SECOND DIVISION, INNER HOUSE, COURT OF SESSION
Lord Justice Clerk Lord Clarke Lady Smith	[2013] CSIH 22 CA132/10 OPINION OF THE COURT delivered by LORD CARLOWAY, the LORD JUSTICE CLERK in the reclaiming motion HEALTHCARE AT HOME LTD, Pursuers and reclaimers; against THE COMMON SERVICES AGENCY, Defenders and respondents: _______

Act: Sandison QC, Watt; Maclay, Murray and Spens LLP

Alt: Clark QC, S Smith QC; NHS Scotland Central Legal Office

21 March 2013

Background

[1] Prior to May 2010, the pursuers and reclaimers (appellants) had the benefit of a contract to provide certain cancer treatment drugs and nursing services to patients in their homes. They had been providing these drugs and services for about eight years. The contract was then put out to tender. The pursuers lost the contract to a competitor, namely BUPA Home Healthcare. The decision to award BUPA, and not the pursuers, the contract was that of the defenders and respondents, whose full name is the Common Services Agency for the Scottish Health Service (National Health Service (Scotland) Act 1978, s 10(1)). The pursuers challenged the defenders' decision, but the commercial judge rejected that challenge. The pursuers have now reclaimed (appealed) that decision.

[2] The Public Contracts (Scotland) Regulations 2006 (SSI no 1), which implement European Directive 2004/18/EC (the Public Sector Directive), create a detailed regime controlling the method by which public contracts are allocated. They provide (reg 4) that the authority issuing the contract must:

"(3) (a) treat economic operators equally and without discrimination; and

(b) act in a transparent and proportionate manner ..."

The contracting authority has to specify (reg 30) the criteria which it intends to employ in assessing the tenders and state the weightings to be applied to each criteria in determining which tender is the most economically advantageous. Following upon the decision to award a contract, the authority must (reg 32) inform all tenderers of its decision in writing and provide a "summary of the reasons" for awarding the contract to a particular tenderer. The reasons require, where practicable, to set out the relative scores obtained by the tenderers. The whole process requires to be appropriately documented.

[3] Specifically, the defenders had invited offers from persons wishing to enter into a single-supplier framework agreement with them for the compounding, dispensing and delivery of trastuzumab (known as Herceptin) to patients' homes together with nursing and support services. Herceptin is used in the treatment of breast cancer. It is administered intravenously in a saline solution. Where the patient is taking the medication at home, a nurse will attend. The medication can be stored in vials and mixed with the saline solution by the nurse at the patient's home (a 'vial service'). Alternatively, the drug can be made up into a solution, ready to administer, at a compounding centre (a 'compounded service'). As already noted, at the time of the competition, the pursuers were already contracted to provide Herceptin to NHS patients, but that was through both a vial service and a compounded service. Under the proposed framework agreement, Herceptin was only to be supplied through a compounded service. The successful party would deliver the drug in solution to the patients' homes and provide nurses to administer it.

[4] The defenders issued an "Invitation To Tender" to interested parties. The ITT served as an application form for tenderers to complete. It was in electronic format with attachments. It set out information about the framework agreement and the way in which the tendering process was to be conducted. It contained directions on how to complete the form. There were blank fields allowing text to be electronically entered and a method of allowing files containing additional information to be attached.

[5] The ITT set out five criteria against which tenders were to be assessed and detailed the relative weightings for each criterion. The scoring method was explained. For some criteria, including "Quality of Service", a maximum score of five was to be given for a "good" answer. This was said to be one which met all requirements and was both comprehensive and unambiguous. For other criteria, including "Risk and Deliverability", the defenders put the score up to six for the best type of answer. This was one which materially exceeded the requirements by providing a creative or innovative response, as well as being comprehensive and unambiguous.

[6] The ITT made it clear (para 3 "Tender Evaluation") that the tenders would be evaluated by a panel consisting of representatives from different specialisms. It stressed (para 4 "Completion and Submission Instructions) that tenderers had to answer all questions fully and accurately. In particular, it was stated (para 4.3, Submission of Tenders) that full details or specifications for the goods and services should be provided and that:

"Attachments must be attached everywhere they are relevant. Do not reference attachments in other sections as they may not be considered.

... All sections must be completed and submitted in full in order for Tender submission to be complete. ..."

As a "condition of participation" (para 7) tenderers had to show that they had an efficient and robust Quality Management System in place which ensured that procedures, from order handling through to administration of medicine to the patient, would be carried out in an efficient and responsible manner.

[7] There were certain "commodity specific" questions (para 8) which required to be answered "fully". These included the following

"8.2 Quality of Service

8.2.1 Dispensing Process

It is essential that the compounding and dispensing process for Herceptin is controlled and meets all requirements detailed in the Specification.

Please detail the process that would be used for compounding and dispensing of the Herceptin. Areas to cover in your response should include but are not limited to manufacturing facilities, clinical checks, compounding checks, communication of interventions back to prescribing centres, dispensing checks and turnaround times for changes.

...

8.3 Risk and Deliverability

8.3.1 Contract Implementation/Change in Business Volumes

... it is essential that there is a smooth implementation process for this contract to maintain standards of patient care and that any increase in patient numbers is managed efficiently and effectively.

The intention for this contract is that any new patients from any Participating Authority will immediately as from the commencement date of the Framework Agreement ... be provided with all required Goods and Services by or on behalf of the successful Framework Participant ... all current patients being transferred into the care of the successful Framework Participant within a six (6) month period, ...

With this intention in mind please detail how your company would implement this contract and manage a change in business volumes on an ongoing basis across NHS Scotland.

Areas which should be covered in the response include but are not limited to: access to stock, initial and ongoing training resources, available delivery resources, recruitment of additional staff, timescales for full implementation, ability to increase delivery capacity and range as well as compounding capacity.

...

8.3.2 Contingency Proposal

It is essential that the successful Framework Participant complies with all of its obligations and meets all of its liabilities in terms of the Framework Agreement to ensure no risk to patients.

What contingency proposals would your company have to ensure that it will comply with all of its obligations and meets all of its liabilities in terms of the Framework Agreement in the event of unforeseen circumstances?

