[2025] EWHC 1889 (Comm)

Unlike the Issues addressed above, Issue 5 does not turn upon any great debate about what the SPA means. Instead, it requires an assessment of the factual evidence given at trial. Issue 7 does, however, require consideration of the true meaning of the contractual provision against which LCG’s alleged knowledge falls to be tested.

The evidence on both issues is to be considered against the language of the definition of ‘Disclosed’ and clause 6.3 (for Issue 5) and paragraph 12.1 of Schedule 5 (for Issue 7) of the SPA. Those provisions are set out in paragraphs 108 and 111 above. The defendants take a point on the true construction of paragraph 12.1 when read in the light of clause 6.3. Their counterclaim said the “and” between paragraph 12.1.1 and 12.1.2 should be read as “or” and, although the point was not pressed in argument, that, if necessary, the SPA should be rectified accordingly on the basis of mutual mistake.

Whereas a determination of Issue 5 in the defendants’ favour would operate to pre-empt an adverse finding against them on Issue 6 (the Breach Issue), because it would result in a qualification to the warranty perhaps otherwise breached, the same cannot strictly be said about Issue 7. On the face of it (though see the arguments summarised below) the language of paragraph 12 of Schedule 5 is that the defendants “shall not be liable” on the warranty claim if the grounds for it were disclosed and were the subject matter of paragraph 12.2 knowledge. As its place within Schedule 5 (‘Limitations on Claims’) confirms, that is the language of a defence to an otherwise established breach of warranty; at least “to the extent” that the facts and matters giving rise to it were disclosed and known. Nevertheless, Issues 5 and 7 are so clearly inter-related that it is sensible to deal with them together and before turning to issues of breach of warranty.

I have noted above that the defendants did not press the counterclaim for rectification of paragraph 12 of Schedule 5 which was pleaded on the basis that the inclusion of “and” rather than “or” reflected a mutual mistake by both parties. The pleading did not elaborate upon that by alleging that it was their common intention, up to the time of signing the SPA, that it should have read “or” and in LCG’s skeleton argument reference was made to a number of earlier drafts of the SPA which (by using the word “and”) undermined the claim for rectification on that basis.

Instead, Mr Sims KC said this was instead a situation for “constructional rectification” (i.e. correction of the provision as a matter of construction) of the kind addressed by the House of Lords in Chartbrook Ltd v Persimmon Homes Ltd [2009] UKHL 38; [2009] 1 AC 1101. The defendants’ argument requires me to conclude that this is such a case, requiring the correction of the word “and” to read “or” through the process of interpretation. In the absence of that conclusion, through the draftsman’s use of the word “and”, paragraphs 12.1.1 and 12.1.2 must be read as imposing two pre-conditions to the existence of a defence to liability, not just one of two alternative pre-conditions as the defendants contend.

The argument for the defendants, in support of such a correction, rests upon their contention that clause 6.3 and paragraph 12 of Schedule 5 to the SPA provide them with two separate defences (in circumstances where no fraud, dishonesty or wilful concealment is alleged, when Schedule 5 as a whole would be disapplied). As I have indicated in my more general observation, about addressing Issue 5 before Issue 6, I do not think it is strictly correct to describe clause 6.3 (i.e. that part of it which confirms that the giving of a warranty is qualified by any disclosure in relation to its subject matter) as giving rise to a “defence”. If that part of clause 6.3 bites then there is no warranty for LCG to sue upon in the first place, as the disclosure has qualified the giving of it. However, the defendants would say that is not material to their analysis. Their starting point is that the language of clause 6.3 identifies two means by which they can escape liability under a warranty. The first is by establishing the relevant matter was “Disclosed”. The second is by establishing what is described in the heading to paragraph 12 as “Purchaser’s Knowledge”.

The essence of their point is that, unless paragraph 12 is read as containing two alternative limbs (“or” rather than “and”) then, despite clause 6.3 expressly referring also to a potential defence under that provision in Schedule 5 (“subject to Disclosure and to the limitations set out in Schedule 5”), paragraph 12 adds nothing to the proviso to liability based on disclosure. That is because paragraph 12.1.1 refers again to disclosure (“Disclosed in the Disclosure Letter or Disclosure Documents”). If reliance upon paragraph 12 not only requires the defendants to establish they have made such disclosure but also (“and”) that LCG had actual knowledge of the facts, matters or circumstances disclosed (see paragraphs 12.1 and 12.2) then the reference in clause 6.3 to Schedule 5, at least so far as paragraph 12 of it is concerned, is surplusage and meaningless. On that basis, the defendants say, disclosure (“Disclosed” for clause 6.3 purposes) gives them a complete defence without the parties needing to consider the impact of paragraph 12 of Schedule 5. Therefore, the correct interpretation, which properly reflects the twin-limbed proviso language of clause 6.3 (I ignore the inapplicable “proviso to the provisos” for cases of fraud etc.) is one that gives paragraph 12 some meaning. It can only have meaning, beyond a defence based on disclosure, if establishing LCG had actual knowledge is an alternative defence (“or”) to a defence based on disclosure.

LCG’s counsel responded by saying that the defendants’ argument does not meet the Chartbrook test and ignores the bargain struck by the parties. Both sides were experienced commercial parties, well-advised and represented by transactional lawyers. There was no proper basis for the court to conclude that they intended “or” rather than “and” (there being no evidence to suggest that) or that LCG would have agreed to it. Instead, they agreed upon language whose plain purpose was to ensure that any relevant knowledge should be apparent from disclosure and that LCG actually had it.

