Edenred (UK Group) Ltd v Her Majesty's Treasury & Ors

This claim arises out of a decision by the First Defendant (“HMT”) and the Second Defendants (“HMRC”) that the new government policy of tax-free childcare (“TFC”) should be delivered by HMT and HMRC working with the Third Defendant (“NS&I”). It is not the policy itself, but the mechanism for its delivery that is the subject of challenge.

TFC is a new scheme to support working families with the costs of childcare. It was first announced by the Government on 19 March 2013, and intended to be introduced in the autumn of 2015, with HMRC the primary budget holder with overall responsibility for its successful delivery. HMRC has been allocated monies by HMT to administer TFC. Under TFC, eligible families will be able to receive 20% of their childcare costs from the Government (the equivalent of basic rate tax relief), up to a maximum of £2,000 per child per year.

TFC will be delivered through childcare accounts. Eligible parents will open a bank account called a “childcare account” into which they make payments (and from which they will also be able to make withdrawals). There will be a separate account in respect of each child, rather than for each household, and grandparents (and possibly others, such as other relatives) will also be able to pay money into the account. The intention is that the parents will use payments into the account to make up 80% of the relevant childcare costs, which will be “topped up” by a transfer by HMRC into the account of the remaining 20%. The parents can then use the total funds to make payment to the registered childcare provider or providers of their choice. Employers are not involved in the scheme. Indeed, TFC will be available to self-employed parents provided that they meet the criteria for eligibility.

The Government estimates that 1.9 million parents could be eligible for TFC, and that the take-up from that constituency is likely to be around 1.2 million over five years, with some 1.6 million separate childcare accounts being opened (because some parents will be opening more than one account).

Instead of inviting tenders for the provision of childcare accounts, which was one option that it considered in the course of an extensive consultation process, HMRC and HMT decided to keep the matter in-house and utilise NS&I, a government department and Executive Agency of the Chancellor of the Exchequer, to provide and administer childcare accounts and supporting services. The main advantage of using NS&I to deliver the services is that it already has a banking infrastructure and customer support network in place that could be utilised and built upon. NS&I is one of the largest savings organizations in the UK, with more than £105 billion invested in its products, and over 26 million customers. Its products range from Premium Bonds to savings accounts, savings certificates and ISAs. Although its customers could choose to invest elsewhere, and to that extent it competes on the open market for custom, no other bank or financial institution offers to the public its unique savings products that involve lending money to the Exchequer. Because of its experience of dealing with millions of customers NS&I is well placed to deal with the levels of support that will be needed for TFC. Moreover, the Government, through HMT, had already invested considerable sums of money in upgrading its banking infrastructure and capability with the express intention that it be used to provide payment and account services to other government departments.

Another perceived advantage of using NS&I is that parents would not have to pay fees to a commercial account provider, or shop around among a number of providers to find the best deal. Instead they would engage with NS&I as a single point of contact to register for the scheme, make payments into their account, and arrange payments to the childcare provider(s). They can also be confident that the money they pay into the account would be secure, as it would be government backed.

HMRC will be responsible for checking eligibility, working out how much money is due to parents, and ensuring that the government contribution is paid into the accounts; if the decision is implemented, NS&I would be responsible for developing the web portal through which parents access TFC, providing the childcare accounts, processing payments into and out of those accounts, and generally administering the scheme (including the provision of customer service and support for the parents through the use of its existing UK call centres). For parents who are unable to access the internet there will be alternative options (e.g. telephone banking).

The demarcation of HMRC’s and NS&I’s roles and respective responsibilities in this inter-departmental collaboration to deliver the TFC policy, and the flows of money between HMRC and NS&I to cover the costs accrued in NS&I (which must be met out of HMRC’s Departmental Expenditure Limit) will be set out in a Memorandum of Understanding (“MOU”) which is considered in more detail later in this judgment. This is normal practice when internal arrangements are made between two government departments to work together to implement a specific policy or policies. It ensures transparency and proper accountability for the use of public funds.

TFC is intended to replace the current system of government support to working parents provided through arrangements referred to as “Employer Supported Childcare” (“ESC”), which will be phased out (although it will still be available for those already using the scheme). ESC does not involve any public contract. The Government provides tax relief to employers who wish to support the costs of childcare for their employees. One of the ways in which employers may do so is by providing vouchers which the parents can redeem with nurseries or other childcare providers. The vouchers are part of the employees’ remuneration packages, and since 2005 they have been free of income tax and national insurance contributions. In essence the employer gives employees money, represented by vouchers, to use to pay for childcare, funded by a tax break.

Commercial childcare voucher providers (CVPs) are not a necessary part of such a scheme, but many employers prefer to engage such a provider to administer their schemes and issue the vouchers on their behalf, instead of doing so themselves. In the vast majority of cases, this is now done online using e-vouchers. It is the employer who engages and pays for the services of the CVP and who plays a pivotal role in the operation of the voucher scheme. The employer is solely responsible for validating an employee’s entitlement to claim the benefit, and for providing the money to make the payments, which it will deduct from the employees’ salaries and pay to the CVP in a lump sum once a month. The role of the CVP is to receive the money from the employer, allocate the correct amount to each relevant employee, issue a voucher or vouchers in that amount to the parents, and make payments to the childcare provider(s) designated by the parents, against production of the voucher. The fees of the CVP are paid out of the savings that employers make in their national insurance contributions, generally charged as a percentage of the value of the vouchers. The fees (and thus the profits) are driven by the competition in the CVP market, which increased considerably when the tax break was introduced in 2005.

The Claimant (“Edenred”) is one of the four largest CVPs in the UK; one of its competitors, Computershare, claims to be the largest, but Edenred estimates that it has a larger market share of around 33% compared to Computershare’s 25%. Its ESC business, run through a subsidiary, Childcare Vouchers Ltd (“CVL”), currently has 15,600 customers (i.e. employers) and provides childcare vouchers for around 170,000 parents per month. Edenred is a founder member of the CVPA, which is a trade association set up in 2011 primarily to represent the interests of CVPs in the UK. Although the childcare voucher business is run by CVL, it is unclear on the evidence to what extent its parent company is involved. It appears that Edenred, or even its French parent, provides the supporting infrastructure (including the web platform). For ease of reference I shall continue to refer to the provider of the childcare vouchers as “Edenred” and that expression, where appropriate, shall be taken to include CVL.

When Edenred has entered into an ESC arrangement with an employer, it will provide the employer with a unique scheme ID, which can then be supplied to and used by parents to log in to Edenred’s platforms. Edenred will create a designated online account in the name of the parents. The parents provide Edenred with their contact details and the name of their child, and Edenred then sends them (electronically, or in hard copy by mail) a pack. This enables the parents to create a password and log-in details to enter Edenred’s computer system. The parents can then use this on-line facility to provide details to Edenred about the identity of the childcare providers they want to pay, how much they wish to pay and how often. They can direct Edenred to pay a set sum each month, in a similar fashion to making a standing order arrangement with a bank.

On receipt of payment from the employer, Edenred will either send the parents a paper voucher representing the amount deducted by the employer and passed on to Edenred, which can be given to the childcare provider and presented by them to Edenred for payment, or else (as now happens in around 97% of cases) the matter will all be dealt with online by the generation of e-vouchers. When the nominated childcare provider redeems the voucher, (which it may do online via an account opened in its name with Edenred), Edenred will cause the payment to be transferred directly into the provider’s bank account.

In operational terms, there are some similarities between ESC and TFC. Both schemes involve money being “ring-fenced” for the provision of childcare services. Other than withdrawals made by (or refunds paid to) the paying party, that is the sole purpose for which money paid into a childcare account can be paid out of it. The schemes also both involve processing inbound and outbound payments, the maintenance of a record of registered childcare providers, the use of a web-based platform, and the provision of a customer support system for the participating parents. However, although (like many other online service providers) CVPs set up and administer on-line accounts for qualifying parents (and for registered child care providers), these are not bank accounts, and the CVPs are not regulated by the Financial Conduct Authority. The value of the vouchers will be seen as a credit and the payments out of the account to the childcare provider as a debit to the parents’ online account. However the money only goes in one direction – from the employer, via the CVP, directly to the childcare provider.

The ESC scheme is much less complex than TFC, as well as being far smaller in scale. It makes no provision for the refunding of money or for dealing with overpayment. From the description of its operation it seems to me that overpayments are unlikely to occur, as payments only emanate from one source – the employer, who decides how much salary will be paid in the form of childcare vouchers. In any event, if the employer mistakenly overpays the CVP provider he can presumably make adjustments to the next monthly lump sum. Under TFC, however, the system must enable adjustments to be made to the accounts in order to ensure that the 80/20 ratio of parent/HMRC contribution is maintained, and that the limits set by the Government for the amount of financial assistance given per child are not breached. If a refund is due to HMRC, the system must ensure that it is repaid to HMRC and not to the parents. The system must also enable parents with more than one child to easily obtain information about all of the childcare accounts they have opened.

Delivery of the new TFC policy by the Defendants directly to parents without the necessary involvement of their employer means that there will be no similar profitable private business opportunity for Edenred or other CVPA members to act as intermediaries in the delivery of TFC. This no doubt explains why in response to the two consultations carried out jointly by HMT and HMRC on behalf of the Government on delivery of TFC, the CVPA lobbied for the continuation and extension of ESC, with the employer remaining at the heart of the system, which would have been the outcome that best protected its members’ commercial interests. The CVPA made a great deal of the likely increases in the administrative, reconciliation and interface costs that the increase in the number of customers from 50,000 businesses (the total number of employers using the services of all the CVPA members) to over 1 million individual parents would cause. However, the Government was not persuaded by those arguments, and Edenred now seeks to play down those factors.

Although Edenred and other members of the CVPA were consulted extensively during the two consultation processes that the Government undertook in respect of the design and delivery of TFC, it eventually decided to go down the route of a banking/savings account and, instead of putting the provision of those services out to tender, to utilise the established infrastructure of NS&I working with the company to which it has outsourced its “back office” operations, Atos IT Services UK Ltd (“Atos”) under the terms of a public services contract which commenced in April 2014 (“the Outsourcing Contract”). A significant feature of this case is that Atos was awarded that contract after a fair and transparent public procurement process complying with the Regulations, about which no complaint is, or could be, made.

That decision was initially taken on 18 March 2014 following the first consultation. Quite understandably, it was not welcomed by the CVPA. They commenced a claim for judicial review. Those proceedings were stayed (and subsequently withdrawn) when the Government decided to carry out a second consultation expressly addressing the option of delivering TFC accounts itself through a public sector provider, and to take a fresh decision as to the appropriate provider in the light of responses to that consultation. The vast majority of parents’ groups who responded to the second consultation preferred a single provider and, of those, the majority preferred NS&I as that provider. After consideration of the responses to the second consultation, and carefully weighing up the various options, the decision under challenge was made. It was announced on 29 July 2014.

In parallel with these proceedings, the CVPA and Edenred again sought permission to bring judicial review. Permission was refused by Laing J on the papers on 2 October 2014, on the basis that the CVPA lacked the necessary standing, and the current claim, brought by Edenred under CPR Part 7, canvassed substantially the same issues. Following a case management hearing before Leggatt J on 6 October, the judicial review proceedings were stayed and a speedy trial was directed of certain aspects of Edenred’s Part 7 claim. That trial took place before me in late November 2014.

Edenred challenges the decision under the Public Contracts Regulations 2006 (“the Regulations”) and Article 56 of the Treaty on the Functioning of the European Union (“TFEU”) on the grounds that these arrangements will involve the conclusion of a public services contract within the meaning of the Regulations or economic opportunity falling within Article 56 TFEU between either or both of HMT and HMRC on the one hand, and NS&I on the other; or alternatively that the arrangements will involve a material variation of an existing public services contract between NS&I and Atos, i.e. the Outsourcing Contract.

The Regulations implement Directive EU No. 2004/38 (“the Directive”). The purpose of the EU procurement regime, and therefore of the Regulations, is to:

“...develop effective competition in the field of public contracts.”

(C-247/02 Sintesi [2004] ECR I-9215 at [35].) Recital (2) to the Directive notes it gives effect to:

“the principle of freedom of movement of goods, the principle of freedom of establishment and the principle of freedom to provide services and to the principles deriving therefrom, such as the principle of equal treatment, the principle of non-discrimination, the principle of mutual recognition, the principle of proportionality and the principle of transparency”.

The Regulations apply to the award of “public contracts”, including “public services contracts” by “contracting authorities”. HMT, HMRC and NS&I are each identified as a “contracting authority” under the Regulations: reg. 3(1)(b) and Sch. 1.

A public services contract is defined as “a contract, in writing, for consideration…under which a contracting authority engages a person to provide services” (reg. 2(1)). It is common ground between the parties that a public services contract for the provision of childcare account services necessary for the operation of the TFC Scheme would be a Part A services contract under the Regulations, to which the full requirements of reg 5(1) would apply.