Areas which should be covered in your response include but are not limited to:-

A major incident alert from a Participating Authority that may necessitate a very quick response and additional support for a particular incident; and/or

An internal (e.g. staffing issue) or external (e.g. utilities failure) problem within or affecting the successful Framework Participant in the continued delivery of contracted Goods and/or Services; and/or

...".

[8] By letter dated 13 May 2010, the defenders informed the pursuers that they had been unsuccessful and that BUPA were to be awarded the contract. Each tenderer's total scores were set out along with the scores for each criterion and sub-criterion. The pursuers' criticisms of the defenders' decision were ultimately limited to the scoring of three sub-criteria, as follows:

			Maximum score	Pursuers	BUPA
Quality of Service
8.2.1	Dispensing Process		8.75	7.00	8.00
...
Risk and Deliverability
8.3.1	Contact Implementation/ Change in Business Volumes		7.00	5.44	6.42
8.3.2	Contingency Proposal		6.25	3.72	5.38
...
TOTAL			100.00	89.06	90.48

The differences between the pursuers and BUPA on the relevant sub-criteria were thus just enough to enable BUPA to edge ahead in the respective total scores.

[9] The letter gave the following reasons for the scoring of the relevant sub-criteria:

"Section 8.2 QUALITY OF SERVICE

8.2.1 ― Dispensing Process

[BUPA] included detail of how interventions would be communicated back to the prescribing centre through all stages of the dispensing process, from receipt of the prescription to compounding and dispensing and the actions taken prior to an intervention being made. [The pursuers] gave an indication that clinical interventions would be communicated back to prescribing centres but did not provide detail on how this would be communicated and agreed with the Participating Authority [the relevant NHS body].

The response from [BUPA] stated that clinical checks would include checks on blood and scan results in accordance with local Participating Authority protocols. [The pursuers] did not make any reference to ensuring compliance with local Participating Authority protocols.

Section 8.3 RISK AND DELIVERABILITY

8.3.1 ― Contract Implementation/Change in Business Volumes

[BUPA] included detail of predicted nursing staff numbers required to support varying degrees of activity. Information provided included trigger points for recruitment, how the staff numbers had been calculated alongside an implementation timeline and how it will be met. [The pursuers] indicated that an increase in business volumes could be met within company resources but gave no evidence on how this would be structured and managed.

[BUPA] included an exit strategy that detailed how the process of transferring patients to an alternative supplier would be managed. [The pursuers] supplied a basic exit strategy with little detail on timescales and how this would be managed.

8.3.2 ― Contingency Proposal

[BUPA] detailed in their response that they could supply compounded Trastuzumab at all points during their contingency plan. [The pursuers] indicated that if their first line contingency for the supply of compounded Trastuzumab was insufficient then a vial service would be provided in order to maintain supply."

[10] By letter dated 20 May 2010, the pursuers' law agents wrote to the defenders with a series of detailed concerns. These included that: (i) for sub-criterion 8.2.1 (Dispensing Process), the defenders had wrongly left out of account the pursuers' answer and attachment to section 8.3.5; (ii) for sub-criterion 8.3.1 (Contract Implementation/Change in Business Volumes), the ability or willingness to predict staff numbers was not a relevant consideration; and (iii) for sub-criterion 8.3.2 (Contingency Proposal), the defenders ought to have seen the provision of a contingency vial service as a strength rather than a weakness and it was not credible that any supplier could provide a compounded service in all circumstances.

[11] By e-mail dated 27 May 2010, a solicitor with the defenders replied:

"8.2.1

Your clients' response to this section of the ITT was not very detailed with regard to communication of results of interventions back to the prescribing centre. The clinical error reporting system, and communication elements were considered to be weak and it was thought that there was a lack of evidence in key areas. With regard to incidents, the cycle of awareness of and reflection on these to minimise repetition is important. This was not considered to be addressed adequately.

Whilst turnaround times were provided, there was no information given regarding what happens when changes are made and the effect this may have on timings.

[BUPA] demonstrated a more comprehensive approach to reporting and reflecting upon interventions. Clinical checks were well defined and the reporting of interventions to participating authorities and communication on a case by case basis was detailed. Reference to the participating authority protocols was made in National Procurement's letter to your clients of 13 May 2010. This reference was to illustrate the comprehensive response given by [BUPA]. We can confirm that [BUPA's] score did not include any element attributable to the ability or willingness to carry out the said check in line with any local Participating Authority protocols. We can also confirm that your client's score would not have been any higher if its response had included any statement that it would be willing and/or able to carry out the said check in line with any local Participating Authority protocols.

[BUPA] clearly defined a replacement process for product and the turnaround time for changes was clear. Audit requirements were also highlighted providing additional quality assurance.

Your letter of 20 May 2010 highlights the response given by your clients to Section 8.3.5 of the ITT. There is no dispute that your client's response to that section was of high quality and you will note that they received the highest score possible for that section. However, the response to each section is evaluated separately and your clients' response to Section ... [8.2.1] was not considered to have the depth and quality of that of the winning supplier.

8.3.1

This section of the ITT was designed to interrogate transfer and implementation issues. One element of this concerns staffing. ... An implementation plan would be necessary in order to deliver the full requirements detailed in the ITT. The plan proposed by your clients was not considered to be comprehensive and was not detailed regarding the ability to scale up compounding capacity. The transfer of existing patients from an existing vial to compounded process is a significant change which was not detailed in the response.

Whilst training was included, insufficient detail on training packages was provided. The exit strategy was considered to lack depth and detail, and did not include timescales.

By contrast, [BUPA] provided a comprehensive and detailed response which outlined a robust implementation plan and exit strategy. A three phased implementation plan was provided which was clear and provided detail and evidence relative to ongoing and initial training, access to stock and recruitment of staff (including details of how the numbers of these had been calculated and how they would be managed). The response also referenced timescales and how they would be met, ability to increase delivery capacity, delivery range and compounding capacity. Additional information was also provided around risk management. The response focused in detail upon local facilities and staffing levels.

