Chapter 5: Procurement Strategy

Preliminary matter

5.1 Before addressing the substantive issues in this chapter, it is necessary for me to make some preliminary comments on the evidence. Mr Kendall worked with tie Limited (“tie”) between September 2003 and May 2006, latterly as its Tram Project Director (“TPD”). He was described by Mr Crosse as the “commercial architect” of the Edinburgh Tram project (the “project”) [TRI00000031_C, page 0001, paragraph 1]. It is clear from Mr Kendall’s own remarks to the Inquiry that he was closely involved in the development of tie’s strategy for the procurement of the project [TRI00000136].

5.2 Mr Kendall attended an interview with a member of the Inquiry team on 19 and 20 May 2016, having been sent in advance a list of topics for discussion prepared by counsel to the Inquiry. A similar approach was taken with other witnesses whose evidence was thought likely to be important to the Inquiry’s terms of reference. For those other witnesses, the Inquiry in many cases produced a draft statement based on the interview, which the witness was then given an opportunity to review or revise, as they considered appropriate. Those witnesses were then invited to sign their statements, which had been revised to take account of their responses, to confirm the truth of their contents. Mr Kendall died before this procedure could be followed with him. The Inquiry has a typed transcript of Mr Kendall’s interview [ibid], which is the only record of his responses to the matters put to him by the Inquiry.

5.3 I have taken the view that it is appropriate for me to have regard to that transcript [ibid]. In doing so, I keep in mind the following points. First, Mr Kendall did not have the same opportunity as other witnesses to see his interview answers presented in a draft statement, or to reflect either on the substance of the statement or on the way in which it was expressed. Second, Mr Kendall did not give evidence in person at the Inquiry’s oral hearings, as many other important witnesses did. He did not therefore have the opportunity to clarify points of importance in response to questions from counsel or from me, nor did I have the opportunity to form an impression of him as a witness. There was no opportunity to put to him any of the evidence given by other witnesses. Third, his interview answers were not given under oath or affirmation. That puts his evidence into a different category from that of witnesses who attended the oral hearings, all of whom at that stage, under oath or affirmation, adopted their witness statements as accurate. I kept these considerations in mind when assessing the evidence, and am satisfied that Mr Kendall’s remarks in the interview are nonetheless of assistance to me.

Introduction

5.4 From an early stage in the project, City of Edinburgh Council (“CEC”) and then tie sought advice on the best means of procuring the main contracts that would be needed to build and operate the tram scheme. For example, in 2001, CEC commissioned a study by Turner & Townsend, management and construction consultants, on appropriate procurement strategies for what was then the proposed North Edinburgh Transit Project. The report of its strategic project review was issued to CEC in September 2002 [CEC01868789]. Between 2002 and 2004, tie
ran a procurement group that was tasked with considering how best to procure the tram system. The procurement group included professional advisers from various disciplines, including DLA, Grant Thornton, Mott MacDonald, Faber Maunsell and Partnerships UK [CEC01853491; CEC00630633, pages 0019 and 0058].

Reports by the National Audit Office, Audit Scotland and Lord Fraser of Carmyllie

5.5 By 2004, there was recognition of a need for more innovative approaches to the procurement of tram and light rail schemes in the UK. A report by the National Audit Office (“NAO”) published that year, which was entitled ‘Improving public transport in England through light rail’, recommended that the Department for Transport (“DfT”) seek to identify more cost-effective procurement methods for such projects [CEC01708649]. It noted the poor financial performance of several existing schemes, and the tendency of scheme promoters to seek to transfer as many of the project risks as possible to the private sector. Such factors were leading to inflated project costs, as the private sector either avoided light rail projects altogether (removing competition) or sought greater margins for taking the risks. The NAO suggested that better sharing of project risk and alternative contract structures could help to reduce the cost of such projects and encourage private-sector investment. In particular, it noted that the private sector was generally well equipped to carry construction risks, but not the revenue risks of operating schemes, the incidence of which was often influenced by factors beyond the private sector’s control. The NAO also noted that the diversion of utilities tended to be an expensive part of tram or light rail schemes, and recommended that the DfT fund projects only where the promoters had adequate proposals for managing the risks associated with such diversions. Mr Kendall told the Inquiry that, following the experience of Carillion Utility Services Limited on the Nottingham project, contractors were not willing to take the risk associated with diverting utilities: in his words, “they lost their shirt and everyone in the market place knew it” [TRI00000136, pages 0062 and 0074]. tie and its advisers were aware of the NAO report and sought, by their procurement strategy for the Edinburgh project, to address the issues that it had raised [see, e.g., CEC00115187, page 0009; CEC01875336, Part 2, page 0038, paragraph 5.2.1; Dr Fitzgerald PHT00000039, page 140 onwards]. The principal lesson that tie took from the NAO report was that risk should be actively managed out of the project. It also noted the need for new procurement structures, with better risk-sharing arrangements than had been used previously. Its objective was to find a cost-effective procurement method “as a means of controlling cost” [CEC00115187, page 0009].

5.6 Outside the narrow sphere of light rail projects, 2004 saw the publication of two reports into the cost overruns suffered by the project to build the new Scottish Parliament building at Holyrood – one by Audit Scotland [“Management of the Holyrood building project”, June 2004 ADS00054] and the other by Lord Fraser of Carmyllie following his public inquiry [“The Holyrood Inquiry”, September 2004 WED00000624]. These were perhaps less directly relevant than the NAO report in shaping a procurement strategy for a Tram project [Mr Fitchie TRI00000102_C, page 0071, paragraph 4.162; Dr Fitzgerald TRI00000036_C, page 0027]. Nonetheless, they made important findings relevant to the procurement by public bodies of major construction projects, and tie aimed to take account of their findings [CEC00630633, page 0008; CEC01875336, Parts 1–2, pages 0007, 0009, 0011 and 0038].

5.7 One of the key issues identified by Audit Scotland was the need to define the client’s requirements before awarding contracts, if price certainty was an objective. On the Holyrood project, contracts had been let before the design was finished to meet a programme that was demanding and, in hindsight, probably unachievable. The consequence was that fixed prices could not be negotiated for many contracts, and price competition at the tender stage was restricted. Further, having awarded contracts with uncertain scope and design, the client was in a weak position to resist contractors’ claims for extra time-related costs. Even where the contractors’ performance might not have been satisfactory, there was little opportunity to attribute delays to those contractors because of delays occurring elsewhere in the overall programme [ADS00054, pages 0008–0009 – in particular key findings at paragraphs 8, 13 and 14]. Audit Scotland emphasised in the following comment that, to achieve price certainty, it was important to allow sufficient time for procurement:

“[t]here must always be sufficient time for procurement to allow the client’s requirements to be adequately defined so that it may obtain fixed and firm prices for the work in a competition” [ibid, page 0011, paragraph 33].

5.8 tie said that it had learned lessons from Audit Scotland’s report. These included:

• the need for realistic programming for design and construction, taking into account all critical project assumptions that could delay the scheme;
• the need to optimise the transfer of design and construction risk to the private sector;
• the need to initiate detailed design at the earliest opportunity to avoid variations; and
• the need to ensure that the decision to award contracts was taken with a clear understanding of the obligations and of any elements yet to be clarified. [CEC01875335, pages 0104–0105; TRS00000053, from page 0020; CEC01705043, from page 0020.]

5.9 On 12 November 2004, tie submitted a response to each of the parliamentary committees considering the Edinburgh Tram (Line One) Bill and the Edinburgh Tram (Line Two) Bill (the “Tram Bills”) [CEC01705043; TRS00000053 respectively]. The issues raised by the committee for each Bill were similar and the responses quoted below were identical. In its responses to the committees tie said:

“[t]he contract structure will prevent any open-ended commitment of funding, as has been a problem on other public projects such as the Holyrood building” [CEC01705043; TRS00000053, page 0002, paragraph 8]

and that

“the risk of capital cost overrun is mitigated by the fact that no commitment will be made to construction until robust contractual arrangements are in place and the affordability of the project is agreed” [CEC01705043, page 0004, paragraph 19; TRS00000053, page 0003, paragraph 16].

Tram operator

5.10 The first element of tie’s procurement strategy to be put into place was its early appointment, in May 2004, of a tram operator, Transdev Edinburgh Tram Limited (“Transdev”), under a Development Partnering and Operating Franchise Agreement (“DPOFA”). In spring 2003, tie had determined that this was an “innovative and critical element of project risk management” [CEC00630633, page 0060], for the principal reasons summarised below.

• Constructing the tram system, on the one hand, and operating it, on the other, involved different risks; by separating those risks into different contracts, tie would allow bidders to focus on those risks that they were suited to bear. That was in turn thought likely to reduce risk premiums, and thus the price that each bidder would demand.
• Early involvement of the operator in production of the design would ensure that the design met the operator’s needs. That would remove the risk of redesign being required at a later stage, as might occur if the operator was appointed only after the design had been produced.
• tiecould draw upon the operator’s knowledge and experience to assist during negotiation of the construction contracts, with a view to keeping construction prices down.
• The operator could assist in planning, from an early stage, the integration of the tram network, especially with bus operations.

