JBW Group Ltd v Ministry of Justice

This appeal raises a short but difficult point on the scope of the Public Contract Regulations 2006. The question is whether the particular contracts under consideration should be classified as public service contracts, in which case the Regulations apply, or as service concession contracts, in which case they do not.

The Ministry of Justice (“MoJ”) put out to tender a number of contracts for bailiff services to be provided to magistrates courts, each contract relating to a particular region in England or Wales. On 29 January 2009 the appellant (“JBW”) tendered for three of these contracts. JBW were unsuccessful and brought proceedings against the MoJ alleging that the MoJ had acted in breach of the Regulations in various ways, such as by leaking information and providing assistance to a rival bidder. They also alleged that there was an implied contract that the MoJ would consider tenders fairly and transparently and in accordance with the terms set down in the invitation to tender, and that the terms of this contract had been breached.

The MoJ sought summary judgment and succeeded before Master Victoria McCloud. She dismissed the claims. The Master held that the Regulations were not applicable because the contracts in issue were service concession contracts which are specifically excluded from the scope of the Regulations. In addition, the Master held that a contract could not be implied in the manner alleged by the claimant.

JBW have appealed against both these conclusions. They also initially contended that the judge must have applied the wrong test for granting summary judgment because she observed in the course of her judgment, with respect to the argument whether the Regulations applied, that there was “a real prospect of an appeal court taking a different view.” However, Mr Knox QC, counsel for the appellant, conceded in argument that it would be pointless for the court merely to rule on whether there was an arguable case that the Regulations applied since there were no material disputes of fact and the court was in as good a position as the trial judge to determine that question. It would be contrary to the overriding objective for the court not to decide the matter. We therefore treat this appeal as though it raised preliminary issues of law on undisputed facts. It has come directly to us because Master Eastman ordered that a “leapfrog” was appropriate.

The Regulations were designed to implement Directive 2004/18/EC on the co-ordination of procedures for the award of public works contracts, public supply contracts and public services contracts. Article 1 of the Directive defines public contracts as follows:

“(a)

public contracts are contracts for pecuniary interest concluded in writing between one or more economic operators and one or more contracting authorities and having as their object the execution of works, the supply of products or the provision of services within the meaning of this Directive.”

A public service contract is then defined as a public contract other than a public works or supply contract which has as its object the provision of services referred to in Annex II of the Directive. Annex II catches these contracts.

Article 1(4) defines a service concession as:

“..a contract of the same type as a public service contract except for the fact that the consideration for the provision of services consists either solely in the right to exploit the service or in this right together with payment.”

Article 17 provides that the Directive does not apply to service concession contracts.

The Directive replaced a series of earlier Directives including Directive 92/50/EEC which regulated the procedures for awarding public service contracts. That Directive did not contain a definition of a service concession although the ECJ held that a concession fell outside the terms of that Directive. In Telaustria Verlags GmbH v Telekon Austria AG [2000] ECR I-10745 the ECJ pointed out that the Commission had originally included public concession contracts in the Directive but these were removed from its scope by the European Council. The current Directive puts beyond doubt that they are excluded. Earlier case law on the meaning of concession in relation to the 1992 Directive remains relevant not least because the first recital of the 2004 Directive states in terms that the Directive is “based on Court of Justice case-law.”

The 2006 Regulations are designed to implement the 2004 Directive. They do not precisely replicate the language of the Directive but both parties accepted that since the Regulations would have to be read consistently with the Directive (and no-one suggested that they could not be so read) we should simply focus on the terms of the Directive itself.

The relevant contractual terms are set out in schedule 3 of the invitation to tender and the contractual specifications in schedule 4.

The vast majority of the work under the contracts consists of the enforcement of warrants of distress issued by the magistrates for the non-payment of fines. The bailiff may levy financial distress, which involves securing payment without confiscation of goods; or he may levy confiscation distress, which involved confiscating and selling goods. For that work the contractor had to identify in the tender the fee he proposed to charge to execute the warrant and it is a term of the contract that he will not exceed that sum. In practice the costs of recovery are born by the defaulters because the bailiff who executes a warrant of distress is entitled under the terms of the warrant to take sufficient to cover not only the unpaid fine but also the costs of recovery.