Mr Booth KC said that, as ‘Disclosure Documents’ in clause 1.1 of the SPA were defined to mean “the documents which have been made available in the Data Room” (and indexed in the annex to the ‘Disclosure Letter’), there was nothing absurd or non-sensical in a provision which protected LCG from being attributed with knowledge of a matter buried deep in the disclosure documents. For that reason, the parties had specifically agreed in paragraph 12.1.2 and 12.2 upon the need for actual knowledge (and awareness) on the part of named individuals, to be attributed to LCG, rather than constructive knowledge only. [I should remark that paragraph 12.2 appears to me to be curiously worded in the way it expresses the attribution of knowledge. However, it is reasonably clear in saying that LCG derives its knowledge from the named individuals, not the other way round and, although it could be read as saying that knowledge on behalf of all those named is required before LCG is to be attributed with it, it is accepted that the knowledge of any one of them would suffice.] This, counsel said, was also consistent with what was required for a matter to be ‘Disclosed’ which meant “fairly disclosed with sufficient detail to identify the nature and scope of the fact, matter or information concerned ….” Paragraph 12.1, read as a whole, also precluded any argument by the defendants that LCG had knowledge of a matter from another source.

Mr Booth KC said that the word “and” was one of the simplest words in the English language and it was the defendants’ argument which resulted in redundant and meaningless language. As he summarised the defendants’ position, the parties might have agreed that clause 6.3 should qualify the terms of each warranty simply by reference to “matters Disclosed”, and ended there, but they did not. They agreed to say something more about what, from the disclosure perspective, would be required to qualify them. He made the irrefutable point that in that further language (of paragraph 12) they had, both before and after the key “and”, used the word “or” when that was appropriate. He likened the current issue to one which requires sticking to a recipe that requires “salt and pepper” rather than only one of those seasonings.

The defendants, who bear the initial evidential burden on this aspect of the case, contend that LCG had knowledge of the facts, matters and circumstances giving rise to the Clawback through Daniel Dowson (one of the individuals identified for that purpose by paragraph 12.2 of Schedule 5). Their case is that those facts, matters and circumstances became known to him through the due diligence undertaken by LCG and that his failure to remark upon them prior to the entry of the SPA reflected his own subjective view that they would not result in a claim and/or were minor matters that could be resolved after completion.

The defendants place heavy reliance upon an ‘Executive Summary’ prepared by Tim Williams (the Company’s Director of Contracts) which explained the Company’s Study Programme for 16-19 years olds (the “SPD”). Mr Williams, who gave evidence for LCG, explained that the SPD, as it evolved over the years, was submitted by the Company to ESFA each year to explain the Company’s approach to planned hours and the “nesting” of qualifications. He said “this is a document I put together each year, to document how and why we do what we do.” He discussed the 2020 version of the SPD with Mr Dowson, after completion of the SPA, in connection with the 2022 Audit.

Mr Lewis repeatedly referred to the significance of the 2018 version of the SPD during his cross-examination. Both he and Mr Williams said the SPD in 2018 was in substantially the same terms as the later version (discussed between Mr Williams and Mr Dowson in February 2022) and that the Company had received a clean audit in 2018 (it was subject to some “internal control recommendations” but for which the funding implications were “zero”).

The defendants say the SPD spelled out the Company’s approach in clear terms and, even though that was found in the 2022 Audit to have been incorrect, no further primary facts, matters or circumstances are or would have been necessary to establish that the Company was misapplying the Funding Rules in respect of Condition of Funding. It is not their case that the SPD was available to LCG in the Data Room for due diligence purposes. Instead, a significant part of their case on this aspect rests upon the court finding that, while LCG was conducting due diligence, Mr Dowson and Mr Williams had a discussion during which Mr Williams would have explained the Company’s approach and thereby bought such facts and matters to LCG’s attention.

The defendants’ counsel submitted that the evidence supports the conclusion that it is more likely than not that the Company’s approach to planned learning hours and Condition of Funding was discussed by Mr Williams and Mr Dowson at the due diligence stage. They said the terms of their later email exchanges on 2 February 2022, concerning the SPD, supported this conclusion, including the absence of any obvious concern expressed by Mr Dowson and the absence of comment by him upon the “nesting” of qualifications.

They also relied upon the terms of an email which Mr Williams sent to Mr Dowson on 3 March 2022, in response to Mr Dowson’s noting some “audit challenges around GCSE & qual set up”. Mr Williams said:

“Yes, like i said to the auditor it is always planned hours and condition of funding! Don't think there is an issue with the "nested" quals, just how you show the hours. But planned hours that are timetabled are planned hours that are timetabled, or have ESFA got a different definition they haven't shared?”

The defendants argued the likelihood is that Mr Williams would have previously explained to Mr Dowson during the due diligence process what the Company did and why it did it.

The defendants also rely upon the fact that information from the ILRs and PDSAT reports was available to LCG in the Data Room. That revealed a statistically significant amount of learners recorded as being on 538 planned hours and also those with planned learning hours which greatly exceeded that required for achieving the recorded aim of an Award. This information was available to Mr Dowson upon whom, as Mrs McLeish (another such named person under paragraph 12.2) confirmed in evidence, LCG relied upon in relation to the due diligence undertaken in relation to English funding.

The defendants point to the difference in approach adopted by LCG during due diligence in relation to the Welsh and English parts of the Company’s business. In relation to Welsh funding, LCG engaged an external consultant (Collette Kelly) to undertake a funding audit but not the English funding. In his witness statement, Mr Dowson referred to the fact that he had with an assistant taken “samples of files and effectively did a sort of mini-audit on those file samples”. Mr Higgins (Mr Dowson’s immediate superior) said in evidence that he would have expected Mr Dowson to have done some sort of audit on the English funding.