Regardless of the precise form of procurement procedure to be followed, reg. 5(1) requires a contracting authority to “publicise its intention to seek offers in relation to the public contract”, by publication of a contract notice in the Official Journal of the European Union (“OJEU”) “as soon as possible after forming the intention [to seek offers]”: regs. 15(2), 16(2), 17(3) and 18(4), and see C/423/07 Commission v Spain [2010] ECR 1-3429. Recital (36) to the Directive explains the significance of the contract notice as follows:

“To ensure development of effective competition in the field of public contracts, it is necessary that contract notices drawn up by the contracting authorities of Member States be advertised throughout the Community. The information contained in these notices must enable economic operators in the Community to determine whether the proposed contracts are of interest to them. For this purpose, it is appropriate to give them adequate information on the object of the contract and the conditions attached thereto…”

Pursuant to reg. 4(3) of the Regulations, a contracting authority must comply with principles of transparency and equality of treatment at every stage of a procurement procedure. This includes the formulation and publication of the contract notice, which must clearly define the work which will be performed under the contract. Hence, in C-340/02 Commission v France [2004] ECR 1-9845 at [34], the CJEU emphasised that: "The principle of equal treatment of service providers, laid down in…the Directive, and the principle of transparency which flows from it…require the subject matter of each contract and the criteria governing its award to be clearly defined."

The importance of a contracting authority’s duties of transparency and equality of treatment means that these obligations are strict, or “hard-edged”. There is no room for any exercise of discretion by the authority, or for any deference by the court to an authority’s margin of discretion: Lion Apparel Systems v Firebuy [2007] EWHC 2179 (Ch), per Morgan J at [26].

Under reg. 47A(1) and (2), the duty to comply with the Regulations and any enforceable EU law obligation is one which is owed by a contracting authority to an “economic operator”. The definition of “economic operator” is contained in reg. 4(1) and includes a “services provider”, i.e. any undertaking that offers services in a market and “who sought, who seeks or who would have wished to be the person to whom a public services contract is awarded” (reg. 2(1)). Edenred plainly falls within that description.

Regulation 47C provides that a breach of duty is actionable by any economic operator which “in consequence, suffers, or risks suffering, loss or damage” (emphasis added). Thus in order to establish it has the necessary status to bring a claim for breach of duty it is unnecessary for an economic operator to demonstrate that it has suffered any loss, but merely that there is a risk of it doing so. However, that does not mean that an economic operator who would have wished to have been awarded the contract in question, but who had no realistic chance of obtaining it (e.g. because it could never have met the pre-qualifying criteria) would succeed in its claim.

If either of the arguments raised by Edenred is correct, it would have been necessary to carry out an open, transparent and competitive procurement exercise as prescribed by reg 5(1) before awarding a contract to anyone to deliver TFC through the provision of childcare accounts and associated services. Edenred claims that if such a procurement exercise had been, or were to be carried out, it would be one of the leading contenders to provide childcare accounts. That is strongly disputed by the Defendants, who contend that Edenred is unable to demonstrate that the decision has caused it to suffer any loss or damage.

In exchange for obtaining directions for a speedy trial, Edenred gave the court an undertaking that if it lost on these aspects of its claim it would be confined to a remedy in damages in respect of the remaining (stayed) causes of action. However, Edenred’s claim at this stage is not confined to damages; the primary relief that it seeks is a declaration that the decision or decisions under challenge were unlawful or ineffective, and orders setting it or them aside. In such event, Edenred would potentially be able to compete for the opportunity to administer childcare accounts under the TFC scheme.

In a case such as the present, where the contract (or contracts) said to offend the Regulations have not yet been entered into, regulation 47I provides that the decision to award the contract “may” be set aside. The court may also order a remedy in damages. A declaration is a discretionary remedy, as is the remedy prescribed by regulation 47I, and one of the matters that the court would have to take into account in deciding whether to grant such a remedy is whether it would achieve any useful purpose. Edenred’s ability to compete and its likely prospects in such a competitive tender are obviously relevant factors.

However, as Edenred’s counsel, Mr Coppel QC, pointed out, the court must also bear in mind the rationale for the discretionary remedy. The Regulations include provisions on remedy which give effect to Directive 89/665/EEC (as amended, in particular, by Directive 2007/66/EC) (“the Remedies Directive”). Recitals (13) and (14) of the amending Directive explain the rationale of the EU remedies regime in the following terms:

“(13)

In order to combat the illegal direct award of contracts, which the Court of Justice has called the most serious breach of Community law in the field of public procurement on the part of a contracting authority or contracting entity, there should be provision for effective, proportionate and dissuasive sanctions. Therefore a contract resulting from an illegal direct award should in principle be considered ineffective. The ineffectiveness should not be automatic but should be ascertained by or should be the result of a decision of an independent review body.

(14)

Ineffectiveness is the most effective way to restore competition and to create new business opportunities for those economic operators which have been deprived illegally of their opportunity to compete…”

Mr Coppel therefore submitted that if the Court concludes that there has been a breach of the Regulations and/or Article 56 TFEU, the (most) serious nature of these breaches should lead to the Court setting aside the decision and thereby ensuring that the unlawful direct award cannot proceed.

Had NS&I not taken the decision more than fifteen years ago to outsource its back office operations and transfer its operational support staff to a private sector service provider, it is difficult to see how this legal challenge could ever have got off the ground. However, in consequence of that decision, all operational aspects of NS&I’s business are now carried out by its private sector partner. The operational delivery of TFC is therefore to be effected by NS&I working with Atos under the Outsourcing Contract.

It is common ground that in order for Atos to provide the services to NS&I that are needed to support the delivery of TFC, it is necessary for changes to be made to the Outsourcing Contract. The intention is that those changes will be implemented by way of an amendment to the Outsourcing Contract (“the Amendment Agreement”) made pursuant to the change control mechanism prescribed in that contract. The key issue in this case is whether those changes amount to a “material variation” to a public contract. The CJEU has held that a material variation constitutes the award of a ‘new’ public contract, which requires a fresh procurement to be conducted: C-454/06 Pressetext [2008] ECR I-4401 and C-91/08 Wall AG [2010] ECR I-2815. Whether the variation is ‘material’ is an autonomous EU law concept which falls to be determined by comparing the contract as originally entered into, and the contract as it would exist after the amendment.

This is not, as might at first appear, simply a case of a disappointed potential bidder wishing to be given the chance to compete on an equal footing with NS&I (and others) for the award of a highly lucrative contract for services. Edenred’s strategy appears to be designed to eliminate NS&I from the competition altogether: at least, that would be the likely effect of accepting its arguments. Mr Coppel candidly accepted that if there were a public tender for the provision of TFC childcare accounts, then NS&I would almost certainly be unable to compete at all, since it is completely dependent upon Atos for the delivery of its operational services. If the Outsourcing Contract cannot be lawfully varied to include the services supporting and implementing the provision of TFC childcare accounts, there would be insufficient time for a fresh public procurement process to be carried out for the provision of those additional services for and on behalf of NS&I, before any public tender took place for the provision of those services to HMRC.

Thus the upshot of a successful challenge by Edenred would be that the Government would be precluded from utilising its own established banking infrastructure and support system, which already services millions of customers, to deliver TFC merely because, in a successful attempt to modernise, drive down costs and become more efficient, instead of retaining its operational staff, NS&I has lawfully outsourced all its operational services to a private sector supplier - using a process which fully complied with the self-same Regulations. On the face of it, that does not appear to be the type of mischief against which the Regulations are aimed. Nevertheless, if the Outsourcing Contract requires a material variation in order for NS&I to be able to deliver childcare accounts, there will have been a breach.

Before considering the legal arguments, it is necessary to say a little more about the parties, the relationship between NS&I and Atos, and the background to the Outsourcing Contract.

The Defendants are all parts of the Crown, within the Executive branch of government, and “Chancellor’s departments” falling under the remit of the Chancellor of the Exchequer. Although they have different functions and statutory powers, they are not separate public bodies which can be regarded as legally distinct from each other in the same way as, for example, individual local authorities, or a parent company and wholly-owned subsidiary. They are more akin to different departments of the same local authority.

Central government is organised into separate government departments, most of which are headed by a Secretary of State or other senior minister, through whom they are ultimately accountable to Parliament. Ministerial departments do not have their own separate legal personality. They are staffed by civil servants. HMT is the government’s economic and finance ministry, which sets the direction of the UK’s economic policy.

Non-ministerial departments (NMDs) are also government departments staffed by civil servants, but they do not have their own minister. They are accountable to Parliament through their sponsoring ministers; in the case of HMRC and NS&I, these are Treasury ministers. Their powers are generally (but not always) derived from statute. Where there is a statutory board, appointments to it are usually made by ministers.

Executive agencies are part of a government department. They are defined business units headed up by a Chief Executive, who is often supported by a management board. Executive agencies carry out executive functions, with policy set by ministers. They operate with a degree of autonomy from ministers and the “parent” department. They typically deliver a service, and may provide services or functions on behalf of other government departments and the devolved administrations. They do not have a separate legal personality. They, too, are staffed by civil servants. Ministers do not concern themselves with the day to day running of executive agencies but are directly accountable to Parliament and the public for the overall performance of the agencies and for their continued existence. Executive agencies publish their own annual report and accounts, but those accounts are consolidated into those of the parent department.

HMRC is the UK’s tax authority, responsible for administering the tax system, collecting tax, and providing families and individuals with targeted financial support. It is a NMD, due to a long-standing convention that politicians should not be directly involved in the tax affairs of a person. Its day to day running is overseen by the Commissioners for Revenue and Customs, appointed by the Queen. In the exercise of their functions, the Commissioners must comply with any directions of a general nature given to them by HMT. HMRC is accountable to Parliament through the Financial Secretary to the Treasury, and ultimately through the Chancellor, as well as through one of the Commissioners for Revenue and Customs in her role as HMRC’s Accounting Officer, Permanent Secretary and Chief Executive.

In broad terms, the Chancellor and other Treasury Ministers set the tax policy that HMRC then administers. However, HMT and HMRC have dual responsibility for the design and delivery of tax (and tax credit) policy through what is known as the “Policy Partnership”. HMT is responsible for tax strategy and for leading tax policy development. It sets general objectives that help to define HMRC’s remit. The Chancellor also instructs HMRC to take a role in the design of tax policy, working with HMT to ensure that it is designed in a way that reflects HMRC’s experience of customers and their behaviour. HMRC is responsible for policy maintenance (including the identification and closure of tax avoidance strategies) and for the delivery of tax policy changes, as well as for the administration of the tax system.

Both departments support each other in their activities. Where HMT leads on an issue, HMRC provides input on technical and operational aspects of work, and helps to work up policy in detail. Where HMRC leads on an issue, HMT provides input to ensure that they take account of wider government objectives or policy. The Financial Secretary to the Treasury is involved on a day to day basis with how HMRC operates, and regularly meets HMRC officials to discuss the administration of the tax system.

NS&I is a NMD which became an Executive Agency of the Chancellor of the Exchequer on 1 July 1996. Its parent department is HMT. NS&I’s primary function is to provide cost-effective financing to the Government by issuing and selling retail savings and investment products to the public, with the overall aim of helping to reduce the cost to the taxpayer of government borrowing. The revenue from the sales of those products is passed directly to the Exchequer. As the Government’s retail debt financing arm, the powers governing the way in which NS&I is structured and managed are derived from specific legislation, including the National Savings Bank Act 1971 which created the office of the Director of Savings, a senior civil servant who is appointed by the Chancellor as NS&I’s Chief Executive. The functions and powers of the Director of Savings are derived from statute. The Chancellor delegates the management of NS&I to the Director of Savings, who is supported in her role by the NS&I Executive Committee, its executive decision-making body.

The Chancellor bears the ultimate statutory responsibility under the National Savings Bank Act for determining the policy and financial framework within which NS&I operates, setting and monitoring its key performance targets, approving its interest rates and the terms and conditions of NS&I products, and agreeing its annual corporate plan; but he may delegate these responsibilities to another Minister within HMT. The Chancellor has delegated ministerial responsibility for NS&I to the Economic Secretary to the Treasury (“the Minister”). All strategic decisions affecting NS&I products require Ministerial consent. The Director of Savings is accountable to the Minister, and through her and the other Treasury Ministers, accountable to Parliament for the performance of NS&I.

As well as appointing the Chief Executive (i.e. the Director of Savings), the Chancellor appoints four non-executive members to the NS&I Board. There are six executive members of the Board, one of whom, Mr Dax Harkins, a member of the Executive Committee, gave evidence. Two officials from HMT also sit on the Board – the Director of Fiscal Policy and the Deputy Director for Debt and Reserves Management. The Board provides advice to the Director of Savings and the Executive Committee on issues within its remit, such as strategy and the deliverability of policies. It also has a supervisory role, in that it scrutinises reporting from NS&I on its performance, and challenges it on how well it is achieving its objectives. However, the Board has no say on policy. As the NS&I framework document of August 2009 puts it, the Board’s focus is upon “getting policy translated into results.” Following the strategic direction agreed upon by the Minister, with input from the Board, it is the Executive Committee that oversees and develops NS&I’s business.