8.3.2

National Procurement does not consider that it has misunderstood your clients' submission. Whilst your clients provided a lot of information in response to this section of the ITT, National Procurement considered that the response lacked focus and clear evidence. Details of a back-up production facility were provided, but the scale of this operation was unclear and the capacity for production was a concern. The plan for local contingency was considered to be vague and lacking in detail. It was unclear what would happen where an incorrect delivery was made and although vials were proposed, these are outwith the scope of the Framework. No details of timescales for remedying incorrect delivery were provided. The business continuity plan template did not provide details of the level of contingency available and contained very little detail about how services to customers would be maintained.

By contrast, [BUPA] addressed multiple tiers of contingency in their response. The proposal included detailed contingency relating to compounding, addressing scale and locations. A detailed pandemic flu plan was submitted. Details of staff shortages were included and details of local action were also provided. Stock shortfall incidents were addressed. An incident reporting plan was also provided which included actions taken after investigation of any complaint or incident and linking it to changes in practice that may result following investigation. A detailed nursing contingency plan was specifically referenced.

National Procurement have confirmed that the evaluation was conducted properly and in accordance with the ITT. The award criteria as set out in the ITT were uniformly applied to all submissions".

Proceedings at First Instance

[12] The pursuers sought an order under regulation 47A(1)(b)(i) of the 2006 Regulations "setting aside" the defenders' decision to award the contract to BUPA. They claimed, in short, that the competitive tendering process had been unfair and that, in any event, the reasons given by the defenders to explain why the pursuers had been unsuccessful had been inadequate. Their complaints were many and varied. The assessment of the various criteria, and the reasons given therefor, were attacked in turn and, in due course, addressed in a similar manner by the commercial judge. At the risk of subsequent repetition, the points which were made have been summarised under five broad categories, some of which are further sub-divided. Those complaints that which were not repeated at the Summar Roll (appeal) hearing are omitted.

[13] 1. Clarity of the ITT: The pursuers maintained that the ITT did not generally have an adequate level of clarity in that the sections were unacceptably subjective and open ended. The ITT had stated that tenderers required to address the issues "including but not limited to" certain listed items. For sections 8.3.1 and 8.3.2, a top score of six had been given to BUPA for answers that were "creative and innovative". The pursuers' responses for section 8.2.1 (Dispensing Process) were said to be poorer than those of BUPA on the topics of: (i) "clinical error reporting"; (ii) how the pursuers would learn from reported errors; (iii) "audit requirements"; and (iv) turnaround times for changes. However, the ITT had not given the pursuers adequate notice that these topics would be considered under this section. The pursuers had received a lower mark than BUPA for section 8.3.1 (Contract Implementation/Change in Business Volumes) regarding general training. However, the pursuers had understood that section to be confined to training required at the commencement of the contract or due to changes in business volumes. BUPA had been marked comparatively favourably for section 8.3.2 (Contingency Proposal) in light of its incident reporting plan. The ITT had not made it clear that this required to be addressed in this section.

[15] The commercial judge noted the jurisprudence of the European Court of Justice in SIAC Construction v County Council of Mayo [2001] ECR I-7725; Concordia Bus Finland Oy v Helsingen Kapupunki [2002] ECR I-7213; AT EAC Srl v ACTV Venezia [2005] ECR I-10109; Emm G Lianakis v Dimos Alexandoupolis [2008] ECR I-251; Evropaїki Dynamiki v European Maritime Safety Agency [2010] EUECJ T-70/05 (Opinion, paras [11]-[14]) and contrasted the approach taken in two decisions in the United Kingdom (Federal Security Services Ltd v Northern Ireland Court Service [2009] NIQB 15 and Mears v Leeds City Council [2011] EWHC 1031 (TCC) when applying that jurisprudence (paras [23]-[25]). He considered that the court ought to ask whether the matter, which was said not to have been encompassed in the ITT, would have been "reasonably foreseeable" to a reasonably well-informed and normally diligent tenderer as encompassed by the particular criterion or sub-criterion (paras [26]-[27]).

[16] The commercial judge was of the view that the "not limited to" formula was unexceptionable and that it would invite as comprehensive an answer as possible (para [61]). Allowing top marks for responses that were creative did not involve the introduction of a subjective element. The judge considered that section 8.2.1 (Dispensing Process) had been sufficiently clear. The defenders had been asking about checks on the quality of work in compounding and dispensing, not only at the time of that compounding and dispensing but also at the various later stages up until the administration of the drug to the patient. It was thus relevant to know how a tenderer proposed to use reports of errors to avoid repetition (para [61]). Regarding "audit requirements", the defenders had simply meant that BUPA had shown that they had a means of checking that an action to correct a mistake had been taken (para [62]). The judge held that it was "reasonably foreseeable" that such a check would fall within the response to that section (para [62]). He took a similar view on "turnaround times". He considered that it was reasonably foreseeable that training of staff and a plan for incident reporting were relevant responses to sections 8.3.1 and 2 (paras [ 74} and [79]) The detail presented was a matter for the tenderer's judgment (para [74]).

[17] 2. Staffing Levels: The pursuers' complaint was that a better mark had been given to BUPA under section 8.3.1 (Contract Implementation/Change in Business Volumes) on staff levels. Consideration of staff levels was irrelevant to whether the tender was the most economically advantageous. Properly understood, staff provision concerned whether a tenderer was capable of meeting the defenders' requirements. It was a qualitative selection criterion and not an award criterion.

[18] The commercial judge accepted that the defenders' concern had not been about whether the pursuers had the ability to perform as such, but the lack of detail from them on how they would perform (para [70]). He observed that, in contrast to the pursuers' tender, BUPA had been clear on how to increase staff levels in a way which would not harm the NHS (para [70]).

[19] 3. "Silo" Method: This was a term applied by the pursuers to the system employed by the defenders whereby, rather than the scorers reading over the entire contents of each tender and determining what material might be relevant to each criteria or sub-criteria, the tenders had been divided up into different sections and each section had been given to a different person to score. Although the commercial judge had recorded a submission that this method was "neither transparent nor proportionate" (para [68]), the criticism before him, in effect, concerned a lack of notice. This was that it had not been made clear that this "silo" method of marking would be employed.

[20] The commercial judge held (para [60]) that it was implicit in the ITT that the "silo" method would be employed. Tenderers had been told to complete the tenders fully (sections 4.1, 4.3) and to attach documents everywhere that they were relevant and not to "reference" attachments to other sections of the ITT, as they might not be considered (section 4.3).

