England and Wales High Court (Queen's Bench Division) Decisions

IN THE HIGH COURT OF JUSTICE
QUEEN'S BENCH DIVISION

		Royal Courts of Justice Strand, London, WC2A 2LL
		14 January 2013

HHJ Moloney QC :

1. INTRODUCTION
2. 1.1 This action arises out of events in April 2008, when the Bank of Scotland ("BoS") refused to provide re-mortgage finance to the Claimant Mrs Carol Gatt "CG" and her husband Michael "MG" (also formerly a Claimant) in respect of their multi-million pound home, Melksham Court in the Cotswolds. On enquiry by their brokers Savills Private Finance, the reason BoS gave for refusal was a problem with MG's credit rating. It emerged that the principal credit reference agencies, Experian, Equifax and Callcredit, were stating on MG's credit reports that his Barclays Bank current account was "delinquent", though not in "default", because he had a credit limit of only £1,500 but an overdraft of some £260,000.

It is accepted that the reports were an accurate reflection of computerised information provided by Barclays to the credit reference agencies under their long-standing and lawful data-sharing agreements. The account in question was in fact a joint account in both MG's and CG's names (though the reports did not say so). The Gatts in turn accept that the account was indeed overdrawn to that extent; but they say that the borrowing was well known to, and actually authorised by, their Barclays Relationship Manager Mr Williams, and that it was therefore untrue for Barclays to imply to the credit agencies and those to whom they passed the data on that MG and/or CG had substantial unauthorised borrowings.

1.2 It is the Gatts' case that this false statement by Barclays, brought about by Mr Williams, had the foreseeable consequence that they were unable to re-mortgage their home to raise finance, which was required to fund their longstanding property-development business, and that that in turn led to the collapse of their business, the loss of their home and other assets, and in MG's case to bankruptcy. (In the course of the bankruptcy proceedings, Barclays bought MG's claim in this action from his Trustee and then dropped it, so that to the Gatts' considerable indignation CG is now the sole Claimant.)

1.3 CG puts her claim in three overlapping ways:

a. breach of contract (she being a Barclays customer in her own right);

b. negligence/negligent mis-statement; and

c. defamation.

At its maximum, depending how much is properly claimable in respect of loss of future business profits, the claim runs to many million pounds.

1.4 The Defendants (to whom I shall jointly refer as "Barclays" or "the Bank" unless dealing with a point specific to one of them) maintain that they are not guilty of any misconduct, essentially because it was true that the joint account had a very large unauthorised overdraft. They further challenge the damages claim in all respects. And they counterclaim against CG for her personal liability on the overdraft and another joint loan of even greater amount; in each case (but particularly the overdraft) she denies liability, asserting that these are her husband's personal borrowings which Barclays has attributed to her without her consent.

3. BACKGROUND UP TO PURCHASE OF MELKSHAM COURT (MARCH 2004)
4. 2.1 Mr and Mrs Gatt were both born in 1946, and married in 1968. In 1990, MG entered a joint property venture with a large Danish company, Larsen & Neilson; he was Managing Director, and later bought the Danes out, but his companies continued to be called the GLN group. From an early stage CG was actively involved in the business as a director and shareholder of some though not all the GLN companies. She had her own office in the group's headquarters in Tunbridge Wells, and carried out executive and administrative duties; MG appears to have been the front man so far as deal-making was concerned. In court they gave me the clear impression of having a strong and equal partnership, though the collapse of their business and lifestyle has plainly taken its toll on MG in particular.

2.2 It is not necessary in this judgment to review in detail the Gatts' business from 1990 to 2004. It is clear that they had a good track record, and played a major part in several successful projects such as leisure parks. Their general method of operation was to locate a development opportunity, purchase the land or an option on it (often with the help of a partner), obtain the necessary planning permissions and other prerequisites of development, and then take their profit by selling on to a larger developer which would carry out the actual building and selling. This middleman role had the advantage of not needing too much capital, and their habit was to fund it by borrowings secured on their home (their principal asset), and paid off when the profit came in, sometimes at fairly long intervals. (For example, in 2005/06 MG declared an income for tax purposes of £1.2 million, which was more than double the Gatts' combined taxable income for the whole of the preceding four years.) Plainly they were not afraid of borrowing, as some people are, but regarded being overdrawn for business and living purposes as a normal condition, confident of being able to repay when their ship came in. Barclays accepts that, up till the time with which this action is concerned, that confidence was well founded; the Gatts sometimes had to ask for an extension of time, but they kept their bankers informed of the progress of their deals, and eventually they always repaid both loan and interest in full.

2.3 As a result, the Gatts were able to live the lifestyle of prosperous, indeed wealthy, people. Before 2004 they lived in Lymden Place, Kent, a house then worth about £2 million, as well as owning their business premises, Copenhagen House in Tunbridge Wells. (For this purpose I shall not distinguish between assets owned by the Gatts directly or through their companies.) In 2004 they bought Melksham Court, Glos., a mediaeval manor house with grounds and outlying cottages, where they could house such things as CG's stable of quarter horses, MG's Ferrari and collection of Harley Davidsons, and a mobile home valued at £240,000. (The business remained based in Kent, though its projects could be anywhere, including the West Country.)

5. BARCLAYS' OVERDRAFT LIMIT APPROVAL SYSTEM
6. 3.1 The best record of the dealings between Barclays and the Gatts in relation to borrowing is Barclays' computer diary "Notepad", which is centrally maintained and cannot be retrospectively altered. It contains a good deal of second-hand hearsay because, while some entries are made by the people who actually dealt with the Gatts, others are made by those to whom they reported. And of course, the Gatts never had access to it, and it also records internal Barclays meetings to which they were not party. Nevertheless I am satisfied that it is in general an accurate contemporary record both of what passed between the Gatts (specifically MG) and Barclays and of the internal Barclays responses.

3.2 Barclays' official process for the approval of increases in overdraft limits is described in the Witness Statements of Mr Williams, who was the Gatts' Premier (i.e. personal banking) Relationship Manager from September 2005 on, and Mr Gurney, who was their Corporate (i.e. business banking) Relationship Manager from 2002 until 2007, when the Gatts ceased actively banking with Barclays as described below. Neither of them could personally approve substantial overdraft increases of the order required by the Gatts. Such facilities had to be "sanctioned" (i.e. approved), initially by an automatic computer system called ERA ("enhanced risk assessment"). If ERA refused approval, the application could be appealed to a credit underwriter, who would consider the representations of the Relationship Manager and, if satisfied, approve the facility, perhaps on condition of the provision of security such as a solicitor's undertaking or a charge on land. He would set a maximum borrowing limit, and a maximum time period after which the sanctioned overdraft limit would lapse unless expressly renewed. These overdraft limits, as to both duration and amount, appear on the monthly bank statements sent to the customer.