Each government department must have an Accounting Officer who is accountable to Parliament for the proper, effective and efficient use of public funds, and the proper custody of assets which have been publicly funded. The Director of Savings is appointed by HMT as Accounting Officer for NS&I. She is therefore accountable for the NS&I annual and product accounts, and may be required to appear before the Public Accounts Committee.

Back in 1997, NS&I invited expressions of interest from potential bidders for a contract to run all its “back office” operations – i.e. the operations required to process its product transactions and provide its customer service. One of the key reasons behind the decision to outsource to an external provider was that it was believed that operational experts within the market would be better placed to achieve the efficiencies and modernisation that NS&I wished to undertake. Following a competition, in 1999 NS&I awarded the contract to Siemens Business Services, which was then renamed Siemens IT Solutions and Services (“SIS”). NS&I transferred its existing operational staff (approximately 4,200) to SIS under a TUPE transfer with effect from 1 April 1999. SIS became responsible for delivering NS&I’s transaction management, customer service, printing, accounting, IT development and management, and all other operational services. However NS&I retained (and still retains) a core body of civil servants, currently around 170 in number, who together with the Director of Savings have responsibility for strategy, branding, pricing and policy decisions (subject to Ministerial approval).

The outsourcing arrangement proved to be very successful, and resulted in substantial cost reductions (estimated at around £530 million over the life of the contract). The technical expertise acquired in consequence of the outsourcing enabled NS&I to grow and improve its services to a significant extent. In 1999 all transactions were completed on paper and there was very little automation. NS&I had no website, and online banking was in its infancy. Now, NS&I’s online customer interactions are around 46 million per year, and the average time to manage a customer transaction has reduced from 9.2 days in 1999 to 3.1 days in 2014. Over the same period the value of investments with NS&I has grown from some £60 billion to over £105 billion.

The contract with SIS was for an initial period of 10 years, extendable (and extended) for a further 5 years. In 2011, Atos purchased SIS, and the staff working on the NS&I activities who had previously transferred to SIS became Atos employees under a further TUPE transfer. Thus at the time when the re-tender process for the Outsourcing Contract commenced, Atos had just become the incumbent contractor.

At the Spending Review 2010, HMT and NS&I agreed that NS&I’s remit should be expanded by leveraging its operational capability and capacity to provide services (specifically, account management, payment processing, and ancillary administrative services) to other public bodies in return for remuneration. This would enable other government departments to benefit from NS&I’s existing expertise in taking, holding and transferring money securely to large numbers of people, as well as from the efficiencies associated with NS&I’s operational activities. It would also allow NS&I to gain a contribution towards the costs of its infrastructure, which it could offset against its Departmental Expenditure Limit. NS&I refers to these types of arrangements as business-to-business or “B2B” services, and they are delivered under the brand “NS&I Government Payment Services”. The amount of income that NS&I can retain from B2B activity is set by HMT at each annual spending review.

The first B2B services NS&I delivered were the administration of the Court Funds Office (CFO) service for the Ministry of Justice, and the delivery of the Equitable Life Payment Scheme (ELPS) for HMT, which were both launched in 2011. The CFO service is a banking and administration service provided to all civil courts in England and Wales, in respect of money paid into court, including payments made on account of awards of damages in civil actions to vulnerable claimants such as children, or people who have suffered catastrophic personal injuries. NS&I’s remit includes the processing of payments into and out of court, and looking after the investments made with that money. Such payments may, of course, be withdrawn by the party paying the money into court in certain circumstances, though the money may also be paid out to someone other than the person who paid it in. The ELPS scheme is a compensation scheme which involves tracing qualifying recipients, validating their identities, and making payment of compensation to them. NS&I was given the power to deliver those B2B services by specific provisions in the Financial Services Act 2010 and Equitable Life (Payments) Act 2010 respectively. MOUs have been entered into by NS&I with each of the relevant government departments to whom those services are provided.

Thereafter, in order to obviate the necessity for the enactment of primary legislation each time NS&I added a new B2B service, Section 113 of the Financial Services Act 2012 gave NS&I a broad general power to provide services to any public body. That section empowers the Director of Savings to enter into arrangements with a public body for the provision by the Director, or persons authorised by the Director, of services to that body, on such terms, including terms as to payment, as may be agreed. There is nothing in the language of s.113 to restrict the services to be provided to deposit-taking, as Edenred submitted, and it would make no sense to construe s.113 so narrowly, given the background against which it came to be enacted, especially the policy of expansion of NS&I’s B2B services. In any event, clause 16 of the Childcare Payments Bill, which has since been enacted as the Childcare Payments Act 2014, puts beyond doubt the power of NS&I to deliver TFC in co-operation with HMRC.

Account functionality for NS&I was provided by SIS (and is now provided by Atos) using the Thaler Banking Engine (version 2). This is a sophisticated banking software application provided by a company called Sopra, which processes and keeps records of sales and repayments, customer account balance details and the history of transactions. It is this banking engine and its supporting functions which provide the core infrastructure that is intended to be used in the provision of B2B services. Between 2008 and 2014 NS&I expended over £53 million in upgrading the banking engine. It is therefore unsurprising that when NS&I invited bids for the current Outsourcing Contract in 2011 it required bidders to use its existing banking engine to deliver the services, rather than some other system, and to upgrade it to the latest version (version 3). Consequently Atos has entered into a sub-contract with Sopra for the delivery and servicing of the Thaler system.

The contract with SIS could not be extended beyond its 15 year term. On 11 July 2011, an industry day was held to generate interest in the re-procurement of the outsourcing contract. It specifically included support for the expansion of NS&I services, and a presentation was given which described NS&I’s general B2B requirements, the current delivery of the ELPS and CFO services, and the potential for growth of B2B services across government. The HMT representative said that:

“For its part, NS&I, along with the rest of the public sector, has made efficiency savings as part of the Spending Review. For its part the Treasury agreed to give NS&I freedom. Freedom to go out and win new business from the rest of government – what NS&I calls its “business to business” agenda. We agreed that NS&I’s operational capacity, its track record on delivery, its management capability to make change happen was strong. And other parts of Government could benefit from that approach. On top of that benefit, the “business to business” will also reduce the cost to the taxpayer of NS&I’s operations”.

80 companies (many of whom attended the industry day) expressed an interest in this opportunity via an online tool that NS&I used for all communications relating to the procurement. None was a member of the CVPA.

In compliance with reg 5(1) the notice commencing the procurement process for the Outsourcing Contract (“the OJEU Notice”) was published in the OJEU on 22 November 2011. It is entitled “UK-London: banking services” and under “Section II: Object of the contract”, the title attributed to the contract by the contracting authority is “outsource services” and the service category is No 7, “computer and related services”. The OJEU Notice stated, inter alia, as follows:

"We pride ourselves on delivering excellent customer service and this, together with the security we can offer to customers, forms the key element of our customer proposition. NS&I outsourced its operational services in 1999 and is now seeking to retender these operational services, including all processing of customer interactions and servicing (eg sales, after sales management and payments including via telephone, internet and mail); service management; IT management and implementation; and other services (eg. complaint handling, channel management, customer management, print and document management, customer market research and analysis, campaign management, compliance, management information etc) and other related ancillary services that support the business operation of NS&I. In addition NS&I now delivers similar operational services (called B2B services) to other public sector organisations. We intend to expand this B2B service during the lifetime of the contract to deliver to other organisations, potentially resulting in significant growth of the outsourced operational services. NS&I intends to structure the contract so that it may be used by other central government departments (including their executive agencies and non-departmental public bodies) and by local authorities." (emphasis added).

The OJEU Notice indicated that the value of the contract was in a range of between £1,250,000,000 to £2,000,000,000, depending on the uptake of B2B services. As Mr Harkins explained, the intention was to provide significant headroom over the expected charge for the core retail service (£660 million) to enable the planned growth of B2B with the addition of services for other government departments. The OJEU Notice also stated that NS&I would be following a competitive dialogue process in which it would have interactive sessions with bidders so that they would be able to build their understanding of the requirements, including B2B services.

Mr Harkins, as the B2B Director of NS&I, was responsible for the procurement of the Outsourcing Contract. He was responsible for leading the overall programme and involved in its day to day detail, including engaging in detailed dialogue with bidders. In his evidence Mr Harkins explained that, given NS&I’s revised strategy and broader remit within government, the provision of existing and future B2B services was a key part of the re-tender for the outsourcing arrangements. As with the SIS contract, the new Outsourcing Contract was procured on the basis that a single provider would act as prime contractor (with other economic operators as sub-contractors to the prime as necessary). The prime contractor (with any necessary assistance from sub-contractors) had to be able to support the full scope of NS&I’s services, including future B2B services, and therefore not only had to be able to provide the back office functions relating to NS&I’s core savings and investment business, but the necessary resources to support the roll-out of B2B services to other government departments.

The first phase of the bidding process, the Pre-Qualification Questionnaire (PQQ) identified those bidders who had the experience, financial strength and capability to deliver the required services in the way NS&I wanted. One of the pre-qualifying requirements to bid for the Outsourcing Contract was a £1 billion annual turnover. Ten organizations submitted responses to the PQQ on 23 December 2011; none of these suggested that they wanted to use partner organisations in their bid (either as part of a consortium, or as sub-contractors) to specifically support the expansion of the B2B services during the lifetime of the contract (initially 7 years). After those responses were independently evaluated and moderated by NS&I’s Executive Committee to select the shortlisted candidates, three candidates, including Atos, were shortlisted and taken through to the next stage.

NS&I then conducted a competitive dialogue with the shortlisted candidates, and issued them with a full draft contract and due diligence material. B2B services were discussed over the three rounds of dialogue during the weeks commencing 1 May, 19 June and 23 October 2012. According to Mr Harkins, the focus of these sessions was on expanding NS&I’s B2B services to other government departments. Following the competitive dialogue process, NS&I issued the final contract and scoring guidance to the bidders and their final offers were submitted on 1 February 2013. The offers were then evaluated and an overall preferred bidder was identified.

As Mr Harkins explained in detail in his third witness statement, the bidders were required to set out how they would run the core infrastructure (including the Thaler banking engine and supporting functions) not only for NS&I’s retail services but also for delivery of B2B services. Atos received the highest score in all three evaluation components, (solution, commercial and financial) all of which included B2B services, as well as the highest score overall by a substantial margin. The recommendation that the contract be awarded to Atos was approved by the relevant minister, and the Outsourcing Contract was signed on 20 May 2013 with a service commencement date of 1 April 2014.

Mr Harkins accepted that NS&I is dependent on Atos to deliver the services that it has outsourced to them. He also accepted that they work closely together, that Atos would be involved in discussions about the strategic and operational development of the business, particularly as regards B2B, and that there is an Atos representative on the Executive Committee, but he stressed that the Director of Savings still has the ultimate control over the decisions on these matters. It is entirely understandable, given the nature of the Outsourcing Contract and the basis on which it was put out to tender, that Atos would be closely involved in such discussions; if a B2B opportunity arose, there would be little point in NS&I pursuing it without the provider of the outsourced functions being closely involved. However, the closeness of the working relationship with Atos does not mean that NS&I has ceased to be a wholly public entity. It remains a government department. It remains publicly accountable for the money it spends.

Edenred (previously known as Accor Services) is part of the multi-national Edenred Group, which was created in 2010 by a demerger from the well-known French hotel group, Accor, and has a presence in 41 countries worldwide. Edenred’s French parent company is listed on the NYSE Euronext Paris, and has a market capitalisation of approximately €5.3 billion. The Edenred Group specialises in providing prepaid corporate services. It designs and manages a wide range of solutions that improve the efficiency of organizations and purchasing power to individuals. The Group’s 2013 consolidated financial statements, the latest full-year results available, show total revenue of €1,030,000,000 and a net profit of €171,000,000 for the year ending 31 December 2013.

Edenred is itself the parent of a number of UK subsidiaries. Thus, for example, prior to the demerger, Edenred acquired a company which specialised in providing luncheon vouchers to private and public sector organisations for the benefit of employees. CVL is the subsidiary which, as well as providing Childcare Vouchers, provides Eye Care Vouchers and Carer Break Vouchers services to corporate organizations, local and national government agencies and the public sector.

The most recent financial statements for Edenred that were adduced in evidence are also those for the financial year ending 31 December 2013. In the profit and loss account for that period its operating income was said to be £1,062,000 and its profit £9,776,000. The balance sheet for 31 December 2013 shows net assets (including pension deficit) of £72,570,000. CVL’s profit for the same period was stated as £8,332,000, but it is not known how much of that profit was derived from Childcare Vouchers and how much from the other aspects of its business.

Mr Patrick Langlois, Edenred’s Managing Director, frankly accepted in cross-examination that not even the Edenred Group as a whole would have met the financial requirement of a £1 billion annual turnover in the PQQ for the Outsourcing Contract with NS&I. However he contended for the first time in his third witness statement (and maintained in his oral evidence) that if the OJEU Notice had made specific reference to TFC or childcare accounts, Edenred would have participated in the bidding process. He added that he was sure that other CVPA members would have done so as well. He said that Edenred had “banking partners” and that it would have “entertained the idea of partnering through a consortium or a joint venture”.