However, in testimony Mr Dowson confirmed he did not undertake any sampling of learners’ files. By an email dated 22 July 2021, noting that LCG had access to the Data Room for just over a week, Tony Davis of LCG informed colleagues (including Mr Higgins, Mrs McLeish and Mr Dowson) that, in relation to English funding, it was “agreed no need to audit English funding following PFA audit in 2018, which had zero error rate, but Burness Paull to capture protection from 2018 to date in the SPA.” Mr Dowson explained that Mr Davis (another individual named – with a misspelt surname – in paragraph 12.2) was responsible for managing the due diligence exercise overall.

The defendants say that LCG was content to adopt a very light touch approach on English funding matters, no doubt comforted by the proposed Funding Indemnity. They point to LCG’s internal document ‘Project Beacon: Summary of Key Points from Internal Due Diligence’. In relation to “compliance”, this referred only to a “significant volume of compliance issues raised by audit work with Colette” (with an “amber’ flag, to denote “a matter of which the team should be aware but is not a major risk”) which referred to the Welsh funding. The suggested next steps in that document were “Audit implementation plan post acquisition, unredacted minutes for any evidence of compliance concerns, indemnities in case of clawback.”

LCG’s position is that the evidence at trial clearly shows that it had no knowledge that a breach of the Funding Rules had taken place as opposed to awareness of the theoretical risk which justified the parties’ agreement upon the Funding Indemnity. I have already considered the basis upon which LCG required the inclusion of the Funding Indemnity in the SPA in addressing Issue 3 above.

LCG’s counsel submitted that the information in the Data Room did not provide LCG with knowledge of matters giving rise to a breach of the Funding Rules either in relation to planned hours or Condition of Funding.

Both on the present issue and Issue 4 (the Vendors’ Knowledge Issue), during the trial Mr Booth KC illustrated the Company’s approach to Condition of Funding by reference to the information given in the PDSAT report in relation to one particular learner (learner reference number ending “RJ9” and initials “KB”). I refer again below to the information in relation to KB and also that provided in the PDSAT report for another learner identified by LCG (learner reference ending “G51” and initial “JH”), in respect of the planned hours issue, in my assessment of the evidence on the Purchaser’s Knowledge Issue.

The essential point made by LCG was that the information in the PDSAT report alone would not have alerted it to an issue with compliance with the Funding Regulations either over Planned Hours in Study Programmes or the Condition of Funding rules. Aa key component to understanding and identifying the Company’s practices in relation to both was the Company’s electronic record known as the ‘Planned Hours Template’. The Planned Hours Template was available to RSM for the 2022 Audit but was not available to LCG in the Data Room.

The Planned Hours Template recorded what was planned for learners on enrolment. The number of hours planned for a learner should have been reflected in the PDSAT record. It was described in the SPD (which referred to it as the “planned hours calculator”) as a document, compiled by reference to the ILR, as follows:

“Once the ILR has been completed for planned learning activities, these are entered into the planned hours calculator by Head Office staff.

This calculated the expected hours during the contract year (excluding planned leave dates) for the learner which is entered onto PICS to ensure accurate planned hours are claimed from ESFA. The calculator takes into account the average attendance based on the previous year’s records.”

This will determine the appropriate funding rate for each learner based on planned attendance. If the learner leaves within the 6-week time frame the hours are reconciled and amended per the guidance.”

The Planned Hours Template was not in the Data Room. Neither were the ILRs/PLRs upon which the Planned Hours Template was based. Without this critical piece of the jigsaw, LCG argued, the information held in the PDSATs did not give rise to any cause for concern. LCG’s submissions in relation to JH and KB (addressed below in the light of the evidence of Mr Dowson) highlight how it is only by comparing the PDSAT report with the Planned Hours Template that the funding issues emerge.

Even if the Planned Hours Template had been available, LCG said it would not have assisted in identifying the issues before the SPA. That is because the Planned Hours Template only contains learner names (and not learner reference numbers) whereas the PDSAT data was anonymised and only contained learner reference numbers. RSM were only able to join the dots as a result of a document identified as the ‘Risk Spreadsheet’ The Risk Spreadsheet was an Excel workbook initially prepared by RSM but as a working document to which Mr Dowson added information during the course of discussions which LCG had with RSM and ESFA aimed at reducing the funding clawback. He attached it to an email dated 1 April 2022 to Mrs McLeish and Mr Higgins and at that stage it identified (as alternatives to the Clawback amount) the potential figures of £982,000 and £763,000. The Risk Spreadsheet obviously post-dated the SPA.

In relation to the planned hours issue, LCG picked up a reference in the defendants’ written opening which demonstrates the obvious difficulties in their suggestion that LCG was aware of matters giving rise to a clawback. The defendants’ opening identified one learner (learner reference ending “LW7”, which Mr Sims KC recognised was an erroneous reference in his cross-examination of Mr Dowson) who was identified in a PDSAT spreadsheet entitled “Learners with high planned hours in-year” (tab ‘21Y-203’). The defendants pointed to total planned hours of 888 for that student (450 planned learning hours and 438 EEP hours and, therefore, recorded as within funding band 5) and yet the core learning aim was identified as an Award with 2 month learning period (3 August to 2 October 2020) identified for it. The defendants said this clearly showed an erroneous nesting approach and was therefore known to LCG.

LCG says there is a striking irony in the fact that this example used by the defendants to assert that it was easy for LCG to see the Company’s error did not in fact involve any such error. Not all nested qualifications involved funding errors. LCG referred to the Risk Spreadsheet where learner “LW7” was not identified as part of the Planned Hours Over-Claim. I note that the other two rows in spreadsheet ‘21Y-203’ for that learner reveal that, in addition to work experience/placement, a second Award was a further learning aim which (in view of the planned learning period for that Award going beyond the actual and planned leaving) may well have been completed.

