Boxxe Limited v The Secretary of State for Justice

The Defendant, the Secretary of State for Justice (‘SoSJ’) applies pursuant to regulation 96 of the Public Contracts Regulations (‘PCR’) to lift the automatic suspension presently in place pursuant to regulation 95(1) PCR in relation to a mini-competition for a call-off contract relating to the provision of digital and audiovisual (‘AV’) equipment for use by His Majesty’s Courts and Tribunals Service (‘the Competition’). The application is resisted by the Claimant Boxxe Limited (‘Boxxe’), the unsuccessful tenderer. Boxxe has applied for an expedited trial. Having considered the applications, and in light of the urgent nature of the application to lift the suspension, I granted SoSj’s application to lift the suspension on Wednesday 8 March 2023, indicating that the detailed reasons would follow. This judgment sets out those reasons.

On 11 August 2022, the Defendant issued an Invitation to Tender (the "ITT") for a call-off contract for the AV. The eventual supplier would be required to provide the equipment specified by the Defendant in the ITT. It would also be required to provide storage, in a secure facility, for the equipment ordered by SoSJ until such time as the SoSJ required it to be delivered to one of its sites. While the equipment remained with the supplier or in transit, the supplier would be entirely responsible for its security and would be liable for any loss, theft, or damage to it. A minimum of three months’ free storage was required to be provided by the Supplier.

On 13 December 2022, SoSJ decided to award the Contract to Specialist Computer Centres Plc (“SCC”). This was notified to Boxxe together with feedback (‘the Decision Notice’) on the same date, and it was downloaded by Boxxe at 16.31. It provided the following information :

Boxxe’s claim is that, on SoSJ’s interpretation and application of the Competition rules, the contract was awarded to SCC not because SCC was the most economically advantageous tender (‘MEAT’). Rather, Boxxe claims that it was awarded to SCC as a result of what it describes as an arithmetic quirk within the pricing model that led to a perverse outcome i.e. the contract being awarded to a bid that was over £1million higher, without, it is said, any rationale.

The claim was issued on 12 January 2023. The challenge to the evaluation of storage pricing is set out over 7 grounds, summarised as follows:

Each of the grounds is resisted substantively by SoSJ. It is also said that the claims have been brought outside the 30 day limitation period required by regulation 92 PCR.

It is common ground that the test to be applied to applications of this nature is American Cyanamid, although, as set out further below, there are nuanced differences between the parties as to how the test is applied in the present context. Therefore, I consider the following (as summarised in a similar context by O’Farrell J in Camelot Global Lottery Solutions Limited v Gambling Commission [2022] EWHC 1664):

I have read and considered the evidence submitted from Mr Edgerton on behalf of SoSJ (three witness statements), and Mr Clark and Mr Hulland (two statements) on behalf of Boxxe. I have also read the letter submitted to the Court on behalf of SCC, the successful tenderer.

At the outset of oral submissions, I informed the parties that the Court would be able to accommodate an expedited trial, if one were to be ordered, in July 2023.

Notwithstanding it is the position of SoSJ that the claim has no prospect of success (and a strike out application has been issued), the substantive merits were not the subject of Mr Paines contention, on behalf of SoSJ, that there is no serious issue to be tried. Notwithstanding the brief submissions by Mr Tankel, on behalf of Boxxe, going to the substantive merits, it would not be appropriate for me to make any findings in relation thereto. Instead, Mr Paines confined his submission to the contention that each ground was out of time. He relied upon the observation of Vos J (as he then was) in Alstom v Eurostar International Limited [2010] EWHC 2747 that, ‘if there were a strong argument that the claims were statute barred, then it would affect my ultimate view as to whether there was…a serious issue to be tried.’ In Camelot, with reference to this passage, O’Farrell J included the phrase ‘clear case’ alongside ‘strong argument’.

