Rubenstein v HSBC Bank Plc

This is a case about the claim of a consumer (someone who is described in the Financial Services and Markets Act 2000 as a “private person”), against a bank, for negligent advice in the recommendation of a financial investment. In August/September 2005 the investor wanted to find a safe place for the proceeds of the sale of his home pending the purchase of another property. He wanted to find an investment, if that were possible, that provided a higher interest rate than a standard bank deposit, but he emphasised that he could not afford to risk his capital at all. He said that the prospective time scale was unlikely to be longer than a year, but in the event he had been unable to find another home three years later, so that, when the market turmoil which surrounded the collapse of Lehman Brothers in September 2008 occurred, he was still invested. His claim has been brought for breach of statutory duty, and in contract and tort. The judge found that the bank was negligent in the advice which it gave, and in breach of various statutory duties, and that the investor relied on the bank’s advice. However, the judge also found that the loss suffered by the investor was not caused by the bank’s negligence or breach of duties: it was rather caused by unprecedented market turmoil, and was unforeseeable and too remote. The investor was therefore awarded merely nominal damages in contract.

The judge gave permission to both parties to appeal. He gave permission to the investor to appeal against his conclusions as to causation, foreseeability and remoteness. He gave the bank permission to appeal his conclusions as to the existence of any contract, as to negligence, and as to breach of the statutory duty as to the suitability of the investment. It is not clear to me that the bank needed permission to appeal. I treat its notice of cross-appeal as a respondent’s notice.

The bank disputed every possible issue at trial: for instance, in addition to the issues which arise on appeal, issues no longer relevant as to whether any advice had been given at all and as to whether the investor had relied on any advice. Those last two issues are no longer live. It is now accepted that the bank provided advice (albeit not it is said pursuant to any contract) and not merely information; and it is accepted that the investor relied on that advice. Nor is it said any longer, as it had been said at trial, that Mr Rubenstein had selected the investment for himself. However it is submitted that the investor got what he wanted and cannot complain, that at any rate in September 2005, at the time of the investment, the investment could have been regarded as safe, and that the judge was in any event right to say that no damages flow from any breach.

The investor resists the cross-appeal and submits that the judge was wrong to hold that no loss flowed from the established breaches. He had been told that the recommended investment carried no risk of capital loss because it was the same as a cash deposit: in the event, he had suffered a loss of capital, for the very reason that it was not the same as a cash deposit, and carried no obligation to return the capital invested but only an aliquot share of a fund invested in products which, apart from cash, varied in value with the market. He was therefore not only misled but suffered loss of a type which should have been foreseen, and was in fact foreseen but not explained. The fact that the size of the loss may have been greater than could have been expected was beside the point.

The major factor on which the bank has relied in resisting the investor’s appeal is that the bank had no duty which extended beyond the investor’s own projection that he would be unlikely to need the investment for more than a year. No loss was suffered within that year, or for well beyond it. Any projection for the year beyond September 2005, when the investment was made, would have been that any risk associated with the investment was minimal. Therefore, whether viewed as a matter of scope of duty (SAAMCO v. York Montague Ltd [1997] AC 191) or in terms of the reasonable contemplation of the parties (The Achilleas [2008] UKHL 48, [2009] 1 AC), or in terms of breach of statutory duty, which it is said must follow the same paths, or whether viewed in terms of causation, foreseeability or remoteness, there could be no liability for any loss which occurred three years later, even putting aside the extraordinary events of that latter time, but especially in the light of them. This is the leitmotiv which flows through the bank’s written and oral submissions on this appeal.

An oddity of these proceedings is that this theme was barely played below, and is absent from the judge’s careful judgment. I do not mean that the bank’s defence or the judge’s judgment do not mention, as part of the events of the case, the investor’s projection of an investment unlikely to last more than twelve months: but no legal point was addressed by reference to that fact. On the contrary, the test put forward was whether, as of the time of the investment itself, the investment was more or less safe (see para 21(iii) of the 35 page defence). The fact was deployed in the defence only in connection with a plea of contributory negligence (see para 22(vi), another point no longer raised on appeal) and only surfaced in the bank’s closing written submissions at trial as a short submission (at para 8.1(iii)), but bound up with numerous other ideas. It is understandable that the judge did not give it separate treatment. In the circumstances, it is harder to evaluate than might otherwise have been the case. However no pleading point is taken.

The judge at trial was His Honour Judge Havelock-Allan QC, sitting as a high court judge in the mercantile court of the Bristol District Registry. His judgment may be found at [2011] EWHC 2304 (QB).

The investor is Mr Adrian Rubenstein, a solicitor. In 2005 he was a customer of HSBC Bank plc (HSBC or “the bank”). Mr Rubenstein is the claimant, and in this court the appellant. HSBC is the defendant, and here the respondent.