In Chartbrook, at [22]-[25], Lord Hoffmann explained that two conditions need to be satisfied if the literal meaning of a contractual term is to be corrected through the process of interpretation. First, it must be clear that there has been a mistake, and, secondly, it must be clear what correction ought to be made in order to cure the mistake. Lord Hoffmann said: “All that is required is that it should be clear that something has gone wrong with the language and that it should be clear what a reasonable person would have understood the parties to have meant.”

Chartbook was a case where the House of Lords concluded that something had gone wrong in the syntactical arrangement of words and the product was grammatical ambiguity. Lord Hoffmann recognised that a “strong case” is required to persuade the court that something must have gone wrong with the drafting. The case was made out there because the literal interpretation of the clause made “no commercial sense”. It was clear from the wider terms of the contract, and also the subject matter of and definitional term used in the clause under consideration, that the clause provided for the possibility of a larger payment being made to the claimant depending upon a contingency as to the sale price of the properties built by the defendant. The interpretation of the clause in accordance with “the ordinary rules of syntax” meant there was virtually no element of contingency and that the payment would be triggered by sales at about a quarter of the price envisaged by the parties at the time of the agreement. The result was “arbitrary and irrational”.

The decision of Nugee LJ in Monsolar IQ Ltd v Woden Park Ltd [2021] EWCA Civ 961; [2022] 2 P&CR 10, at [31]-[33], confirms that the test for correcting the words of a contract through the process of interpretation is quite a high one. He contrasted the language of a provision which is commercially unattractive or unreasonable for one side from one that produces “irrational, arbitrary, nonsensical or absurd results”.

Applying the higher test is necessary to ensure that the court does not fall foul of the prohibition in Arnold v Britton against relying upon its (or one party’s) perception of commercial common sense in order to relieve a party of “bad” or onerous aspects of the bargain which the ordinary and natural meaning of the contractual words shows they have reached. Instead, and consistent with parties sometimes using the term “constructional rectification”, the court is required to consider the contractual language from the perspective of both parties to see whether it reveals a drafting mistake which, evidently, cannot be said to reflect their intentions at the time of the contract in the light of the commercial consequences which appear to flow from it.

In my judgment there are obvious problems with the defendants’ reliance upon the Chartbrook principle which indicate that their interpretation of paragraph 12 cannot be correct.

The first problem, addressing their suggestion that on LCG’s case the reference in clause 6.3 to Schedule 5 is unnecessary, is that this ignores the fact the reference is justified regardless of its implications for Issues 5 and 7. As I remarked in exchanges with counsel, it is a reference to Schedule 5 generally and not just to paragraph 12 of it in particular. Absent fraud etc., a claim upon a General Warranty is subject to all of the limitations in Schedule 5, including those which have resulted in Issues 1, 2 and 11. This means there is no real scope for invoking the presumption against redundant words. Even if that were not the case, the presumption (i.e. invoking a general requirement that the court should give meaning to the chosen words if it can) is not of great weight when authorities such as Arnold v Britton reflect the court’s awareness that, in the real world, some contractual language may not only be inapt but unnecessary.

As for the defendants’ suggestion that, when clause 6.3 is read as being subject to paragraph 12, in particular, LCG’s interpretation is nonsensical, I do not agree. The natural and ordinary meaning of the respective provisions (with paragraph 12.1.2 being read as an “and” rather than “or”) reveals them to be consistent with one another.

The key point is that the exercise of disclosure (and it is important to bear in mind that for present purposes the relevant disclosure is that identifying facts, matters and circumstances which relate to the General Warranties rather than about the Company’s business and affairs falling outside their scope) should lead to actual knowledge on the part of LCG. To express it another way, given that the parties specifically agreed that LCG’s knowledge (through any of the named individuals) had to be actual knowledge rather than constructive knowledge, the defendants should not be permitted to argue that disclosure which might only have resulted in constructive knowledge on the part of LCG should suffice. Subject to what I say below about the meaning of ‘Disclosed’, the introduction of the “or” in place of “and” might provide the basis for such an argument. As Mr Booth KC observed, it would also provide the defendants with scope for argument that LCG had knowledge of a matter from another source even though they themselves did not disclose it.

I have suggested that the defendants’ interpretation of paragraph 12 gives rise to the opportunity for them to argue that LCG had constructive knowledge of a relevant matter even if each of those named in paragraph 12.2 lacked actual knowledge of it. As I observed in exchanges with counsel, the exercise of disclosure can in very general terms (free from any particular contractual requirements or exactness) be said to be all about giving the party provided with the fruits of it the opportunity to find out things; or what a lawyer would describe as notice or constructive knowledge. Both sides referred to the Court of Appeal decision in Infiniteland Ltd v Artisan Contracting Ltd [2005] EWCA Civ 758; [2006] 1 BCLC 632 where Chadwick LJ, at [69], approached his decision upon the adequacy of the warrantor’s disclosure in that case by noting the well-established distinction between actual knowledge, on the one hand, and such imputed or constructive knowledge on the other.

In Infiniteland, the seller warranted that “the contents of the Disclosure Letter and of all accompanying documents…….fully, clearly and accurately disclosed every matter to which they related”. The seller’s disclosure letter contained general disclosure which referred in quite general terms to “all matters from the documents and written information supplied by us to your reporting accountants, Pridie Brewster” and “all matters contained or referred to in the following documents supplied by us to you in the green lever arch files…”. Chadwick LJ, said, at [70], with my underlining:

“It would have been open to the Purchaser to refuse to accept disclosure made in general terms by reference to what had been supplied to its reporting accountants; and to insist that it would only accept disclosure which was specific to each individual warranty. But the Purchaser did not choose to take that course. It was content to rely on its reporting accountants to identify from the documents supplied to them – and to report on – the matters about which it needed to be informed. That is the effect of the terms in which disclosure was made under the disclosure letter; and, for whatever reason, those were the terms upon which the purchaser was content to accept disclosure. In those circumstances, as it seems to me, the disclosure requirement was satisfied in relation to such matters as might fairly be expected to come to the knowledge of the reporting accountants from an examination (in the ordinary course of carrying out the due diligence exercise for which they were engaged) of the documents and written information supplied to them.”