Whilst accepting that the test before me is whether there is a serious issue to be tried, Mr Paines urged me to determine the question of whether the claim(s) were out of time finally. Mr Tankel submitted that it would not be appropriate to consider the matter finally, and that in circumstances where the law was unclear, as he contended it was, I should determine that there was a serious issue to be tried, and leave a conclusion on the issue, either way, to be determined as part of the forthcoming strike out application. He also contended that I should take into consideration Boxxe’s application to extend time in respect of Grounds 1, 2, 3, 4 and 7 should they be considered to have been out of time. That application was made on the very eve of the hearing of the suspension application, and was not before the Court for determination at that hearing. Boxxe had proposed that that application be heard at the same date as the strike out application.

Regulation 92 PCR states:

As set out above, the Decision Notice was issued and received in the afternoon of 13 December 2022. Put shortly, if 13 December is counted within the 30 day calculation, the period expired on 11 January 2023 and (subject to Boxxe’s point relating to knowledge for the purposes of Grounds 5 and 6), the claims issued on 12 January 2023 were out of time. If 13 December 2022 is excluded from the calculation (such that the first day of the 30 days is 14 December 2022), the claims were issued in time.

Whilst the burden lies upon Boxxe to demonstrate there is a serious issue to be tried (rather than to disprove there is no serious issue to be tried, to the extent there is a real rather than semantic difference between the two in the present context at least), it is convenient to summarise SoSJ’s case first.

SoSJ contends that a period ‘beginning with the date when’, in ordinary parlance, means that the day upon which the trigger event occurs is included. Courts do not count in fractions of a day, and in the present case Boxxe knew or ought to have known about its claim when it was notified on 13 December 2023. SoSJ rely upon a line of authority which Mr Paine contended was applicable to ‘issue of proceedings’ cases such as the present one. Central to his submission was the dicta of Chadwick LJ in Zoan v Rouamba [2000] 1 WLR 1509. At paragraphs 23 and 24, the following observations were made:

In Trow v. Ind Coope (West Midlands) Ltd. [1967] 2 Q.B. 899, referred to by Chadwick LJ in the passage above, writs were issued at 3.05pm on September 10, 1965. They were served on the defendants on September 10, 1966, at 11.59 a.m. and 12.49 respectively. The Rules stated that ‘a writ… is valid… for 12 months beginning with the date of its issue’. The first question related to whether account could be taken of the time of day on which the writs were served. The Court of Appeal unanimously determined that this was not the case. As Lord Denning MR put it,

In Wang v University of Keele [2011] ICR 1251, the issue was when the right to bring an employment claim expired. The relevant legislation provided that the complaint was required to be presented ‘before the end of the period of three months beginning with the effective date of termination’. After a comprehensive review of the case law, including the two cases referred to above, Judge Hand QC sought to summarise the position, which usefully included the following points:

The principal issue in dispute was when the notice of termination was effectively given, and thus when the termination became effective for the purposes of calculating the time in which the complaint had to be brought. Having found that the termination was effective on 3 February 2009, Judge Hand QC concluded that ‘the period of three months beginning with the effective date of termination’ ended on 2^nd May 2009. It is clear that, although not saying so in terms, his interpretation of the wording of the statute was that the date on which the termination was effective was to be included in the calculated period of three months.

In Stevenson v General Optical Council [2015] EWHC 3099, the period for bringing an appeal under the relevant provisions of the Opticians Act 1989 was considered (a case in which, it is fair to point out, the claimant was not legally represented). The relevant section required the appeal to be brought, ‘within the period of 28 days beginning on the day on which the decision was served on you’. HHJ Dight, sitting as a Judge of the High Court, concluded that the day on which the decision was handed to the prospective appellant optician was to be included within the calculation, relying upon paragraphs 23 and 24 of Zoan, set out above. He also noted that the editors of the White Book in relation to CPR Rule 2.8 (on ‘Time’ within the broader Rule 2 dealing with ‘Application and Interpretation of the Rules’), refer to Zoan and stated that:

That note remains in the present edition of the White Book.