3.3 The sanctioned overdraft limits on duration and amount were also logged on to the Bank's computer system ("marked up"). As described to me by Mr Hill of Barclays' credit reference agency department and explained further below, the marked up limits on each account form part of the shareable data which participating banks automatically provide to the credit reference agencies every month, thus giving rise to the statements complained of in this case.

3.4 I have described the above systems as "official", because it is a central part of the Gatts' case that whatever Barclays' internal procedures may have been, so far as they, the customers, were concerned the bank's practice was very different. They describe a much more traditional and informal approach, under which MG would keep in regular touch with his "bank manager", often over lunch; he would describe the state of business and his credit needs, and the bank manager would usually give oral approval there and then. MG told me that he regarded his bank managers as having the personal authority to do this without higher approval – that was their job. In those circumstances, he maintains, it would be thoroughly unfair and misleading for the Bank publicly to describe as "delinquent" or "unauthorised" a resulting overdraft of which the bank was fully aware, and which it had taken no steps to call in, just because some back-office paperwork (as he would no doubt see it) had not been complied with. This is an important issue to which I shall return.

7. THE SOLE OVERDRAFT AND THE JOINT LOAN ACCOUNT (MARCH 2004 TO FEBRUARY 2007)
8. 4.1 The Gatts did not sell their former home Lymden Place immediately on buying Melksham Court in 2004. Melksham Court was heavily mortgaged to Bank of Scotland. Perhaps as a result, MG's overdraft on his sole bank account increased steadily, the limit rising from £30,000 in May 2004 to £165,000 in June 2005. The increased limits were of course put in place by the Bank, but MG habitually exceeded them and then retrospectively obtained the Bank's agreement to an increased limit. In September 2005, Mr Williams became the Gatts' personal-banking Relationship Manager. Because the prospective purchaser of Lymden Place required more time to raise his finance, sanction was given in September 2005 to increase the overdraft limit to £200,000 on the Gatts' solicitors confirming that contracts had been exchanged and that they held a deposit and would repay the overdraft from the proceeds of sale. After various discussions which it is not necessary to summarise here, this was put in place in December 2005; almost immediately MG exceeded it and requested a further increase to £250,000 which was granted in January 2006, initially for 2 months only. That limit was in turn exceeded in May 2006.

4.2 The Gatts rely on this pattern of events, which was to recur subsequently, as a course of dealings showing that the Bank did not hold them strictly to the terms of their authorised overdraft limits. It might impose interest penalties or even stop debits, but it did not actually call the loan in or even threaten to do so.

4.3 What the Bank did do was decide to seek better security, over the Gatts' equity in real property which was their sole free asset and was jointly owned; so the sole overdraft needed to be transformed into a joint loan secured on their joint land. The joint loan, as offered in July 2006, was to be in the amount of £325,000, for the period of 6 months from drawdown, and secured on the office building Copenhagen House. A joint loan account (different from the Gatts' existing joint current account) was used for this purpose, and the loan plus fees were debited to that account in August 2006 and credited to MG's sole account to pay off the overdraft. This was done before any charge was put in place, and in fact Copenhagen House was never charged.

4.4 By late 2006, two separate plans were under consideration to regularise the position between the Gatts and the Bank. One was the "secured" loan, already paid over, but with security now to be granted over Melksham Court rather than Copenhagen House. The other, more fundamental one, originally proposed by Mr Williams and favoured by CG in particular, was to consolidate everything into one comprehensive remortgage of Melksham Court (by Barclays' mortgage arm the Woolwich), repaying Bank of Scotland, securing all the Barclays borrowings, freeing up some further capital, and reducing the overall level of payments.

4.5 A yet further complicating factor was that, although the joint loan of August 2006 had put MG's sole current account back into credit, he had not stopped making further borrowings on it over and above the joint loan. His overdraft limit in August 2006 was back down to £30,000. By November 2006 the limit remained the same but the overdraft was £106,000. The limit was increased to £150,000 (for four months) in December 2006, but by February 2007 the actual borrowing had reached £207,000. (In other words, in less than 3 years MG had borrowed and spent about half a million pounds, apparently on day-to-day living for himself and CG, though it is fair to say that he had maintained a continuous dialogue with the Bank while doing so.) MG's proposal in December 2006 was that the security over Melksham Court, already intended to cover the joint loan which had cleared his "first overdraft", should also extend to cover his "second overdraft"; once again, it was inherent in this plan that as before his sole liability should become a joint liability with his wife so that it could be secured on their joint property.

4.6 In February 2007, two important developments took place. First, Mr Williams submitted a mortgage application to the Woolwich, seeking to borrow £2.38 million on Melksham Court against a valuation of £3.4 million; the loan was to be interest only at a low tracker rate. Almost immediately the application was rejected by the Woolwich, which had deep concerns about the Gatts' finances. Mr Williams attempted to persuade the Woolwich to reconsider, but having received information from the Gatts' accountants the Woolwich persistently declined the application, the final refusal being in April 2007. (Significantly, on 8 March 2007 the Woolwich wrote separately to both MG and CG explaining that one possible reason for refusal was credit scoring, and telling them how for £2 they could get a copy of their Experian reports.)

4.7 Second, and centrally to this case because it is the inception of the particular overdraft to which the credit reports complained of refer, on 9 February 2007 Barclays received confirmation from the Gatts' solicitors that they had executed a second charge on Melksham Court. (In fact this charge was never registered, apparently because of the possibility of a remortgage.) Upon that assurance, the "second overdraft" on MG's sole current account was cleared on 26 February 2007, by the mechanism of the transfer of £228,000 from the Gatts' joint current account, putting that account into overdraft. The borrowing was thus transferred to the joint current account, which had a sanctioned overdraft limit of £250,000 for four months, i.e. until June 2007, after which time it would revert to its original level of £1,500. (I shall return later to the question of how far CG personally knew and approved of these steps.)