In the present case the defendants gave disclosure by the Disclosure Letter (as defined in the SPA) dated 29 October 2021. It included the following statements (with my underlining):

“This letter is the Disclosure Letter defined and referred to in the SPA and constitutes formal disclosure to the Purchaser for the purposes of the SPA of the facts and circumstances which are or may be inconsistent with the General Warranties. Such facts and circumstances will, to the extent Disclosed, be deemed to qualify the General Warranties accordingly.

…………

By way of general disclosure, the following matters are, to the extent Disclosed, disclosed to the Purchaser:

………

2.

All matters contained in the documents included in the Disclosure Documents, an index of which is appended to this Disclosure Letter at Annex 1

………..”

Annex 1 is a detailed 46-page list of documents identifying further folders containing many files.

The Disclosure Letter then went on to state (before setting out a table identifying certain warranties by number): “The following specific disclosures are made in relation to the General Warranties”. Warranty B5.2.2 (including the Key Warranty) was not the subject of any specific disclosure.

Recognising that the Disclosure Letter uses the defined term ‘Disclosed’, the underlined passages demonstrate the potential for the type of situation encountered in Infiniteland to arise, so that there might be scope for an argument that LCG’s ability to find out from the Disclosure Documents that, for example, the Key Warranty was qualified by something revealed in those documents.

However, the reason why the SPA is to be read as precluding the scope for such an argument (because “actual knowledge and awareness” under paragraph 12.2 is required) emerges from the use of the defined term in the Disclosure Letter. ‘Disclosed’ means “fairly disclosed with sufficient detail to identify the nature and scope of the fact, matter or information concerned ……” The defendants’ counsel recognise that this type of language imposes an obligation to bring a matter to LCG’s attention, through a positive statements about the true position, rather than leaving LCG to work things out for itself: see Triumph Controls UK Ltd v Primus International Holding Co [2019] EWHC 565 (TCC), at [329]-[335], where O’Farrell J considered some of the authorities on this point.

It is because that kind of disclosure is required of the defendants, if any particular warranty is to be read as qualified by it, that paragraph 12 should be read without any alteration by substituting the word “or” for “and”. Linking actual knowledge and awareness to such fair and detailed disclosure appears to make sense and, in my judgment, there is certainly no basis for second-guessing the drafting on Chartbrook grounds.

Whether a relevant matter (qualifying Warranty B5.2.2 in particular) was disclosed and LCG had knowledge of it turns on an assessment of the evidence.

As I remarked during Mr Booth KC’s opening submissions, the defendants’ position (see Issue 4: the Vendors’ Knowledge Issue) is that, as at the date of the SPA, they were not aware that the Company was in breach of the Funding Regulations and that obviously highlights the need for careful interrogation of the basis for them saying LCG became aware of it through the process of due diligence. The defendants’ position on Issue 4 is consistent with the Disclosure Letter (bearing the same date as the SPA) not making any specific disclosure in respect of Warranty B5.2.2.

On the face of it, the fact that the Company had a clean audit in 2018 (which is relied upon heavily by the defendants on Issue 4: the Vendors’ Knowledge Issue) might be said to run against the general thrust of their case on Issue 7. Mr Dowson was also asked about the 2018 audit. He said he recalled looking at the 2018 audit report during due diligence. He took comfort from it. He agreed that it would have been conducted by ESFA on a random basis by reference to a sample of the Company’s learners.

Clearly, making the representation in Warranty B5.2.2 but qualifying it by clause 6.3 and paragraph 12, shows that the parties expressly recognised that LCG might come to know (through a process of discovery by undertaking due diligence) what the Company then did not. The references above to the Company’s Welsh funding illustrate that the defendants cannot therefore be accused of adopting two fundamental and inconsistent positions (although also not mentioned in the Disclosure Letter, there has been no need in these proceedings to investigate whether the Company was already aware of the amber warning compliance issues on Welsh funding mentioned in LCG’s internal due diligence report).

In her evidence Mrs McLeish accepted that, after the outcome of the 2022 Audit, she did wonder in hindsight whether the breaches of the Funding Regulations should have been picked up during the due diligence exercise; though she did emphasise that it was “the last thing on my mind” in early meetings when her focus was on avoiding a potential maximum clawback of £2.9m which Mr Dowson had identified. However, she said “[i]n hindsight, we did look back and question our processes in the data room, this issue could not have been identified by the data.” She contrasted the full audit of a sample of learners which RSM undertook in the 2022 Audit by reference to an attendance register and the actual hours of learner participation.

Mrs McLeish’s comment upon what Mr Williams had said in his email to Mr Dowson of 3 March 2022, in connection with the 2022 Audit, was “a very broad statement that is very high-level overview of what was underneath the planned hours and the condition of funding issue and the extent to which it went.”

Mr Dowson thought that, during the due diligence process, he had restricted himself to the funding section of the documents in the Data Room. Mr Dowson said there was “no learner specific paperwork” within it to enable a “mini-audit” to be carried out. He said he had access to the PDSAT reports but not the ILRs from which the PDSAT information was extracted. He said that the PDSAT information indicated about 30% of the students were shown as part-time.