5.11 tie considered that the NAO report pointed strongly to the early involvement of the operator of the light rail system “as a means of improving … procurement and achieving a stable and affordable system” [ibid, page 0060]. tie’s approach was a rejection of the simple private funding model used for other public transport infrastructure projects, under which a concession company would design, build, finance and operate the infrastructure in return for the revenue streams it generated [ibid, page 0058]. By separating the construction risks and operating risks into different contracts, and seeking to allocate them to the parties best able to bear them, tie was following the advice of the NAO. Furthermore, tie noted the financial underperformance of UK tram and light rail schemes procured under full private finance initiative (“PFI”) or public–private partnership (“PPP”) structures, in which all of the revenue (“fare box”) risk had been transferred to the private sector. In such schemes potential concessionaires lacked confidence in patronage modelling, leading them to discount revenue streams heavily when negotiating contract prices and to seek significant risk premiums [ibid, page 0064]. To avoid these problems, under the DPOFA, tie proposed to leave much of the revenue risk with the public sector. A degree of control over the public-sector exposure to that risk would be introduced via a “pain/gain sharing mechanism” [CEC01875335, page 0062].

5.12 tie’s Pre-Qualification Guide for the proposed competitive procurement of the DPOFA contract identified further objectives in the early appointment of an operator:

• to help to support tie’s economic and technical case in the parliamentary process for approval of the Tram Bills;
• to support tie in the procurement of tram infrastructure, vehicles and equipment;
• to identify the “Edinburgh tram network as an attractive, innovative and deliverable scheme with strong, committed private sector involvement” [DLA00004903, page 0013].

5.13 tie considered that awarding separate contracts for operating the tram system and for constructing it would:

“offer a fundamentally more attractive commercial package to bidders for the respective contracts and should, as a consequence, deliver a better value for money solution to tie and the Council” [CEC00630633, page 0048].

5.14 tie reported an “enthusiastic” response to its market consultation on the DPOFA contract. Pre-qualification submissions were received from six candidates, the majority of whom were described by tie as “market leading operators” [CEC01875336, Part 3, page 0053, paragraph 5.6.3]. They included Transdev SA,  a public transport operator headquartered in France, and on 14 May 2004 tie appointed Transdev under the DPOFA [CEC01862694].

Vehicles and infrastructure

5.15 Having decided on the separate appointment of the operator under the DPOFA, tie next had to consider how best to procure the tram system infrastructure and vehicles. In its updated Preliminary Financial Case of September 2004, tie reported on work done in this regard by its procurement group [CEC00630633, page 0066, paragraph 6.4 et seq.]. The procurement group considered six different options in detail, with varying degrees of risk transfer to the private sector. The least risk transfer would be achieved with a “Traditional Procurement Option”, under which tie would separately procure contracts for each of the design, infrastructure, vehicles and system integration elements. The greatest risk transfer would be achieved with a “Full Consortium Option”, under which tie
would conduct a single procurement for one consortium to provide all those elements [ibid, pages 0069–0072, paragraphs 6.4.2–6.4.3]. It favoured one of the options with an intermediate amount of risk transfer: the “Infrastructure and Integrator Consortium Option”. This involved two separate procurements: one for the tram vehicles, and another for the design, infrastructure works and systems integration. The tram vehicle supply contract would ultimately be novated to the infrastructure contract (the “Infraco contract”), with the result that while there were two procurements, novation would establish a single consortium with responsibility for all the design, infrastructure, vehicles and systems integration elements.

5.16 tie anticipated that its procurement model would transfer substantially all the construction and integration risk to the private sector, with the revenue risk (under the DPOFA) being substantially retained by the public sector [ibid, page 0011]. The procurement group did not consider it appropriate for tie to retain much of the construction and integration risk, given the scale and complexity of the project and its lack of resource for, or experience of, major project management, as tie was essentially a procuring body [ibid, page 0068].

5.17 There were, however, certain matters over which the procurement group thought that tie should retain control, even at the potential cost of lessening the risk transfer, for the benefits that that would bring. The first of these was the choice of tram vehicle. There were relatively few tram vehicle suppliers, and requiring infrastructure bids to include the supply of vehicles risked severely reducing the number of bidders. It could also compromise choice if, for example, the preferred vehicle supplier was tied to a sub-optimal infrastructure bidder, and vice versa.

5.18 The second matter was design. Design was known to be a sensitive issue, since parts of the line would pass through Edinburgh’s World Heritage Site. As the planning authority, CEC was known to have concerns about the design of the tram network. The procurement group therefore thought that there was merit in considering a preliminary package of targeted design work ahead of letting a main infrastructure contract. The aim of this would be to develop designs that were likely to satisfy planning requirements, reducing risk and wasted design work, and speeding up the overall timetable.

5.19 The third matter was diversion of utilities, which was recognised as being a time-consuming and high-risk element of the project. If tie were able to gain greater certainty on the requirements for utility diversions, that could help to achieve the timetable and reduce risk for the main infrastructure contractor [ibid, page 0068].

5.20 The procurement group noted that, under its preferred approach, there was the potential for maximum risk transfer if the vehicle contract and advanced designs were novated to the infrastructure contractor. There was also scope for cost certainty. This would come, in particular, from advance design and utilities work reducing uncertainty, and thus risk, for the infrastructure bidders. The procurement group considered that advance design and utility diversion work should increase market appeal, but noted that market consultation was needed to confirm that [ibid, page 0071]. tie did, however, understand that there would be considerable demand from the construction industry to undertake delivery of the scheme and that its proposed strategy would be well received [CEC00115187, page 0023].

5.21 It is therefore clear that:

• in its conception, the procurement strategy for infrastructure and vehicles aimed to transfer most of the construction risks to the private sector because tie was unsuited to manage them;
• tie continued to take seriously the recommendation of the NAO that light rail procurement should seek better allocations of risk to achieve value for money;
• design and utilities were identified as potential sources of risk, in relation to which advance work bytie might reduce risk and therefore cost; and
• tie would conduct market consultation on its strategy while developing it and before implementing it.

Draft Interim Outline Business Case (May 2005)

5.22 By the time of issue of its draft Interim Outline Business Case (May 2005), tie had analysed the procurement options in detail and devised a strategy that it considered was suitable for market testing [CEC01875336, Part 1, page 0009]. It acknowledged that its strategy was “a unique approach” that differed from market norms. While there was not a fixed template for light rail projects, tie’s strategy differed from previous approaches in the following respects:

• early introduction of the operator, under the DPOFA;
• separate contracts for operation and construction of infrastructure;
• sharing of revenue risk between the operator and the public sector;
• early involvement of a designer;
• utility diversions undertaken in advance of infrastructure works; and
• separate selection of infrastructure and vehicle providers.

ibid, Part 2, pages 0041–0042, paragraphs 5.3–5.3.6.]

5.23 tie considered that, through its strategy, it would take more control during the development phase (that is, prior to commencement of construction) than the public sector had done on previous projects. Design work and utility diversions would be carried out, under contract to tie, prior to the procurement of the Infraco contract. The aim was to give the infrastructure bidders a better-defined basis for their bid. That greater certainty would reduce risk and thus the cost of the project [ibid, Part 2, page 0040, paragraph 5.2.3].

5.24 In the draft Interim Outline Business Case tie identified affordability constraints for the project:

• the limited support of £375 million without indexation confirmed by the Scottish Ministers;
• the lack of significant support from CEC to fund capital expenditure; and
• uncertainty about CEC sources of income to fund the project.

5.25 In light of these constraints it had become:

“all the more important to achieve as much certainty as possible on the likely price for the different elements of the network before entering into commitments” [ibid, Part 6, page 0130, paragraph 8.5.2].

5.26 tie’s aim was “to achieve as close to fixed prices as possible” [ibid, Part 6, page 0131, paragraph 8.5.4]. The extent to which fixed prices could be achieved would depend on the extent of detailed design that tie was able to provide as part of the tender documentation for the Infraco contract. However, tie’s intention in relation to that contract was to be

“in a position to proceed to detailed negotiation of contracts with a large
measure of certainty on price and therefore affordability” [ibid, Part 6, page 0132, paragraph 8.5.4].

5.27 This approach – of developing a more certain basis for pricing, with a view to fixing prices – was consistent with the themes emphasised by Audit Scotland in its report on the Holyrood project.