9. THE JOINT OVERDRAFT (FEBRUARY 2007 TO APRIL 2008)
10. 5.1 No more than 2 weeks later, the overdraft limit on the joint current account was exceeded, and Barclays began refusing to pay direct debits on that account. It would appear that at this point the Gatts finally realised that Barclays' patience was exhausted, because in May 2007 they:

a. remortgaged Melksham Court to the Northern Rock;

b. moved their current account banking to the National Westminster; the sole and joint current accounts became dormant, though some payments were made towards the joint loan account.

At the same time, on 10 May 2007, MG wrote a letter of complaint to Mr Williams, claiming that Barclays had not fulfilled its obligations to the Gatts in respect of the Woolwich mortgage, referring to "the overdraft which we should not have incurred in the first place", and concluding: "you have not got a second charge over the property as you have not fulfilled the obligations you agreed to".

5.2 In June 2007 the higher overdraft limit on the joint current account expired, and the old limit of £1,500 became applicable. From this time on, Barclays' monthly returns to the credit reference agencies, and the resulting reports, took the form of which the Gatts now complain, though they did not become aware of it until their attempt to remortgage Melksham Court in April 2008. (The reports only named MG, though the account was of course also in CG's name, because she had opened the account in her sole name before the date on which customers were deemed by Barclays' standard terms to consent to such disclosure of otherwise confidential data; MG became a joint party to the account after that date, so the data could be disclosed in relation to him but not her.)

5.3 During the latter part of 2007 (during which, as is well known, the "credit crunch" began to manifest itself) Barclays continued to seek repayment or at least better security. One significant incident was a meeting between MG and Mr Williams at the Spa Hotel, Tunbridge Wells, following which MG sent Mr Williams an email dated 25 September 2007 about the steps he was taking to repay. It discussed the possible sale of Melksham Court, and two development projects nearing completion in Heathfield and Wrotham, but went on:

"As discussed at this meeting I would be grateful if you could show the joint account overdraft being authorised as at the moment it shows £240,000 overdrawn against a £1,000 facility [in fact £1,500]."

This is significant in several respects. It proves that, contrary to MG's later denials, he at least was well aware in 2007 that the overdraft was on the joint account. It demonstrates MG's apparently sincere belief, before the matters complained of, that the overdraft was "authorised" notwithstanding what was stated on the bank statements, and his desire that that should be a matter of record. But on the other hand, the Bank took no action on his request, not even a reply so far as is known, and both the bank statements sent to the Gatts and (unknown to them) the credit reports continued to show the overdraft vastly in excess of the limit. Of course, the limit could not have been raised without sanction from above, and discussions continued between MG and Mr Williams about regularising the situation.

5.4 Following a meeting with the Gatts on 4 December 2007, Mr Williams submitted a proposal to his superiors along familiar lines; both loans (the joint loan and the joint overdraft) were to be consolidated into a single loan secured by a second charge on Melksham Court, provided it could be valued at £3.5 million or more so as to give adequate clearance above the Northern Rock mortgage. His proposal was sanctioned, and indeed renewed in late April 2008, but never carried into effect.

11. THE CREDIT REFERENCE REPORTS AND DEMAND FOR REPAYMENT (APRIL 2008 ONWARDS)
12. 6.1 While MG was negotiating with Barclays about security for the loans, the Gatts also engaged Savills Private Finance to assist them in remortgaging Melksham Court for a larger sum, which would of course diminish Barclays' potential security (unless some of the funds were used to repay it). On 10 April 2008, CG filled in by hand Savills' mortgage advice questionnaire. There are various serious errors and omissions in this document, including:

a. the false statement that they had never had a mortgage application refused, when less than a year before the Woolwich had done so amid considerable discussion;

b. the statement that MG had unsecured loans amounting to £40,000, and CG none, when in fact they owed Barclays alone over £500,000 on the loan and overdraft then outstanding, and had other substantial liabilities including credit cards and finance agreements;

c. the statement that the purpose of the remortgage was "to get extra moneys for improvements to increase value of property", when it appears that the improvements had already been carried out, and were reflected in the valuation being put forward, and it is their case before me that the true purpose of the remortgage was to raise seed capital for pending business projects.

6.2 On the basis of this information, Mr Elmes of Savills approached the Bank of Scotland for the loan. But on or about 23 April 2008, BoS told Mr Elmes that there was a problem with MG's credit rating, and directed him to Experian and Equifax. At this point the fact that the reports referred to the overdraft as being delinquent, and the limit being only £1,500, came to the Gatts' attention. MG at once emailed Mr Williams, telling him of the problem, and saying "you and I know is incorrect as it is authorised". Mr Elmes also contacted Mr Williams and asked him to correct the situation both with the credit reference agencies and with BoS directly.

6.3 On 29 April 2008, Mr Williams wrote Mr Elmes the following letter (for showing to BoS):

"Re Mike and Carol Gatt

Further to our telephone conversation concerning our mutual clients above, please take this as confirmation that even though on the credit searches it shows as an agreed facility of £1,500 the actual agreed overall facility is in fact £500,000.

This is an error on behalf of Barclays bank and we are currently looking into why this is so, but it should not reflect in any way on Mike & Carol Gatt.

If you require anything further then please call me on the number below."

Not surprisingly, the Gatts invite me to treat this letter as a clear admission by Barclays that the joint account overdraft had in fact been authorised and that the credit reports were wrong to suggest otherwise. Mr Williams, whose position is obviously an embarrassing one, told me that he was trying to help the Gatts as much as possible, both for their sakes and the Bank's (which might hope for repayment if the remortgage went through). To that end, he was referring to the fact that the Bank was indeed ready at that time to authorise lending of over £500,000 to the Gatts, as mentioned at 5.4 above, provided it was secured on the house (a condition the letter did not mention). His letter does not actually use the terms "overdraft" or "authorised".

6.4 From this point on, all matters relevant to liability have taken place, and the rest of the story can be quickly told. The Bank did offer the Gatts the consolidation loan, in the amount of £690,000, but the Gatts never took it up, complaining that the interest rate was higher and the term shorter than had been agreed. (They also complained that it was unacceptable to make CG liable for MG's debts, an issue to which I shall return.) The Gatts attempted to remortgage elsewhere, but had no more success, the credit reports remaining in the same form. In August 2008 the Bank demanded repayment of all loans and overdrafts, and in September 2008 the Northern Rock complained of substantial mortgage arrears. The Gatts say that, because of their inability to raise finance on their home, they were unable to fund their pending development projects, so their income dried up and their predicament worsened; I shall deal with these damage issues below. These proceedings were issued in March 2010. There followed the repossession of Melksham Court by the Northern Rock, its disposal at what the Gatts regard as a considerable undervalue, and MG's bankruptcy referred to above.