Mr Dowson confirmed he looked at the PDSAT report as part of that process, which identified each student’s learning aim, the start and end date of the programme and the number of hours allocated to the student (including many allocated with 538 learning hours and therefore just below the full-time funding band 5). He said that there are multiple rows for a particular student and that he probably did not scroll through one to see how the planned learning hours (distinct from EEP hours) were apportioned to each learning aim. He accepted that those rows revealed a nesting approach to qualifications for some learners. However, he said that the PDSAT report did not reveal the hours attributable to the individual qualification. I should note here that there are 5063 rows, relating to 1,508 students, in the relevant spreadsheet within the PDSAT report: “All 16 to 19 funding model learners and learning aims” (tab ‘21Y-101’).

Therefore, Mr Dowson said, the very high proportion of students with the learning aim of an Award would not have alerted him to a breach of the Funding Regulations in terms of planned hours. As he said:

“Yes, I mean there was crudely 1,500 learners on the PDSAT report and two diplomas, there was not 1,498 funding errors identified.”

In rejecting the suggestion that the PDSAT information would have disclosed the planned hours issue, Mr Dowson said:

“It wouldn't have directly shown that, no. To see that you would have needed to see the make up of the qualifications planned for that learner and what those planned hours were attributable to. So MPCT have a document called a planned hours calculator, which says the programme is made up of – or that student's programme is made up of those X number of qualifications, to realise those 863 hours included other qualifications I would have needed access to the planned hours calculator.”

The Company’s “planned hours calculator” (as it is also described in the SPD) referred to by Mr Dowson was the Planned Hours Template.

When asked about the information for a particular learner in the PDSAT report (learner reference number ending “H31” with planned learning hours of 437 and EEP hours of 426), Mr Dowson also made the point that the person was planned to be on the programme for almost a year – 12 August to 31 July – even though the core learning aim was identified as an Award (column L) which was planned for just 2 months (columns M and N). He accepted that could have been a prompt for questioning why the person was proposed to be on the programme for a year but, looking at the end date and assuming about 40 teaching weeks in a year with about 20 hours per week as learning hours, it would be reasonable to assume that the planned hours of 863 was reasonable. He said: “the learner planned end date is very deceptive in this regard” and “if it was set correctly, it would have identified the issue and the learners would have appeared on one of the other exception reports which the ESFA monitor.” (Mr Dowson’s reference to an “exception report” - he also referred to it as an “error report” – was to a spreadsheet which is part of the PDSAT report which identifies any incorrect data iwhich requires corrective action to be taken.) The PDSAT report did not show the number of weekly learning hours.

Mr Dowson said that, for the purposes of the 2022 Audit, RSM made reference to a separate spreadsheet – namely, the Company’s Planned Hours Template (as a third source of information alongside the ILR and PDSAT report) – in order to reach their conclusion on the Planned Hours Over-Claim.

With that further information available, ESFA did not determine that the Company’s entire nesting approach was incorrect. Instead, they determined that the Company’s approach to nesting for a proportion of the learners was incorrect. As RSM confirmed (see paragraph 63 above) the 2022 Audit identified “a number of learners who were only enrolled onto one learning aim plus work experience which typically was planned for two months which they completed.”

In order to identify those learners, RSM looked at other information beyond the PDSAT report; namely the ILRs/PLRs and the Company’s Planned Hours Template. As shown by the relevant information within it for KB (mentioned below in relation to Condition of Funding to which his case is directly relevant) the PDSAT does not itemise the number of planned hours for each of the individual qualifications the learner is undertaking. There are five rows in the PDSAT relating to KB. Each row reflects a different recorded learning aim but the number of specified planned hours is the same in each of the rows (as is the number of planned weeks in the year and the average weekly planned hours). As explained below, KB was shown as part-time (funding band 4b) in the PDSAT report when the Planned Hours Template showed him having planned hours of 888 (therefore full-time and funding band 5) comprised of 540 qualification hours, 349 non-qualification hours and 144 work experience hours.

The same point about the absence of a breakdown of hours between different learning aims applied to learners in respect of whom RSM identified a breach of paragraph 119 of the Funding Regulations. That breakdown of planned hours between each learning aim (which Mr Dowson indicated would be required if the total planned hours and planned end date was to be second guessed) was missing from the PDSAT report. In their closing submissions LCG’s counsel identified JH, who formed part of the Planned Hours Over-Claim, to illustrate this.

JH was shown (at rows 1876 to 1881 of the spreadsheet) as having a start date of 16 November 2020 and a planned end date of 31 July 2021. However, the “learner actual end date” is recorded as 3 March 2021. Therefore, instead of staying with the Company for the 8½ months indicated by the planned end date, he left after 3½ months. That early departure identified him in the 2022 Audit as a learner impacted by the planned hours issue. Only the Planned Hours Template showed a breakdown of the planned hours between his different learning aims (indicating a total of 795.5 hours). In the absence of a breakdown in the PDSAT report of the total planned hours (there recorded as 666) between the different learning aims in each row – (a) Functional Skills English Entry Level 2 (row 1876), (b) Functional Skills Maths Entry Level 2 (row 1877), (c) Work Experience (row 1878), (d) an Introductory Award in Employability Skills Level 2 (row 1879), (e) Functional Skills Maths Entry Level 3 (row 1880), and (f) Functional Skills English Entry Level 3 (row 1881) – there was no basis for LCG to question the planned end date as being inaccurate or unrealistic. As Mr Booth KC and Mr Adamyk submitted, on the assumption that the 8½ months would include 6 weeks’ leave, the 666 hours indicated about 30 teaching weeks on the basis that the learner planned to attend for approximately 22 hours per week. That information alone would not have flagged up a breach of paragraph 119.