5.28 The draft Interim Outline Business Case was the final iteration of the business case before tie concluded a contract with Parsons Brinckerhoff (“PB”) for design and technical services in relation to both lines 1 and 2 on 19 September 2005, known as the contract for System Design Services (the “SDS contract”), as mentioned in Chapter 6 (Design (to May 2008)) [CEC00839054]. Apart from design, the technical services to be provided by the System Design Services (“SDS”) contractors included the majority of the “extensive advance survey work” that tie planned to carry out: the information derived from such survey work would aid detailed design [CEC01875336, Part 3, page 0058, paragraph 5.7.5]. In the draft Interim Outline Business Case, tie explained the proposed interaction between the advance design works, the utility diversions and the Infraco contract. It also explained that early designs would facilitate advanced utility diversions, which would in turn reduce programme and cost risks for the infrastructure bidders and, therefore, their prices. This strategy would thus reduce uncertainty, risk and cost in two respects: by removing uncertainty about the design itself, and thus the cost and time of building to it; and, through facilitating advance utility diversions, by removing uncertainty about the time impact that these diversions would have on the programme for constructing the infrastructure. The designer’s initial focus was to be the on-street section between  Ocean Terminal and Haymarket (via Princes Street) [ibid, Part 1, page 0010, paragraph 4].

5.29 The procurement of the Infraco contract was to take place in parallel with the production of the design. Tenderers for that contract were initially to be given the preliminary design and, later (as it became available), the detailed design for those parts of the scheme that were perceived to be critical for cost and programme. Two shortlisted bidders for the Infraco contract would receive a significant design update as a basis for refining their proposals during the Best and Final Offers stage [ibid, Part 4, pages 0069–0070, paragraphs 5.12.3–5.12.4]. The SDS contract’s overall objectives would include:

• tackling critical design elements as early as possible;
• reducing project risk for the Infraco bidders; and
• generating design solutions that an Infraco bidder could competitively price. [ibid, Part 3, page 0054, paragraph 5.7.]

5.30 tie anticipated that the overall design process would take between two and two-and-a-half years. That estimate was based on the experience of Mr Kendall and tie’s professional advisers [TRI00000136, page 0089]. tie expected that, by the time that the Infraco contract was signed, the design of utility diversions would be complete; planning permissions for the most critical sections (that is, between Haymarket and St Andrew Square) would have been granted; and the design would be “60–70% complete” [CEC01875336, Part 3, page 0054, paragraph 5.7.1]. Mr Kendall explained his view that this meant that the scope of the project would be designed and that, insofar as design was incomplete, it would be in matters of “detailed engineering” [TRI00000136, page 0089]. Mr Kendall explained that the decision to produce design in parallel with procurement of the Infraco contract was taken to save time [ibid, page 0075]. tie expected that it would reduce the procurement timetable by at least a year and acknowledged that it was keen to adhere to an ambitious procurement timetable, but emphasised that this would not come “at the expense of increased cost or risk” [CEC01875336, Part 3, page 0054, paragraph 5.7]. In September 2005, in a progress report to the parliamentary committee considering the Tram Bills, tie noted that the programme for the appointment of both the Infraco contractors and those for the tram vehicle supply and maintenance contract (the “Tramco contract”) was of a “compressed duration”, to minimise construction cost inflation [CEC00380894, page 0014, paragraph 8.2].

5.31 One can readily understand the desire for timely progress to avoid, so far as possible, increased cost attributable to inflation. However, as Audit Scotland had noted in relation to the Holyrood project, the pressure of a programme brought with it a risk that the design might not be as complete as one would ideally want when the time came to award contracts. Further, incomplete design was itself a source of increased risk and cost, as tie’s strategy recognised. Despite tie‘s assertion that its “ambitious timetable” would not come at the expense of increased cost or risk, it inevitably introduced a tension with the objective of managing risk out of the project before the Infraco contract was awarded. The design contract was to be novated to the infrastructure contractor when the Infraco contract was signed, and the infrastructure contractor would be required to adopt the design contractor’s design. Variations to the design could be introduced, but tie considered that this would be done only with its agreement and at the risk of the infrastructure contractor. tie explained that, in this way, all design risks would be transferred to the private sector, without the risk premiums that would be charged due to uncertainty if the design were to be carried out after contract signature [CEC01875336, Part 3, page 0056, paragraph 5.7.1.3].

5.32 In relation to utility diversions, tie proposed to retain and manage the significant risks itself. It would appoint a single contractor, approved by all the affected utility companies, to implement the major utility diversions. It would do this in advance of procuring the infrastructure contractor, with the objective of completing “significant” utility diversion works before the infrastructure works began [ibid, Part 3, page 0059, paragraph 5.8]. By appointing a single contractor to divert utilities for all the utility companies, tie would avoid the risks that would otherwise arise from dealing separately with each utility company (a risk that tie said would typically arise if diversions were left to the infrastructure contractor). Further, diverting all utilities at a single site at one time under a single contract would help to minimise cost and disruption to the public [ibid, Part 3, page 0059]. tie explained that utility companies, whose engagement was needed for accurate scoping, programming and pricing of utility diversion works, were generally unwilling to engage with prospective contractors who were still at the stage of competing with other bidders for a contract. This meant that, if utility diversions were handled as part of the main Infraco contract, the work necessary to scope, programme and implement utility diversions generally came only after the Infraco contract had been awarded. This tended to delay the infrastructure works but, further, meant that infrastructure contractors’ bids would carry risk premiums because of the uncertainty about the scope and timing of utility diversion works [ibid, Part 3, page 0059]. The diversion of utilities under tie’s management, in advance of the Infraco contract being awarded, would therefore save programme time (and related cost) and reduce risk pricing by the infrastructure contractors [ibid, Part 3, page 0060]. To determine the scope of the utility diversions, tie would be dependent on advice from the design contractor and from the utility companies. The design contractor would determine the area of the track bed, under which utilities would have to be diverted or protected [ibid, Part 3, page 0059]. Some utility diversions would remain the responsibility of the infrastructure contractor. That would be the case where the diversion depended on that contractor’s design – for example, the re-siting of utilities necessary to accommodate overhead line supports [ibid, Part 3, page 0059, paragraph 5.8.1].

5.33 tie recognised that its strategy left it with important risks to manage during the period up to award of the Infraco contract. These included risks associated with development of the detailed design, the obtaining of planning consents, the diversion of utilities, and management of the programme [ibid, Part 5, page 0093, paragraph 6.5.1]. There were also risks associated with the planned novation of the design and vehicle supply contracts to the infrastructure contractor. These included that, due to supervening events, the parties would become reluctant to implement novation [ibid, Part 2, page 0043, paragraph 5.4.1 and Part 3, pages 0046–0047, paragraphs 5.4.5–5.4.5.2 inclusive]. The procurement strategy thus placed an important responsibility on tie to identify and manage the risks retained by the public sector. tie acknowledged that responsibility in the following passages in the Interim Outline Business Case:

“[T]he public sector is exposed to significant but manageable risks during the period of scheme development. The introduction of the SDS Contractor and USFA [utilities] Contractor in the proposed procurement strategy reduces risk to an extent, but, as in all projects of this type, the major responsibility for identifying and managing potential risks during this period will remain with the project team and their advisers. tie has assembled a team with significant experience in the tram industry and, together with the TSS Contractor, the Operator [Transdev], and its other advisers, believes that it has the necessary skills to manage risk during this period.” [ibid, Part 5, page 0095, paragraph 6.5.1.]

“The theme of the overall strategy is to ensure that risks are aggressively managed and in particular that tie’s stakeholders are not asked to commit to either contractual or financial obligations until each stage has been thoroughly analysed and approved.” [ibid, Part 1, page 0017.]

5.34 The success of the procurement strategy was therefore dependent upon tie’s ability to manage the risks inherent in the procurement process. One important aspect of this was for tie to manage the design and utilities contracts to reduce, so far as possible, uncertainties (and thus risk and cost) affecting the Infraco contract. Another was to ensure that any remaining risk was properly understood, analysed and presented to CEC, so that their decision whether to undertake the contractual and financial obligations of the Infraco contract was properly informed about the nature and scale of the risk involved in doing so. tie held itself out as having the experience needed to deal with these issues. To the extent that design and utility diversions remained incomplete when the Infraco contract was awarded, risk would remain with the public sector. It was tie’s responsibility to analyse that risk, and accurately report on it to CEC, before CEC was called upon to decide whether to enter into the Infraco contract. Critical in this management of risk was the timing of the award of the Infraco contract. In his evidence to the Inquiry, Mr Kendall noted that, before the Infraco contract could be awarded, the design and utility diversions would have to be sufficiently advanced to reduce the risk of delay to the Infraco works. tie had, he said, a complete understanding that it would “cause a huge financial problem if Infraco was awarded too soon” [TRI00000136, page 0084]. He noted that the decision on when to award the Infraco contract was entirely under tie’s control. The significance in the overall strategy of delivering the utilities diversion programme before awarding the Infraco contract was emphasised in the following two passages of his evidence. In the first he said that, if he had been managing the procurement process, he would not have awarded the Infraco contract

“no matter what it meant, until such time as I was confident that I could deliver the utilities diversion programme, or have modified the Infraco to reflect the fact that I couldn’t divert those and come to a different answer”

even although he acknowledged that delay in the procurement would mean an increase in project costs due to inflation [ibid, page 0122]. The second passage related to the possibility that the planned strategy of removing a significant element of risk premiums from the tenders would fail because of delays in design and the diversion of utilities. In that situation he observed:

“The first thing that you knew that you had to do to manage that risk was to award the MUDFA [Multi-Utilities Diversion Framework Agreement] as fast as possible and to get it underway [sic] and I don’t believe that was done … I would not have awarded the Infraco until I was dead certain, based on everything I knew at the time, not what I knew in theory, but what I knew at the time, to award that Infraco, until it was clear that I had got headroom, space, time, to be able to finish it.” [ibid, page 0123.]