13. MRS GATT'S CLAIM IN CONTRACT
14. 7.1 Although MG's claim is no longer being pursued for the reasons above stated, it is helpful to consider whether it would have been well founded, since CG's claim, at least in contract, can hardly be stronger. Unlike CG, MG had consented to the disclosure of his credit information to the CRAs, by virtue of the standard terms applicable to new customers at the time he became a Barclays account-holder. (The Gatts deny receipt of these; there is no direct evidence either way, but I find it to be more probable that the Bank did send MG its current terms in accordance with its usual course of business, though it may well be that MG paid them no special attention and does not now recall receiving them.) It is easy to see how he might have a good contract claim, if in breach of an obvious implied term Barclays disclosed information about him that was in fact false and damaging. Here, the allegedly false statement specifically complained of is that the overdraft limit on the joint current account at the material times was only £1,500; taken in conjunction with the true statement that the sum borrowed was over £250,000, a damaging inference clearly arises.

7.2 So was it false, that the authorised overdraft limit was only £1,500? Under the bank's official procedures, it was perfectly true. The higher limit of £250,000 had been sanctioned in February 2007 for four months only and expired in June 2007, at which point it went back to £1,500. This lower limit was recorded on the bank statements sent to the Gatts thereafter. MG at least was aware of the problem, because he mentioned it in his email of 25 September 2007 to Mr Williams, to which he does not appear to have got an answer.

7.3 The Gatts' case is that it was authorised, informally but sufficiently, by a combination of representations from Mr Williams that all was well and the Bank's failure to treat it as unauthorised, for example by demanding repayment. MG's email is certainly contemporary evidence that it was his belief in 2007 that the overdraft was authorised in that sense and that the official limit should be raised to reflect this.

7.4 Mr Williams told me, and I accept, that he would never have told MG that a loan or authorised overdraft would certainly be granted, or even that he could authorise it of his own authority. The Notepad system is clear evidence that he, and his colleagues in a similar position, always put MG's requests for increased overdraft limits through the sanction procedure, often with some delay; and MG was often involved in that procedure, for example by providing a solicitor's undertaking or whatever other security might be needed. As to the failure to enforce the loan, it is notable that, as the 25 September 2007 email itself makes clear, the bank was by then pressing for repayment, and MG was seeking to reassure it by explaining his anticipated future profits.

7.5 The Gatts place great reliance on Mr Williams's letter of 29 April 2008 to Mr Elmes (and on a later letter of 4 August 2010 to their MP in which the Bank's customer relations department sought to defend that letter). They rely on it as, in effect, an admission that the overdraft was in fact authorised. Even if it had said so, I do not think an admission written at that time and in those circumstances would prevail over the clear evidence of what the actual position was. But, as stated at 6.3 above, the letter refers to an "agreed overall facility" not an "authorised overdraft", and was probably a rather disingenuous reference to the proposed consolidated loan rather than an admission as to the authorised overdraft limit in place. It was, I have to conclude, a misleading document, over-favourable to the Gatts, but written at their request with the intention of helping them out of their difficulties. It would be wholly unfair to the Bank now to interpret it in the manner for which the Gatts contend.

7.6 My conclusion is that there is a clear distinction, which was accurately reflected in the information sent by Barclays to the CRAs, between an unauthorised overdraft and one that is in default. This overdraft was not authorised in any sense; nobody at Barclays had agreed to a higher limit. On the other hand, MG was in discussions with the bank about it, as a result of which it was not being treated as in default. In these circumstances, MG could not have succeeded in a claim for breach of contract in respect of the publication of this information to the CRAs; it was true information falling within the permitted categories for disclosure.

7.7 If this is a correct analysis, then what is CG's position under her contract with the Bank? It was clearly obliged not to disclose any confidential financial information about her to the CRAs, whether true or false. Except in the refined sense that this information related to a joint account to which she was a party (a matter which nobody who saw the credit report knew) the Bank complied with that duty. The Bank's contractual obligations to her cannot have extended so far as to preclude it from disclosing information about her husband, when he had contractually consented to that disclosure. I conclude that insofar as her claim is based on breach of contract it must fail. (I shall deal below with the issues of damage that might arise if this or my other conclusions on liability were wrong.) I should add that of course the 2^nd Defendant Mr Williams was not in contractual relations with the Gatts, though he might be liable personally to them in respect of the tort claims considered at (8) and (9) below.

15. MRS GATT'S CLAIM IN NEGLIGENCE
16. 8.1 This claim, which has much in common with the contract claim considered above, is based on the principles set out by the House of Lords in Spring v. Guardian Assurance [1995] 2 AC 296. As is well known, in that case the HL held that the writer of a job reference owes a duty of care towards its subject, which if breached can give rise to a claim for pure economic loss. Whether such a duty applies in a particular case, particularly in relation to a new fact-situation, is a delicate question, addressed by Lord Bridge in Caparo v. Dickman [1990] 2 AC 605 at 617 by reference to the need to find both a relationship of proximity and that it is fair, just and reasonable to impose the duty; he commended extension by careful analogy with existing classes of duty.

8.2 In the present case, given the importance of credit rating in the modern world and the analogies (more than just semantic) between job references and credit references, I would have no great difficulty in recognising such a duty owed by a bank to its customer to whom the reference related; indeed for the reasons above stated it is likely also to be a contractual duty. Whether the duty extends to the spouse of the person given the credit reference is a much more difficult question. In the ordinary case, the answer is almost certainly not. But on the facts of this case, where to the Bank's knowledge she was a joint holder of the same account and a co-director of the family business which largely depended on her husband's credit, so that both proximity and foreseeability of damage are present, I conclude that such a duty was owed to CG as well as MG.

8.3 However, the existence of such a duty is a largely academic matter unless it has also been breached, that is, unless the Bank has fallen below the standard of reasonable care in relation to the disclosure of this information. At this point, it is only necessary to repeat my conclusions of fact at (7) above, and to refer to the further discussion at (9) below. The information was disclosed by the Bank pursuant to its obligations under the data-sharing agreement, and with the husband's contractual consent. The information was true and not misleading. The process by which the information was disclosed was largely automatic, and I do not find any employee of the bank, or the structure of its systems, at fault for failing to prevent this disclosure. I therefore conclude that if the bank (or Mr Williams personally) owed CG any duty of care in negligence, in respect of its disclosure of credit information about her husband to the CRAs, it fulfilled that duty and did not breach it.