Although I do not doubt the sincerity of Mr Langlois’ beliefs, in my judgment that is no more than wishful thinking coloured by hindsight. Mr Langlois’ evidence, particularly in his witness statements, was prone to hyperbole (particularly as regards the description of the impact that the decision to deliver TFC through NS&I would have on Edenred’s business) which led me to approach it with a degree of caution. Of course, as the OJEU Notice made plain, the tender of the Outsourcing Contract was not a tender for NS&I’s savings, investment and accounts products but for the IT and other operational support needed by NS&I to be able to carry on its business and offer B2B services to other government departments. That is why there is no reference to Premium Bonds or other products or services offered by NS&I on the face of the OJEU Notice. I accept it is possible that if childcare accounts had been specifically mentioned, Edenred might have been interested in participating in a tender, but that interest would have been restricted to delivery of the childcare accounts. It would not have been interested in providing the required services to support NS&I’s core retail business or any other B2B opportunities. It has no expertise whatever in the banking sector. It could not possibly have bid for the Outsourcing Contract by itself even if it had wanted to (which it would not have done). Its best chance would have been to try and interest a potential prime contractor to include it as a sub-contractor.

There was no evidence before me that a bank, or any other company with the wherewithal (financial and administrative) or the established track record to deliver all the services required by NS&I and which met the PQQ criteria, would have entertained the idea of joining forces with Edenred simply because one of the types of B2B services to be provided would (or might) be childcare accounts. Indeed it is far more likely that such an entity would have preferred to tender in its own right and keep all the profit for itself, as commercially Edenred would have had nothing of sufficient interest to offer it in return for a share of the profits.

Although Edenred had experience of running the ESC scheme and thus of receiving money to be paid to childcare providers from employers, apportioning those sums to relevant employees and making payments out to the providers, keeping records of registered childcare providers, and dealing with queries from parents, those are all administrative and logistical tasks that business processing outsourcing companies (“BPOs”) such as Experian, Liberata, Capita and Atos, as well as any bank, were well placed to carry out without assistance, input or advice from Edenred (or any other member of the CPVA). That experience in and of itself would be insufficient incentive for such an entity to bring Edenred on board even as a sub-contractor, particularly given the differences (including the vast difference in scale) between the ESC and the TFC schemes. I am reinforced in this view by the fact that none of the bidders for the Outsourcing Contract proposed bringing in an additional party to give advice in respect of the CFO or ELPS schemes, and Atos has not proposed bringing in anyone specifically to help it with the setting up and administration of childcare accounts.

Moreover, even if Edenred had wanted to bid as part of a consortium, there was no room for manoeuvre so far as a change to the existing banking engine was concerned. Mr Harkins accepted in cross-examination that, at the behest of HMRC, another instance (which he termed a “sort of duplicate version”) of the existing banking engine was being created for TFC, with the same functionality that has already been developed, which could be replicated and potentially used by a different TFC account provider in the future. It was pointed out by Mr Coppel that just as Atos had sub-contracted with Sopra for the Thaler banking engine, Edenred could have gone out and done the same. Whilst of course that is theoretically possible, Edenred would not have done so. Edenred (and its competitors) had invested considerable amounts of money, (though nothing like as much as NS&I) in developing their own online platforms – Mr Langlois mentioned a figure of £10 million. It is fanciful to suppose that they would willingly write off that kind of investment in order to familiarise themselves with, utilise, and pay someone else to upgrade, develop and maintain the Thaler software or a replicated version of it, in order to deliver childcare accounts, particularly if the only part of the Outsourcing Contract in which Edenred would have been interested in delivering was that which enabled delivery of TFC by NS&I.

Even if, which I do not accept as remotely likely, Edenred had managed to interest another party or parties to join in the bidding with it, then given the nature and strength of the competitors the chances of such a consortium even getting past the first stage of the bidding process are so remote as to be described as fanciful.

It is contended by Edenred that the proposed MOU between HMRC and NS&I is a “public services contract”. Alternatively, if it is not, Article 56 of TFEU applies because it creates an economic opportunity in the form of a licence or exclusive right to engage in an economic activity which is of cross-border interest. On behalf of the Defendants, Mr Moser QC’s response was that the arrangements made between HMRC and NS&I do not engage the Regulations or Article 56 at all, but alternatively if they do, they fall within one of the recognized exceptions exemplified by C/480/06 Commission v Germany [2009] ECR 1-4747 (known as the “Hamburg Waste” exception). In my judgment, Mr Moser is right on the first point, and thus I need not consider the exceptions.

The definition of a public service contract is a matter of Community law. It must be a legally binding agreement made in writing with a “person” who has a distinct legal persona from the contracting authority. The most recent EU Directive on public procurement, Directive 2014/24/EU, which incorporates the principles developed in a considerable body of CJEU case law, states that:

“(5)

It should be recalled that nothing in this Directive obliges Member States to contract out or externalise the provision of services that they wish to provide themselves or to organise by means other than public contracts within the meaning of this Directive. The provision of services based on laws, regulations or employment contracts should not be covered….

(31)

… the sole fact that both parties to an agreement are themselves public authorities does not as such rule out the application of procurement rules. However the application of public procurement rules should not interfere with the freedom of public authorities to perform the public service tasks conferred on them by using their own resources, which includes the possibility of cooperation with other public authorities.

It should be ensured that any exempted public-public cooperation does not result in a distortion of competition in relation to private economic operators in so far as it places a private provider of services in a position of advantage vis-a-vis its competitors.” (Emphasis added).

This reflects, in particular, the decision in C-84/03 Commission v Spain [2005] ECR 1-139 in which a challenge by the Commission to the legality of Spanish legislation on public procurement was upheld. One of the grounds of challenge was that the codified law excluded from its scope all co-operation agreements concluded between public bodies. The Commission argued that in order to show the existence of a contract, it must be determined whether there has been an agreement between two separate persons. Thus in principle, inter-administrative co-operation agreements between two (different) public bodies could fall within the procurement Directive, and the blanket exclusion in the Spanish legislation was unjustified. The CJEU agreed, on the basis that it was sufficient, in principle, if the contract was concluded between a local authority and a person legally distinct from it.

However, the fact that each of HMT, NS&I and HMRC are denoted as “contracting authorities” in the Regulations does not mean that any arrangements made between them will necessarily constitute a public contract (or a contract at all). None of the cases to which I was referred dealt with a situation in which one government department had made arrangements with another government department to share or utilise the latter’s resources. The definition of a “contracting authority” in the Regulations is extremely wide and embraces any minister of the Crown, any government department, each House of Parliament, and the Assembly for Wales. That is because different branches of Government may contract with private bodies in many different ways – for example if the chief executive is a statutory office holder, such contracts will normally be made in that person’s name. It does not follow from this that any arrangement entered into between two government departments which involves the provision of services by one to the other over and above the minimum value set by the Regulations has to be advertised in the OJEU and made open to public tender, any more than an arrangement made between two separate departments of the same local authority.

The evidence of Christopher Hall, the Director of Complex Transactions within the Crown Commercial Service, which is an Executive Agency of the Cabinet Office, was that it is increasingly common for government departments to share common services rather than duplicate them. Thus, for example, legal services to most government departments are provided by the Treasury Solicitor’s Department, rather than by each department employing its own lawyers. Likewise, if a specialist asset such as a computer system has been built to support a particular activity, it is better value for the taxpayer’s money for the Government to make the widest possible use of that asset, if the activity in question is engaged in by more than one government department. Mr Hall’s evidence, which I accept, is that these intra-government arrangements are not about competition with the private sector, but about government organizing itself in an efficient manner to ensure that public money is well spent.

In the light of the doctrine of Crown indivisibility I am not convinced that two government departments such as HMRC and NS&I could contract with each other, as they are both limbs of the Executive. Of course, as different personifications of the Crown, they can each contract with private parties, and with outside public bodies, but that is a different question from whether they have sufficient independent legal personality to contract with each other, or whether in truth they are all parts of the same public body. However, even if they could potentially contract with each other, (for example, as Mr Coppel contended, because the Revenue Commissioners and the Director of Savings are statutory office-holders) it would be wrong in principle to treat them as separate public bodies when considering, for the purposes of the Regulation or the Directive, whether an arrangement made between them to make use of, or to develop existing resources, is internal to government.

It is the norm for one department to “charge” the other for such services – the department procuring the services will pay the department providing them, out of its own budget. To ensure transparency and accountability of government funds, it is therefore common practice across government for the relevant departments to enter into a MOU specifying in detail the services to be provided, and the flows of money between them. These MOUs are not contracts, but set out what services those who are accountable to Parliament for the expenditure of public money can expect to receive in return for the charges levied on them by the provider of the services, and what those charges are. An MOU has to be capable of being torn up and replaced at a moment’s notice with no legal repercussions, in order to respond to changes in policy.

The MOU relating to the TFC scheme is to be entered into between the Director of Savings and HMRC at the same time as the proposed amendment to the Outsourcing Contract. The introduction to the MOU states that, as NS&I and HMRC are both government departments, there is no requirement for a contractual arrangement between them. The MOU is intended to outline the requirements underpinning the TFC service and the required standards to which the service delivery must conform. The Director of Savings will provide the TFC service in accordance with these requirements and service levels via [NS&I’s] Outsourcing Contract with the Provider [Atos], which underpins the MOU.

Paragraph 3.1 specifies that the MOU is not legally binding on the parties and does not contain representations on which either party may rely. It therefore contains no binding legal obligations. Paragraph 3.2 explains that it sets out how the Director and HMRC will work together in co-operation to discharge their statutory roles under the Childcare Payments Bill (when enacted) and effect the delivery of TFC via childcare accounts. The term of the arrangements runs from the effective date “and will expire at the direction of HM Treasury in accordance with its processes and governance”. That means that HMT, which is not even a named party, can bring the MOU to an end immediately. A contractual arrangement cannot usually be terminated unilaterally by someone who is not even party to it (assuming, as one must for the purposes of this argument, that Edenred’s contention that NS&I has a distinct legal personality from HMT is correct).

Thus even if NS&I and HMRC could contract with each other, a matter on which I need express no final view, it is clear that they have not done so. The Director of Savings (or NS&I of whom she is the Chief Executive) has no binding or enforceable obligation under the MOU to deliver the services to HMRC. HMRC has no binding or enforceable obligation to pay for them.

Edenred relies on the level of detail in the MOU (including the fact that HMRC’s operational requirements are set out over some 63 pages in Schedule 2), HMRC’s right to make deductions from the monthly service charges if the specified service levels are not met, the detailed charging structure set out in Schedule 5, and the provisions for interest to be paid by HMRC on any late payment of charges and for any profit from efficiency savings to be shared. There is no denying that in many respects the MOU resembles a detailed commercial contract. However, that is not what it is. It is a document that reflects the internal arrangements, including charging arrangements, made between government departments working together to deliver a policy set by HMT.

Although NS&I produced a tariff to HMRC, that reflected how much the TFC services would cost NS&I to deliver under the previously procured Outsourcing Contract with Atos (as amended by the Amendment Agreement). Ms Suzanne Newton, the Director for the TFC programme in HMRC, explained in her evidence why the original estimate of those costs increased as the detail and timetable for delivery of TFC developed. Mr Moser submitted that the tariff was not set commercially by NS&I, not only because it was dictated by the pre-existing terms of the Outsourcing Agreement, but because it had to be agreed to by its parent department, HMT, as well as by HMRC. I agree that the charging arrangements between NS&I and HMRC cannot be characterised as the equivalent to a commercial price freely fixed after an arms’ length negotiation. Although NS&I is a government department, and not a public company wholly owned by the Government, nevertheless, as in C-295/05 Asemfo [2007] ECR 1-2999, it was plain on the evidence that once HMT had decided with HMRC to utilise NS&I to provide childcare accounts, NS&I had no choice but to do so, and it was not entitled to fix freely the tariff for those services. HMT dictates to NS&I how much profit it can keep. As in Asemfo, NS&I can be described as an instrument of the general state administration, through which banking and payment services are provided. Members of the public cannot require NS&I to open particular types of bank account; it does so purely at the behest of the Chancellor, the Minister or HMT.

Case C/2206 Correos [2007] ECR 1-12175, a case relating to the Spanish post office, relied upon by Mr Coppel, addresses a very different scenario, not least because EU law only permitted certain strictly defined types of postal service to be reserved to a monopoly provider, and the Spanish legislation that was the subject of the reference in that case attempted to reserve the provision of postal services provided to the Spanish Government to a single state-owned provider, without drawing a distinction between reserved and non-reserved services. The services in that case were undoubtedly of a contractual nature, the Directive and the Regulations were applicable, and the only potential exception to the rules (that which was recognized in C107/98 Teckal [1999] ECR 1-8121) did not apply because the relevant entity did not carry out the essential part of its activities with the public authority which controlled it. Although the CJEU distinguished Asemfo, it did so on grounds which are of no relevance here. This case is not and never has been a Teckal case, and the Defendants have never sought to rely on that exception. This is not a case about a public body awarding services to a wholly-owned subsidiary. It is a case about co-operation between different departments of the same public body.