Market consultation on the procurement strategy, and the Draft Outline Business Case (March 2006)

5.35 As mentioned in Chapter 9 on procurement up to preferred bidder, on 6 October 2005 tie published a Prior Information Notice, inviting experienced tramway infrastructure contractors to participate in a consultation on its proposed procurement strategy for the tram infrastructure works. An information memorandum [CEC01866826] sent to consultees set out tie’s proposed strategy consistently with the way in which it had been described in the draft Interim Outline Business Case. The memorandum noted that, to achieve its procurement objectives, tie had harnessed experience, by recruiting to the project team individuals who had a breadth and depth of experience of other light rail projects, by engaging Transdev (which had experience of procuring and operating such schemes) and by selecting advisers with direct relevant experience. At the time, members of tie’s project team who had experience of light rail projects included Mr Kendall, the TPD, who had worked on the Croydon Tram project [TRI00000136, page 0082]. Although Transdev and other advisers had direct relevant experience of other light rail projects, it is not apparent which individuals, other than Mr Kendall, within the project team had the necessary breadth and depth of experience mentioned in the memorandum.

5.36 Following responses from eleven interested parties, tie invited six potential Infraco bidders and five potential Tramco bidders to have discussions on 2 and 3 November 2005 in Edinburgh, at sessions chaired by Mr Kendall, supported by members of the tie team and tie’s advisers. In advance of the discussions, the invited parties were sent a Project Information Memorandum, setting out the background to the project and tie’s current thinking on the procurement strategy, together with a list of specific questions. Separate lists were prepared for potential vehicle suppliers and lnfraco bidders. The questions were compiled by the project team and advisers, in consultation with officials in the Scottish Executive. As the TPD, Mr Kendall reported on the consultation to the meeting of the Tram Project Board (“TPB”) on 22 November 2005 [TIE00090571 page 0003; TRS00002067, page 0002, item 3.2]. He reported that all the infrastructure consultees were concerned over whether funding would be available for the Tram project, and he advised the board that:

“committed funding is critical to having a successful tendering process. TPD’s position is that there will be no tender release unless and until he is satisfied that funding is committed, subject to contract” [TIE00090571, page 0003].

5.37 In his evidence to the Inquiry, Mr Kendall noted the positive market response to the procurement strategy, but added that:

“there was still really a quite small degree of interest in the Infraco market place for this project because there was a high degree of scepticism as to whether it would ever happen and whether it would ever actually be funded.” [TRI00000136, page 0084, see also page 0126.]

5.38 Mr Harper (Mr Kendall’s immediate successor as TPD) gave evidence to similar effect [TRI00000043_C, page 0023, paragraph 78].

5.39 In its draft Outline Business Case of March 2006, tie also discussed the market consultation [CEC00380898, page 0043, paragraph 6.2.2]. The consultees were reported as having generally welcomed the overall approach that tie
had taken in developing the procurement strategy, and as having recognised the rationale for it. The Infraco consultees considered the separation of tram operations and revenue risk from the Infraco contract to be attractive and an important driver for good, value-for-money bids. They generally understood and supported the rationale for early utility diversions. All saw the benefit of seeking early planning consents relating to the core network. Those consultees with major in-house design capability were said to have been “slightly disappointed” that significant elements of design would be undertaken prior to the award of the Infraco contract. The consultees did not consider the proposed novation of the design contract to be problematic: such arrangements were common practice and PB was well known and respected. The general rationale for separate procurement of infrastructure and tram vehicles was widely accepted, although two infrastructure contractors were reluctant to accept the risks associated with vehicles and vehicle integration. Respondents noted that running bid processes in parallel for the vehicles and Infraco contracts had the potential to be complex and expensive; generally, infrastructure contractors were likely to prefer to know the identity of the preferred vehicle supplier at an earlier stage, to help to remove uncertainty on integration risks. All Infraco bidders were said to have been interested in the availability and commitment of public-sector funds for the project (in light of recent experience in England – presumably a reference to the losses suffered by private-sector companies in running tram operations). Bidders were said to have been concerned about prolonged uncertainty on the point, but comforted by the assurance from tie that clarification would be provided prior to procurement in 2006. This assurance was fulfilled on 26 January 2006, when CEC approved in principle a contribution of £45 million for the project, subject to a satisfactory business case, and noted the willingness of the Scottish Executive to consider indexation of its “in principle” commitment of £375 million to take account of construction price inflation [CEC02083547, pages 0004 and 0008, paragraphs 3.14 and 6.1]. As indicated in the introduction in Chapter 1 (Introduction and Overview), in February 2006, the Scottish Ministers announced that their proposed grant would be increased to approximately £500 million in line with indexation, subject to a satisfactory business case. The Tramco bidders were said to have welcomed the opportunity to concentrate on tram vehicle delivery outside the complications and risks of a consortium structure, and were content with the proposed novation of the vehicle contract to Infraco [CEC00380898, page 0046]. It is therefore clear that tie’s procurement strategy was generally well received by the market. The only area in which there was arguably a poor response was in relation to the Infraco contract; on the evidence, that did not appear to relate to the procurement strategy itself but to the perceived uncertainty about the extent to which the project would be publicly funded.

5.40 tie continued to report the objective of achieving a fixed-price contract for the infrastructure [ibid, page 0072, paragraphs 7.3.4–7.3.5]. Its description of anticipated design progress by the date of award of the Infraco contract was broadly consistent with what it had said in its draft Interim Outline Business Case in May 2005 [CEC01875336, Part 3, page 0054, paragraph 5.7.1]. By the time of contract award, detailed design would not be complete but would be significantly advanced: the majority of consents would have been obtained; and outstanding design work might include non-critical areas, amendments required by consenting authorities but yet to be completed, and value engineering by the infrastructure contractor [CEC00380898, page 0058, paragraph 6.7.1.1].

5.41 Both CEC and the Scottish Ministers had said that they would not approve the start of physical utility diversion works until a draft of the Final Business Case (“FBC”) had given sufficient comfort that the capital cost estimates for the project overall were robust and that the project was affordable [ibid, page 0004]. This was consistent with the strategy of not committing to expenditure until there was greater certainty about costs, but it perhaps introduced a tension: early completion of utility diversions was an important factor in achieving price certainty, but CEC and the Scottish Ministers were seeking reassurance about cost certainty before allowing those diversions to proceed. Assessments about affordability made before utility diversions had begun would inevitably have to rest on assumptions about the scope of those diversions and the time needed to carry them out. At this stage tie acknowledged potential overlap in the programmes for utility diversions and the infrastructure works. However, the majority of utility diversion work was scheduled to commence in early 2007 and end in summer 2008, resulting in significant diversion works being completed prior to the commencement of Infraco works. Accordingly, potential conflicts between the utility diversions and infrastructure works would be minimised, and any remaining time overlap could be managed to avoid conflicts on the ground. The scope and time risks of overlap remained with tie, so its ability to manage them was critical [ibid, page 0061]. This acknowledgement of overlap in the programmes for the utility diversion works and the infrastructure works is reflected in tie’s Memorandum of Information for the Infraco procurement (6 March 2006), which noted that MUDFA works would be carried out

“in advance of and in conjunction with the implementation of the Infraco Contract. … The successful Infraco will be required by the Infraco Contract to liaise with the MUDFA contractor and co-ordinate the works stipulated under the Infraco Contract with the utilities diversion works being carried out under the MUDFA” [CEC01781572, page 0009, paragraph 2.9] (emphasis added).

5.42 While there was a recognition that work on utilities diversion would be carried out in conjunction with the Infraco works and that there would be a need for liaison between the two contractors to co-ordinate the Infraco and MUDFA works, that did not mean that any overlap in the respective programmes of the two contractors would result in any conflict at any particular site or that access to Infraco sites would be delayed. It simply meant that there would be a need for co-operation in the programming and management of the Infraco and MUDFA works to ensure that no such conflicts or delays occurred. In my view, the procurement strategy, at least up to the stage of the draft Outline Business Case of March 2006, was not explicit about the extent to which utilities would be diverted, and design completed, by the time of the Infraco contract being awarded. However, it was implicit in the procurement strategy’s objective of negotiating a fixed price for that contract that the degree of completion would be sufficient for the infrastructure bidders to fix their prices to a substantial extent. The MUDFA award was scheduled for early June 2006. On appointment, the MUDFA contractor was to undertake a series of pre-construction activities including working with the SDS contractors to optimise the design of the utilities, minimise disruption to the City of Edinburgh and maximise construction productivity [CEC00380898, page 0010]. tie noted that four contractors pre-qualified for MUDFA and would be invited to submit tenders [ibid, page 0062, paragraph 6.8.3].