17. MRS GATT'S CASE IN DEFAMATION
18. 9.1 Bearing in mind that the Gatts were not legally represented at the trial, and that the law and practice of defamation are less familiar to them than the commercial and contractual matters with which they have dealt in business, I have approached their pleaded case on the basis of the legal issues that the matters relied on would disclose to a libel lawyer, rather than on its strict wording; I hope that in doing so I have also sufficiently respected Barclays' rights. Again, it is helpful to begin by considering what MG's position would have been if his claim in libel had been before me. (All the same evidence has been heard, and he has been able to respond to it, so this is not an unfair or unreasonable exercise.) He would certainly have been able to establish the three essential elements of the cause of action, at least against the Bank (I shall consider the personal liability of Mr Williams separately):

a. There is no doubt that the Bank published the words complained of to third parties, directly to the CRAs and indirectly (but with such inevitability as to render the Bank a primary publisher) to the member institutions who took up the reference from the CRAs.

b. There is no doubt that the words referred to or identified MG, who was expressly named.

c. And, after more reflection, there is no doubt that it is defamatory of someone to say that he has a delinquent or unauthorised overdraft, especially if it is to the tune of £250,000 on a £1,500 limit. The defamatory meaning, which includes the inferences that a hypothetical reasonable reader would draw from the literal facts stated, would be to the effect that "he had shown serious irresponsibility in financial matters by overdrawing money from his bank in grave excess of the limits it had allowed".

9.2 When one turns to the libel defences upon which the Bank relies, however, a different picture emerges. For the contractual reasons stated at (7) above, MG would be taken to have authorised the publication, and therefore could not have sued upon it; (but this would not apply to CG's claim). More significantly, while it would not be right for me to reach a conclusion on justification as against MG, there would obviously be a serious risk that the words complained of would be found to have been substantially true; I have held above that they were literally true, and there is much in the history of his dealings with the Bank, set out above, which would support the truth of the defamatory inferences spelt out in the meaning at 9.1.c above. (The words complained of are clearly statements as to fact, not expressions of judgment, comment or opinion, so the defence of fair comment/honest opinion cannot apply.)

9.2 When one turns to the libel defences upon which the Bank relies, however, a different picture emerges. For the contractual reasons stated at (7) above, MG would be taken to have authorised the publication, and therefore could not have sued upon it; (but this would not apply to CG's claim). More significantly, while it would not be right for me to reach a conclusion on justification as against MG, there would obviously be a serious risk that the words complained of would be found to have been substantially true; I have held above that they were literally true, and there is much in the history of his dealings with the Bank, set out above, which would support the truth of the defamatory inferences spelt out in the meaning at 9.1.c above. (The words complained of are clearly statements as to fact, not expressions of judgment, comment or opinion, so the defence of fair comment/honest opinion cannot apply.)

9.3 Most importantly, and a matter on which I am in a position to make the same finding in respect of both MG's ineffective case and CG's effective one, there is the defence of common-law qualified privilege. It is well established that such a privilege will apply to non-malicious communications between people who share a legitimate, common and corresponding interest in the subject-matter of the communications, provided that it is in the public interest for such a privilege to be recognised. In the old case of Macintosh v. Dunn [1908] AC 390, the Privy Council declined to accept such a privilege in relation to a so-called "trade protection society" which gathered credit information and sold it to its subscribers for profit; the PC held that "those who engage in such a business, touching so clearly very dangerous ground, should take the consequences if they overstep the law". But very soon afterwards, in London Association for Protection of Trade v. Greenlands [1916] 2 AC 15, the House of Lords distinguished Macintosh in relation to information provided by the secretary of a mutual association of traders to a member who had inquired about the financial standing of a business with which he was considering dealing. For present purposes the key dictum is that of Lord Atkinson at 37:

"…the association…was not created or worked solely from motives of pecuniary gain. It was formed…for the protection in trade or business of each of its members."

9.4 It is necessary for me to remind myself that the credit reference agencies themselves, Experian, Equifax and Callcredit, are not parties to this litigation. I am not in a position to, and do not, make any findings one way or another as to which side of the Macintosh/LAPT line any of those agencies might fall if sued in this or another case. (It may well be that some categories of business they do would be protected, and some not.)

9.5 What this case concerns, as the uncontradicted evidence of Mr Hill referred to at 3.3 above shows, is not so much the ordinary relationship between a CRA and its general subscribers as the special relationship between a large number of financial institutions which have agreed to share certain categories of customer performance data with one another, through the medium of the CRAs. He describes how the Information Commissioner, whose statutory role is to oversee the working of the Data Protection Act, works closely with a body called SCOR (Standing Committee on Reciprocity) to regulate the operation of credit referencing. SCOR is a cross industry forum made up of representatives from credit industry trade associations, credit industry bodies and CRAs. Its members include the British Bankers' Association and the Council of Mortgage Lenders; Barclays in turn is a member of those two well-known bodies. SCOR has put together a set of principles, the Principles of Reciprocity, which governs the sharing of personal data between the CRAs and their financial industry subscribers. The essential principle is that of reciprocity; each bank or other financial institution which wishes to subscribe for access to the pool of information available must itself agree to provide regular performance data about its own customers into the pool maintained and administered by the CRAs. (As stated at 5.2 above, this only applies to those customers who have joined a given bank after its standard terms were amended to authorise such disclosure; that is why the reports here only mention MG not CG.)

9.6 So, in this case, Barclays as a subscriber to the system put the information now complained of into the pool; and when the Gatts through their brokers applied to BoS for a mortgage, BoS (also a subscriber, also a contributor to the pool) was able to access the information from Barclays through the medium of the CRAs. Because the data originated from Barclays' own client records, it was inherently likely to be accurate and reliable, though of course error in a particular case cannot be excluded. In the particular circumstances of this special and well-regulated system (which may well be very different from other classes of credit reference business):

a. the elements of mutuality, reciprocity, and self-protection emphasised in LAPT are clearly present;

b. and the elements of profit, narrow self-interest, and inherently unreliable sources emphasised in Macintosh are clearly absent.

In the modern world, it is plainly in the public interest that such authoritative credit information can be obtained and relied on by banks and other financial institutions, provided it is done in a lawful and duly-regulated manner which respects the rights of the general public and the individuals affected. On the evidence provided by Mr Hill, the passing of the present information by Barclays into the CRA pool, and its onward transmission by the CRAs to BoS and/or any other subscribers who may have accessed it for the purpose of deciding whether or not to accept an application from the Gatts for finance, plainly took place on occasions of common-law qualified privilege. This occasion protects the publications equally whether complaint is made by MG and/or CG, subject to the question of express malice considered next.