Finally, in Access for Living v London Borough of Lewisham [2021] EWHC 3498 (TCC), in which Mr Paines acted for the Defendant, and Mr Burton KC acted for the Claimant, it was common ground between counsel that the time for calculating the 30 day expiry under Regulation 92 was inclusive of the day on which the economic operator first knew or ought to have known that grounds for starting the proceedings, as recorded (by implication) in paragraph 13 of Jefford J’s judgment.

In the face of these authorities, Mr Tankel points out, rightly, that the question is one of construction. In the context of that exercise of construction, he argues that the Court should bear in mind the distinction which appears to exist in the authorities between the approach to the issue of a claim and to the service of a claim. Mr Tankel points out that it was the latter and not the former being considered in Zoan. In support of this distinction, he relies upon McGee on Limitation, 9^th Edn (2022), the well known practitioner textbook. Chapter 2 deals with ‘The Running and Expiry of Time’. At 2.005, the text states:

The position in relation to issue is then contrasted with the textbooks’ approach to service, in which reference to at least part of the line of authority relied upon by Mr Paines is made:

No reference is made to Zoan or the line of other cases relied upon by Mr Paines which, Mr Tankel argues, suggests their inapplicability to the issue of proceedings.

Mr Tankel also points out that several of the provisions within the Limitation Act 1980 use similar language to Regulation 92 of the 2015 Regulations. Regulation 92 refers to the period expiring “within 30 days beginning with the date when...” Sections 10B(2)(b) and 10B(3)(b) of the Limitation Act 1980 refer to the expiration of a period “15/30 years beginning with the date on which...”. It is right to note that these provisions have been recently inserted into the Limitation Act by the Building Safety Act 2022. It is instructive to look at the entirety of Section 10B, in this context:

The new section of the Limitation Act, which faithfully reproduces the language of section 150 of the Building Safety Act 2022, uses both the language of ‘from’ and ‘beginning with’. Did the drafters intend the distinction drawn by Zoan? Or are the concepts being used interchangeably? If the former, it is not immediately obvious what purpose was sought to be achieved by requiring the 30 year period to be calculated excluding the date of accrual of the cause of action and requiring the 15 year period to be calculated inclusive of that date. Irrespective of the confusion potentially caused by this very recent inclusion within the Limitation Act 2022, it should be noted that the general language of the pre-existing Limitation Act 1980 is generally that actions shall not be brought after the expiration of a certain period ‘from’ the date on which the cause of action accrued.

Mr Tankel also relies upon European Council Regulation 1182/71, which governs the calculation of time periods in certain EU instruments, including Directive 2014/24/EU. Article 3 of Regulation 1182/71 provides that:

“Where a period expressed in days, weeks, months or years is to be calculated from the moment at which an event occurs or an action takes place, the day during which that event occurs or that action takes place shall not be considered as falling within the period in question.”

This is, unsurprisingly, then reflected in the PCR. As set out in Regulation 2(3), this method of calculation is the prescribed method of interpretation for any reference to a period of time in Part 2 of the PCR (containing the substantive requirements). Mr Paines points out, and Mr Tankel fairly accepts, that this does not apply to Part 3, in which Regulation 92 sits, but Mr Tankel maintains that as a matter of construction, it remains relevant. In this context, he emphases the requirement to promote certainty, as identified in Uniplex (UK) Ltd v NHS Business Services Authority [2010] PTSR 1377, in which it was stated:

Finally, recognising it as a point which is no more than forensic, Mr Tankel points out that on 20 December, in its letter before action, Boxxe sought an extension to the standstill period until midnight on 10 January 2023. In its substantive response, the SoSJ voluntarily extended the standstill date to 23.59 on 12 January 2023, which is the expiry of the limitation period when calculated in the manner Boxxe contend is correct. There is some force in the suggestion that the date was probably put forward on the basis that, at least at that time, SoSJ had themselves calculated the limitation period by excluding the date the Decision Notice was issued.

The proper construction of Regulation 92 is an important issue of law, and in the context of the very short 30 day period in which to bring a claim, the inclusion or exclusion of a day may be a matter of significance. Indeed, given that the law clearly requires fractions of a day to be ignored, the effect of sending a decision notification at the end of a business day, may in practice mean that the 30 day period becomes 29 days.