Mr Coppel submitted that the question whether a transaction is a “contract” in this context has to be resolved by looking at its substance rather than at its form; it will suffice if it has the economic consequences of a contract and there are binding legal obligations on the parties. Parties cannot conceal the award of what in substance is a public contract by structuring their arrangements in order to circumvent EU law duties: C-29/04 Commission v Austria [2005] ECR I-9705 at [42].

Mr Coppel also placed reliance on what was said by Hickinbottom J in R (Midlands Co-Operative Society Ltd) v Tesco Stores Ltd [2012] EWHC 620 (Admin) at [107]:

“… in considering whether the procurement provisions apply, one must have look at the whole of the arrangements between the contracting authority and the contractor; and, in particular, whether there is in reality a multi-stage award procedure which comprises in substance a unity which includes an obligation to perform works and is consequently subject to the procurement rules...”

However, this is neither a case of a “multi-stage award procedure” nor of a deliberate structuring of arrangements to get around the Regulations. Of course the way in which an arrangement would be treated as a matter of domestic law is not an end of the matter. One needs to look at the economic realities. However even under EU law, the arrangement has to be binding.

Edenred submitted that in substance there is a public contract here by reason of the fact that the obligations that will be contained in the Amendment Agreement to the Atos Outsourcing Contract are repeated and reflected in the Schedules to the MOU, and that in consequence of this and the fact that they are intended to be executed simultaneously, the Amendment Agreement and MOU should be treated as one economic transaction. However, I do not consider that the fact that the Amendment Agreement underpins, and is necessary to enable NS&I to deliver the operational services denoted under the MOU, and that the two arrangements will be made simultaneously, means that in reality HMRC is procuring Atos, rather than some other private entity, to deliver those services to it. I reject Edenred’s suggestion that NS&I is just a front for Atos. That is not the position either legally or as a matter of fact. NS&I will not be providing Atos’s services; it will be providing its own services as a bank to HMRC, using its back office functions that have already been outsourced to Atos following a legitimate public procurement process.

The fact that NS&I will be delivering the services for which it undertakes responsibility under the MOU, through human and other resources provided by Atos under the Outsourcing Contract, does not transform the MOU into a binding contract, still less does it mean that “in substance” Atos will be entering into a substantial new contract with a new customer, HMRC, with NS&I acting as a front. In substance, as well as in law, Atos delivers its operational services under the Outsourcing Contract and the proposed amendment to that Agreement to NS&I; Atos provides the operational means by which NS&I is able to fulfil its responsibilities to its B2B counterpart, in this case HMRC.

Edenred’s argument based on the timing and content of the Amendment Agreement and its reflection in the MOU really begs the question whether the proposed Amendment Agreement is a material variation to the Outsourcing Contract or not. In my judgment, if it is not a material variation, then the fact that instead of using its own civil servants to deliver the services NS&I will be using staff and other resources outsourced legitimately to Atos would not engage, let alone offend, the Regulations.

Mr Coppel relied upon what, at the time of the trial, was Clause 16(2) of the Childcare Payments Bill (now the Childcare Payments Act), which provides as follows:

“16 Account providers

(1)

Childcare accounts may be provided by any of the following—(a) the Commissioners for Her Majesty’s Revenue and Customs, (b) a person or body with whom the Commissioners have entered into arrangements for the provision of childcare accounts, (c) if the Treasury so determine, the Director of Savings (“the Director”).

(2)

If the Director provides childcare accounts, the Director must in doing so act in accordance with any arrangements made between the Director and the Commissioners with respect to the provision of childcare accounts.

(3)

Arrangements made between the Commissioners and a person or body within paragraph (b) or (c) of subsection (1) may include provision for the making of payments by the Commissioners to the person or body in respect of the provision of childcare accounts (and accordingly nothing in section 15(8) or (9) affects the inclusion of such provision in the arrangements)…”

Mr Coppel submitted that Clause 16(2) gave rise to a binding statutory obligation on the Director of Savings to comply with the terms of the MOU and that was good enough to transform the MOU into a “contract” for the purposes of the Regulations. I am not persuaded by that argument. Of course, a State cannot get round the public procurement rules by enacting a statute conferring monopoly rights to carry out relevant work on a private body instead of entering into a contract with that body; but that is not what was planned and it is not what has happened. I accept that regard must be had to public law obligations as well as private law obligations when determining whether an arrangement constitutes a “public contract,” but in my judgment the provisions of the Childcare Payments Act take that matter no further. This is not a case where the statute is the source of mutually binding obligations.

The Childcare Payments Act does not, and does not purport to, transform the nature of the MOU or make it an enforceable agreement. Clause 16 merely sets out who may provide childcare accounts. It puts beyond doubt the capacity of NS&I to do so, though it already has such capacity under s.113 of the Financial Services Act 2012. However the Director of Savings (representing NS&I) can only provide childcare accounts if HMT (not HMRC) so determines. That is what the statute says. Assuming that HMT has so determined, Clause 16(2) directs the Director to act in accordance with any arrangements she makes with the Commissioners of HMRC (her executive counterparts within that department). Those arrangements will either have contractual force, or they will not; those statutory provisions will not transform them into mutually binding obligations, thereby superseding what is expressly stated in Clause 3 of the MOU. Clause 16(2) says nothing about HMRC electing to procure services from NS&I – on the contrary, it uses the language of mutual arrangement, consistent with the MOU being about co-operation. Clause 16 does not impose any legal obligations on HMRC or confer on it the right to enforce NS&I’s obligations; at most it puts beyond argument the power of HMRC to pay the person who provides childcare accounts. Even if Clause 16(2) were to create an enforceable public law obligation on the Director, it is not enforceable at the behest of the other party to the MOU, HMRC. Nor does it transform the arrangement with HMRC reflected in the MOU. If anyone can enforce it, presumably it is HMT or the Minister – but they are already entitled to tell NS&I how to implement matters of fiscal policy. In short, if the MOU itself is not a contract, which it is not, the statute cannot and does not make it one.

The La Scala case upon which Mr Coppel relied (C0399/98, Ordine degli Architetti delle Province Di Milino and others v Comune di Milano [2001] ECR 1-5409) is not authority for the proposition that for the purposes of EU law, the enforceable obligation to provide the services in question may be derived from legislation requiring only one of the parties to comply with the provisions of an arrangement which is expressly agreed by the parties to it to be non-contractual and non-binding. In that case, the relevance of the legislation, which bound both parties, was that it dictated that the contract for the works had to be entered into by the municipal authority with a specified person.

The issue in La Scala was whether the procurement Directive precluded national (and regional) legislation in Italy which provided that, when implementation of an urban development plan requires construction works in order to provide community facilities (in this case, a theatre) the developer holding the building permit is to be responsible for carrying out those works, at his own expense, in return for exemption from payment of the amount due to the municipality in respect of the building permit, (unless the municipality decides to collect the payment instead of opting for direct execution of the works), without requiring any tendering procedure for the award of the works contract. The CJEU held that the Directive did prohibit such legislation, even though under the legislation the authority was obliged to contract with that developer (and no-one else) for the carrying out of the works – and thus on the face of it no-one else could compete. The developer was not obliged to do the work himself. Therefore in order to comply with the Directive the national legislation had to make provision for the municipal authority to require the developer, under the agreements concluded with him, to put the works out to public tender.

Whilst it is true that the legislation in question obliged the developer to carry out the building works, there was an enforceable contract for the works between the developer and the municipal authority, and the contrary was not suggested. The case did not involve any issue as to whether the contract gave rise to any enforceable legal obligations, nor did it concern the question whether a statutory obligation to perform obligations to another party, which that party is unable to enforce as a matter of law, engages the Directive or the Regulations.

Given that the MOU is not a contract as a matter of domestic or EU law, or in substance, the Regulations do not apply to it. Nor does Article 56 TFEU. The TFEU imposes directly effective obligations on public authorities to advertise and carry out a transparent procurement process for “economic opportunities” which do not take the form of contracts at all but, for example, take the form of a licence or exclusive right to engage in an economic activity which is of cross-border interest: “Such an authorisation is no different from a service concession in terms of the obligation to comply with the fundamental rules of the Treaty and the principles flowing therefrom, as the exercise of that activity is liable to be of potential interest to economic operators in other Member States”: C-221/12 Belgacom [2014] 2 CMLR 23 at [25-26], [33-34] and [44].

Edenred contended that the MOU has, at the very least, granted to “NS&I/Atos” a monopoly right (by way of authority or permission) to provide childcare voucher services under the TFC Scheme, even if it is not a “public contract” falling within the scope of the Regulations. The first reason why that argument must fail is that the MOU has granted no rights to Atos. Atos’ rights and obligations derive from the Outsourcing Contract, as amended, and are owed to NS&I. Secondly, NS&I does not derive its authority to provide the services from the MOU – it derives it from HMT or, at least to the extent that authority includes capacity, from statute. The MOU reflects an arrangement between NS&I to co-operate with HMRC in the implementation and delivery of government policy by providing certain types of bank accounts, which is what NS&I is empowered to do by statute, especially if so directed by HMT. NS&I neither needs nor obtains “permission” from HMRC; like HMRC, it is told by HMT that this is the way in which it has decided that TFC is going to be delivered. In any event the TFEU is not aimed at, and does not apply to what are, essentially, resource-sharing arrangements made internally between two government departments, any more than the Directive does. See also AG Quidnet Hounslow LLP v Hounslow LBC [2012] EWHC 2639 (TCC), discussed in paragraphs 121-124 below.

In substance and in reality what has happened here is that the Government has decided to deliver TFC itself, internally, rather than through an external provider. The arrangements between HMRC and NS&I do not constitute a public contract and there was no “opportunity” that was required to be offered to the market. The fact that the implementation of this decision is not confined to HMRC, but involves HMRC making use of another government department, NS&I, which is part of HMT, to set up and administer the childcare accounts, does not change the character or the substance of what is planned from an essentially in-house implementation of policy into the sort of external arrangement that would attract the requirements of the procurement Regulations. Nor does the fact that NS&I had already outsourced its back-office functions to Atos – unless the delivery of the services would not fall within the ambit of the existing outsourcing arrangement but would require what is essentially an entirely fresh outsourcing contract.

If NS&I was still using a body of civil servants to run its existing banking operations it would be absurd to suggest that another government department making use of NS&I’s infrastructure instead of building one from scratch itself would create the kind of unfairness at which the Regulations (and Article 56 TFEU) are aimed or offend the policy behind them. NS&I’s outsourcing arrangement with Atos does not alter the essential character of the arrangement with HMRC from internal to external. In my judgment, the Regulations are only engaged if delivery of TFC would involve a material variation of the Outsourcing Contract. I therefore reject the first ground of challenge.

EU law prevents significant new work from being added to an existing public contract by way of variation without a new tender process being held. If such variations were permitted, contracting authorities would be able to award what are in substance new contracts for services that had never been subject to proper advertisement or open, transparent competition, and economic operators could win procurements by bidding in an unrealistic fashion, and then racking up their prices once the contract was won.

A variation is material under EU law if it “…demonstrate[s] the intention of the parties to renegotiate the essential terms of that contract”: Pressetext at [34]. The test is applied with regard to the object and purpose of the EU procurement regime, namely, opening up public contracts to competition and preventing the distortion of competition. Thus, a variation is material under the principles explained in Pressetext if the change:

Where amendment is envisaged and expressly permitted under the terms of the tendered contract, the amendment will not constitute a material change to the contract (Pressetext, at [40], [60] and [84]) save where the amendment to the contract is of such magnitude that it is essentially the award of a separate, second contract (Commission v France, at [36]). In the present case the procurement of the Outsourcing Contract did expressly cater for contract changes to include the provision of future B2B services, up to a financial ceiling of £2 billion, so the key issue is whether the contract changes made by the Amendment Agreement to cater for the delivery of the necessary outsourcing services supporting childcare accounts was tantamount to the award of a separate, second contract to Atos. In my judgment, given the terms and scope of the OJEU Notice, it was not. Objectively there is no intention of NS&I and Atos to renegotiate the essential terms of the Outsourcing Contract; on the contrary, the intention is to implement those terms exactly as was envisaged when the contract was made.

The first question to be determined is whether the support services to be delivered by Atos to allow NS&I to deliver childcare accounts fell within the scope of the services described in the outsourcing contract opportunity that was advertised for tender, and awarded to Atos at the end of that fair and transparent competitive process, or whether the scope of the contract is extended to encompass new and different services. The services here are not new. They plainly fall within the scope of what was advertised, and what any reasonable bidder would understand to have been advertised for tender. The subject matter of the advertised contract and the criteria governing its award was clearly defined. The OJEU Notice made express reference to B2B services and explained what they were. It made it clear that NS&I intended to expand the B2B service during the lifetime of the contract to deliver to other organisations, potentially resulting in “significant growth” of the outsourced operational services. It made it clear that the contract was to be structured in a way that would enable other central government departments to make use of the services that were being put out to tender and it made it clear that “the contracting authority is purchasing on behalf of other contracting authorities”. If any bidder was in doubt about what NS&I did, and what sort of services it might be required to support if NS&I arranged to provide them to or for or with another government department, the Notice gave details of how it could access further information on-line including the PQQ and the Prospectus.