9.7 Qualified privilege is defeated if the Claimant can prove that the Defendant was actuated by express malice in publishing the words complained of. In this context, what must be shown is a dominant and improper motive, i.e. other than the purpose for which the privilege was given; and generally this is proved by showing that the Defendant knew that his words were untrue, or at least did not believe them to be true (see Horrocks v. Lowe [1975] AC 135). Negligence is not enough, unless it rises to the point of reckless indifference to truth. Where the Defendant is a corporation, it must also be shown that a particular employee or agent both participated in the publication and had the required malicious state of mind; Broadway Approvals v. Odhams [1965] 1 WLR 805. Here, the only such person would have to be Mr Williams.

9.8 Given the complex and largely automated chain of events by which the words complained of came to be published, it would be impossible for either MG or CG to prove that Mr Williams was personally liable for the publication at all. The only way it could have been prevented would have been for the £250,000 overdraft limit to have been renewed and extended for a further 10 months or more after the initial 4 months expired in June 2007 (or alternatively for the debt to be cleared in some other way). Mr Williams had no power to do that. The most he could have done was to use his best endeavours to persuade those above him to sanction the higher limit or otherwise regularise the Gatts' borrowings; and all the evidence, in particular the Notebook records, points strongly to his working to that end, in particular by advocating the consolidated loan which would have cleared the unauthorised element of the overdraft. He was aware of the credit reference system in general, but there is nothing to suggest that he gave specific consideration to it in relation to the Gatts, let alone that he consciously engaged in acts or omissions with the intention of influencing the content of their credit references.

9.9 Even assuming that some act or omission of Mr Williams caused or contributed to the publications complained of, was he malicious in the above sense at that time? (The Gatts do not in fact suggest that he was; but for the reasons stated at 9.1 above it is right for me to consider the possibility.) Again, his letter to Mr Elmes is relied on, to show that he knew or believed that the words complained of were false; if with that state of knowledge he had caused them to be published, there would be strong evidence of malice. But, for the reasons set out at 7.5 above, my interpretation of that letter is different. It appears to me to show that Mr Williams was shocked at the credit report and its possible effects, and that he was prepared to go to the outer limit of truthfulness in order to undo it and help the Gatts to get their loan. If he was a malicious person who wanted to falsify the Gatts' credit record to their detriment (and no reason has ever been suggested why he should want that) then he would certainly not have written such a letter.

9.10 On this point, I should also add my own assessment of Mr Williams, based on his oral testimony before me and the other evidence, in particular Notebook and his few contemporary emails that were available. He was plainly embarrassed by his part in the problems the Bank found itself in, and in particular the ill-advised terms of the Elmes letter. But he did not give the least impression of a person actuated by malice towards MG or CG. On the contrary, they had won his approval; he was one of the few people in the Bank to retain faith in their ability to pay their way out of their excessive borrowings, and but for him I suspect the Bank itself would have acted against them, even before the credit reference problem emerged. It follows that there is no case on malice, and any defamation claim based on these publications must fail on the ground of qualified privilege.

9.11 Although that is a sufficient answer to CG's claim, I should also consider some specific issues in relation to which her position may differ from MG's, in particular the questions of identification/reference and defamatory meaning. In doing so I should bear in mind in her favour the fact that the readers of the credit reports of which she complains will have read them in the context of the joint mortgage applications made by the couple, since it was those applications which gave rise to the wish to check MG's credit. Those readers will therefore have known not only that CG existed, was married to MG and was a joint owner of Melksham Court, but probably also that (as stated in their questionnaire provided to Savills which formed the basis of that application) she was the Finance Director of the GLN group which he managed. (They will not, however, have known that she was a joint account-holder of the overdrawn account.) These facts are capable of giving rise to a "legal innuendo" on both the above issues; that is, a reader who knew those facts might as a result understand the words complained of differently from a reader who was not aware of them.

9.12 Unlike negligence by statement (or even perhaps malicious falsehood) in which a person does not necessarily have to be personally identified or referred to, so long as there is a causal mechanism by which they may suffer damage as a result of the publication, the tort of defamation does have such a requirement. But here, CG is not referred to or identified expressly, or even by innuendo implication; the words refer only to one person, who is named as MG. Even a reader who knew that CG was married to MG and was Finance Director of their company would not think that these words, apparently relating to his sole personal account, were about her as opposed to him, or even about her as well as him.

(To be clear, if to take an entirely different example I were to say that Mr A is a thief, then of course it is possible that those who know his wife may think to themselves: "If he's like that, what does that tell me about her?" But that would be an independent speculation on the hearer's part, and would not show that my original statement had identified or referred to her personally. The position would be different if the allegation against Mr A was of such a nature as to imply that his wife was also likely to be culpably involved. But it is entirely possible, indeed common, for one party to a marriage to run up large personal debts of which the other is not even aware, let alone to blame for.)

9.13 For rather similar reasons, even if CG could prove that the words referred to her, she would be unable to establish that the words were defamatory of her, that is, such as to lower her reputation in the minds of ordinary reasonable people of the kind likely to read the words (in this case, bankers evaluating the mortgage application). The hypothetical reasonable reader "can read in an implication more readily than a lawyer and may indulge in a certain amount of loose thinking, but he must be treated as a man who is not avid for scandal" (Gillick v. BBC [1996] EMLR 267 per Neill LJ at 272-3, as part of his longer and now classic summary of how to determine defamatory meaning). Even such a reader, who knew CG was MG's wife and business colleague, could not reasonably read the credit report as meaning: "that woman has an unauthorised overdraft" or even: "that woman shares her husband's blame and irresponsibility for having an unauthorised overdraft". At worst, they would think it meant: "that woman is married to a financially irresponsible man with an unauthorised overdraft". Although that might be a misfortune, it is not an allegation which reflects on her personal reputation.

9.14 I therefore conclude that CG's claim in defamation must fail for lack of identification and/or defamatory meaning, as well as on the qualified privilege ground set out above.

9.15 Finally I should deal with the issue of limitation. The claim here was issued in 2010, some 2 years after the Gatts became aware of the credit reports. There is a 1-year libel limitation period, which may be be extended by the Court under s.33A of the Limitation Act 1980 in an exceptional case if it is equitable to do so. Given that trial has now taken place and has exposed the fatal flaws in CG's libel case, set out above, there is no prejudice to her in refusing to extend the limitation period, and no reason in equity to do so. The libel claim therefore also fails on limitation grounds.