I therefore reject the defendants’ argument in closing submissions that the information in the PDSAT was “the only information needed to deduce an impermissible nesting approach contrary to para 119 FR.” The Planned Hours Over-Claim reflected the Company claiming learning hours for qualifications that were not included as learning aims in the relevant ILRs. The PDSAT report indicated the total number of hours planned for a learner but, without the ILR to identify the learning aim(s), LCG would not know whether this reflected a permitted nesting of qualifications or the breach of paragraph 119 of the Funding Regulations identified by RSM. The point is illustrated by the fact that not all nested qualifications were found in the 2022 Audit to involve funding errors.

In relation to the Condition of Funding Over-Claim, Mr Dowson also said he did not know that the Company was foregoing an element of funding in order to satisfy the Condition of Funding rules. Looking at the particular learner H31, for example, he said the PDSAT information did not contain the information as to whether or not the learner had previously obtained maths or English GCSE (or a grade 3 in either or both). Therefore, it could not be seen whether the Condition of Funding rules applied to that learner.

Asked about another learner (learner reference number ending “X62”), whose planned hours were 538, he said that this did not indicate that the hours had been reduced to below 540 in order for the learner not to be subject to the Condition of Funding rules. He said:

“There is legitimate reasons for using 538 hours, that could be a genuine plan that you were intending to deliver. It is not an impossibility that those hours were just genuinely planned hours.”

“There is no indication of the reason why they would be on at 538 hours in this document.”

“….. to make that assumption, I would need to derive initially that the 530 hours -- 538 hours, has falsely input and make an assumption that the GCSE grades upon entry were grade 3, I think there is an incredible leap between those.”

and

“To know if you would need to do GCSE, you would need to know the GCSE grades on entry, which is not on this report.”

I note that when Mr Booth KC asked Mr Lewis about whether a PDSAT report showing 537 learning hours and 1 EPP could reflect a genuine learning plan, Mr Lewis answered “not knowing what I now know, no” before saying “I don’t know.” Those answers, given in the light of the 2022 Audit, lend support to Mr Dowson’s first answer above.

Mr Dowson was also asked by Mr Sims KC about the PDSAT information for KB who was recorded in the PDSAT reports as being within funding band 4b and shown as having total planned hours of 538 made up of 537 learning hours and 1 EEP hour. I have noted above that the Planned Hours Template shows that KB enrolled with the Company as a full-time learner. It was suggested by Mr Sims KC that the 538 hours was a pretty obvious pointer to someone carrying out due diligence that KB’s hours had been reduced to comply with the Condition of Funding Rules. Referring to the fact that the 2022 Audit picked up that the Planned Hours Template actually showed KB’s planned hours to be 888, Mr Dowson answered:

“I don't think so. They wouldn't have -- the 888 hours doesn't exist anywhere on the PDSAT report either. The 538 would exist on the PDSAT but not the 888.”

On the basis that the ILRs and Planned Hours Template were not in the Data Room, Mr Dowson’s evidence demonstrates why LCG did not have (or, even if it was the relevant test, could not properly be attributed with) knowledge of breach of the Condition of Funding rules prior to entering into the SPA. A reference to the PDSAT report containing KB’s details supports his evidence. The five rows for KB (rows 533 to 538) each identify his ’16 to 19 funding band’ as “450 to 539 hours (Band 4b)” based upon total planned hours of 538. Reference to the details for the next learner (learner reference number ending “J27”) – rows 539 to 541 – record total planned hours of 888 and the resulting funding band of “540+ hours (Band 5)”. There was no reason for LCG to question the accuracy of KB’s hours, alongside the recorded hours for other learners, especially when the PDSAT report contained no indication as to whether or not KB had previously achieved grade 3 maths or English GCSE.

As Mr Dowson said in relation to the Condition of Funding breach:

“……. The ILR should show the accurate planned hours in which you are intending to deliver for a learner. If you are intending to deliver higher planned hours, then you should declare it within the document.”

and

“I didn't see anything wrong with learners being planned to do 538 hours. If I knew the reasons why they were doing that, I would have raised an issue.”

Mr Dowson knew about the Condition of Funding rules in 2021. He confirmed that LCG was not delivering GCSE courses at the time of the SPA and offered part-time courses to students who would have been affected by them.

I accept Mr Dowson’s evidence that he could not have known that more than 538 planned hours were to be provided to a learner when he did not have access to the ILR or the Company’s Planned Hours Calculator. I have referred above to the SPD which referred to the Company’s Planned Hours Calculator being used “to ensure accurate planned hours are claimed from ESFA.” I am satisfied by Mr Dowson’s evidence that LCG did not have knowledge to the contrary.

Neither does the evidence support the conclusion that LCG acquired knowledge of the relevant matters through being told about them by Mr Williams at the due diligence stage.

In his witness statement, Mr Dowson referred to a conversation he had with Mr Williams during the due diligence exercise. He said they would have discussed planned timetables including in relation to the provision of maths and English. In his witness statement he said he could not recall if they specifically discussed the Company’s approach to planned hours or Condition of Funding. In testimony, Mr Dowson said he did not recall having any conversation with Mr Williams during due diligence about the Company’s approach to the nesting of qualifications. Mr Dowson said that, even if there had been such a discussion with Mr Williams explaining the Company’s approach set out in the SPD, then it would have identified that the Company adopted the nesting approach to qualifications but “nesting as such is not against the funding rules.”

Mr Sims KC suggested to Mr Dowson that communications with Mr Williams at the due diligence stage involved Mr Williams explaining (in much the same terms as later set out in the SPD sent to him on 2 February 2022) the Company’s approach to the Condition of Funding rules so that full-time learners were recorded as being planned to do 538 hours. Mr Dowson responded:

“That didn’t happen ….. I am saying that didn’t happen. I did not have that conversation.”