5.43 tie continued to acknowledge a risk that either the preferred infrastructure contractor or PB would be reluctant to implement the novation of the design contract. That might lead to tie deciding not to insist upon novation. In terms of the SDS contract tie was entitled, but not obliged, to require that contract to be novated to Infraco. The refusal of either party to agree to tie’s request for novation might also lead to the termination of the design contract or to the cessation of negotiations with the preferred infrastructure bidder in favour of the reserve bidder. However, tie’s view appears to have been that a failure of novation was unlikely, and that Infraco would benefit from novation – something that tie expected to be reflected in their bids [ibid, page 0051]. The risk of failed novation was one that tie believed that it could manage [ibid, page 0059].

5.44 In response to the market feedback, tie modified its strategy by deciding to identify the preferred vehicle supplier at an earlier stage of the procurement process. This meant that infrastructure bidders would know the identity of the preferred vehicle supplier, and could therefore tailor their bids accordingly. In this way, the risk of the tram contracts not being novated to the infrastructure contractor would be mitigated. tie continued to acknowledge a risk that novation of the tram vehicle contract might fail, with the possibility that either the Infraco contract or the Tramco contract or both would have to be re-tendered, but it considered that to be unlikely [ibid, page 0052]. Seven tram vehicle suppliers submitted returns to a pre-qualification questionnaire, of which four were selected for the invitation to tender process. tie aimed to identify a preferred bidder before the end of December 2006 [ibid, pages 0010 and 0067].

5.45 Following a comparison exercise, tie concluded that its procurement strategy could deliver similar contractual risk transfer to that of a PPP structure as well as potentially better value for money. This better value would be achieved through tie managing down the pre-construction risks instead of paying a private-sector contractor to take them on. tie’s strategy was similar to the PPP option in terms of the proposed risk transfer and risk management approaches.

“Both options would be based on a planned series of advanced contracts which directly reflect the lessons learned from previous (largely PFI) light rail projects, with the aim ultimately of facilitating a fixed price contract for the infrastructure, under which the private sector Infraco was responsible for the key risks associated with that infrastructure (construction, system integration, maintenance and continuing system availability) but which mitigated wholly or substantially the pre-construction risks which often carry large price premiums under PPP structures e.g. design, planning, land purchase/access and utilities diversions.” [ibid, page 0072, paragraph 7.3.4.]

5.46 tie acknowledged that the management of these interlocking contracts, to establish the best possible “platform” for a fixed-price Infraco contract, would be a challenge (albeit one that would apply equally under a PPP structure) [ibid, page 0072, paragraph 7.3.5]. tie’s case for its “enhanced” conventional procurement strategy included its assertions that it had

“assembled the means to carry out its own ‘due diligence’ on all aspects of the project ahead of the Infraco contract, in effect, simulating the rigorous analysis of contractual and management arrangements that would normally be undertaken by the senior lenders under a PPP approach” [ibid, page 0072, paragraph 7.3.6].

5.47 tie also acknowledged that success would be dependent on the quality of tie’s team and its ability to implement the procurement strategy and actively manage the risks inherent in that strategy. tie was confident that, as a direct response to that challenge, it had assembled the necessary level of expertise and experience

“within the tram project team and the group of specialist advisors who form part of that team (including Transdev as the future operator) is a direct response to this challenge” [ibid, page 0072, paragraph 7.3.5].

5.48 While tie recognised the importance of its team being appropriately qualified for the task, it also considered that an in-house management team with the necessary expertise, supported by specialist advisers, was the required solution. That is apparent from the above quotation and the following passage in the Outline Business Case.

“The formation of a highly competent and experienced team is a necessary prerequisite for the successful execution of the advanced conventional procurement strategy being followed by tie and the active management of risk that entails … The conclusion in management terms is that an in-house management team is the correct way to resource this complex project offering the advantages of knowledge retention, flexibility and control. This facet is reinforced by Ian Kendall’s ability to source experienced and skilled managers with expertise which is specifically relevant to the project.” [ibid, page 0093, paragraph 9.2.1.]

5.49 tie‘s specialist advisers included Scott Wilson Railways Limited, appointed by tie as the provider of Technical Support Services for the Tram project on 25 July 2005, as a result of which tie could obtain key resources in areas including utilities and planning approvals [CEC01651410; CEC01580826, Parts 1–2]. tie’s emphasis on the advantages of knowledge retention, flexibility and control arising from an in-house management team should be viewed in the context that most of the management team, including Mr Kendall, were retained on a consultancy basis.

5.50 As part of its procurement strategy tie claimed to have

“[d]eveloped a series of value for money risk transfer mechanisms to be implemented for the Vehicle and Infrastructure contracts which will, in tie’s view, be effective in incentivising the private sector in a manner similar to PFI whilst minimising the funding costs and risk premia which might be borne by the public sector” [CEC00380898, page 0071, paragraph 7.1].

2006: Removal of Mr Kendall and delays in procurement

5.51 In May 2006, Mr Kendall was removed summarily as tie’s TPD and replaced, on an interim basis, by Mr Harper. According to Mr Kendall, Mr Harper had no prior experience of a Tram project [TRI00000136, pages 0207–0208]. He did, however, have experience of project management. On a visit to Edinburgh prior to his appointment, Mr Harper met the principal members of the project team and formed the impression that they had “lost direction and focus, and were extremely demotivated”. Mr Bissett appeared to him to be the only person who understood how the structure and project worked [TRI00000043_C, pages 0002–0003, paragraphs 6 and 8]. He considered that tie lacked commercial and procurement expertise and on that basis brought in Mr Gilbert, whom he described as “the key project team member on procurement” [ibid, page 0008, paragraph 23]. From this time on, those primarily responsible for implementing the procurement strategy were different from those who had devised it. Mr Harper thought the strategy too complicated and that it was overly optimistic to consider that it would all work as planned. He did not consider it practicable to revisit the strategy at that stage, although he did ask Mr Gilbert to lead a review of it [ibid, pages 0003–0005 and 0014, paragraphs 9, 15, 46 and 49].

5.52 Around this time, delays had begun to affect the implementation of tie’s procurement strategy. A readiness review undertaken in May 2006 recommended re-phasing the procurement programme in response to delays [CEC01881455; CEC01793454, page 0008]. A paper by Mr Gilbert to the TPB dated 18 September 2006 noted that slippage in the design programme presented problems for implementing the procurement strategy. Since endorsement of the procurement strategy through acceptance of the Outline Business Case, the following had occurred:

• delivery by the SDS contractors of the assured preliminary design had slipped by three months;
• issue of the Tramco tender had slipped by three months;
• award of the MUDFA contract had been delayed by four months; and
• the Infraco tender period had been reduced by one month to enable the FBC to be informed by returned tenders.

5.53 The paper also noted that, since utilities diversions could not start before the draft FBC had been approved, the risk of delay to the Infraco works had increased [CEC01688881, page 0045; TRI00000038_C, pages 0080–0081, paragraph 214]. Mr Gilbert explained to the Inquiry that tie’s objective was to find an approach
under which the principles of the procurement strategy could still be met, but on an accelerated timescale [ibid]. The mitigations proposed in his paper included:

• agreeing with the Infraco bidders the price-critical information that they needed in relation to critical design, performance and consents;
• agreeing a priority design programme with the SDS provider to deliver that price-critical information;
• developing a plan for the phased delivery of consents;
• conducting the bid process as an ongoing negotiation;
• undertaking advanced works prior to award of the Infraco contract, to take pressure off the critical path in the early stages of the Infraco works; and
• seeking CEC’s agreement to a limited mobilisation of Infraco in advance of full approval to award the contract [CEC01688881, page 0046].

5.54 Mr Harper’s recollection was that tie “generally addressed the key issues [mentioned above] by reprogramming and by getting the SDS to commit to doing things differently”. He was dissatisfied with the SDS provider, but stated that it was difficult to take steps to enforce their obligations without bringing all the risk back to tie [TRI00000043_C, pages 0015–0016, paragraphs 51–52].