[21] 4. Manifest Error re Vial Service: The pursuers complained that the defenders' scorers had fundamentally misunderstood their offer. The pursuers had explained in section 8.3.2 (Contingency Proposal) that, if both their primary and alternative supplies of compounded Herceptin failed, a replacement vial service would be provided. However, one member of the tender assessment panel thought this was being offered as an option when there had been an incorrect delivery.

[22] The commercial judge noted that one scorer's evidence had been that the vial service was being proposed as an option where there had been an incorrect delivery. However, he found that this was not reflected in the evidence of the other scorers, which he preferred.

[23] 5. Clarity of Reasoning: The pursuers argued that the defenders' reasons in their letter and e-mail had been inadequate and contradictory. In the e-mail, the defenders had referred to a "lack of evidence in key areas" in section 8.2.1 (Dispensing Process), but these areas had not been specified. The letter had said that BUPA's adherence to "Participating Authority protocols" had been a relevant matter, yet the e-mail had said that this had made no difference to the scores. The letter had referred to the "turnaround time for changes", but these changes had not been specified. The evidence had contradicted the statement in the e-mail (re section 8.3.1 (Contract Implementation/Change in Business Volumes)), that the pursuers had received lower marks concerning the transfer from an existing vial to compounded process.

[24] The commercial judge found that the reasons given by the defenders were adequate. The quality of the reasons could not be assessed "by examining individual phrases out of their context" (para [92]). The reference to a "lack of evidence in key areas" in the e-mail ought to have been understood in the context of the e-mail's first paragraph which addressed the quality of the system advanced for detecting errors. The reference to "Participating Authority protocols" implied simply that the pursuers had been marked down for not mentioning their willingness to adhere to them. The e-mail had explained that willingness to adhere to protocols had not affected the scores. The e-mail qualified the letter rather than contradicting it (para [94]). The reference to "turnaround time for changes" was clear in the context of a sub-criterion addressing the quality of the compounding and dispensing processes (para [95]).

[25] The commercial judge held that there had been no variance between the evidence of the scorers and the reasons given in the e-mail regarding section 8.3.1. The reference in the e-mail was a record of a point which one of the scorers had considered to be a significant component in the lack of detail which had caused each scorer to assess BUPA's response to be better than that of the pursuers (para [99]).

Submissions

PURSUERS

[26] The pursuers essentially renewed the complaints made to the commercial judge. There were also two new contentions. First, the pursuers submitted that the "silo" method had been applied in a flawed manner and ought not to have been employed at all. In any event, the way the offers were assessed by individual scorers had, in practice, resulted in unfairness. Secondly, in relation to section 8.3.1 (Contract Implementation/Change in business volumes), the criticism that the pursuers' implementation plan had not been sufficiently comprehensive and detailed had been left undeveloped.

[27] 1. Clarity of the ITT: The commercial judge had misdirected himself regarding the appropriate standard of clarity. The principle of equal treatment implied an obligation of transparency. The award criteria had to be formulated in such a way as to allow all reasonably informed and diligent tenderers to interpret them in the same way (SIAC Construction v County Council of Mayo (supra), paras 41-42). This test had been recast, after the commercial judge's decision, to mean that the formulation had to be such as allowed all tenderers exercising ordinary care to know the "exact scope" of the criteria and thus to interpret them in the same way (Commission v Netherlands [2012] 3 CMLR 11, paras 66, 88 and 109). The implications of this test were that, if views might differ as to the meaning of a criterion between reasonably well-informed and normally diligent tenderers exercising ordinary care, there was a lack of clarity (Clinton v Department for Employment and Learning [2012] NICA 48, also decided after the commercial judge's decision).

[28] The pursuers had produced evidence from members of their staff as to what his or her understanding of each of the impugned parts of the ITT had been. Evidence was led on what a reasonably well-informed tenderer might understand from the ITT. None of that evidence had been rejected by the commercial judge. He had not stigmatised the interpretations of the ITT as unreasonable or as unrepresentative of what the hypothetical reasonably well-informed tenderer exercising ordinary care might have concluded.

[29] The commercial judge's approach had tolerated a range of legitimate views amongst reasonable tenderers. In effect, he had asked whether the interpretation of the ITT adopted by the defenders might have been shared by the (or a) reasonably well-informed tenderer. What he should have asked was whether all reasonably well-informed tenderers would have interpreted the ITT in that way. It followed that, unless the pursuers' interpretation, adopted in its responses to the ITT, could be said to be unreasonable, the pursuers ought to succeed.

[30] The "not limited to" formula in the ITT and the provision for giving extra credit for creative responses had allowed the defenders to expand the terms of the relevant sections of the ITT in a subjective and open-ended manner. This had given the defenders an unlawful "unrestricted freedom of choice" as to whom to select (see Gebroeders Beentjes v Netherlands [1988] ECR 4635, para 26).

[31] The commercial judge may have been correct (at para [61]) to say, in relation to "clinical error reporting" (section 8.2.1 (Dispensing Process)), that this was asking about checks on the quality of the tenderer's work right up until the administration of the drug. However, it had not been clear to all tenderers that this information ought to have been included again at section 8.2.1, given that earlier sections had asked for information on quality management systems. The evidence of the pursuers' employees had demonstrated that reasonable tenderers might have thought it was unnecessary to cover "audit requirements" in this section. The judge had erred in being satisfied that such cover was "reasonably foreseeable" (para [62]). This was the wrong test and it had been applied again by the judge in relation to the pursuers' employees' understanding of "turnaround times" (para [63]).

[32] The pursuers had given additional information regarding training in response to section 8.2.3 (Nursing service) and section 8.3.5 (Product handling awareness). It was reasonable for the pursuers to understand that section 8.3.1 had not required all aspects of training to be repeated.

[33] Section 8.3.2 (Contingency Proposal) had not clearly required a reference to an incident reporting plan given that this was relevant to an earlier section 7.5.5 (Quality Management System). Again, the commercial judge had contented himself with finding that an incident reporting system was reasonably foreseeably relevant (para [79]), but that was not enough. BUPA had been marked comparatively favourably for section 8.3.2 (Contingency Proposal) in light of their apparent ability to draw lessons for the future from contingencies that had occurred. It had not been sufficiently clear from the ITT that this was what required to be addressed.