In addition to executing distress warrants, the bailiff will sometimes be required to execute clamping orders. There is a fixed fee in relation to enforcing these orders. Again, the costs are generally borne by the defaulters, but the financial arrangements are different than in the case of distress warrants.

Finally some courts also require bailiffs to effect arrests pursuant to financial warrants and breach of community penalty warrants. In this case the fees, which are fixed by the contract, are paid directly by the MoJ through the Court Service. However, the execution of distress warrants is by far the most significant part of the contract covering more than 90% of the work, and the parties accepted that in order to analyse the true character of this agreement, it was necessary to focus on the arrangements relating to those duties.

The following terms of the contract are, in my view, particularly relevant. Clause 1 of the contract requires the contractor to provide the services detailed in the Specification. Clause 1 of the specification states that the services are provided to the Ministry of Justice. The contractor must perform the services detailed in the specification (cl.5) and to the service levels specified (cl.6). In the event of default resulting from the fact that the services are not provided in accordance with the contract, the MoJ has certain remedies available (cl.7). These include withholding money from the contractor in certain circumstances until the default is rectified (cl.7.1.2) and ultimately even terminating the contract (cl.17).

The specification itself also confers certain powers upon the MoJ. If the contractor fails to meet the contracted performance requirements in two consecutive quarters, the MoJ can divert 20% of the warrant volumes to the reserve contractor.

Clause 5 of the specification requires the contractor to “work strategically with the Department to assist in achieving ongoing increase in performance and government targets” and to agree an annual service enhancement to support the attainment of continuous improvement and best value.

Clause 6 sets out in some detail the operational protocol. It permits the MoJ to impose certain restrictions on how the bailiff’s functions are performed, such as specifying the days and times when certain orders can be enforced. Detailed information has to be provided if warrants are not executed, and the specification spells out the minimum inquiries which must be carried out before the warrant can be returned as unexecuted. The contractor must provide the Department with information to allow the Department to monitor performance (cl.13).

The payment arrangements are in large part determined by the bailiff’s legal powers. Rule 52.8 of the Criminal Procedure Rules which is headed “Execution of magistrates’ court distress warrant” provides, so far as material:

“(2)

The warrant shall authorise the person charged with the execution of it to take as well any money as any goods of the person against whom the distress is levied; and any money so taken shall be treated as if it were the proceeds of the sale of goods taken under the warrant.

(3)

The warrant shall require the person charged with the execution to pay the sum to be levied to the court officer for the court that issued the warrant. …

(12)

The person charged with the execution of any such warrant as aforesaid shall cause the distress to be sold, and may deduct out of the amount realised by the sale all costs and charges incurred in effecting the sale; and he shall return to the owner the balance, if any, after retaining the amount of the sum for which the warrant was issued.”

The specification itself provides that the contractor must pay monies received from the defaulters into a client account separate from other monies. The contractor must then remit periodically the monies referable to the warrant, but it can retain the contractually agreed fees due to it for recovering the fines.

Paragraph 6.26 expressly provides that “the contractor shall apply monies received from the defaulters in relation to any given warrant to the court penalty first, with its own fees to be paid afterwards.” To that extent the interests of the contractor are subordinated to those of the Department.

We are concerned with tracing the boundary between public service contracts and service concessions and determining on which side of the line this contract falls. That is a question of EU law. The concepts must be given an autonomous EU meaning. Counsel took us to various authorities of the ECJ (now the Court of Justice of the European Union.) Each counsel claimed that these authorities, properly analysed, plainly determined the outcome in his favour. It is necessary, therefore, to consider the authorities relied upon in a little detail.