Notwithstanding what appears to be a formidable line of authority, as presented by Mr Paines, I consider that it is inappropriate for me to give in to the temptation to decide the matter finally in the context of this application for the lifting of the suspension. The matter is not so clear cut as to conclude that there is not a serious issue to be tried. The overall time for Mr Tankel to have made detailed submissions on the point, given the need to address the other factors, was short and in fairness to him, he did not attempt to cram the proverbial (and imperial) quart into the pintpot. I consider that it is appropriate that the ultimate determination of the issue should follow fuller argument and submission, which will take place no doubt relatively shortly in any event in the context of the strike out application which has already been issued.

In these circumstances, it is not necessary for me to consider whether it would have been appropriate for me to take into account the fact that there is an extension of time application in the background, even though it was not before the Court. However, had it been necessary for me to do so, I would have concluded that it would not have been appropriate for me to do so. The consequences for Boxxe had I acceded to SoSJ’s arguments on limitation and determined there was no serious issue to be tried would merely have been the consequences caused by the timing of its extremely late decision to apply for an extension.

It is also not necessary for me to consider whether the date on which Boxxe knew or ought to have known grounds 5 and 6 is later than 13 December 2022, and it is appropriate to leave that question to the strike out application.

In cases of this nature, arguments around the adequacy of damages usually revolve around claimed loss of reputation, the effect on the viability of the unsuccessful tenderers business, loss of other contracts, redundancies etc. That is not the case here. There are three grounds upon which Mr Tankel seeks to persuade me that it would not be just in all the circumstances that Boxxe is confined to its remedy in damages. These are:

Boxxe’s own claimed loss is relatively modest. It is particularised as £421,817.86, a loss relating to a 3 year contract. This equates to lost profit of around £140,000 a year, in the context of a business which reported gross profits in 2021 of £17.6m.

However, Boxxe relies upon the losses of its key sub-contractor, Involve, which it is said will suffer losses of approximately £1.5m. It is said that on the current state of the law, Involve’s standing to bring a claim in its own right is highly uncertain (Mr Tankel referred to Camelot v The Gambling Commission [2022] EWHC 1664 (TCC) at [60] to [77], in which O’ Farrell J, whilst accepting there was a serious issue to be tried, identified the difficulties that may be faced by sub-contractors, at least in some circumstances.

Whilst Mr Tankel urges upon me that the impact upon the interests of third parties, who do not have a reliable remedy in damages in their own right, is a materially relevant factor that the Court both can and should take into account in its overall assessment, that is not a submission which assists him on the question of the adequacy of damages. On that question, it is clear that the position of Involve is irrelevant and should not be taken into account regarding the adequacy of damages. This conclusion is consistent with the view taken by Sir Antony Edwards-Stuart in Circle Nottingham Limited v NHS Rushcliffe Clinical Commissioning Group [2019] EWHC 1315 (TCC) in which the Court had to consider whether to take account of the losses which may have been suffered by the group within which the unsuccessful tenderer company (an SVP) sat. The learned judge said this:

The position is even starker here. Whether or not the case is uncertain, Involve could have started a claim as a second claimant to this action, which would have added little by way of additional effort, and Involve’s claim of around £1.5m would have been pleaded as particulars of its loss. Faced with this, and assuming Involve was then also a respondent to SoSJ’s application to lift the suspension, no doubt Mr Paines would have argued that damages were an adequate remedy for Involve too. What would be wrong in principle is for Involve to determine not to seek to claim damages, and that then be prayed in aid by Boxxe as relevant in some way to the question of the adequacy of its damages.

Mr Tankel points out that by paragraph 56(4) of the Defence, SoSJ avers that the alleged breach is not sufficiently serious in Francovich terms to justify an award of damages. Boxxe maintains that the sufficiently serious threshold is more than made, but Mr Tankel argues that the fact that SoSJ has chosen to make this averment gives rise to the risk that, even if Boxxe’s allegations of breach are made out, there will be no award in damages. In short, it is said that SoSJ cannot have it both ways: contending both that damages will be an adequate remedy, and that as a matter of principle damages should not be available. He also points out that the risk of failing to resist the lifting of suspension and then receiving no award of damages became a reality recently in the case of Braceurself Limited v NHS England [2022] EWHC (TCC).