[34] 2. Staffing Levels: The credit given for BUPA's explanation of how they would obtain staff ran contrary to the relevant legal principles. The words "award criteria" did not include criteria which were not aimed at identifying the most economically advantageous tender but were linked to the evaluation of the tenderer's ability to perform the contract. A contracting authority could not take into account under the heading of an "award criteria" the tenderer's experience, manpower, equipment or ability to perform (Emm G Lianakis AE v Dimos Alexandoupolis (supra) at paras 30, 32). The defenders' concerns about staffing levels amounted to a consideration of the pursuers' manpower. The commercial judge had accepted evidence that the defenders' concern had not been whether the pursuers would perform, but the lack of detail as to how it would perform. This amounted to the same thing.

[35] 3. "Silo" Method: Commission v Netherlands (supra) emphasised that "all the conditions and detailed rules of the award procedure" (at para [109]) had to be absolutely clear. Both the criteria for selection and the manner in which those criteria were assessed had to be of sufficient clarity such that each reasonable tenderer would have the same understanding of how the competition would be run. As it was reasonable for the pursuers not to have realised that the "silo" method was being used, the commercial judge ought to have held that the pursuers did not have adequate notice.

[36] BUPA had also been unsure on how the offers would be scored. Their own staff had said that the "silo" method would be used, but had agreed with a suggestion in cross-examination that they had "guessed" this (Day 6 Transcript p 898). The commercial judge's observation that clarification could have been sought by the pursuers was nothing to the point (Clinton v Department for Employment and Learning (supra) Girvan J at para [36]).

[37] As the scorers had assessed tenders individually, they had been left to interpret the different criteria and sub-criteria in different ways. They had not been provided, for example, with model answers. There had been no consistency between the scorers on whether to take account of relevant material appearing in other sections. The result had been that scoring became effectively randomised. The subsequent scorers' meeting had been conducted in an open-textured way with no record of consensus being arrived at and recorded. The absence of a record meant that there was a lack of transparency on how the scoring had been conducted. The "silo" method ought never to have been employed as the object was to select the supplier offering the best service; not the one whose assumptions on how their offer might be assessed might be proved to be correct.

[38] 4. Manifest Error re Vial Service: The evidence of one scorer had demonstrated that he had fundamentally misunderstood the pursuers' offer concerning the provision of a vial service. In that at least one scorer had been labouring under a misapprehension, this was a manifest error. At the very least, the lack of clarity as to what effect that misunderstanding had on the outcome had amounted to a lack of transparency. Although the commercial judge had been entitled to prefer other evidence to that of the one scorer, it did not follow that the latter's evidence had been wholly rejected. The judge had accepted that the scorer had been under a misapprehension, albeit that he found that this was not shared by the rest of the panel.

[39] 5. Clarity of Reasoning: It was not enough to dispose of the pursuers' complaint (under section 8.2.1 (Dispensing Process) relative to protocols, to hold that the e-mail had been a correction of the earlier letter. The defenders had required to provide something by way of evidence to substantiate the fact that their expressed willingness to adhere to the protocols had had no effect on scoring, given the earlier comment to the contrary.

[40] The criticism that the pursuers' implementation plan (section 8.3.1 (Contract Implementation/Change in Business Volumes)) had not been detailed enough had not been adequately explained. The commercial judge had accepted that a "very detailed" plan would be "theoretical" given that much would depend on the requirements of new Participating Authorities (para [73]). In that context it was incumbent upon the defenders to explain why a lack of detail had been penalised.

DEFENDERS

[41] 1. Clarity of the ITT: An ITT could be worded at a certain level of generality so as to invite a range of different responses. This ought not to be characterised as creating an ambiguity. Some tenderers might respond with a brief answer, others with a detailed one. That did not mean that the document prompting those responses was unclear. It was desirable that ITTs provoked competition amongst tenderers in relation to their interpretation of the stated requirements and the submission of innovative responses. There was, in any event, an inevitable "open texture" in language.

[42] It may be that a given criterion was habile to include a number of relevant matters. A particular question could be understood to cover one matter only, or two matters. It was open to interpretation whether both were included. That possible range did not mean that there were two different meanings to the question. Rather, the range was part of the meaning itself. Only when someone could reasonably read the question as excluding one of the two matters could there truly be an ambiguity. It was enough for the challenge to fail on this point that the reasonable well-informed tenderer exercising ordinary care would contemplate that an expanded meaning encompassing both matters was possible.

[43] The criteria must not be so vacuous as to allow the contracting authority an unrestricted freedom of choice. However, that still left the contracting authority with a discretion on how precisely it fixed the criteria. This approach was in accordance with the European jurisprudence. The criteria had to be such as to allow the "reasonably well-informed and normally diligent tenderers" to interpret them in the same way (SIAC Construction v County Council of Mayo (supra) paras 41 - 42). The obligation was to "allow" all such tenderers to interpret the criteria uniformly. The effect of the pursuers' submissions was that what was required was that all such tenderers would interpret the criteria uniformly, but this amounted to an unwarranted gloss.

[44] The reference in Commission v Netherlands (supra) to "the exact scope" of the criteria did not introduce a radical new step. The Court had cited EVN v Austria [2004] 1 CMLR 22, which had in turn referred back to the words "in the same way" in SIAC (supra) (para 57). The case had involved the omission of a significant detail and it was this that had rendered it impossible for tenderers to know the "exact scope" of the criterion and "thus to be able to interpret it in the same way" (para 58). The Court had found violations where a contracting authority had effectively kept something hidden such as the existence of sub-criteria and weightings (Emm G Lianakis AE v Dimos Alexandoupolis (supra)). By contrast, where the authority had looked for certain "elements" of an offer, the Court had held that these elements did not amount to undisclosed criteria where there were "indissociable elements" of the published criteria (Evropaїki Dynamiki v European Maritime Safety Agency [2010] EUECJ T-70/05, paras 51, 53).