In around the middle of August 2005 Mr Rubenstein, then 38 years old, was anticipating the receipt of £1.25 million on the completion of the sale of his matrimonial home in London, and approached the bank about a safe place for those proceeds. He and his wife intended to rent for a period before looking to buy another home. They wanted to deposit the proceeds where they could be readily accessible, in case of a repurchase. They could not afford to risk a loss of capital. They had two young children. It was not suggested by the bank that Mr Rubenstein was other than a private, retail, customer.

Mr Rubenstein had researched the bank’s deposit rates, but those on offer were capped at a maximum of £1 million. He thought he might do better than that if he made a direct approach. His initial proposal was for an interest bearing deposit with ready access for no more than about 6 months. The bank introduced him to Mr Matthew Marsden, a financial adviser employed at its private client department in Cardiff. He was a fully qualified independent financial adviser (IFA), but was only permitted to recommend financial products which were on the bank’s approved list.

Mr Rubenstein and Mr Marsden first spoke on the telephone on 22 August 2005. The judge rejected Mr Marsden’s evidence that Mr Rubenstein was looking for something other than a deposit, and found rather that Mr Rubenstein did not rule out a deposit, but was looking to improve on the rates identified in his internet research. Mr Marsden’s immediate reaction was to mention an investment in the AIG Premier Access Bond (PAB). He said it was an insurance product and said he would send some literature.

On 23 August 2005 Mr Marsden e-mailed Mr Rubenstein. His e-mail was headed “Account information”. Attached to the e-mail were AIG’s product brochure and key features document (see below). The e-mail quoted a gross equivalent rate basis for £1 million plus of 5.78% (on a fee basis) and 5.12% (on a commission basis). The corresponding rate for a deposit at HSBC was 3.93%. The difference between a fee basis and a commission basis was explained by Mr Marsden in a further e-mail later that day. The message informed Mr Rubenstein that he could “pay for the advice” in one of two ways: either by way of a fee at £190 plus VAT per hour, with a minimum of £1500, with any commission received rebated in the form of a higher return, or, as he implied, by way of commission. This e-mail attached the bank’s “key facts” about their services and the cost of them.

On 24 August 2005 Mr Rubenstein e-mailed in response as follows:

“I’ve done a few sums to try to compare the impact of the two payment mechanisms. Underlying all this is the fact that we are very unlikely to need this account for more than a year; probably less.

Over the course of that year, the commission-based rate removes about £8500 gross interest. If I’ve done my sums properly, that equates to about 37 hours’ work based on the charge-out rates you quoted. Are you really likely to spend that kind of time in setting up and managing the account?”

Mr Marsden replied on the same day:

“It is very likely that the minimum fee will be charged and that that will be all. As once the account open it is effectively an instant access account so it is unlikely that you will need further advice.”

That, and Mr Marsden’s next e-mail are important documents.

On 24 August, on speaking to his wife (who was then in New York) about the proposed investment in the PAB, Mr Rubenstein was asked by her to get an assurance from the bank that there was no risk to their capital, ie that the investment was as good as cash. In the light of their conversation, Mr Rubenstein sent the following message to Mr Marsden:

“We can’t afford to accept any risk in the investment of the principal sum. Can you confirm what – if any – risk is associated with this product?”

Mr Marsden replied almost at once as follows:

“We view this investment as the same as cash deposited in one of our accounts.

I understand this question as no doubt you have read a statement such as “fall as well as rise”.

The reason for this is that as the bond is wrapped in insurance bond legislation (for tax reasons) it is looked after by the Financial Services Authority (FSA). As a result you have to be informed of all the risk factors.

Putting it into context if you put all the money into an account with HSBC, the guarantee of your money is underpinned by the financial security of the institution. If HSBC were to fail however you would get money back in line with the depositor protection scheme which would only guarantee 33K of the 1.25m.

In the context of the FSA, the fall as well as rise pertains to the risk of default (as above with HSBC) of one of the accounts held within the fund. As the enhanced variable fund has a minimum security standard of A (HSBC is AA) the risk of default of one of the accounts is similar to the risk of default of Northern Rock.

I hope this clears up any queries you may have.

Matthew

PS The brochure explains the strength of AIG on page 4 or 5.”

These two emails sought to persuade Mr Rubenstein that the product Mr Marsden was recommending (AIG’s enhanced variable rate fund or EVRF) was the same as an instant access deposit account with HSBC, and that the only risk was the ultimate risk of default, which was no risk at all. Mr Marsden omitted to explain that, quite apart from the risk of default, the essential risk of the EVRF was that it was an investment in the market, and subject to market fluctuations. Mr Marsden advised that once Mr Rubenstein was invested, he would need no further advice. The EVRF was the only recommendation that Mr Marsden made to Mr Rubenstein. There was no discussion of the other alternative funds available within AIG’s PAB, such as the more conservative standard variable rate fund (or SVRF).

At trial, Mr Marsden gave evidence that he regarded Mr Rubenstein as a rate chaser, who cared less about the security of a deposit than the rate that was on offer; and that was the submission which was put before the judge on behalf of the bank by Mr Stephen Cogley QC, who represents the bank again on this appeal. However, the judge disagreed. He observed (at para 89):

“This was not investment capital: it was capital needed for a new property to live in. True he said that he was looking for a better interest rate than was available from the deposit accounts he had seen advertised: but he made it quite plain on 24 August that he and his wife could accept no risk to the capital. No risk to capital (which I interpret as meaning “the minimum possible risk”) was to be the sine qua non of the investment.”