One of the earliest cases was European Commission v Italian Republic [1994] ECR -1409, in which the installation and running of a computer system for the operation of the Italian national lottery was contracted to a third party. The payment was a percentage of the gross receipts. One of the issues before the ECJ was whether the Italian Government had awarded contracts in breach of the Public Service Contracts Directive then in force. The Italian Government claimed that this was a concession and therefore fell outwith the scope of the Directive. The court rejected that submission. It held that the contract was of a technical nature; that there was no transfer of responsibility to the concessionaire for the various operations inherent in running the lottery; and that the fact that the annual payment was related to revenue did not convert the contract into a concession.

In Arnhem and Rheden v BFI Holdings BV [1998] ECR I-6821 Advocate General La Pergola identified two particular criteria which distinguished concessions and service contracts caught by the 1992 Directive. The first was that the recipient of the service in a concession is a third party which receives the service rendered; and the second was that the remuneration derives wholly or in part from the provision of that service to the beneficiary. He added that “the concessionaire automatically assumes the economic risk associated with the provision and management of the services.”

This approach was followed in Telaustria where the contractor was given the right to produce telephone directories and electronic databases of subscribers. The authority took a 40% stake in the operation. The court held that it was a services concession since the contractor obtained the right to exploit for payment its own service. This concept of exploiting the service underpins the definition of concession in the current Directive. It is also pertinent to note that Advocate General Fennelly in his opinion rejected an argument that the concept of service concession should be construed narrowly as an exception to the general rule. Concessions were not, properly analysed, an exception to the rules; they were simply not covered by them.

In R (on the application of the Law Society) v Legal Services Commission [2007] EWHC 1848 (Admin) paras 61-69, Beatson J followed this dictum in Telaustria in concluding that the contracts between the Legal Services Commission and firms of solicitors under which the latter provided legal services to the public were not concession contracts. As in the Italian lottery case the volume of work available, and therefore the remuneration, was unpredictable since the contract did not guarantee levels of work. However, this was not enough to create a concession.

In Parking Brixen GmbH v Gemeinde Brixen and Stadtwerke Brixen AG (Case C-458/03) [2005] ECR 1-08585 the public authority granted a contractor the right to manage a public car park in consideration for which he was remunerated by sums paid by third parties for the use of the car park. The ECJ confirmed that an important distinction between a standard public services contract and a concession is that under the former the provider is remunerated directly by the contracting authority whereas in the latter his remuneration comes from third parties using the service. The court noted (para 40) that:

“That method of remuneration means that the provider takes the risk of operating the services in question and is thus characteristic of a public service concession….”

Not surprisingly on the facts the ECJ found that this was a concession and fell outside the terms of the Directive. There were the two interrelated aspects of third party payment and the risk inherent in running a service of this kind.

Parking Brixen was followed in Wasser and another v Eurawasser Aufbereitungs and another, (Case C-206/08) [2009] ECR 1-08377. The contract under consideration in that case involved the distribution of drinking water and the disposal of sewage. The terms of the tender were that the successful tenderer would supply the services on the basis of private law contracts in its own name and on its own account to user residents, and it would be paid directly by those users. It could fix the prices but subject to certain limits set by local municipal rules.

The court referred with approval to the grounds for distinguishing a service contract and a concession identified in Parking Brixen. The court held that receiving remuneration from third parties was one means of exploiting the service and necessarily meant that the provider was taking the risk of operating the service. In view of that, it did not find it necessary to consider precisely what constituted “the right to exploit”.

The court also rejected a submission to the effect that if the risks involved in running the service were small (as was alleged to be the case here) there would be no concession even if there was a transfer of a service. The court noted that in certain sectors of activity, in particular the public utilities, rules of public law often limit the degree of risk. Nonetheless, as long as there is a transfer of all or at least a significant share of such risks as arise in the operation of the service, that will suffice to establish that a concession had been transferred.

Shortly before the Wasser case the ECJ had given judgment in Oymanns GbR, Orty, Case C-300/07 [2009] ECR 1\4779. The claimants were an orthopaedic footwear company who submitted a tender for the manufacture and supply of footwear suitable for diabetic foot syndrome. The services provided were divided into the provision of footwear for different groups, and tenderers had to submit prices for the cost of footwear for each group. Payment for the services was made by a social security scheme to which the patients would make some contribution. The quantity of shoes supplied was not fixed and depended upon the number of patients who had the appropriate documents, including a medical prescription, choosing to contact the successful tenderer. The orthopaedic footwear had to be individually tailored to the patient and advice had to be given both prior to and after its supply about its use.