(I should note, however, that had her claim otherwise appeared to have merit I would have extended the limitation period, because of the factor, which I would have viewed as decisive, that since the Bank would have been facing litigation on all the same facts and matters under the other causes of action in any event, the additional prejudice to it from extending the limitation period would plainly have been less than the prejudice to CG of shutting out an otherwise worthwhile claim.)

19. DAMAGE ISSUES
20. 10.1 Although for the reasons given above it is my primary finding that CG's claims all fail on liability, so that no question of damages or other compensation arises, I did hear evidence on the issues of causation and amount of loss, and it is right for me to summarise that evidence and state my conclusions on it in the alternative to my decisions on liability, and on the basis that the credit reports were false and misleading as the Gatts contend. CG's case on damage is similar in respect of each of the three causes of action, and relies on proving each of the following elements to the civil standard:

a. that but for the false or misleading entry on the credit reports, the Gatts would have been able to remortgage Melksham Court and raise about £400,000 capital;

b. that with that money they would have been able to fund some or all of a series of development projects which they had in the pipeline;

c. that without that money they would have been unable to fund those projects; and

d. that, if so funded, those projects would have made them a sufficient profit in a sufficient time to restore their financial affairs to balance and permit them to retain their home, their other assets and their lifestyle until the present day.

10.2 The Bank challenges each of those elements, and contends in summary that even if the credit reports had never been published, by April 2008 the Gatts' affairs were already beyond redemption, partly because of their excessive and irredeemable existing level of borrowing, and partly because of the "credit crunch" and serious economic downturn that were well under way by that time, and which have continued until the present day so as to render unviable the property development business model which had served the Gatts so well over the previous 20 years.

10.3 It also takes an overarching point of law, that CG cannot pursue these damages claims because they are claims for "reflective loss". That is, any losses of profit caused by inability to fund promising deals would, at least in the first instance, be losses of profit sustained by one or more of the companies in the GLN group, since it is the companies which engage in the development projects, not the Gatts personally. Further, it is likely that if the deals had been profitable then the companies would, as before, have passed the profits back via a large salary for MG not CG; though she would have benefited from that salary, it would still be his not hers, and therefore she cannot sue for the loss of it any more than she can for losses caused to the companies in which she may have been a shareholder.

CG challenges this as an artificiality. She says she and her husband are a team, who share everything; their business is a joint business and its profits fund their joint lives. If the business has been destroyed by the Bank's misconduct, so that she loses her home and her assets, how can it be said that she has not suffered personal loss, at least to the extent of half the joint losses? It is clear, however, that the law is as the Bank states; see for example Prudential Assurance v. Newman (No.2) [1982] Ch 204, CA. For their own good reasons, including no doubt tax and limited liability, the Gatts have always conducted their business not as a partnership but through a corporate structure, and have channelled the profits primarily through MG rather than CG as the disparity in their taxable incomes demonstrates. They cannot now step back from this when it suits them and claim that CG was personally and directly entitled to her share of the profits, so as to have a personal claim for them against the Bank. Although the damages claims would fail on these grounds alone, I shall now consider the other issues they raise.

10.4 Would the Gatts have been able to raise substantial funds by remortgaging Melksham Court? This depends partly on the amount of free equity available as security at that time, and partly on their ability to satisfy the new lender that they would be able to service the increased loan. I note that in 2007 the Woolwich, which as an arm of Barclays had full information about the Gatts' affairs, had flatly declined to do so, and they had had to go to Northern Rock, whose high-risk lending strategy at that time is a matter of judicial notice. I also note the serious errors and omissions made by CG personally when making the 2008 mortgage application, set out at 6.1 above. CG cannot put forward a damages claim on the implicit basis that but for the credit report the errors and omissions in her mortgage application would not have been detected. The court must work on the assumption that the Gatts would have told the proposed lender everything material to their application, or at least that the lender would have found it out before reaching its decision.

The Gatts maintain that the property was worth some £4 million in mid-2008, leaving ample security even if £3 million, i.e. the existing mortgage plus an extra £500,000, was borrowed. Neither side has presented formal expert valuation evidence, but in early 2007 the Woolwich had only valued it at just over £3 million. It is well known that by mid-2008 property prices had already fallen substantially from their peak, though of course it was not then known for sure what the future would hold. Any lender would demand a valuation comfortably higher than the total of its loan and any prior charge, and I am not persuaded that it was then worth more than £3.5 million excluding the cottage.

As at that time, there was secured on the property not only the £2.5 million Northern Rock mortgage but also a second charge of £450,000 in favour of Saphire Properties to secure a separate debt of MG, which would have to be postponed in the event of a remortgage. The mortgage alone demanded repayments of £130,000 a year, i.e. over £200,000 before income tax. (In addition there was a separate mortgage of over £300,000 on a cottage which the Gatts had, in effect, sold off separately from the original estate, thus reducing its overall value.) Then of course there was over £600,000 of unsecured borrowing from Barclays on the joint loan and joint current accounts (only one of which had appeared on the offending credit reports). Even if described as "authorised", this would have raised questions in the lender's mind as to their ability to repay. And there were several lesser but still substantial borrowings (credit cards, vehicle finance and so forth) that although of no great significance on their own would not have been reassuring to a lender in the context of the other matters.

On the balance of probabilities, CG has not persuaded me that but for the problem with the credit reports she and her husband would have succeeded in remortgaging Melksham Court in mid-2008 so as to raise the extra finance they required. I consider it likely that, in the more cautious lending environment then prevailing, lenders would have taken the Woolwich's view of their prospects rather than the Northern Rock's.

10.5 Would they have been able to use money so raised to fund proposed development projects? On this point, I am with CG. The Gatts produced both contemporary documentary evidence, and the oral evidence of their then business associate Gavin Thomas, whose evidence I found persuasive, to show that at that time they had several worthwhile schemes in hand, which required the injection of medium-sized capital sums if they were to be brought to the point at which the Gatts could sell on at a profit in accordance with their usual business model. They included the following:

a. Newport, Glos: option concluded, £50,000 required, £1.5 million profit anticipated;

b. Dursley: purchased subject to planning; £60,000 required; £1.1 million profit anticipated;

c. Clapham Road, London: terms agreed; £75,000 required; £1 million profit anticipated;

d. Chippenham, Wilts: agreement in principle; £100,000 required; £12 million anticipated (but not for many years).