It is important to bear in mind that childcare accounts are just another type of bank account into which money is deposited and from which payments are made. They may be more complex in some respects but that does not alter their essential nature. Like NS&I’s core products, they will be opened by members of the public and deposits will be made into them by members of the public. Unlike the retail savings accounts, where the deposits are essentially loans to the Government and interest is paid for and on behalf of HMT, the deposits will be topped up by payments from HMRC – but in both cases money is flowing into the accounts from the Government. Like some of NS&I’s core products the accounts can only be opened for a restricted category of persons – in this case, children with parents who meet the criteria. Like the CFO accounts, payments out may be made to someone other than the person who deposited the money, though the categories of payees will be restricted. In the case of CFO the payee may be the recipient of awards of damages or costs, whereas in childcare account cases it will be a registered childcare provider. As in the ELPS scheme, NS&I will also be handling money paid by a government department, in this case HMRC, to be paid out to a restricted category of payees. However the childcare accounts appear to me to be closer in nature to the core business of NS&I than either of those pre-existing B2B services.

Under the Outsourcing Contract, Atos will be providing the staff and call centres to NS&I to run and service these new bank accounts, just as it provides staff and call centres to NS&I to run and service its other products; and it will be providing the necessary IT, and developing the necessary software which will be based on the existing banking engine, to meet the more complex requirements of accounts with the features that I have already described. The banking, financial information systems and financial transaction processing and clearing-house services that are required of Atos under the Amendment Agreement fall within the 50 different specific types of service that are listed in the OJEU Notice using common procurement vocabulary; they did not become “new and different” services because they relate to a new type of bank account or because the bank account is a B2B service being provided by NS&I to HMRC.

In reliance on Commission v France and C-423/07, Commission v Spain, Edenred contended that the childcare account services were not advertised and that the OJEU Notice was insufficiently specific. Mr Coppel pointed out that the original tender in Commission v France for the first phase of the project (the feasibility study) embraced the possibility of further clearly defined services being provided in the second phase, and yet, although there was specific reference to the nature of the particular works in the second phase, that was not good enough to avoid the need for a further procurement exercise. In the present case, he submitted that the future B2B services adverted to in the OJEU Notice were far too generic, and even less clearly defined than the specific services in Commission v France. The same was true of Commission v Spain, which related to the construction of two new links in the A-6 motorway from Madrid to La Coruna. The Commission’s objections in that case were chiefly based on the fact that the successful tender included works that were not even mentioned in the specifications that the Spanish Government had advertised. Moreover the wording of the specifications was not such as to lead participants to suppose that they could propose additional works on a scale and in a location comparable to the scale and location of the additional works proposed by the successful tenderer, which already held the concession for work to a different section of the same motorway.

Mr Coppel relied in particular on paragraph 70 of the judgment:

“…it does not satisfy Directive 93/37 when, without any transparency, a public works concession contract is awarded which includes works referred to as “additional” which of themselves constitute “public works contracts” within the meaning of that directive and the value of which exceeds the threshold laid down therein”.

In my judgment that is just another way of expressing the principles in Pressetext that the Regulations and Directive cannot be evaded by making a variation to an existing public contract even if variations are allowed. A public authority cannot justify awarding an existing private contractor what is in substance a new public services contract to do works that were not part of the original specification, on the basis of a provision in the first public services contract that envisages that additional works of an entirely unspecified nature might be required in the future.

However, those arguments blur the distinction between the B2B services, which are to be delivered by NS&I to other government departments, on the one hand, and the outsourced support services which are to be delivered by the successful bidder, in respect of NS&I’s core business and any existing and future B2B services, on the other. The latter are the services with which this case is concerned, and they were very clearly (and in my view sufficiently) defined in the OJEU Notice. The nature of those services has not changed. They were part of the original specification. They were properly advertised. No actual or prospective bidder could have been in any doubt as to what it was required to deliver to NS&I.

In Commission v France the services to be provided at phase 2 were clearly different from those to be provided at phase 1. All that the bidder was being invited to bid for was the services encompassed in phase 1, and thus it could not be treated as also having successfully bid for the services expected to be included in phase 2. Likewise in Commission v Spain the works were to completely different parts of the same motorway. Cases of this nature are extremely fact-sensitive. The Outsourcing Contract is not of a similar nature. It does not relate to any particular phase of a development. It relates to all operational services that NS&I requires to support its business. The 50 different types of services falling within the ambit of the general description of “outsource services” are separately and independently identified, ranging from 66110000 (“banking services”) to 79512000 (“call centre”) to various specific types of software packages (financial analysis and accounting software package, statistical software package, Customer Relations Management software package). The situations addressed by the CJEU in Commission v France and Commission v Spain are simply not analogous.

In this case, therefore, the Amendment Agreement will not vary the services that were put out to tender by NS&I, by introducing new and different services. The nature of the operational services that Atos will be providing to support the delivery of childcare accounts, is essentially the same as the nature of the services which are supplied by it to NS&I for existing banking, accounting and payment products and which would have to be supplied for any new product delivered by NS&I, whether or not it was a new type of savings account to raise money for HMT or a bank account to be utilised by another government department – or a payment service offered to another government department akin to the ELPS.

If as part of its core services NS&I offered a new type of savings account to the public, an amendment to the Outsourcing Contract to cater for setting up and administering that new account would not have to be advertised and tendered separately. Mr Coppel accepted this, on the basis that those were NS&I’s core services. However the same must be true if a future B2B service is a banking, account and/or payment service which is what NS&I’s business entailed, and was known to entail, at the time of the OJEU Notice. I do not accept that in order to comply with EU law the OJEU Notice would have had to spell out in detail that the bidder might have to provide the 50 specified types of administrative service in support of the delivery by NS&I of identified and specified B2B banking, account or payment services to identified government departments. Nor do I accept that provision of those administrative services in support of such B2B services entailed additional or new works.

On reading the OJEU Notice it was plainly envisaged that the successful bidder would be required to provide all the specifically described types of “back office” services that NS&I was offering to outsource, not only to support the delivery of NS&I’s core retail business, but also in order to support and enable NS&I to provide bank accounts and payment services to other government departments under existing and future B2B contracts up to the envisaged financial ceiling of £2 billion. There is no dispute that the value to Atos of the additional services under the Amendment Agreement would fall comfortably within the financial headroom allowed in the OJEU Notice. Any potential bidder knew, or had the means of finding out by registering an interest, what the existing B2B services were, and thus what future B2B services were likely to be – but they were not the subject of the tender; the operational and technical support services to enable NS&I to deliver them were.

The bidder would have known that, if successful, it would be expected to deliver “processing of customer interactions and servicing (eg payments including via telephone, internet and mail); service management; IT management and implementation; complaint handling…..customer management, print and document management and other related ancillary services that support the business operation of NS&I” in respect of any bank account or payment service that NS&I was required to provide to another government department during the future term of the Outsourcing Contract, especially if (like the core services) this involved interaction with members of the public. That was good enough. There was an open competition for the provision of these services and that competition was transparent.

If the B2B service that NS&I was going to provide was not a banking or account or payment processing service then Edenred’s objections may have had some force, but any bidder for this Outsourcing Contract would have known that it was bidding for the opportunity to provide exactly the same kind of administrative support to NS&I if NS&I were to open and service bank accounts for another government department, as it would have to provide if NS&I opened new types of savings accounts to raise money for HMT. There was no necessity to spell out what sort of bank account it might be and identify each and every other government department it might be delivered to.

Edenred’s argument was tantamount to a contention that the Outsourcing Contract could not lawfully be used in the manner in which it was plainly intended and flagged up in the procurement process that it should be used, in accordance with the strategy agreed in the 2010 Spending Review, namely, to enable NS&I with the support of the successful bidder to deliver bank account and payment services to other government departments, just as it had provided CFO and ELPS under the previous Outsourcing Contract for the MOJ and HMT. In other words NS&I could never offer its banking, account or payment services to another Government department without going through a fresh public procurement process for the outsourcing support necessary to enable it to do so. Mr Coppel did not shrink from adopting that proposition, though he tried to soften it by suggesting that there were other ways in which the Government might have legitimately achieved the same objective. However, I do not accept the proposition.

If a bystander had asked one of the entities that attended the industry day, or read the OJEU Notice, “suppose that in future, during the term of the advertised contract, NS&I had been tasked with providing bank accounts in the names of children that would be used by HMRC and parents to jointly fund payments for their childcare. If you were awarded this contract, would you be required to provide all the administrative services specified in the OJEU Notice to enable those accounts to be opened, operated and serviced?” the answer would plainly have been “yes, of course”. The entity might well have added “that is the whole reason why, as we understand it, NS&I have said this contract could be worth up to £2 billion, because if we win, we will have the opportunity to deliver to NS&I the back office support for future B2B services of that type”. Therefore, if Edenred had read the OJEU Notice or attended the industry day, it would have known enough to appreciate that if in future the Government decided to deliver childcare accounts through NS&I instead of putting them out to tender, the only way that Edenred would be able to get involved would be by making (or participating in) a bid for that contract.

If, contrary to those findings, there has even been a variation of the services to be provided from those that were advertised, the variation is not material. The Amendment Agreement meets none of the characteristics of the three examples of material variation given in Pressetext. The services are not new or different in any material respect. The Amendment Agreement does not introduce conditions into the Outsourcing Contract which, had they been part of the initial award procedure, “would have allowed for the admission of tenderers other than those initially admitted or would have allowed for the acceptance of a tender other than the one initially accepted.” It is worth noting that the language used by the CJEU in Pressetext is “would have allowed for” not “might have allowed for”. If the services in the Amendment Agreement to support the delivery of childcare accounts had been expressly included in the initial award procedure, would this have allowed for the admission of any other bidder, or for the acceptance of a tender by someone other than Atos, or attracted additional participants in the bidding process? In my judgment it would not, because it would not have widened the range of potential bidders beyond those who expressed an interest in the first place.

I fail to see how the express introduction of a reference to administrative services to support the delivery of childcare accounts by NS&I at the inception would have had any bearing on the tender process at all. Any other hypothetical tenderer would still have had to be able to deliver all the services required by NS&I and to meet the financial PQQ requirements. Edenred could do neither. It knew that. NS&I were advertising for a prime contractor (with any necessary specialist support to be provided by way of sub-contracting). It is wholly unrealistic to suppose that adding childcare accounts to the other types of accounts that needed to be serviced and supported would somehow have engendered new bids for the entire Outsourcing Contract from entities who were only interested in being involved in providing the childcare accounts themselves, and whose core business was not an IT or operational or technical support business.

The original bidders who could provide the remaining services (as well as the operational services needed to enable NS&I to provide childcare accounts) did not need a CVPA member on their team to be able to bid for this contract. They would not have been interested in bringing a CVPA member into a consortium or joint venture company to make a bid for services that they would obviously be better placed to deliver by themselves. If they were entitled to, and prepared to sub-contract some of the services to someone like Edenred, then they would probably do that after they had won the bid (though Atos has not) but it is unrealistic to suppose that they would have been prepared to sub-contract something they could easily do themselves. It would have been uncommercial. There is no evidence that any entity who did not bid would have leapt at the chance of bidding together with a CVPA member as part of a consortium but was dissuaded from throwing its hat in the ring because it was unaware that childcare accounts were part of the package, and the inclusion of childcare accounts would have made all the difference provided it could get a CVPA member to join in and deliver those services.

In short, if childcare accounts had been on the agenda and mentioned from the start as a potential or actual B2B service for HMRC, I am not satisfied that anyone else who could have done so would have bid for the Outsourcing Contract or would even have been attracted to make a bid. The truth is that all the original bidders were the types of entity that could provide the services required under the Amendment Agreement without calling on a childcare voucher provider. The same entities would have bid, and the same three entities would have been shortlisted.

Mr Coppel’s position was that it was enough for Edenred to show that if the services in the Amendment Agreement were included from the outset, then hypothetically other bidders (not necessarily Edenred) would have been admitted or would have been allowed to have been admitted or would have wished to have been admitted. However, in my judgment the examples of material variation given by the CJEU have to be interpreted as examples of scenarios in which, in substance, a new contract has been concluded, unfairly conferring a competitive advantage on the existing contractor over someone else who would have participated in the process. There would be no such unfairness, and no distortion of competition, if no-one else would have bid or if the complainant’s putative bid would never have got off the ground, which is the case here.

Mr Moser pointed out that in Pressetext itself, the CJEU held that an amendment to the basic agreement renewing, for a three year period, a waiver of the entitlement to give notice before the term of the contract expired was not material. It held that there was no economic incentive for the relevant state (Austria) to change to another service provider during the waiver period, and that the contractual partner was able reasonably to assume that in the waiver period there would be no equivalent offers under more favourable conditions such as to justify the expenditure entailed by making such a change (emphasis added). The contractor was therefore not favoured over other possible service providers, by reason of the amendment, since there was no risk of a distortion of competition. Thus if there is no risk of a distortion of competition, the change cannot be regarded as material.