[45] The use of a test of "reasonable foreseeability" by the commercial judge was consistent with the Court of Justice's approach. That formula had been used in Mears v Leeds City Council (supra) as deriving from the SIAC test. It was a useful way to underline that what was being undertaken was not an orthodox exercise in the interpretation of a contract by trying to find one true meaning. Rather, it was a way of asking what a reasonably diligent tenderer, exercising ordinary care, might take the relevant criterion to mean, and how they might decide to respond. Clinton v Department for Employment and Learning (supra) was distinguishable on the basis that it concerned a qualitative selection, rather than an award, criterion. As a difference in understanding of the former criterion might lead to outright exclusion from the competition, there may be good reasons to take a stricter approach. The commercial judge's reasons for rejecting the pursuers' individual criticisms were correct under reference to the evidence given.

[46] 2. Staffing Levels: The defenders' concern had not been whether the pursuers could carry out the tasks required by the Framework Agreement but whether they could do so properly. It was wrong to say that the concern was connected to manpower and must therefore amount to a qualitative selection criterion.

[47] 3. "Silo" Method: The reasoning of the commercial judge (at para [60]) was correct. It ought to have been obvious that a "silo" method for scoring was likely to be employed, given the very substantial amount of material. The pursuers' offer filled two ring binders. To require each scorer to read every part of every tender would have been impractical. Provided criteria and sub-criteria were disclosed in sufficient detail so as not to be vacuous, the obligation of transparency was discharged.

[48] The provision of model answers or other means of constraining the scoring of offers was hazardous, as it might be claimed that these amounted to the application of undisclosed criteria. The commercial judge had observed in a "postscript" to his Opinion that the scorers' views differed on whether they ought to have taken into account for each section material which they were aware had been submitted in other sections. However, he noted that there was no evidence suggesting that this had had any adverse effect on the pursuers; and, there was evidence suggesting that it had not (para [109]).

[49] The complaint that there had been no record of the tendering assessment panel's consensus was not pleaded, had not been advanced in the Outer House and ought not to be entertained. In any event there was no legal obligation to have such a record.

[50] 4. Manifest Error re Vial Service: There was no finding by the commercial judge that any part of Mr Morrison's evidence could be relied upon. Rather, he had preferred Dr Fraser's evidence, as he was entitled to, as to what the scorers' understanding had been.

[51] 5. Clarity of Reasoning: The reasoning of the commercial judge (at paras [96]-[99]) had been correct and supported by the evidence. The need to provide evidence in relation to "Participating Authority protocols" went beyond the obligation to give reasons.

Decision

1. Clarity of the ITT

[52] In SIAC Construction v County Council of Mayo [2001] ECR I -7725 the European Court of Justice set out the test to be applied as being whether the criteria have been formulated:

"in such a way as to allow all reasonably well-informed and normally diligent tenderers to interpret them in the same way" (para 42).

It is doubtful whether any gloss upon or addition to these plain terms would advance any understanding of their meaning. If criteria "allow" every reasonably, well-informed and normally diligent tenderer to arrive at the same interpretation, that is sufficient. The test does not involve a guarantee that all tenderers will respond in an identical manner. It seems inevitable that there will be a range of responses. Provided the criteria enable all to reach the same interpretation as to what services they are being invited to perform, each may, in accordance with that common interpretation, elect to respond in a manner different in both detail and content. Indeed, in a tendering process as complicated as that under consideration, it would be surprising if the tenderers did not do so. If that happens, that does not, of itself, demonstrate that the ITT lacked the requisite clarity.

[53] This approach is entirely consistent, and borne out by, the decisions that have followed SIAC. Thus, in Commission v Netherlands [2012] 3 CMLR 11, the Court had been dealing with how the supplying and management of automated coffee machines would fulfil criteria of sustainable purchasing and socially responsible business and how they contributed to the goals of the contracting authority. In condemning these criteria as, in essence, excessively vague, the court said that the criteria must be formulated:

"such as to allow all reasonably well-informed tenderers exercising ordinary care to know the exact scope thereof and thus to interpret them in the same way" (para 88)

and that the criteria:

"must be drawn up in a clear, precise and unequivocal manner ... so that ... all reasonably informed tenderers exercising ordinary care can understand their exact significance and interpret them in the same way ..." (para [109]).

The Court of Justice explained that these dicta were derived respectively from EVN AG v Austria [2004] 1 CMLR 22 and Commission v CAS Succhi di Frutta [2004] ECR I‑3801. It is important to understand them in context.

[54] In EVN v Austria (supra) the issue was whether, in a contract for the supply of electricity, a criterion that the electricity be from renewable sources, but which did not require specification of the period over which the renewable energy would be obtained, was too uncertain. The Court held that such an omission would be fatal to the fairness of the competition if:

"that omission made it difficult or even impossible for tenderers to know the exact scope of the criterion in question and thus to be able to interpret it in the same way" (para 58).

Put in this way, the question is whether the tenderers' path to understanding the meaning of a criterion is obstructed by the contracting authority's formulation of its terms. The Court continued, in relation to the tenderers' ability to interpret a criterion, as follows:

"In as much as that requires a factual assessment, it is for the national court to determine, taking account of all the circumstances of the case, whether, ... the award criterion ... was sufficiently clearly formulated to satisfy the requirements of equal treatment and transparency ..." (para 59).

[55] The goals of equal treatment and transparency are what has to be achieved. The formulation of the criteria is to be judged according to whether it meets those goals. That is illustrated by Commission v CAS Succhi di Frutta, from which the second formulation in Commission v Netherlands (at para 109) is derived verbatim (at para 111). The European Commission had made complex arrangements to obtain fruit juice and jam; for which it was to pay the tenderer in raw apples. The winning tenderers offered to accept payment in apples, which failing peaches. The Court, not surprisingly, found there had been a lack of transparency because the price criteria had been altered clandestinely from apples to apples or peaches.

[56] The test is then simply whether the criteria are sufficiently clear that the tenderers can all understand their scope, interpret them "in the same way" and respond accordingly. If the court considers that the manner in which a successful tenderer has responded was in accordance with what the reasonably informed tenderer would have understood from the formulation of the criteria, it will be difficult to argue that the criteria were not transparent. The fact that an unsuccessful tenderer, who may normally be reasonably well-informed and act diligently, responded in a different way does not, per se, demonstrate that the criteria did not meet the required legal test.