In each case Mr Thomas attributed the inability to raise finance to the effect of the adverse credit rating on a particular bank manager being asked to lend on that project, but there was no direct evidence that that was in fact the case. The Gatts put it rather differently before me, that they were going to raise the money themselves on their house and then invest it in these projects. (They told the proposed lenders that their objective was to do improvements on the house and repay borrowing, which is another reason why I am not persuaded that they would have got the remortgage if they had told the truth.) Given their past history in business, I accept that it is likely that, if they had been able to raise extra finance in the amount of £400,000 or so as they wanted, they would probably have invested most or all of it as "seedcorn" in some or all of the projects listed above.

10.6 Did they need to raise money in order to fund those projects? It occurred to me in the course of the trial that if, as the Gatts claim, they lost valuable assets over and above the house as a result of the Bank's actions, and if as they also claim these projects were such good investments, then should they not have realised those assets by way of mitigation in order to pursue the projects? However, on reflection it appears to me that this is a false point, and that by mid-2008 these "assets" are likely already to have lost much of their original value or to be pledged to the hilt, often through the finance agreements used to acquire them. I would not regard this point alone as fatal to CG's case on damages.

10.7 Would the projects have saved the Gatts' financial position? Although I accept Mr Thomas's evidence as to the existence of the four projects and the anticipated returns as at mid-2008, that does not mean I accept that, had the Gatts been able to pursue the projects, they would on the balance of probability both have matured into profit and done so in time. On this question I note that Mr Thomas had not been willing to put his own money into any of these projects when his co-venturers the Gatts had to drop out; he said he had better, or at least preferable, prospects of his own. This suggests to me that they may not have been quite such certainties as the Claimant's evidence was suggesting.

More to the point, I note that in each case the returns were dependent on the general property market, which we now know continued in decline for several years so that the prospects for new developments worsened rapidly. In addition, any profits that might be realised were some way down the line; the evidence does not permit me to be sure, but I consider it unlikely that any would have produced a profit in the Gatts' hands until at least a year later than April 2008, maybe much longer. Meanwhile, however, the wolves were at the door, in the form of both Barclays itself and the Northern Rock whose payments were £75,000 in arrears by late 2008. It is possible that, if the Gatts had been able to fund the projects, they could have used them to persuade their creditors to wait as they had often done before; but it is more likely that, in the financial conditions then prevailing, their bluff would finally have been called.

I therefore conclude that it is more probable that like so many others the Gatts, who had prospered on the rising market but had spent their profits and continued borrowing beyond their means, would not have been able to survive the falling market even if these credit reports had not been published, and even if they had continued to speculate with more borrowed money into 2009. In short, they would probably have lost their net assets (which were in fact very few) in any event.

21. THE COUNTERCLAIM
22. 11.1 The final question is in some respects the most difficult. The Bank seeks to recover from CG her liabilities under the joint loan account and the joint current account. As to the former, after an initial denial she has accepted that she did assume that liability, in particular by signing up for it to be charged on the jointly-owned home, even though that charge was never in fact perfected. But as to the latter, she denies liability, saying that she never consented to the debt on MG's sole account being transferred to their joint account in February 2007.

11.2 The Bank's primary contention is the simple one, that either signatory to a joint account has the authority of the other to bind them to any transaction on the joint account, in this case to the making of a large transfer out which had the effect of putting the account into overdraft. In the ordinary case, this is certainly true; a bank cannot be expected to look behind such transactions, especially into the dealings between husband and wife. The present case may however be different, because the Bank here was actively involved in the transaction by which the sole account overdraft was transferred to the joint account, and indeed had an interest in it since the effect of the transfer was to improve the Bank's position in respect of the debt. It is therefore appropriate to look into the evidence about the decision to transfer the overdraft to the joint account.

11.3 CG accepts that she discussed that plan with Mr Williams at a meeting in February 2007. At that time the Bank was considering refinancing the Gatts by means of a remortgage (the deal which the Woolwich eventually vetoed). CG says that she told Mr Williams that she would consent to the overdraft going into the joint account, but only if the remortgage was granted. She was not told that the transfer went ahead later that month, and did not find out about it until later (her witness statement says in May 2007). She did not see the joint account bank statements and letters which would have shown her what had happened.

11.4 Mr Williams points out that at this time there was a plan, quite separate from the remortgage, to secure the existing joint loan and the sole overdraft by a second charge on the house; this would require the overdraft to become a joint loan, which was the reason the overdraft was transferred to the joint account. There was no reason to link this to the grant of a remortgage and he does not recall CG doing so. (By the time the overdraft was moved to the joint account, the Woolwich had already declined the remortgage, though negotiations were continuing.) In February 2007 the Gatts' solicitors informed the Bank that the second charge had been signed by both MG and CG, and MG consented to the transfer of the overdraft to the joint account.

11.5 On this issue, I prefer the evidence of Mr Williams to that of CG. It makes no commercial sense to link the joint overdraft to the remortgage, and I do not accept that CG ever made it clear to Mr Williams that she would not consent to the joint overdraft unless the remortgage was granted. In those circumstances, MG's authority to commit them both to transactions on the joint account remained unfettered. Moreover, at this time MG was engaged in extensive discussions with the Bank about their joint affairs, as had always been the case with CG's clear consent; and past history showed that CG had recently consented to a very similar transaction in relation to MG's previous sole overdraft when it was transferred to the joint loan account. I further note that, very soon afterwards, the Gatts did remortgage, and moved their active banking elsewhere, and I am confident that this would have followed a review by them both of their financial affairs and the overall state of their dealings with Barclays. Finally, and more generally, I note that CG has always emphasised her equal status with MG in business and financial affairs as Finance Director of the group, and that they both accept that their joint strategy was to fund their business and lifestyle through borrowing, secured if necessary on their joint property. In those circumstances, I do not accept that her husband took this important step without her knowledge and consent. (He later protested after the crisis broke that he had been unaware of the joint overdraft either, but it is clear from his email of 25 September 2007 that he was fully aware of it.)

11.6 I therefore conclude that when MG authorised the Bank to debit the joint account with the amount of his sole overdraft he had CG's authority, both ostensible and actual, to do so, and that therefore CG is personally liable to repay Barclays all sums due under both the joint loan account and the overdrawn joint current account, and that the counterclaim succeeds.