5.55 A further readiness review in September 2006 described the timescales in tie’s procurement plan as “tight but deliverable” [CEC01629382, page 0008, paragraph 10]. The review team noted that tie intended to issue the Infraco invitation to negotiate (“ITN”) documentation to pre-qualified bidders on 3 October 2006, with the principal sections being issued initially and further design information issued at the end of October. The review team noted a number of risks that would arise if the documentation was issued either too early or too late. Those arising from the documentation being issued too early included the documentation not being of a high enough quality to achieve robust pricing, due to ill-defined requirements, and an extended negotiating period. Those of the documentation being issued too late included withdrawal of bidders, or at least additional costs, due to uncertainty. The review team concluded that the impact of risks due to a late release outweighed those of an early release, and that the Infraco ITN documentation should not be delayed. The risks of an early release could be mitigated by, among other things, the early adoption of the recommendation that tie listen to the Infraco bidders’ concerns and address them in later information releases. The review team reported again on 22 November 2006, after a follow-up review. Among other things, it noted that Mr Harper, the TPD who had replaced Mr Kendall on an interim basis, was to leave towards the end of the year. Although it noted that the search for a replacement was under way, the review team was “concerned at the loss of continuity and possible loss of momentum”
and advised that his prompt replacement was essential at this critical stage. The review team considered that there should be a review of the levels of experienced negotiating resource within tie to maximise the chances of delivering the integrated contract structure within the planned timetable at best value [CEC01791014, page 0006].

5.56 As mentioned in Chapter 8 (Utilities), tie appointed Alfred McAlpine Infrastructure Services Limited as the MUDFA contractor in October 2006, with the first stage pre-construction works to be carried out between October and December 2006, and the second stage between 2 March 2007 and 27 June 2008 [CAR00005833].

Draft Final Business Case – November 2006

5.57 By the time that tie produced its draft FBC in November 2006, it was part-way through the implementation of its procurement strategy: MUDFA had been let and tenders for the Tramco contract were under evaluation. The SDS contractor had submitted preliminary designs in July 2006, from which cost estimates had been built up [CEC01821403, page 0155, paragraph 10.40]. tie and its advisers had completed a detailed review of the cost estimate of £592 million for line 1, subdivided into £500 million for phase 1a and £92 million for phase 1b. tie reported a “relatively high” degree of confidence in these cost estimates [ibid, pages 0013–0014, paragraphs 1.56–1.57]. As will be noted in Chapter 9 on procurement up to preferred bidder, the unexpected outcome of the publication of the Official Journal of the European Union notices was the receipt by tie of only three responses to its pre-qualification questionnaire for the Infraco contract tender. Amec subsequently withdrew but Mr Harper considered that two bidders were sufficient to provide good competition between bidders [TRI00000043_C, page 0026, paragraph 90]. Tenders for the Infraco contract were due to be returned by 9 January 2007 and would thereafter be the subject of negotiation [CEC01794929; CEC02083466, pages 0005 and 0007, paragraphs 3.22 and 4.2].

5.58 As was noted in paragraph 5.30 above, in the draft Interim Outline Business Case in May 2005, tie had anticipated that the detailed design would be 60–70 per cent complete when the Infraco contract was signed. It made broadly consistent remarks in March 2006, in the draft Outline Business Case. Departing from that, tie now said that the detailed design was expected to be 100 per cent complete by the date of signature of the Infraco contract [CEC01821403, page 0085, paragraph 7.53]. This change is surprising, given the delays affecting completion of the design by that stage. It is not clear why the change was made. In this regard the draft FBC said that

“[t]he creation of the Infraco contract as a lump sum contract transfers the pricing risk to the private sector. Finalisation of the Infraco contract price on the basis of SDS Detailed Design significantly reduces their scope and performance risk pricing premium that would otherwise be necessary under conventional design and construct or PFI approaches” [ibid, page 0098, paragraph 7.125(b)] (emphasis added).

5.59 A layperson might infer that a 100 per cent complete design would eradicate all design-related risk pricing from contractors’ prices. However, Mr Crosse (Mr Harper’s replacement as TPD) told the Inquiry that it would not [PHT00000021, page 37–46]. The basis for his assertion that no design is ever 100 per cent complete seemed to be that changes will be required to suit the contractor’s requirements and construction methods. However, where changes to design were “tailored to suit contractors’ technologies and approach” it appears to me that they would be at the risk of the contractor and would not have cost implications for the client who had provided a design that was described as 100 per cent complete and did not require changes at the insistence of the client. In any event, Mr Crosse said that tie’s management did not actually believe that the design would be 100 per cent complete [ibid, page 40]. He was unaware whether that view was shared with CEC. In these circumstances it is difficult to understand the change in the business case to say that design would be 100 per cent complete by the date of award of the Infraco contract. That is particularly so where it appears in a document that might be used to support the case for public funding. Such a statement would affect the assessment of risk retained by the public sector, which, in turn, could undoubtedly influence CEC in its decision on whether to proceed with the project. Having said that, the version of the final business case in October 2007 (“FBCv1”) specifically acknowledged that design would not be 100 per cent complete. As will be mentioned in later chapters, CEC had approved the FBC before deciding to proceed with the Infraco contract.

5.60 The procurement timetable was still under pressure. tie noted that its programme for delivery of phase 1 of the Tram project was “based on the assumption of ‘right first time and on-time’ delivery of activities with very little float within the programme”. This required close attention by tie:

“[t]he criticality of much of the design activities mean [sic] the need for on-time delivery is particularly true for SDS design work and the project team are currently actively pursuing improved performance in this area and critically reviewing these elements of the programme. Key risks are delivery of design for construction for the Utility Diversion works, traffic modelling and junction designs which form the basis of the Traffic Regulation Order process. Also essential is the timely delivery of Detailed Design for structures to ensure these key risk items in the Infraco contract can be de-risked and priced competitively.” [CEC01821403, page 0164, paragraphs 11.3–11.4.]

5.61 Concerns were expressed about how realistic the programme was. In its comments on the draft FBC in March 2007, officials in Transport Scotland observed, in relation to there being only little float in the programme, that

“[i]t appears that the programme provided describes only a ‘Best Case’ scenario with no real feasible mitigation of delay or additional time for any secondary works required” [TRS00004145, page 0009].

5.62 The programme durations for design, procurement, approvals and commissioning, in the view of officials in Transport Scotland, looked “very compressed. The lack of float or mitigation opportunities and ‘right first time’ planning would appear optimistic.” [ibid, page 0010.]

In response to these comments tie stated that it was in the process of reviewing its programme [TRS00004274].

5.63 On 21 December 2006, CEC approved tie’s draft FBC, and the continued negotiation of the Infraco contract and the Tramco contract [CEC02083464]. The report on which its approval was based noted that to maintain control over the capital cost of the project, enabling works, including utility diversions, should be authorised to proceed on a timetable that would not disrupt the main infrastructure programme, and that negotiations with the “bidders should continue with a focus on achieving a high proportion of fixed cost” [CEC02083466, page 0012, paragraph 4.32].

Final Business Case – October and December 2007

5.64 In October 2007, tie produced a first version of its FBC (“FBCv1”) [CEC01649235, Parts 1–11]. On 25 October 2007, CEC approved FBCv1 and endorsed tie’s procurement process. In doing so, CEC noted that the Auditor General for Scotland had reported that procedures were in place actively to manage risks associated with the Tram project, and that tie had “implemented a clear procurement strategy aimed at minimising risk and delivering successful project outcomes” [CEC02083538; CEC02083535, pages 0005–0007]. It also noted that the procurement strategy had previously been endorsed by the Office of Government Commerce (“OGC”) reviews.

5.65 FBCv1 contained a forecast of the capital cost of the project, said to have been based on firm rates and prices from the Infraco and Tramco bidders [CEC01649235, Part 1, page 0007, paragraph 1.4]. tie forecast the cost of phase 1a (Airport to Newhaven) at £498 million and phase 1b (Roseburn to Granton) at £87 million, reducing to £82 million if both phases were built concurrently [ibid, Part 1, page 0016, paragraph 1.65]. tie expressed a high level of confidence in these estimates, and said that approximately 99.9 per cent of the costs included in them were based on the rates and prices for firm bids received for the main contracts (Infraco, Tramco, MUDFA and SDS) [ibid, Part 1, page 0016, paragraph 1.66]. tie also noted that it had a high level of confidence in its risk allowance, which it assessed statistically at 90 per cent, meaning that there was a 90 per cent chance that the actual cost would be below the risk-adjusted level. This assessment reflected:

“the evolution of design and the increasing level of certainty and confidence in the costs of Phase 1a as procurement has progressed through 2006/2007 and fixed price bids for the Infraco and Tramco contracts have been received” [ibid, Part 1, page 0017, paragraph 1.68].

5.66 In summary, tie noted that:

“the cost estimate reflects substantial external validation from the procurement process for the major contracts and contains a sensible level of risk contingency” [ibid, Part 1, page 0017, paragraph 1.72].