Mr Moser used that example from Pressetext as the foundation for a submission that it was not good enough to meet the test of materiality for Mr Coppel to establish that, purely hypothetically, if childcare accounts had been specifically adverted to at the time of the procurement of the Outsourcing Contract, someone other than the entities who did bid might have come forward and expressed an interest in bidding for the Outsourcing Contract, either on their own or in conjunction with someone else. In support of that proposition he also relied on the decision of Coulson J in AG Quidnet Hounslow LLP v Hounslow LBC [2012] EWHC 2639 (TCC). That was a case about Article 56 TFEU, rather than the Regulations, where the issue was whether the economic opportunity, though entirely falling within the territory of one state, would have engaged cross-border interest.

At [69]-[72] the judge discerned a conflict between two strands of European authority. One strand, exemplified by C-108/98, RISAN Srl v Comune di Ischia [1999] ECR 1-5219 held that the mere fact that there might be third parties in other member states who might have been affected was insufficient to engage Art 56. The rationale underpinning the other strand was that if an undertaking in another member state might be interested in bidding for the work, but it does not know about the opportunity to do so, it will be deprived of that opportunity. Yet in C-231/03 Coname [2005] ECR 1-7287, the CJEU deliberately refrained from reconciling the two strands in favour of preferring the latter over the former, as the Advocate General in that case had suggested; and the subsequent case C-245/09 Omalet [2010] ECR 1- 13771 is authority for the proposition that the approach in RISAN Srl remains correct.

Coulson J resolved that conflict by resorting to first principles, and at [74] he found that there was no evidence that any undertaking in any other member state was in fact interested in the development of the relevant site, therefore Art 56 was not engaged. Thus it had to be shown that there was a potential tenderer in another member state who would have made a bid, and not just that it was hypothetically possible that someone in another state might have wished to tender for that project. Mr Moser submitted, by analogy, that for the purposes of the Directive it must be necessary for Edenred to satisfy the court that it, or another CVPA member, would have bid for the Outsourcing Contract if it had been made clear that the operational services specified in the Amendment Agreement supporting the delivery of childcare accounts was one of the future B2B services that NS&I might provide. In order to have bid, of course, it would have had to have satisfied the PQQ requirements.

There is much to be said for the approach taken by Coulson J of requiring evidence that someone beside the original bidders would have bid for the contract, because the EU procurement rules are designed to protect against real, not hypothetical, distortion of competition. However I do not need to decide the point, because even if one approaches the question on the basis that a hypothetical bidder has been shut out of the bidding process by the absence of reference to the subject-matter of the proposed amendment, it seems to me that in principle that must necessarily be a realistic hypothetical bidder – i.e. the evidence must demonstrate that there would be someone else who would have been ready, willing and able to bid and who would have wished to have done so if the opportunity had been made clear, but who did not do so because it was not.

I do not understand the CJEU in Pressetext to be saying that a variation is material if someone other than the original bidders might have wanted to bid for the Outsourcing Contract if childcare accounts had been mentioned, however unrealistic that person’s aspirations might have been. That interpretation would lead to the highly unsatisfactory conclusion that an amendment which in fact would have made no material difference to the number of actual bidders or the outcome of the bidding process had it been included at the onset would be regarded as a material amendment necessitating a separate procurement process (which in those circumstances would be a waste of time and money).

I have already made the finding that Edenred could not have bid by itself, as it had insufficient financial resources, and that I am not satisfied that any other economic operator who could have delivered the services in which Edenred was not interested, and who would have met the PQQ financial threshold, would have joined forces with it to make a bid, or would have wanted to do so. That rules out Edenred itself as a potential bidder. So far as evidence about other potential CVPA bidders is concerned, it was Mr Langlois’ view that one of Edenred’s competitors would have wanted to bid because it had expended millions of pounds in 2008 in acquiring a subsidiary whose business was providing childcare vouchers: but that is pure speculation, and even if it is reasonable speculation, it tells the court nothing about that unidentified competitor’s ability to make a bid, or to meet the PQQ threshold, as opposed to its desire to do so. Such a CVP would have faced exactly the same problems as Edenred in getting another economic operator to join forces with it.

The witness statement of Mr Julian Foster, the Managing Director of Computershare Voucher Services Ltd, served in the first judicial review proceedings, contains an equally bare assertion that his company would have participated in the procurement process as part of a consortium or as a sub-contractor, but there is no evidence relating to Computershare’s ability to meet the financial requirements in the PQQ or to find a partner or partners who could deliver all the other services that NS&I required. Mr Foster was not a witness in the Part 7 proceedings, although his statement was treated as being in evidence. The evidence in respect of Computershare is even less satisfactory than that in respect of Edenred. There are no financial statements in respect of Computershare, and as with Edenred, there is no evidence that any third party who had the capability to bid for the remaining services under the Outsourcing Contract would have been ready, willing and able to join forces with Computershare so as to enable Computershare to deliver the childcare accounts, instead of bidding in its own right and keeping all the profit if successful. If Edenred’s experience of delivering ESC would not be a sufficient incentive then neither would Computershare’s.

Thus no reliable evidence was placed before the court that there was in fact any detriment to any other putative tenderer or any distortion of competition by reason of the fact that childcare accounts were not specifically mentioned when the procurement exercise for the Outsourcing Contract took place. Indeed there is no reliable evidence that there would have been any other bids for that contract if they were mentioned, and it seems to me to be inherently unlikely that mention of one further species of bank account would have made any difference to the cadre of actual or potential bidders. Therefore Atos was not being placed in a position of competitive advantage over Edenred in that regard.

The final example of material amendment given in Pressetext concerned a change to the economic balance of the contract in favour of the contractor in a manner not provided for in the original contract. The Amendment Agreement did not change the economic balance of the contract in favour of Atos, let alone in such a manner. At paragraph 70 of its skeleton argument, Edenred alleged that the charging provisions under the Amendment Agreement would differ from those under the Outsourcing Contract but at that stage, no detailed reasons were provided. At the start of the trial Edenred made an application for further disclosure relating to this issue, which I allowed. Edenred then put in a “Supplementary Note on Economic Balance” towards the end of Day 2, on which Mr Moser took some instructions overnight and to which he raised strong objection the next morning. After hearing argument I decided to admit the Note, and allow the Defendants to provide a written response to it.

Edenred’s argument that Atos’ profit margin was greater under the Amendment Agreement was fatally undermined when it was pointed out by the Defendants (and subsequently accepted by Edenred) that it had mistaken the calculation of mark-up for the calculation of profit margin. This led to the basic error of dividing profit by cost to calculate profit margin, instead of dividing profit by the charges. As the Defendants were able to demonstrate, the (projected) profit margin shown in the baseline financial model for the Amendment Agreement is consistent with that in the baseline financial model for the main contract.

I gave Edenred’s legal team the opportunity to digest the implications of the admitted error, and put in some further written Reply Closing Submissions after the trial had ended. I also gave permission to the Defendants to respond in writing to any points raised in Edenred’s Reply Closing Submissions which Mr Moser felt that he had not sufficiently addressed in oral argument, and they did so. For the reasons set out in that response, Edenred’s remaining arguments on the economic balance point were just as fundamentally misconceived and equally based on factual misapprehensions.

Atos does not stand to make a significant profit under the Amendment Agreement above and beyond the profit it receives on its core services, as alleged. The contractual charging mechanism remains the same under the Amendment Agreement as it is under the original Outsourcing Contract. Without trespassing in this judgment more than is absolutely necessary into matters of commercial sensitivity which have been disclosed within a confidentiality ring, suffice it to say that the services to be provided under the Amendment Agreement will be charged on the original basis set out in Schedule 5, and not on some more advantageous “cost-plus” basis as alleged by Edenred. Thus it is simply wrong for Edenred to assert, for example, that Atos will be paid for the overheads and a 10% profit on top. Edenred has fallen into the further error of confusing Atos’ cost model projections with the charges for its services. The financial model for the TFC variation forecasts projected cost and thus projected price, but it is not the basis upon which NS&I will be charged by Atos.

The basis of charge remains that set out in Schedule 5. There are three elements; (i) a fixed operational charge, (ii) a variable FTE (full time estimate) operational charge and (iii) debit card and postage charges. The charges include a fixed element, so Atos’ profit cannot be guaranteed. The projected profit will only be achieved if the costs that are actually incurred are exactly in line with the calculations in the baseline financial model. Any increase in costs will be absorbed by Atos, not passed on to NS&I. Thus any changes in the cost base will directly affect Atos’ profitability. If the costs rise, Atos takes the risk of the erosion of its profit margin. Under the variable FTE charge, Atos also takes the financial risk of variations in volume (e.g. if more software licenses are required than has been estimated, or if salaries rise). Atos’ profit margin is susceptible to exactly the same risks under the main Outsourcing Contract and the Amendment Agreement.

The original contract and Amendment Agreement contain entirely normal provisioning for costs over-runs to deal with unforeseen items. A proportion of that contingency provisioning is attributable to a Time and Materials Contingency Budget for implementation, for which HMRC is responsible. That is provided so that HMRC can make provision in its own business case; if it is not needed, it will not be charged. The balance relates to risk provision for Atos against unforeseen items, the amount of which reflects the nature of the relevant activities and, in the case of TFC, the lack of historic evidence of volumes, speed of uptake and profile of traffic which makes it prudent to err on the side of caution. The existence of these contingency budgets does not, and cannot, confer any economic advantage on Atos under the Amendment Agreement or tilt the economic balance in favour of Atos.

In summary, Atos does not stand to gain any greater financial advantage from providing the supporting services for childcare accounts under the proposed Amendment Agreement than it does under the main Outsourcing Contract.

Edenred also sought to make a point about some payment services (known as HOPP) that Atos had already been providing for the Home Office at the time of the Outsourcing Contract, though it seemed to me that it was little more than a prejudice point which did not take it any further on the substantive issues or assist me to decide them. On the evidence of Mr Harkins, which I accept, those services (which Atos was already contractually obliged to deliver to the Home Office) were incorporated into the Outsourcing Contract as a B2B service as a short-term practical measure whilst the Home Office implements its strategy to close them, and the charging structure for them was different because it was inherited from the previous contract between Atos and the Home Office. I am not persuaded that these historic payment services have any impact on the question of material variation.

For all the above reasons, the Amendment Agreement does not amount to a material variation and I therefore reject the second ground of challenge. I find that there has been no breach of the Directive, the Regulations, or Art 56 of the TFEU. The decision that TFC should be delivered by NS&I working in co-operation with HMRC was therefore lawful despite the fact that Atos will be providing the operational services necessary to enable NS&I to deliver those aspects for which it is responsible, namely, childcare accounts and supporting services.

In the event that, contrary to my above findings, there was a breach of the Regulations or Art 56 TFEU because childcare accounts were not separately put out to public tender by HMRC (or included in the procurement of the Outsourcing Contract by NS&I) the next question is whether Edenred can establish that the breach has caused it loss. The parties were agreed that this has to be assessed as the loss of a chance, and on well-established principles that means that Edenred had to satisfy the court that there was a real, as opposed to fanciful, prospect that if it bid for a contract to deliver childcare accounts, it would have been awarded the contract. A real prospect means just that; it does not mean more than 50%, and previous cases have encompassed awards of damages where the lost chance has been evaluated as low as 17%.

After the second consultation and before the decision was taken to use NS&I to provide the accounts, a document was drawn up by senior civil servants for consideration by the Minister and further discussion on the various options for delivery of TFC. The authors were Ms Newton and Ms Elizabeth Russell, the Director of Personal Tax, Welfare and Pensions in HMT with responsibility for TFC within that department. Both gave evidence at trial.

The document, which is dated 14 July 2014, identifies five delivery options for TFC: NS&I, HMRC, a single private sector provider, a limited number of multiple private sector providers, and an open market. It highlighted the positive and negative aspects of each option and stated that there was “no unambiguously superior option”. The private sector options were said to have performed well on many of the criteria, but less favourably on speed of delivery (likely to be spring 2016 at earliest) and ease of build criteria. The HMRC option had “some but not all” of the positive features of the NS&I approach, but could not be delivered until autumn 2016 as, unlike NS&I, HMRC could not leverage its current infrastructure and would instead have to build a solution from scratch. If the matter was put out to tender, the procurement process could take until late spring or summer 2015 to complete.

Existing voucher providers were described as “bullish” that they could build accounts in 3-6 months from the date of any award at the end of a procurement process, but they had provided no evidence in support of that claim. No such supporting evidence was provided by Edenred in the present case. Although Mr Langlois expressed optimism about Edenred being able to scale up its operations to meet the requirements of TFC, based upon its expertise in delivery of childcare vouchers under ESC, and on the expertise and resources of its wider corporate group, (he referred in particular to the experience of another Edenred subsidiary in handling large volumes of prepaid transaction processing services) he did not address timescale, other than to estimate that it would take around 6 months to establish the necessary interfaces between Edenred’s and HMRC’s systems (which is just one aspect of what would be required). There was no supporting evidence to enable the court to evaluate whether that time estimate was realistic. No evidence was even adduced from Edenred’s parent company to confirm that the necessary finances would be made available.