[57] This court endorses the approach in Federal Security Services v Northern Ireland Court Service [2009] NIQB 15 (McCloskey J at para [52]) to the effect that the criteria must be formulated in such a manner as to allow all reasonably well informed and diligent tenderers to interpret them uniformly. If such a tenderer could, "understandably and plausibly" (ibid para [53]), have construed the criteria in different ways then the criteria must be deemed insufficiently transparent. However, that is a long way from a proposition that the mere fact that a tenderer, who might normally be regarded as reasonably well informed and diligent, construed the criteria in his own particular way is destructive of the process. For such an outcome, the court has to be satisfied that the interpretation was open to the hypothetical tenderer and not simply that the unsuccessful tenderer had been reasonably well informed and diligent and in fact reached that interpretation.

[58] The commercial judge asked himself at various points whether the approach of the defenders to the ITT had been "reasonably foreseeable". This court has considerable doubts about the usefulness or appropriateness of this formulation, which is more familiar in other branches of the law and appears to add nothing to the established jurisprudence in this field. Its use may only serve to confuse. Having commented (para [26]) that a requirement that criteria be formulated with such precision as to preclude two reasonable persons reaching different interpretations lacked realism, the commercial judge did go on (para [27]) to employ the concept of "reasonable foreseeability", but he did so in a limited context. This was as part of an equation which asked whether what the successful tenderer did do was in line with what the hypothetical tenderer would have foreseen as being required by the criteria. The judge appears to have been conducting a similar exercise to that carried out in Mears v Leeds City Council [2011] EWHC 1031 (TCC) (Ramsey J at para 122). The question is whether what the successful tenderer did was what the hypothetical reasonably well-informed tenderer would have anticipated was required by and embraced within the criteria and whether, conversely, what the unsuccessful tenderer omitted to do fell short of what the hypothetical tenderer ought to have seen was entailed in the terms of the criteria.

[59] For each allegedly irrelevant or undisclosed element which the pursuers complained about, BUPA had believed that element to be within the ambit of the criteria. The commercial judge occasionally referred to evidence on the reasonableness of the respective beliefs of the parties and others about what was needed (eg para [60]). At least BUPA must have thought that those elements which they included might be encompassed, given that they were addressed by them in their response to the ITT. If the court accepted, as the commercial judge did, that these responses were in accord with what the reasonably informed tenderer would consider were required to meet the criteria, any challenge to the criteria would be bound to fail. It could not succeed on the basis that another tenderer did not, in fact, think that the content of the response of his competitor was what was needed.

[60] In this area of commerce, it is of considerable importance that decisions of the courts on the validity of a tendering process are taken with all due expedition. It is desirable that parties know, without delay, whether or not the contract is going to proceed. Unless there is a strong reason to suppose that it will cause injustice, such decisions ought to be capable of being taken in the absence of detailed oral testimony. If it is otherwise, a swift decision will be almost impossible (see Federal Security Services v Northern Ireland Court Service (supra) McCloskey J at para [5]). It is therefore, at best, of marginal relevance for a judge to examine the views of the unsuccessful tenderer or his agents and employees on the meaning of the tender and to determine, where they differed from the approach of the contracting authority, whether those views were reasonably well-informed and arrived at after ordinary care. It is also relatively pointless for a defender to select one or more tenderers, whose interpretations matched the contracting authority's approach, and to invite the court to determine whether they were reasonable. The court's decision will involve it placing itself in the position of the reasonably informed tenderer, looking at the matter objectively, rather than, as occurred here to a degree, hearing evidence of what such a hypothetical person might think (see Clinton (t/a Oriel Training Services) v Department for Employment and Learning [2012] NICA 48, Girvan LJ at para [33]). Although different from an orthodox exercise in contractual interpretation, the question of what a reasonably well-informed and normally diligent tenderer might anticipate or understand requires an objective answer, albeit on a properly informed basis. Just like those other juridical creations, such as the man on the Clapham omnibus (delict) or the officious bystander (contract), the court decides what that person would think by making its own evaluation against the background circumstances. It does not hear evidence from a person offered up as a candidate for the role of reasonable tenderer. In a disputed case, the court will, no doubt, need to have explained to it certain technical terms and will have to be informed of some of the particular circumstances of the terms or industry in question, which should have been known to informed tenderers. However, evidence as to what the tenderers themselves thought the criteria required is, essentially, irrelevant.

[61] The formulation in the ITT allowed all reasonably well-informed and normally diligent tenderers to interpret the criteria in the same way and as they were applied by the contracting authority. What BUPA submitted was entirely in line with what the hypothetical tenderer would have done and did not stray beyond that. Whatever might have been in the minds of the pursuers (and their agents and employees) subjectively as meeting the criteria, the pursuers, along with that tenderer, ought also to have realised, from the contents of that ITT, that material, such as that formulated and submitted by BUPA, was required to meet the criteria.

[62] The formula of "including but not limited to" is entirely understandable. It does not require the consideration of "expert" tenderers. It would convey to the hypothetical tenderer the idea that, although he should cover all the matters specifically alluded to, he should add other material which the ITT inferred was relevant. As the words themselves say, he would understand that he was not limited to the specified items. A process which provides extra marks for creative thinking is unexceptional, as is one which allows a tenderer to include novel methods of working which the contracting authority might find attractive, provided that they fall within the reasonable ambit of the specified criteria as it would have been understood by the hypothetical well-informed tenderer.

[63] It was clear from the terms of the section 8.2.1 (Dispensing Process) that the defenders were seeking information on checks from compounding to injection; hence the reference to the communication of interventions. Equally, the existence of audit checks and the turnaround times for changes were obvious aspects of any response to this section, even if the pursuers had not sought to address them under this heading. The training of staff was a matter expressly required in relation to contract implementation (8.3.1) and was not restricted to that required at the commencement of the contract or in the event of increasing business volumes. The existence of a contingency plan was a factor which, if submitted, would be relevant to the section on contingency proposals (8.3.2). All of the material produced by BUPA under these headings would, and should, have been in the contemplation of the hypothetical tenderer when determining what range and detail of response the ITT required.