5.67 Throughout FBCv1 there were various references to the Infraco contract price being “fixed” – for example:

• mention in paragraph 1.68 of the receipt of fixed-price bids for the Infraco contract and the Tramco contract;
• the claim that one of the principal attributes of the procurement strategy was the “[l]ump sum price for delivery into service of the tram system” [ibid, Part 6, page 0112, paragraph 7.99];
• the reference to “the award of a single turnkey fixed price contract” as one of the key benefits of the procurement strategy [ibid, Part 6, page 0114, paragraph 7.110];
• the assertion that the “creation of the Infraco contract as a lump sum contract transfers the pricing risk to the private sector” [ibid, Part 6, page 0117, paragraph 7.126(b)]; and
• the statement that fixed prices had been agreed for phase 1a [ibid, Part 8, page 0166, paragraph 10.53].

5.68 Although negotiations on the Infraco contract were not complete, in the following passage tie claimed that its procurement strategy had achieved reductions in risk and cost:

“tie’s Procurement Strategy has resulted in it taking a greater degree of control over the process during the early ‘development’ phase, compared to what the public sector has done on other projects. This has resulted in tie progressing the overall project sufficiently in advance of seeking bids from Infraco bidders such that it was able to offer the private sector Infraco and Tramco bidders a better defined basis on which to bid and a less onerous risk allocation (and in particular reducing the extent of design and approval uncertainty at bid stage). Therefore the private sector were able to price their bids with a greater degree of accuracy and certainty than has been achieved on other projects. In this way, tie believes it has significantly reduced the cost of the overall project, having considerably de-risked certain of the elements of the project that fall to the private sector to deliver. This is shown by the minimal risk allowance included in the Infraco and Tramco bids.” [ibid, Part 5, page 0096, paragraph 7.7.]

5.69 tie’s claims about the benefits delivered by its procurement strategy included the following.

• “Delivery of preliminary design and key elements of the detailed design has resulted in a reduction in risk pricing in the Infraco tenders.”
• “Shorter period from letting Infraco contract to completion of the system – this also reduces the overheads incurred by the Infraco.”
• “Early design of utilities has enabled commencement and completion before commencement of Infraco works, which again reduces overall programme duration.”
• “Reduction in risks associated with utilities diversion ‒– early completion of utilities diversions will result in a reduced likelihood that utilities works will disrupt progress of the main infrastructure works.” [ibid, Part 5, page 0106, paragraph 7.60.]

5.70 In view of slower than expected delivery of designs, tie acknowledged the necessity of its close management of design activities and of giving priority to those designs required to ensure that progress on site was maintained “to meet the execution programmes” [ibid, Part 9, page 0188, paragraph 12.4]. In the following passage, tie recognised the consequences associated with the failure to deliver designs on time.

“Key risks are delivery of design for construction for the utility diversion works and traffic modelling and junction designs, which form the basis of the TRO process. Also essential is the timely delivery of Detailed Design for structures to allow these key items in the Infraco contract can [sic] be de-risked and priced competitively …” [ibid, Part 9, page 0188, paragraph 12.5.]

5.71 tie acknowledged that the detailed design would not be 100 per cent complete by the time that the Infraco contract was signed. Its response to this was as follows:

“by identifying key risk areas and prioritising SDS activities, tie is completing several key elements of the Detailed Design in time to inform the Infraco bids on price-critical items. This has enabled the Infraco bidders to firm up their bids based on the emerging Detailed Design and thereby reduce the provisional scope allowances and design risk allowances that they would otherwise have included.” [ibid, Part 5, page 0104, paragraph 7.53.]

5.72 The key benefits claimed by tie for the MUDFA strategy included increased confidence in the overall programme (by taking the design, negotiation and implementation of utility diversions out of the project’s critical path). It would also significantly reduce price uncertainty for the Infraco contract: tie’s approach avoided the need for Infraco to apply to its tender price risk premiums normally associated with contracts where contractors had responsibility for undertaking the above tasks. Although, under tie’s strategy, the public sector retained the risks associated with these tasks and had to manage them, tie observed that:

“the cost of the risk to tie under this approach is considerably lower than would be the case had Infraco managed the utility diversions directly … because Infraco would have found it difficult to quantify the risks in advance of bidding, and the knock-on effects of those unquantifiable risks to Infraco’s programme would be considerable” [ibid, Part 5, page 0109, paragraphs 7.80–7.81].

5.73 In FBCv1, tie summarised the public sector’s risk exposure, which it considered involved “significant, but diminishing and manageable, risks during the remaining period of scheme development” [ibid, Part 9, page 0180, paragraph 11.58]. While the strategy of engaging SDS and MUDFA contractors directly had reduced risk, tie acknowledged that “the major responsibility for identifying and managing potential risks during this period remained with the project team and their advisors” [ibid].

5.74 In December 2007, tie produced the second version of the FBC (“FBCv2”), in largely identical terms to FBCv1 [CEC01395434, Parts 1‒11].

Discussion

5.75 When tie had to devise a procurement strategy for the Tram project, it did not have the option of using a tried-and-tested procurement strategy. There was no generally accepted procurement method for tram or light rail projects, and the approaches taken on previous schemes had failed to produce a satisfactory outcome [Mr Fitchie TRI00000102_C, page 0053, paragraph 4.69]. Accordingly, the only realistic option for tie was to devise its own procurement strategy afresh. This was a challenge that tie took seriously. It engaged professional advisers, looked in detail at different procurement strategy options, and selected the one that it considered to be most suitable for the Edinburgh project. tie’s objective was to procure the project in a cost-effective way, through the appropriate management of risk: allocating risks to the parties best able to manage them; and, where possible, managing risks down prior to the award of contracts, to minimise the premiums charged by the contractors for bearing the risks to be transferred to them.

5.76 In my view, tie’s strategy was coherent and logical, at least theoretically. Uncertainty and the associated risk inherent in any project have a direct effect on the prices charged by contractors for building it because of the addition of risk premiums. Accordingly, a strategy of reducing uncertainty and its associated risk should result in a tender price that is exclusive of any additional premiums for the element of uncertainty that has been removed. Various features of tie’s procurement strategy were intended to reduce elements of uncertainty and their associated risk. These included:

• producing a consented design in advance of procuring the Infraco contract, so that the consented design was available as a basis for fully informed pricing by the infrastructure bidders.
• diverting utilities in advance of the infrastructure works, to remove the uncertainty that undiverted utilities would otherwise cause for the infrastructure contractor in negotiating with the various utility companies whose apparatus had to be diverted and in programming their works. This approach also provided potential infrastructure contractors with an informed basis for pricing their bids and reduced the need for risk premiums.
• separating operational activities from the Infraco contract, so that the infrastructure contractor could focus on construction costs and risks with which it was familiar and removing any need for high risk premiums to protect them against unfamiliar risks.

5.77 tie took the prudent step of consulting the market about its proposed procurement strategy, and received essentially positive feedback. Where there was criticism, tie modified its strategy to address it (eg in deciding to identify the preferred tram vehicle supplier at an earlier stage, so that the Infraco contract bidders could assess any risks relating to the particular vehicle supplier). There was no evidence before the Inquiry that indicated that there was any significant criticism of, far less opposition to, tie’s overall procurement strategy at the time that it was proposed. Officials in the Scottish Executive and officials in Transport Scotland, as the case may be, endorsed it [TRS00004145, page 0005].

5.78 While aspects of the strategy were not in themselves uncommon (such as the advanced diversion of utilities or the novation of design contracts to construction contractors [Dr Enenkel TRI00000161_C, page 0009]), the overall strategy was new, at least in the context of UK light rail schemes. It was untested in that context. There was no experience of how it would actually work in practice and it was in the practical implementation of the strategy that problems emerged. Mr Kendall, who devised much of the strategy but was removed from tie before it was fully implemented, emphasised that its implementation required appropriate experience and expertise. He said, colourfully,

“[t]o deliver it required intimate knowledge and expertise. You couldn’t be a mug punter and never done this before, come along and expect this to be done optimally, you couldn’t and I don’t know that expertise was there after I left.” [TRI00000136, page 0144.]

5.79 tie made much of the expertise that it had built up in its project management team. The expertise was needed because, as tie was plainly aware, the strategy left it managing the design contract prior to its novation and the utility diversion contract, as well as the risks associated with them. tie acknowledged that the management of the various contracts would be “a challenge” [CEC00380898, page 0072, paragraph 7.3.5].

5.80 Perhaps unsurprisingly, those who came after Mr Kendall, and who therefore inherited the strategy, tended to be more critical of the strategy itself than of the expertise of those who implemented it. Mr Harper, who succeeded Mr Kendall as TPD attie, said that

“[i]n a perfect world it was probably an effective and sophisticated strategy for procurement, but we live in the real world. It was always based on everything working to the plan but people do things differently and forget the plan.” [TRI00000043_C, page 0006, paragraph 18.]