The court concluded that it was a mixed supply and services contract but went on to consider whether, if the provision of services was regarded as the more important element, it should be regarded as a service concession or a service contract (which would, in the circumstances, be a framework agreement). The court emphasised that the legal classification depended on a careful analysis of the factors in any particular case and continued (para 71):

“…. it flows from the above-mentioned definition of a service concession that such a concession is distinguished by a situation in which a right to operate a particular service is transferred by the contracting authority to the concessionaire and that the latter enjoys, in the framework of the contract which has been concluded, a certain economic freedom to determine the conditions under which that right is exercised since, in parallel, the concessionary is, to a large extent, exposed to the risks involved in the operation of service. On the other hand, the distinguishing characteristic of a framework agreement is that the activity of the trader who has concluded the agreement is restricted in the sense that all contracts concluded by that trader during the given period must comply with the conditions laid down in the agreement.”

The court in that case held that the successful tenderer would not enjoy the degree of economic freedom which was the mark of a concession holder. Nor was it exposed to a significant risk connected with the provision of the services. Accordingly this was a contract to which the Directive applied. The court recognised, however, that this did not mean that the business was risk free. It said this (para 74):

“It could certainly be remarked that the trader in such a case exposed to a certain risk in as much as insured persons may not avail themselves of its products and services. However, that risk is limited. The trader is spared the risk connected with the recovery of payment and the insolvency of the other party to the individual contract since, in law, the statutory sickness and insurance fund alone is responsible for paying the trader. In addition, although the trader may be sufficiently equipped to provide its services, it does not have to incur inconsiderable advance expenditure before an individual contract with an insured person is concluded. Because the tenderer did not bear the principle burden of the risk associated with the carrying on of the activities the court concluded that this was an agreement and not a concession.”

The assumption in this case seems to be that the freedom to exploit rights conferred by the contract necessarily creates the risk - it is “in parallel” as the court put it.

The issue arose again in the recent case of Stadler v Zweckverband fur Rettungdsienst und Feuerwehralarmierung Passau (Case C-274/09). This concerned the provision of rescue services by the contractor to the Passau municipal association. The contract was unusual in a number of respects. The contractor could charge a usage fee upon all the persons and bodies which called upon the service. The amount of the fee was agreed not with the contracting authority but with certain social security institutions. In the event of disagreement the question would go to arbitration. The payments by third parties did not go directly to the contractor; they were paid to a central settlement office which in turn paid the contractor. Most of the users were covered by compulsory insurance but some were either subject to private insurance or were uninsured. The contractor took the risk that they would not be able to meet their liabilities. In the light of the previous case law the court concluded that the arrangement constituted a concession. The court considered that it was immaterial that the payment was made via a third party body: it observed that “the fact remains that the remuneration obtained by the provider of the services comes from persons other than the contracting authority which awarded it the contract.” Furthermore, the court applied the Wasser case in finding that (para 33)

“where the remuneration of the provider comes exclusively from a third party, the transfer by the contracting authority of a “very limited” operating risk will suffice in order for a service concession to be found.”

The court answered the questions posed in the following way:

“…where the economic operator selected is fully remunerated by persons other than the contracting authority which awarded the contract concerning rescue services, where it runs an operating risk, albeit a very limited one, by reason inter alia of the fact that the amount of the usage fees in question depends on the result of annual negotiations with third parties, and where it is not assured full coverage of the costs incurred in managing its activities in compliance with the principles laid down by national law, that contract must be classified as a ‘service concession’ within the meaning of Article 1(4) of Directive 2004/18.”