In the face of this argument, Mr Paines first takes a pleading point, asserting that Boxxe has not pleaded that the breaches are a sufficiently serious breach for the entitlement to state liability damages. This is right. That averment is a necessary part of Boxxe’s claim to an entitlement, and it would not technically have been necessary for SoSJ to assert any more than the absence of a positive case by Boxxe. That said, it is also right that SoSJ has gone further in its Defence and positively averred that the breaches alleged do not comprise sufficiently serious breaches of the PCR 2015. This positive averment places the matter in issue in the proceedings.

Pleading point aside, therefore, Mr Paines relies upon the judgment of O’Farrell J in Alstom Transport UK Limited v Network Rail [2019] EWHC 3585 (TCC) in which she stated, ‘As a preliminary observation, in the context of a procurement challenge, although each case must be examined on its merits, if a breach of EU-based law is not sufficiently serious to satisfy the Francovich conditions for an award of damages, it is unlikely to be sufficiently serious to justify setting aside the contract under challenge’. In other words, if the outcome is that the breach is not sufficiently serious for the purposes of an award of damages, and by extension probably not sufficiently serious to justify setting aside the contract under challenge, it ought not be a factor which should weigh (or weigh heavily) in preventing the letting of that contract under challenge in the context of an application to lift the automatic suspension. There is some force in this observation, and it seems to me that it cannot be right (as would follow from Mr Tankel’s submission) that every time in which the question of whether the breaches satisfy the Francovich conditions is in issue in proceedings, damages are automatically rendered inadequate for the purposes of considering whether to lift a suspension.

Mr Paines also submits that in circumstances where the Court is not in any position to determine at this early stage of the proceedings whether the breaches will fall to be characterised as ‘sufficiently serious’ or not, this potential obstacle to success on a damages claim can be treated no differently from any other obstacle (including the merits of the claims of breach). This is not a good point: the consideration of adequacy of damages must be predicated on the assumption that breach is established. The Francovich issue is different: the point is, having established breach, causation and loss, the unsuccessful tenderer is still potentially left without an effective remedy, because the breach is not sufficiently serious.

Finally, Mr Paines states that, although his client should not be required to do so, SoSJ is prepared to give an undertaking, should the suspension application succeed, that if it is ultimately established as a matter of fact that a breach or breaches took place but for which, had they not occurred, the contract would have been awarded to Boxxe rather than SCC, SoSJ will not pursue its pleaded averment that the breach(es) do not meet the Francovich conditions. Mr Tankel fairly accepted that the Francovich point would be a point no more were this undertaking to be given. I consider that this is a sensible way in which to cut through the issue on the facts of this case, and the offered undertaking should be given if I consider it otherwise appropriate to lift the suspension.

Mr Tankel argues that the assessment of damages will be difficult in this case because:

Neither of these points are good. As to the first, the ‘range of options’ forms no part of Boxxe’s pleaded case. In any event, even if it did, courts deal every day with counter-factuals. It is a matter of assessing the evidence and coming to a conclusion on balance of probability. Once the counter-factual outcome is determined, the entitlement either follows or it does not. The need for such an assessment and judgment does not mean that damages (if an entitlement is established) would be an inadequate remedy. As to the second, it is clear from the evidence of Mr Edgerton, which I accept, that HMCTS has every intention of spending its allocated budget. Moreover, by the time of the ultimate trial, there will be good evidence of what HMCTS has in fact spent and a clear basis will exist to determine what profit Boxxe would have made had it been successful (should that arise).

In these circumstances, and in light of the undertaking to be given by SoSJ as set out above, I find that damages will be an adequate remedy, and it is just in all these circumstances that Boxxe should be confined to its remedy in damages.