2. staffing levels

[64] The defenders were entitled to have regard to the issue of staffing levels in the context of section 8.3.1. The commercial judge relied on testimony from the scorers to this effect. However, looked at objectively from the viewpoint of the hypothetical tenderer, it can be seen that that tenderer would, or ought to, have anticipated the importance, or at least the relevance, of how the tenderer would perform the contract by reference to the recruitment of non NHS staff. This was not something which was concerned with the tenderer's ability to perform by reference to their own manpower but how they would execute their obligations. The judge concluded on the evidence that BUPA's ability to find additional staff without "poaching" them from the NHS was a material and relevant consideration. The court agrees with that conclusion, which entitled the judge to find that the defenders' concern was not simply about an ability to obtain adequate manpower.

3. "Silo" method

[65] Because of the features of the ITT identified by the commercial judge, all reasonably well-informed tenderers and normally diligent tenderers applying their informed minds to the matter ought to have appreciated that the so-called "silo" method of marking was likely to be employed. This conclusion is reached in the absence of the fact that the pursuers could, in any event, have sought clarification from the defenders on this point. The ITT had made reference to the existence of a panel (section 3.1) to score the tenders. Given the different areas of expertise needed to score the completed forms, it was most likely that different scorers would be looking at different parts of the forms. The ITT stated clearly that each section was to be completed "fully" (4.1 and 4.3). It was made clear that: "Attachments must be attached everywhere they are relevant. Do not reference attachments in other sections as they may not be considered". Such language would make it clear to all tenderers that different sections would be scored separately and potentially by different scorers (ie those possessing an expertise in the particular field). All tenderers were thus put on notice of this likelihood. This was particularly transparent given that some sections, to which material might be attached, carried no score, whilst other sections, to which the material was of importance, were to be given significant weight.

[66] The use of the "silo" method was an appropriate practical response to the daunting task of assessing the tenders. There is nothing in the method used by the defenders that would have prejudiced the pursuers or otherwise have led to unequal treatment. As to the consideration by some, but not all, scorers of material submitted in other sections of the ITT, there is no reason to alter the commercial judge's finding that there was no evidence of this having put the pursuers at a disadvantage.

4. Manifest error re the vial service

[67] The commercial judge appears to have been neutral on whether the particular scorer correctly recalled his understanding of the scoring of the alternative vial service. This is consistent with the judge's later observations, albeit in the different context, on the adequacy of the defenders' reasons. He said that the reference in the e-mail of 27 May 2010 to a vial service "appears to reflect Mr [M's] understanding" (para [101]), but that "I do not think it matters if Mr [M's] view was not shared by other scorers" (para [102]). Any misunderstanding on the scorer's part is of no significance for the purpose of this challenge. The scorer had said that the "main concern" was that the pursuers were proposing the use of a vial service "specifically when there was an incorrect delivery" (para [82]). In cross-examination he corrected this to "for example when there was an incorrect delivery" (para [83]). Thus his position came to be that the use of a vial service for an incorrect delivery was only one aspect of his concerns regarding the pursuers' contingency plans. This is borne out by what the scorer recorded at the time. The commercial judge noted that this included that it was "Unclear what would happen with an incorrect delivery - looks like vials would be used" (para [83]). The scorer's comments on section 8.3.2, however, also made reference to the pursuers' plan for a local contingency being "very thin". The use of vials was given only as an example in that context. He mentioned the lack of detail on timescales for incorrect delivery and the keeping of low stock levels. In summary, he concluded: "Overall - too much information provided and not really focused upon what was asked". That having been recorded, the 66.66% represented a mark of 4 out of 6, being an answer that fell short of a "good" one.

[68] The tender assessment panel had a general concern over the use of a vial service (see minutes of meeting of 28 April, quoted by the commercial judge at para [85]). This resulted in three scorers reducing their mark to 66.66% of the total points available. In these circumstances, given that any misunderstanding by the scorer was just one aspect of his concerns and that three scorers arrived at the same 66% without any misunderstanding, it cannot be said that there was any material error in the overall assessment.

5. Clarity of REASONING

[69] The obligation upon the defenders was to provide the pursuers with "a summary of the reasons" why they were unsuccessful and "the characteristics and relative advantages" of BUPA (Reg 32(2)(d),(e) (supra)). The commercial judge was correct in his view that it is inappropriate to assess the quality of the reasons by examining phrases in isolation. Looking at the reasons as a whole, they were more than adequate to explain why BUPA had been successful and why their offer was considered to be comparatively better than that of the pursuers. In particular, they detailed, both in terms of numerical scoring and verbal commentary where BUPA had, albeit marginally, outshone the pursuers in terms of quality of service and deliverability.

[70] The commercial judge's reasoning, on the detailed complaints advanced by the pursuers, is sound. The reference to "key areas" in the e-mail referred back to the system of reporting errors. The e-mail's reference to the protocols was a clarification of what might have been a misinterpretation of the content of the earlier letter. The reference to turnaround time for changes was in the context of the replacement process.

[71] The question of whether the transfer from a vial service to a compounding service was of any importance to the scorers (as suggested by the e-mail) raises a slightly different point. However, there is no reason to alter the commercial judge's finding that the relevant part of the e-mail accurately reflected the concern of one scorer. The other scorers had said that they had not questioned the pursuers' ability to change services, but there is no contradiction between that and the view that the methodology of change ought to have been explained.

[72] Regarding the question of "Participating Authority protocols" (Section 8.2.1 (Dispensing Process)), there was no obligation on the defenders to substantiate the correction to the letter in the e-mail. The duty was to provide an explanation for the decision, not to produce evidence for it.

[73] The new complaint regarding the "undeveloped" criticism in section 8.3.1 (Contract Implementation/Change in Business Volumes) is rejected. It is true that the commercial judge found that a "very detailed" implementation plan would have been theoretical in content. However, that does not mean that no detail of it whatsoever should have been given.

[74] Ultimately, the goals, which the legislative provisions and the case law in this area seek to achieve, are those of equal treatment and transparency in the tendering process. The court is satisfied that those goals were achieved is this case.

[75] For all of these reasons, the reclaiming motion must be refused and the court will adhere to the commercial judge's interlocutor of 1 May 2012.