Mr Vajda contends that on any sensible view of this arrangement, it is plain that the conditions necessary to establish a concession are met. Payments for the service are made by the third party defaulters and the risk of non-payment is transferred from the MoJ to the contractor. The risk may be relatively small, in that it is known in broad terms what the typical recovery rate is, and the tender price can be fixed accordingly. But Wasser and Stadler show that this matters not provided a significant part of whatever risk there is transfers. Mr Vajda submits that the arrangements in this case clearly constitute a concession; they comply with the principles enunciated in the Parking Brixen, Wasser and Stadler cases.

Mr Knox submits that Mr Vajda’s analysis fundamentally distorts the true nature of the relationship. The contract terms, when read with the specification, demonstrates that the MoJ retains a significant influence over the running of the contract. This is not what one would expect if the contractor was being put in possession of a concession which he could exploit. The true beneficiary of the service is the MoJ which, through the court service, receives the fines recovered. Looking at substance rather than form, the payment is made by the MoJ albeit by reference to the sums seized by the contractor from the defaulters to cover the costs of recovery. The fact that the remuneration is tied to the success of the operation does not change the true nature of the consideration under the contract. There is no contractual relationship between the contractor and the defaulter. Nor can it be said that there is a business to exploit; the contractors can neither increase the client base nor can they alter the price they charge the defaulters. They cannot, for example, allow extra time to pay if the defaulter pays them a sum of money. The contractor necessarily bears the risk of unsuccessful execution of the warrant, and it is true that this is a risk which is not borne in a classic service contract. But the mere transfer of some risk is not of itself enough to convert the arrangement into a concession.

The submissions advanced below were essentially the same as those submitted before us, although the Stadler case had not been decided by then. The Master found that this was a concession, accepting the arguments submitted by the MoJ. In her view the contract was akin to the facts in Parking Brixen and Wasser. It satisfied the two-fold criteria of remuneration from third parties coupled with the transfer of risk. Whilst it might be what the judge termed “an ugly use of language” to refer to the defaulters as beneficiaries of the service provided by the bailiffs, nonetheless they were “on the receiving end of the service and benefit from the protections inherent in provision of lawfully regulated modes of enforcement.” In reaching this conclusion the Master did not, however, give any specific consideration to the specific terms of the contract between the MoJ and the contractor.

As the EU authorities show, the paradigm case of a service contract is where the applicant performs a service for the authority and is paid an agreed fee for that service. It is important to emphasise that such a contract is not necessarily risk free. Like any contracting party, the contractor may find that he has struck a bad bargain; the cost of providing the service to the authority may prove to be greater than the remuneration received.

The paradigm case of a concession is where the applicant is put in possession of a business opportunity which he can exploit by providing services to third parties and charging them directly for those services. The contractor then bears the risks of running the business which are typically greater than those involved in performing a contract for a fixed fee. The Parking Brixen and Wasser cases are classic examples of such contracts, albeit that the risk in the latter case was limited. The fact that there may be some regulation of the price which the contractor can charge the third party for the service does not of itself prevent the arrangement constituting a concession.

We are concerned with a case which does not fit neatly into either category. This contract lacks many of the principal features of a typical concession. I would identify four interrelated aspects in particular. First, it is in my view difficult to say that the effect of the contract is to place the bailiff in a position to exploit a service. Typically that right confers upon the contractor the opportunity to develop and expand the service so as to maximise profits. He is able by his own endeavours to increase the take up of the service and to fix the price to be charged to third parties, although some control over price is not inconsistent with a concession. Neither factor is in play here: the contractor can only deal with defaulters identified by the MoJ, and the fee he can recoup from the third party defaulters is fixed as a result of the negotiations with the MoJ.

Second, a concession usually involves a direct contractual relationship with third party clients or customers of the service who are charged directly by the contractor. That element is absent here also. Indeed it is highly artificial to describe defaulters as third party beneficiaries or customers of the service at all. They are forced into a relationship by compulsion.