That response was consistent with his response that (knowing about the Condition of Funding rules) he would have raised it as an issue if he knew that this was what the Company was doing.

Mr Williams said in his witness statement that he did not recall speaking to Mr Dowson during due diligence. In cross-examination, he confirmed this and said “I can’t recall speaking to Daniel during due diligence about condition of funding.” Mr Williams said if he and Mr Dowson had such a conversation then his response would have been as stated in the SPD.

I accept the evidence of Mr Dowson and Mr Williams on this point. That evidence, from the two parties to a “conversation” alleged by the defendants to have taken place on the balance of probabilities, undermines the suggestion that LCG acquired actual knowledge of matters giving rise to a breach of the Funding Regulations as a result of things said by Mr Williams during due diligence.

In any event, Mr Dowson said that the SPD which Mr Williams later sent him in February 2022 did not directly point to the conclusion that there had been a breach of the Funding Rules and “[y]ou would need to extract further information.”

In my judgment, the terms of the SPD did not mean it was reasonable to assume that there would be an audit-based clawback.

In a section headed ‘How is Condition of Funding for Maths and English Checked?’, the SPD stated:

“On analysis of starting points for the assessments, MPCT has made a decision not to set learners the unachievable task of attaining a qualification grade of 4/C or above. Therefore, we calculate the planned hours and forego a banding of funding, to ensure the learner meets the condition of funding, with the appropriate qualifications achievable being followed.”

Mr Sims KC suggested to Mr Dowson that the document clearly indicated that the Company was providing additional hours for which it was not receiving funding and was foregoing a band of higher funding, to which it was otherwise entitled, in order not to fall foul of the Condition of Funding rules. Mr Dowson did not share that interpretation of the document. In addition to pointing out that the Company’s decision to provide 539 hours of funded learning was mentioned in another section (not relating to Condition of Funding) he said he interpreted the statement as the Company saying it delivered a part-time programme on a part-time basis when it might have been a full-time programme delivered on a part-time basis; and that the change to a part-time programme required recalculation of the hours. He said: “I think if they just mindlessly put it to 538, that would not require a recalculation.”

There is also force in Mr Dowson’s observations that even the SPD, later brought to his attention by Mr Williams once the 2022 Audit was underway, did not clearly identify the facts, matters and circumstances giving rise to the Clawback. He said the document did not clearly state that the Company was not providing maths and English GCSEs (the reference to maths and English at “relevant level” could imply that this extended to GCSE as well as “functional skills” maths and English) or that, in order to meet the Condition of Funding rules, the Company would not receive full-time funding for learners whose planned hours were in fact in excess of 540 hours. As he said, the SPD indicated there was a question to ask about that.

That observation highlights the difficulty for the defendants. Even if Mr Dowson had been prompted by the documents in the Data Room to ask such a question before LCG entered into the SPA, the agreement is clear in putting the burden squarely upon the defendants to inform LCG about the relevant matters rather leave it to LCG to make appropriate inquiries once put on notice of a potential issue.

Their case that LCG had the requisite knowledge to qualify Warranty B5.2.2 falls to be tested against the meaning of the word ‘Disclosed’ in the SPA and the fact that the Disclosure Letter gave no “specific disclosure” (to quote from it) in respect of that warranty. I have addressed above the impact of the meaning of ‘Disclosed’.

In Triumph Controls, at [335], the six principles which O’Farrell J derived from the authorities included the following:

“……….

iii)

The adequacy of disclosure must be considered by careful analysis of the contents of the disclosure letter, including any references in the disclosure to other sources of information, against the contractual requirements,

iv)

A disclosure letter which purports to disclose specific matters merely by referring to other documents as a source of information will generally not be adequate to fairly disclose with sufficient detail the nature and scope of those matters. For that reason, disclosure by omission will rarely be adequate.

v)

However, it is open to the parties to agree the form and extent of any disclosure that will be deemed to be adequate against the warranty. That could include an agreement that disclosure may be given by reference to documents other than the disclosure letter, such as by list or in a data room.”

……….”

In my judgment, the defendants’ argument does involve seeking to impute to LCG knowledge of the funding errors when, on my assessment of the evidence, LCG had no actual knowledge of them and (compare the concept of “disclosure by omission”) the defendants did not speak up about them. That is not good enough to satisfy the terms of clause 6.3 and paragraph 12 of Schedule 5.

As LCG’s counsel submitted, the defendants’ case is to be considered against what the defendants might quite easily have done and what in my judgment was required, by ‘Disclosure’, to be done. The defendants could have volunteered a brief explanation of their approach to planned hours and Condition of Funding, giving an indication of the approximate percentage(s) of affected learners over recent years, and indicated their belief that it was compliant with the Funding Regulations but that LCG should be aware of the risk that it might not be. Their explanation would probably have required a little more clarity on both points than the one bearing upon Condition of Funding that was later given in the SPD provided to Mr Dowson.

The defendants not having done that, there is no evidential basis for their suggestion that LCG the relevant matters were disclosed or that LCG had knowledge of them for the purpose of paragraph 12 of Schedule 5.

It is very significant, as LCG points out, that at no point during the 2022 Audit did Mr Lewis, who was still involved in the business at that time and aware of the discussions between LCG and RSM, suggest that LCG (even without such clear disclosure having been made) had been made aware of the planned hours or Condition of Funding issue before entering into the SPA.

My decision on Issue 5 is that matters relevant to the warranty claims made by LCG were not disclosed by the defendants.

My decision on Issue 7 is that LCG did not have actual knowledge and awareness of the facts, matters or circumstances giving rise to those warranty claims such as to result in the defendants not being liable on them.