5.81 He considered that it was “overly optimistic that it would all work in line with the strategy on what is a complex tram scheme. It was too complicated an approach for me” [ibid, page 0013, paragraph 45]. His own preference was for a procurement model in which the same entity was responsible for design, construction and utility diversions, but he recognised that this approach would come with a large risk premium [PHT00000021, page 0046]. That preference is all very well, but it must be kept in mind that the main objective of the strategy was to reduce such risk premiums to make the project affordable. Mr Gilbert, who was recruited by Mr Crosse to run the procurement, and who most directly faced the challenge of implementing the procurement strategy, said

“[t]he core of the strategy was to avoid or ameliorate risk by managing and designing it out by driving certainty at each phase of the pre-construction phase. That meant that to be successful the strategy was contingent upon certain things happening in complete form in a specific order and schedule. This is perhaps its flaw and area of weakness in that it is inflexible to changing circumstances and external events. Such inflexibility creates problems in delivering projects with a high degree of complexity. Projects of this degree of complexity require flexibility to enable the emerging issues to be managed effectively … The inflexibility in the strategy also meant that the programme was very sensitive to design progress and associated project and external design approvals.” [TRI00000038_C, page 0140, paragraph 346.]

5.82 He concluded:

“I would not pursue this strategy on future projects. In theory, it was a great strategy but it did not take much cognisance of the real complexity of the project.” [ibid, page 0142, paragraph 351, see also page 0009, paragraph 21.]

5.83 I accept the description of the strategy as having been “idealised”. (See also paragraphs 6.59 and 10.8 below, where I discuss this description in different contexts). To put it another way, it was good in theory. In my view, the “idealism” of the strategy is best demonstrated by the aspiration (reported in the draft FBC of November 2006 [CEC01821403]) that the detailed design would be 100 per cent complete by the time that the Infraco contract was signed; and that the utilities would be diverted sufficiently in advance that conflict with the Infraco works could be avoided. These were laudable aspirations, but it could not be assumed that they would be achieved. That is particularly so given the competing, perceived time pressures to award the Infraco contract and proceed with its works to avoid either an increase in costs due to inflation or loss of interest in the project from the Infraco bidders. So far as the aspirations were not achieved, it was obvious that the risk that tie had aimed to manage out of the project would still exist. That was risk that tie would either have to carry itself or persuade its contractors to accept (almost certainly in return for an additional risk premium in the contract price).

5.84 To the extent that tie retained the risk, it would require accurately to identify and quantify that if it were to give CEC a properly informed basis for their decision on whether to proceed with the project. It is my view, explained in Chapter 21 (Risk and Optimism Bias), that tie failed to do so, and consequently failed to give CEC such a basis for their decision. Witnesses acknowledged this. For example, Mr Bissett acknowledged that the delay in production of design, and the overlapping of design and utility works with the construction period, created difficulties for the project [TRI00000025_C, page 0007, paragraph 19]. That resulted in

“a lot of effort put into designing an additional process to complete the design that would protect the public sector from the difficulties that the overlap could otherwise create” [ibid, page 0058, paragraph 160].

5.85 He noted that the overlap of design work and utilities work with the construction period “proved to be difficult in practice”, and expressed the opinion that “[t]he implications and risks of failure to avoid an overlap were probably under-estimated, in terms of cost and programme delay”. Nonetheless, Mr Bissett was still of the view that “the right strategy was in place” [ibid, page 0086, paragraph 248]. In contrast, Mr Gilbert considered the procurement strategy to have been flawed, because it did not take into account “the complexity of the scheme”, particularly as it involved carrying out major construction work on busy roads in a congested city centre and “the number of stakeholders that needed to be consulted and whose approval was required” [TRI00000038_C, page 0004, paragraph 9].

5.86 I prefer the evidence of Mr Bissett to that of Mr Gilbert in this regard. I do not consider that there was any inherent logical flaw in the strategy that tie developed. Nor do I consider that tie should be criticised for choosing it, particularly given the exhortations of the NAO that new approaches to procurement were needed. The strategy that tie developed was creative and had the potential to save considerable cost, allowing the project to be delivered within the constraints of the budget. The difficulty did not lie in the strategy but was caused by the failure to implement it. In my view, tie was much too optimistic about the prospects of its procurement strategy achieving its objectives on the timescales that it had to meet, and therefore about the extent to which risk would be either managed out of the project or transferred to the private sector. That over-optimism continued in the face of delays in production of the design and utility diversions, and influenced tie’s view that it was capable of managing the challenges arising from those delays.

5.87 Mr Gilbert referred to tools, including the Procurement Route map, developed by Infrastructure UK (now the Infrastructure and Projects Authority) for understanding project complexity. He said that these

“allow a better assessment of the capabilities of both the team and supply chain, identify gaps and then adjust the strategy to deal with them. These tools allow you, in general, to work out the best approach and strategy to align market and project objectives. They also allow you to clearly understand the strengths and weaknesses of both the market and the delivery team, enabling you to address those aspects and refine your strategy from the outset.” [ibid, page 0004, paragraph 10.]

5.88 This is no doubt good advice that will be relevant to projects being planned following the development of such tools. It reflects my own view that any project using a new procurement strategy, untested in the context in which it is being used, must guard against over-optimism that it will work as intended. In all projects it would be prudent to use any available tools that test assumptions and they should be used to devise and refine the procurement strategy.

Conclusions

5.89 tie devised and implemented a procurement strategy that was novel for a tram or light rail scheme in the UK. In doing so, it followed the recommendations of the NAO and sought to avoid problems that had been encountered on previous UK light rail schemes.

5.90 tie took appropriate professional advice in developing the procurement strategy and consulted the contractor market before implementing it. The market reacted positively to the proposed strategy. To the extent that there were criticisms, these were minor and tie refined its strategy to address them.

5.91 The strategy sought to keep costs down so far as possible by managing risk out of the project and transferring remaining risks to those parties best suited to bear them. Risks of very great significance to the cost of the project were those relating to design and to the presence of utilities that would interfere with construction of the tram system. Through the early appointment of a design contractor and a utility diversion contractor, tie sought to remove, or at least substantially mitigate, those risks prior to the award of the main Infraco contract, and thereby to reduce the premiums that bidders for that contract would otherwise have included in their bid prices. The inevitable consequence of delay in completion of the design and utility diversions was that there was greater uncertainty in relation to pricing of the Infraco contract because of the re-introduction of risk, which it had been the objective of the procurement strategy to remove.

5.92 tie’s objective was to mitigate the risks sufficiently that the main infrastructure contractor would undertake the infrastructure works, and accept all design and construction risks associated with them, for a largely fixed price that would allow the project to be completed within the approved budget. That strategy required tie
to manage the design and utility diversion contracts so that sufficient progress was made both to mitigate the relevant risks prior to the award of the Infraco contract and accurately to assess, quantify and report upon any remaining risk so that CEC understood the likely costs before committing to construction. It also aimed to avoid open-ended commitments of funding by ensuring that robust contractual arrangements were in place, and the affordability of the project agreed, before any commitment was made to construct the scheme. tie recognised that this placed significant responsibility on it, but considered that it had sufficient experience and expertise to discharge it.

5.93 tie was over-optimistic in its assessment that it would be able to manage the difficulties affecting the design and utility diversions. As a consequence, it failed properly to identify or quantify the risks that those difficulties created and, in turn, failed to provide CEC with a properly informed basis for their decision on whether to proceed with the project. It was also over-optimistic about its programme, which was based upon being right first time with very little float. In particular, its assumption that its planning proposals would be right first time failed to recognise the iterative process of planning and the sensitivity of the World Heritage Site.

5.94 I do not consider that it is appropriate for me to be prescriptive about the procurement strategies to be used for future tram or light rail projects in Scotland. Each project should be considered on its own merits, and there are many factors that will influence the appropriate choice. In any given project such a choice will be a matter for those engaged on it, taking account of the particular circumstances of that project and with the benefit of appropriate professional expertise and an understanding of the risk appetite of those promoting and funding the project. It would not be sensible to inhibit commercial creativity by scheme promoters in the way in which they go about this.

5.95 Having said that, I can appropriately offer the following guidance.

• There is much to be said for the strategy that tie developed – in particular, its objectives of managing risk down prior to the award of contracts and transferring risks to the parties best able to manage them, with a view to keeping costs down.
• Project promoters should be alive to the fact that implementing a new, untested strategy is itself a risk, and should not assume in advance of its conclusion that its objectives will be achieved. Rather, it should be assumed that unexpected problems will arise.
• Project promoters should be alive to the tension between achieving the risk management objectives of a procurement strategy, and awarding contracts on a particular timescale: while there will usually be cost consequences of delay in awarding a contract, there are also likely to be cost consequences of awarding it before the risk transfer objectives have been achieved in full.
• Project promoters should be alive to the risk that, because of supervening events, they will fail fully to achieve the risk management objectives of their procurement strategy and, in that event, should be scrupulous in re-assessing accurately the risks that are present as a result.

5.96 It is essential that in any strategy that requires risks to be managed down prior to the award of major contracts, those managing the risks have the appropriate expertise and experience either to ensure that the risks are managed down to the predetermined level, or to recognise, accurately quantify and report upon any remaining risk.