Third, as Oymanns makes clear, the concessionaire typically has considerable control over the manner in which the service is provided; the authority takes a back seat. Again, that is not so in this contract. The MoJ has retained detailed and close control over the way in which this contract is performed. Not only are the services tightly defined, but so are the service levels which must be attained under threat of sanctions if they are not. There are obligations to work strategically with the department; and the operational protocols regulate in some detail when visits should take place and when enforcement might be inappropriate. In addition certain specific information has to be provided when warrants are not able to be executed. The contractor is constrained in various quite detailed ways in the manner in which he can operate the contract. He is bound by the protocols The contractor’s rewards may even be reduced for unsatisfactory performance. All this amounts to the retention of considerable control by the MoJ.

Finally, in a classic concession, although the contracting authority has an interest in the service being performed for the benefit of third parties, it does not itself directly benefit from its performance, as in this case.

At the same time, it is not a classic service contract. The remuneration from the MoJ is fixed by reference to such costs as the contractor is able to recover from the defaulters. Accordingly although the contractor has no direct relationship with the defaulters, they do, albeit unwillingly, in fact pay for the service because of the operation of rule 58 of the CPR. There is no payment from the MoJ directly.

Moreover, the contractor bears such financial risks as are involved in running the service. These arise for a number of reasons. The total remuneration is unknown in advance not only because the number of defaulters is unpredictable, which according to the Italian lottery case would not of itself be sufficient to constitute a concession, but also because the number of those who will avoid payment altogether is unknown. Furthermore, because the costs are subordinated to the fines, the contractor takes the risk not only of being unable to recover anything from the defaulters, but also of recovering insufficient to cover both the fine and the costs of recovery. These risks include, but go beyond, those necessarily involved in any service contract of being unable to provide the service at the agreed price. Even if it can be said that the relevant risks are small - since statistics provide a good indication of the extent of those risks and they can be catered for in the price offered in the tender - they are precisely the same risks as those to which the MoJ would be subject if it were to perform the contract for itself.

I do not accept that the solution to the question can be found, as Mr Vadja submits it can, simply by saying that since the risks are transferred and the defaulters are in fact paying for the action taken against them, the EU authorities compel the conclusion that the contract is a concession. I accept that a contract can amount to a concession even though some of the classic ingredients I have set out above are absent. Stadler shows that a concession may exist even though the fees are fixed by a third party other than the beneficiaries of the services, and even though they are not paid directly by the recipients of the service.

However, in none of the EU cases where there was found to be a transfer of a concession does the public authority appear to have retained the degree of control over the performance of the contract which is found here. Nor did the public authority itself benefit from its performance in the way that the MoJ does here. I accept, as the Master below noted, that even in a concession case the authority will necessarily benefit because it is relieved of the obligation to provide the service itself. But the benefit which the MoJ derives from this contract goes beyond that since the MoJ receives the unpaid fines.

Further, the EU cases are envisaging a situation where the risks are of a kind which are inherent in running a business for third party beneficiaries. Recalcitrant offenders subject to court orders who pay for a service they do not want under compulsion of law do not happily fit that description. On any view, therefore, this is not a typical concession if concession it be.

Accordingly, whilst I accept that a literal reading of the authorities - and the answer to the questions posed in Stadler will provide an example - might suggest that payment by third parties plus transfer of risk necessarily means that the arrangement is a concession, I believe that is too simplistic an approach. The Court of Justice has not yet had to engage in a case of this nature where the risks of running the service are incurred but where there is no real opportunity to exploit a service in any meaningful way.

I confess that I have found this a very difficult question. Taking all the relevant factors into account as the EU case law requires me to do, and bearing in mind that this is an autonomous concept of EU law, I have concluded that this is a concession and not a public service contract.

In reaching this conclusion I have born in mind the following considerations. First, there can be no doubt that in so far as the undertaking of risk is concerned, the risks transferred here are all those involved in running and managing the bailiff service. The MoJ is released even from the costs incurred in unsuccessfully failing to execute a warrant. Second, there is no direct payment by the MoJ for the performance of the service. The fact that this is an unwilling payment by third parties because the CPR 58 empowers the bailiff to distrain for the cost of enforcement does not alter that fact. Third, whilst it is true that the MoJ benefits from the performance of the service in a different and additional way to that found in a normal concession, it does not alter the fact that a service is also provided to third parties. Fourth, although the beneficiaries are not willing recipients of the service, that is equally the case in other circumstances where a concession has been found to exist e.g. those who have to take advantage of rescue services in Stadler. It ought not to preclude a concession arising.