Mr Paines submits that this should lead inexorably to the granting of SoSJ’s application. In particular, the existence of a potential trial date should not be taken into account as any part of the assessment of the adequacy of damages. In doing so, he draws to my attention the judgment of Joanna Smith J in Kellogg Brown & Root Limited v (1) Mayor’s Office for Policing and Crime (2) Metropolitan Police Service [2021] EWHC 3321 in which 8 carefully reasoned sub-paragraphs, Joanna Smith J rejected the submission that the availability of a swift trial was effectively a pragmatic solution which should be brought into account in considering the justice of the case in the context of adequacy of damages.

Mr Tankel, however, argues that the question of the adequacy of damages should not be determinative. He urges upon me the approach of Sir Antony Edwards-Stuart in Circle Nottingham Limited, in which he concluded that having found that it would not be unjust to confine the claimant to its remedy in damages, that is usually the end of the enquiry. However, the learned judge continued at paragraph 19:

In this context, Mr Tankel urged that I should apply a modified American Cynamid test of the type used in judicial review proceedings, in which once it is established that there is a serious issue to be tried, the Court moves immediately to consider the overall balance of convenience (although he rowed back somewhat from this in oral argument, accepting that the adequacy of damages remained a ‘weighty’ issue).

In my judgment, there is nothing on the facts of this case which allows me to find that my conclusion that damages are an adequate remedy is not determinative of the success of the application to lift the stay. I respectfully agree with the reasoning of Joanna Smith J in Kellogg, and whilst the ability of the Court to accommodate an expedited trial may provide a pragmatic solution where matters are finely balanced, that balancing exercise does not arise in circumstances where it is clear that damages are an adequate remedy. It must not be forgotten that O’Farrell J’s decision in Draeger Safety UK Ltd v London Fire Commissioner [2021] EWHC 2221 (TCC) was made in the context of her finding that damages were arguably not adequate.

That said, and not least because both parties have engaged in the evidence submitted, and in written and oral argument, on the questions of the adequacy of damages to SoSJ and the balance of convenience, I will briefly set out my views in respect of those aspects of the parties’ respective cases.

There are four bases upon which SoSJ contend that damages would not be an adequate remedy for them. These can be summarised as failure, functionality, funding and increases in price. Mr Edgerton’s evidence is that the purpose of the Competition was to upgrade and replace existing, end of life, AV equipment, and that the new equipment would be both more reliable, and have greater functionality, than the existing equipment. There is debate between the parties as to the precise anticipated failure rate, and the impact of failure on the Court service. However, whatever the precise extent of failure, there is no doubt that a delay of 6 months (assuming a July expedited trial was accommodated) to the implementation of the project would be likely to have a real life impact in avoidable failures and delay to the improvement of functionality on the Courts Service which would not adequately be compensated for in damages. However, even were this not the case, of acute concern to HMCTS, as explained by Mr Edgerton and which I accept, in the present case is the fact that if the suspension is continued, SoSJ will lose the first year’s funding (£7.4m inc VAT) and so simply will not be able to buy the first year’s tranche of equipment under the Contract. Further negotiation with HM Treasury might be possible, but it would be novel and uncertain and would itself cause substantial delays. I also accept that if there were to be a rise in prices (which have not been held open beyond 31 March 2023), there is a possibility that the Competition itself may have to be abandoned. The effect of this would be to jeopardise, and at the very least, significantly delay the realisation of the benefits in terms of reliability, functionality and enhanced quality, and this impact on the Court service would not be compensatable in damages.

In these circumstances, I find that damages would not be an adequate remedy for SoSJ.

Given my findings above, it is obvious that the course of action which is likely to carry the least risk of injustice is to lift the automatic suspension:

For these reasons, the application to lift the suspension succeeds, on terms that if it is ultimately established as a matter of fact that a breach or breaches took place but for which, had they not occurred, the contract would have been awarded to Boxxe rather than SCC, SoSJ will not pursue its pleaded averment that the breach(es) do not meet the Francovich conditions.