risk - reconfiguration; f) Revenue risk. Capital costs risk - overrun Capital cost risk - overrun is simply the risk of cost overruns in the capital costs of the project. Possible causes of such overruns, in addition to the usual incomplete budgeting risks, have been identified as difficulties in access to the roadside verges and ground conditions. The risk is stated in terms of minimum and maximum percentage variations from the annual capital cost input data The assessments for each strategy are set out in Figure 7.2 below: Project 2 3 PSC 3 PFI 4 PSC 4 PFI % % % % % Minimum n/a -5% n/a -5% n/a Most Likely n/a 0% n/a 0% n/a Maximum n/a +50% n/a +50% n/a Non project Minimum -5% -5% -5% -5% -5% Most Likely 0% 0% 0% 0% 0% Maximum +15% +15% +15% +15% +15% Figure 7.2: Capital cost risk - overrun assessment 32/40

33 7.5.4 The risk workshop assessed capital cost risks separately for Project and Non Project costs In the PFI Strategies 3 and 4, it considered that risk of capital cost overruns for Project costs would be transferred entirely to the private sector and there would therefore be minimal retained risk to the HA However where the costs are retained by the public sector, which includes all Non project capital costs and Project costs in the PSCs, the risk is retained by the HA. Project costs were viewed as carrying a substantially higher risk than Non project costs, with the potential for up to a 50% overrun. Historically, the HA has never completed a capital project for a lower cost than the original estimate. The risk for Non project costs of overruns was less at 15%, and both on the upside could benefit from up to a 5% underspend. Capital costs risk - transition Capital cost risk - transition is the risk relating to the transition and migration from the HA s existing largely bespoke, non-standard and multiple standard equipment to a new, modern, commercial equivalent. The risk is stated in terms of minimum and maximum percentage variations from the annual capital cost input data The basis of the risk assessment is that the risk of transition cost overruns in the 3 and 4 PFI options is transferred entirely to the private sector and there is therefore no retained risk to the HA. For the public sector options, 2, 3 PSC and 4 PSC the risk is retained by the HA and it may suffer a maximum 2.5% overspend on Project capital costs as a result The workshop agreed that this risk is reflected within the 50% assessment in Figure 7.2 above. HA management and administration cost risk HA management and administration cost risk is simply the risk of cost overruns in these costs. Possible causes of such overruns, in addition to the usual incomplete budgeting risks, have been identified as difficulties in managing the transition from existing contractual arrangements to new arrangements. The risk is stated in terms of minimum and maximum percentage variations from the annual HA management and administration cost input data The assessments for each strategy are set out in Figure 7.3 below: 2 3 PSC 33/40 3 PFI 4 PSC 4 PFI % % % % % Minimum -5% -5% -10% -5% -10% Most Likely 0% 0% 0% 0% 0% Maximum +15% +15% +25% +15% +25% Figure 7.3: HA management and administration cost risk assessment