The most powerful arguments against this conclusion are two interrelated points: first, the MoJ has preserved much greater control over the performance of the contract than is normally the case where the right to exploit a service is granted; and second, that the scope for exploitation is extremely limited. As to the latter, however, it can be said that in cases like Wasser and Stadler there was little opportunity to improve the client base. It is inherent in the nature of the service being performed.

I see the force of the point that the MoJ seeks to retain real controls over the way the bailiff’s powers are exercised. But I have concluded that this is not enough to outweigh the contrary considerations so as to cause me to characterise the arrangement as a service contract, even when combined with the inherent restrictions on the ability to exploit the service.

It follows that I would dismiss the appeal on this point.

The argument here was that by offering the contract out to tender, the MoJ was impliedly entering into a contract which would oblige it to treat all tenders equally and with transparency and in accordance with the terms of the tender document.

Mr Knox accepted that if he had succeeded in establishing that there was a service contract, this would add nothing to his case. It would then be unnecessary to imply any contract. Initially he suggested that even then the implied contract argument might entitle him to bring a claim for six years rather than within the much stricter three month period permitted under the Directive. However, in reply he resiled from that position and conceded that it would be inconsistent with the purpose of the Directive to imply any such contractual right.

That concession was, in my view, rightly made and is consistent with the decisions of two first instance judges, Morgan J in Lion Apparel Systems Ltd v Firebuy Ltd [2007] EWHC 2179(Ch), para 212 and Flaux J in Varney and Sons Waste Management Ltd v Hertfordshire County Council [2010] EWHC 1404, paras 232-235 citing Monro v HMRC [2009] Ch. 69.

However if, as I have found, the Regulations are not applicable, the same argument cannot be advanced. I reject a submission of Mr Vajda that it would be illogical to find that an implied term can be excluded if the arrangement is analysed as a service contract but not if it is a concession. The reason it would be excluded in the first situation is that it is unnecessary and would, if implied, be inconsistent with the statutory scheme. Those arguments do not apply where the arrangements constitute a concession. Nor do I accept an argument he advanced, which was accepted by the judge below, that by excluding concessions from the scope of the Directive and hence the Regulations, the draftsman intended that provisions of a kind found in the Regulations positively ought not to apply to them. I would not be prepared to read the effect of the exclusion in that way. A tendering authority is not obliged to comply with the Regulations where a service concession is in play, but there is in principle no reason why it could not choose to do so and I do not see how it could be illegal for it to do so. The parties could expressly agree to contractual terms mirroring the Directive and the Regulations if they so wished, and therefore there is no reason in principle why implied terms could not cover that same ground. Having said that, the difficulties of implying terms akin to those found in the Regulations, terms necessarily premised on the assumption that this was the common intention of the parties, in circumstances where the MoJ has throughout been acting on the assumption that the Regulations did not apply, is obvious.

When considering the implied contract question, two issues arise for consideration: first, is there any implied contract? Second, if so, what is its scope? As to the first issue, I would be prepared to accept, in line with the well-known judgment of Bingham LJ, as he then was, in Blackpool Aero Club v Blackpool Borough Council [1990] 1 WLR 1195 that the MoJ would in principle be under an obligation to consider the tender. Also, contrary to the submissions of the MoJ, I would have no difficulty in implying that any such consideration should be in good faith. Mr Vajda contended that this was an obligation under public rather than private law, but I do not see why this should preclude the obligation arising in private law also. Indeed, if a tender is not considered in good faith, I do not think that it can sensibly be said to have been considered at all.