There was a summary of the responses to the consultations: parents’ groups favoured a single provider on the basis that parents were unlikely to have the time and resources to make informed decisions between account providers. Security of funds was considered very important. Their main priorities were that TFC should be simple, work well from the outset and be introduced by autumn 2015. The majority of those who responded from the childcare industry who expressed a preference also indicated a preference for a sole provider model, as it would be a significant simplification from having to deal with a large number of voucher providers. They felt this would help to overcome administrative issues (such as late payments by CVPs). CVPs, in stark contrast, favoured the open market option. That is entirely understandable.

Ms Newton said in her second witness statement that it was very unlikely that the Government would move to an open market option, even with regulation of that market, where anyone could offer TFC accounts provided that they met set criteria. This model would not meet parents’ desire for simplicity because they would have to choose between a large number of providers, and it would require fees to be paid to the account providers by parents, rather than by the Government, which ministers have since decided they do not wish to happen. In addition it would make it very complicated for HMRC to organize the 20% top-up payments among different providers.

The evidence of Ms Newton and Ms Russell was that, if delivering the policy with NS&I was found to be unlawful, HMRC would have to carefully weigh up the options of delivering all of TFC within HMRC (which Mr Coppel accepted would not engage the Regulations) or tendering for one, or a very small number of suppliers (no more than 5) to provide the accounts. Ultimately the decision would be taken at Ministerial level.

I accept that there was a real, rather than a fanciful prospect that if NS&I were ruled out, the decision would have been made to go down the route of a public tender instead of using HMRC, although delivery via HMRC would not necessarily have been ruled out of the running and may still have been preferred. It was submitted by Mr Coppel that HMRC was not a viable option on timescale alone. That was a point he put in cross-examination to Ms Russell, but she did not accept it. She agreed that speed of delivery was a factor in the decision to go with NS&I but she said “I wouldn’t say that the timing was the main issue”. She later accepted that at the time when the decision under challenge was made, the anticipated extra year in delivery of TFC made HMRC a non-runner when compared to the timing of the other options. However, she added this:

“I think that one of the reasons that we chose the NS&I …our ministers chose the NS&I option, was some of the feedback coming out of… from parents and out of focus groups and the other consultation we did about favouring an in-house, a Government provider. So whilst at this point clearly with NS&I also on the table, that was the superior of our options, if NS&I had not been on the table, we may have looked more closely at the HMRC option”.

When it was put to her that the autumn 2016 timing issue and the disadvantage of it being outside HMRC’s expertise were very serious disadvantages nevertheless, Ms Russell agreed, but said that the procurement route had “not dissimilar timing” and agreed that the estimated difference was six months. Ms Newton’s evidence in her second witness statement was also that “there is no presumption that Ministers would choose to tender for an account provider for TFC - it may be that HMRC deliver the accounts in-house instead.” I anticipate that much would depend on whether the private sector route could be shown to be really likely to provide a significant advance on the delivery date and whether speed of delivery outweighed the other perceived advantages of keeping the matter in-house.

There is some evidence of who might have been interested in providing TFC if the matter had been put out to public tender in 2014 instead of making the decision under challenge. 15 potential suppliers, from a range of different organizations including Edenred and other CVPs, a high street bank, and BPOs attended a supplier day in response to a “prior information notice” published by HMRC in June 2014 during the second consultation. The smaller CVPs who attended the supplier day said that they would be most interested in bidding for TFC if there were 5-10 suppliers. The larger ones, including Edenred, did not say on that occasion whether they would bid (though it was assumed by the authors of the 14 July 2014 document that they would). That assumption was well founded, because in its response to the second consultation, whilst favouring an expansion or adaptation of ESC, Edenred indicated that it would be interested in bidding for the work if the decision was made to use one or more private sector suppliers to deliver TFC. Ms Russell and Ms Newton’s non-evidenced expectation was that the larger CVPs would prefer a multi-provider solution of 4 providers to partition the market. Follow-up one to one interviews were offered for the following day. Four or five of the attendees followed up in one-to-one telephone conversations and a further two submitted further information in writing in response to the one-to-one questions. Edenred did not take up the offer of a one-to-one discussion.

Ms Newton’s second witness statement correctly identified the alternative to an in-house option as being a tender for either one or a small number of suppliers to provide childcare accounts and she gave some evidence about who might bid in such event, but she gave no evidence about the likelihood or otherwise of ministers choosing between the single or multiple provider route. I consider, on the basis of all the evidence that I have seen and heard, including the government’s published response to the second consultation, that the option of tendering for multiple providers would not have commended itself to the ministers making the decision in HMT/HMRC, for all the same reasons that led to their ruling out the open market option, which apply equally to a small number of providers. Perhaps more significantly, it would also have been contrary to the general thrust of the preferences expressed in the consultation by those who would be the ultimate consumers of the service, namely, parents and childcare providers. The decision under challenge favoured a single provider. In the light of that evidence I consider there was no realistic prospect of a multiple provider tender even though it was correctly identified as being one of the remaining options. Given the express wishes of parents and childcare providers for simplicity, and the clear preference expressed in the consultation for one single provider over a small number of multiple providers, the complexity for HMRC of delivering top-up payments to more than one provider, the recognition that single provider models would be easier to integrate into a single scheme, and the wish to deliver childcare accounts sooner rather than later, any public tender for TFC would have been for a single provider which could deliver the accounts as soon as possible in or after autumn 2015. Given that it is unrealistic to suppose that the multiple provider option would have been available, Edenred’s chance of being awarded a TFC contract as one of the providers chosen as part of such a bid is negligible or non-existent.

I therefore have to evaluate whether Edenred had a real rather than fanciful prospect of being awarded the tender as the sole provider. Whilst Edenred has a substantial parent company there was no evidence adduced from that entity. There is no financial or accounting evidence to give the court any idea of how much money from its own business ventures it might be able to afford to invest in Edenred for this project, assuming that it was prepared to back it (as to which there was no evidence from the parent company either). I do not lose sight of the fact that the Claimant in this case is Edenred, not its parent, nor the group to which it belongs, nor CVL, the Edenred subsidiary that actually runs the childcare vouchers business.

Despite its track record of delivering accounts on the ESC scheme, Edenred itself has no experience of dealing with complex accounts or millions of customers. On Mr Langlois’ evidence, since 2014, 220,000 parents have conducted “at least one transaction” on their account and Edenred handles 110,000 telephone calls per annum relating to childcare vouchers. Edenred would need to scale up its operations – including increasing its capacity - by a factor of 10, with all the additional increases in human and other resources that entails. At present it only employs 86 full time equivalent employees on its CV business. TFC has different operational requirements from ESC. The magnitude of the challenges facing Edenred or any CVPA member was something that the CVPA emphasized in its responses to the consultations. Ms Russell’s evidence was that based on the scale of the challenge for Edenred it is far from clear that they would have even met the pre-qualification criteria for bidding, let alone been the successful bidder for a government contract that attracted the interest of 15 companies in the pre-market engagement exercise. She points to the fact that Edenred did not even engage in follow-up conversations after that event. I agree that this is hardly the behaviour of a really keen would-be bidder; but of course, Edenred was still advocating a variant of ESC as the preferred means of delivering TFC at that time.

In evaluating whether there was a real, rather than fanciful, loss of a chance here I must consider Edenred’s potential rivals. A clearing bank would be the obvious type of provider that could deliver the kind of service the Government was looking to NS&I to provide, and which would have the advantage of experience of dealing with complex accounts and millions of customers. It would have the further advantage of being independently regulated, unlike the CVPs. It would be the nearest equivalent to using NS&I. The 15 entities which expressed an interest at the supplier day included a high street bank. Of course there could be no pre-disposition towards a particular type of bidder, but if a bank were to bid, its ability to deliver what the Government wanted would easily eclipse that of Edenred.

Mr Coppel submitted that the fact that Edenred is not a bank is immaterial, because TFC does not require a bank, it just needs a provider who can operate accounts. He pointed to the fact that the Government itself thought that the CVPs could deliver TFC – that is why they entered into such extensive discussions with them on that topic. In basic terms, he submitted, what was required was just a more complex (and more expensive to set up and deliver) variation on the service they were delivering already to employers under ESC – the handling of multiple online accounts and payments into and out of such accounts. I consider that this plays down the differences between ESC and TFC. Mr Coppel also made the point that even NS&I could not deliver TFC without making changes to its existing operations to cater for its greater complexities and making the significant financial outlay required, and so anyone who tendered for TFC would face the same challenges. That is true, but it would be easier for some to overcome those challenges than it would for others. Mr Langlois’ oral evidence was that the financial outlay was no obstacle as Edenred could rapidly raise millions from its parent company or its shareholders, but as I have already said, there is no evidence from Edenred’s parent company let alone the shareholders that they would have been willing to sink millions into such a venture and no evidence that the matter had even been put before the board of the parent company for consideration.

I agree that the delivery of TFC does not have to be by a bank. However, it by no means follows from this that Edenred would be as attractive or as viable a provider as a bank or a BPO were they to bid, and on the evidence of the types of entity displaying an interest even at a supplier day of which fairly short notice was given, it is inevitable that bids would be made by such entities.

I am not persuaded that if the tender had been for a single provider, which is the only realistic course that would have been taken if delivering TFC in-house through HMRC had been ruled out, Edenred would even have made a bid (on the assumption in its favour that it would have met the qualifying criteria for such a bid). Despite Mr Langlois’ optimism, which as I have already said in the context of his evidence about bidding for the NS&I Outsourcing Contract, I treat with a degree of caution, I do not accept that it could have scaled up its operations to deliver TFC all by itself. Mr Langlois made a lot of claims in his third witness statement about Edenred’s experience, its IT platform, the experience of other members of the Edenred Group in prepaid business solutions, the group’s experience as working successfully as members of a consortium or joint venture but there was nothing in that evidence to convince me that Edenred had any particular expertise or other unique selling point that would make it a leading contender for the award of the TFC contract. The fact that it was experienced in making payments to childcare providers on a much smaller scale is not such a feature because all its rivals would have experience in making payments.

In terms of bidding as part of a joint venture or partnership, Edenred would have faced exactly the same kind of problems as it would have done in seeking to make a bid for the Outsourcing Contract if it tried to interest another entity in joining forces with it. There was no evidence from any of its CVP competitors that they would have done so. A BPO would not have been interested in joining forces with Edenred to deliver services that it could deliver by itself. It was the BPOs who expressed a strong interest in the single provider option after the supplier day. There was no, or no reliable evidence that anyone else would have teamed up with Edenred, despite Mr Langlois’ reference to Edenred having (unidentified) “banking partners”. Even if Edenred had made a bid, it would have stood no realistic chance of succeeding against a bank or against a BPO.

Edenred may well have succeeded in persuading HMRC that it could deliver childcare accounts. However, it would have faced very significant problems in scaling up to the necessary extent. There is no reliable evidence that it could have built the necessary infrastructure and been ready to deliver within 6 months of an award or that it could have done so alone. There is no evidence that any other entity would have participated in a joint venture with it, other than Mr Langlois’ supposition. It is fanciful to suppose that Edenred would have persuaded HMRC that it was better placed than a bank or BPO to deliver TFC merely on the back of its successful track record of delivering ESC. It must be borne in mind that although the possibility of delivering TFC through voucher providers was something that was considered early on, NS&I was selected instead because it could build on its existing banking and service infrastructure, it had experience of dealing with millions of customers and millions of accounts already, and it had UK call centres already in place (even if it needed to provide another one). Moreover, being backed by the Government, it provided the most secure option. Realistically, therefore, Edenred bidding for a single provider award by HMRC would not have stood a chance against an independently regulated private sector organization such as a clearing bank, or even against a BPO. In the alternative scenario where Edenred would be bidding for the provision of support services to NS&I so as to enable NS&I to provide the accounts to HMRC, its chances of successfully outbidding Atos or another BPO are even more remote.

Therefore, even if there had been a breach of the Regulations or of Article 56 TFEU, Edenred has failed to discharge the burden upon it of establishing that the breach caused it to suffer any loss. In those circumstances I would not have exercised my discretion to grant declaratory relief or to set aside the decision to award the contract to NS&I under Regulation 47I. It is not the practice of the court to grant legal remedies that have no practical purpose, or to make declarations merely to express disapprobation or to act as a deterrent. I do not consider that it would accord with the underlying principles of encouraging and promoting fair and transparent competition to set aside a decision that did not, in fact, deprive the person complaining about it of any valuable economic opportunity and which, if it had to be re-made, would confer no benefit on that person.

For all the above reasons, this claim must be dismissed. I would like to conclude by expressing my thanks to both leading counsel and to the parties’ legal teams for the clear, helpful and attractive manner in which this case was presented and for their co-operation in ensuring that the timetable was met.