However, Mr Knox does not contend that there has been a breach of this limited duty. The question is whether the implied obligations can extend beyond that limited requirement to embrace the much fuller set of duties relied upon by Mr Knox. I see no conceivable basis for concluding that it can. There is simply no basis on which it can be contended that these terms necessarily have to be implied to give efficacy to the contract; and nor can there be a common intention that they should given that the MoJ has always been denying that the Regulations apply. Moreover, as Mr Vajda pointed out, the specific power conferred on the MoJ to depart from the terms of the tendering document is itself inconsistent with the EU principle of transparency which would require strict adherence to the published terms.

Mr Knox relied upon the fact that there are fundamental EU principles of transparency and equality, and he submitted that these would mould the nature of the implied term. However, I agree with Mr Vajda that there is no proper basis for assuming that EU principles can alter the way in which terms are implied at common law. It is common ground that these principles are not engaged as a matter of EU law, since there is no cross-border element in the arrangement. In effect Mr Knox is seeking to use the implied term as a means of expanding the reach of EU law and that is not, in my judgment, a legitimate exercise.

It follows that I would reject this ground of appeal also.

I have read the excellent judgment of Elias LJ in draft. He has fully set out the facts and issues, and his reasons for dismissing this appeal. I agree with those reasons, and do not think that there is anything which I can usefully add, save in relation to one point. That point is whether we should refer the issue raised on this appeal to the Court of Justice of the European Union (‘CJEU’), because it raises an issue of law as to the application of a Directive to an agreement, which, as Elias LJ has said, raises a very difficult question – see para 52 above.

Although both parties urged us to decide the issue raised on this appeal, rather than referring it to the CJEU, there may be thought to be an argument that we should refer the issue pursuant to Article 267 of the Treaty on the Functioning of the European Union (formerly 234 of the EC Treaty).

As pointed out by Arden LJ in Cooper v Attorney-General [2010] EWCA Civ 464, [2011] 2 WLR 448, para 10, Article 267 not merely entitles the national court to refer to the CJEU a question ‘concerning the interpretation of directives of the Community where the national court consider[s] that a decision on the question was necessary for it to give judgment and the question was not acte clair’, but it actually ‘impose[s] an obligation on a court of final appeal to request a preliminary ruling in those circumstances’. Although this is not a court of final appeal, there may seem to be a powerful argument for saying that, if we all take the view that the point is very difficult, we should not impose on the parties and the Supreme Court the cost and time of an application for permission to appeal, which would inevitably lead to that court referring the issue to the CJEU: rather we should refer the issue ourselves.

The reason I would reject that argument is that I do not believe that the legal principles applicable to the facts of this case are unclear: it is the application of those principles to the unusual facts of this case which raises the difficulty identified by Elias LJ. The question of whether to refer an issue often gives rise to conflicting considerations. From the perspective of the national court and the parties, where an issue is not acte clair, there may be an ultimate duty to refer, but, through nobody’s fault, a reference leads to significant delay and expense, and not infrequently can produce a decision which in practice takes the particular case little further forward. From the perspective of the CJEU, there is the conflict between the need for any point of uncertainty to be referred, to ensure consistency of approach across the Union, and the practical concern of the court being swamped with references, leading to increasing pressures and delays.

If I had concluded that the legal principles applicable to determining whether an agreement was a public service contract or a service concession for the purposes of the Directive were unclear, then, albeit with reluctance, I would probably have taken the view that we should refer the issue in the present case to the CJEU. However, in the light of the guidance given by the CJEU (mostly in its previous incarnation as the European Court of Justice) and the Advocates General in the cases discussed by Elias LJ in paras 23- 25 and 27-36 above, it seems to me that the principles have been identified and clarified.

The difficulty of applying those principles to the facts of an unusual case such as this does not to my mind justify our referring any question to the CJEU. The primary purpose of Article 267, as I see it, is to ensure uniformity of legal principles and approach across the Union, not to ensure that application of those principles and that approach produces the same result in every case. That would be an impossible objective in terms of both achievability and in the burden it would place on the CJEU. It may be that, if the facts of this case were common, one might take a different view, but, as Elias LJ explains in paras 40-47, the agreement in this case is very unusual, so it cannot be said that our conclusion would have significant repercussions in relation to other cases.

I agree with both judgments.