The sale of the Rubinstein home was due to be completed on 22 September 2005, and there was no further contact until then. On that day Mr Rubenstein rang Mr Marsden, having decided on the basis of the emails exchanged to invest in the EVRF. He exercised no independent choice of his own, other than to accept the EVRF over an ordinary bank deposit, which, on the basis of the judge’s findings, he would not have done without Mr Marsden’s advice. Mr Marsden sent him an e-mail with the necessary forms to complete. One was the PAB application form, a second was a transfer request form, and the third was the bank’s fee agreement form (“FEEPAY”). Mr Marsden told Mr Rubenstein to fill in the box for the EVRF.

Mr Rubenstein completed the forms, stating that he wished to make withdrawals of “£27,300 p.a.” from the fund on a monthly basis. This was to pay the rent on their rented home. He signed the “Investor Declaration” on the PAB application form, and dated it 23 September 2005. The declaration stated:

“I/We have read and understood the Premier Access Bond Brochure and Key Features document, agree to the terms of this offer, declare that the details given herein are true and complete to the best of my/our knowledge and belief, and I/we ask AIG Life to accept my/our investment in the Premier Access Bond.”

The judge was to find that, however much Mr Rubenstein may have read, he did not properly understand the structure of the investment. This was disputed by HSBC, but the judge resolved this dispute in favour of Mr Rubenstein (para 102). (Footnote: ¹) Mr Rubenstein “did not understand that his capital might be at risk because a large number of investors wanted to withdraw their funds” (ibid). He was looking for an investment which posed the minimum possible risk to his capital and relied on Mr Marsden’s response as providing reassurance that the EVRF met the requirement (para 105). I will refer to the structure of the investment below. Mr Marsden did not understand the structure of the investment and thus the nature of the risk either (at paras 94 and 113).

On or about 26 September 2005 Mr Rubenstein transferred the £1.25 million to AIG and returned the signed FEEPAY form to HSBC. The latter began:

“Please provide the following services to me. I understand that a fee will be charged for these services at the rate of £190 + VAT per hour and where the fee relates to advice on packaged services it is subject to a minimum charge of £1500 + VAT unless specified…”

The form as sent to Mr Rubenstein had already had entered in type by HSBC a “Y” for yes in the box next to the option described as “Advice on packaged products (e.g. investments, pensions and life products – minimum fee £1500 + VAT. This will include any technical support required for these products.” The form continued: “Payment will be required upon completion of the provision of our advice and/or services. Please send me an invoice and arrange to debit my HSBC bank account…”. It was common ground, however, that the bank did not process the charge of £1,500 plus VAT and it was never debited from Mr Rubenstein’s account. The judge made no finding as to why that was. Mr Marsden was to give evidence at the trial that he had not provided Mr Rubenstein with any advice, only with information. However, the judge rejected that evidence, and there is no appeal on that issue. Similarly there is no appeal against the judge’s rejection of the submission that Mr Rubenstein selected the EVRF for his investment by himself.

Mr Marsden recommended the EVRF to Mr Rubenstein without discussion of any other product, and having rejected the idea of an HSBC deposit account in his own mind, having formed the view, which the judge found was mistaken, that Mr Rubenstein had already rejected that solution. He did not conduct any “Know Your Customer” analysis, and was, as the judge found, in breach of a number of both procedural and substantive rules of the FSA’s Conduct of Business Rules (“COB”). Those rules included those requiring the bank to take reasonable steps: to communicate with its customer in a way which is clear, fair and not misleading, to ensure that its customer understands the nature of the risks involved, and, where a packaged product is recommended, as in this case, either to recommend the most suitable from the range of such products or to make no recommendation at all.

The judge found that a contract (for advice) had been made between Mr Rubenstein and HSBC and that it did not matter when it was made. HSBC accepted that a contract had been made (albeit it submitted that it had been for execution services only), but said that it was not made until 26 September, after the investment. (The investment was in fact made on the same day as Mr Rubenstein returned the FEEPAY form, both on 26 September.) The judge said that the precise timing did not matter because in any event the advice was given in contemplation of a contract on the basis that a fee would be payable (at para 69).

The premier access bond or PAB was a product issued by AIG Life, a UK division of ALICO (American Life Insurance Company), itself a wholly owned subsidiary of AIG (American International Insurance Group Inc), one of the largest insurance groups in the world. The PAB had two variable rate interest funds, the standard fund (SVRF) and the enhanced fund (EVRF), as well as a number of guaranteed funds. It is unnecessary to consider the guaranteed funds, but the SVRF was an important and more conservative variable rate fund. The bond was held within an insurance package for tax effective reasons, but only such as provided life cover for about 101% of the bond’s value.

The judge set out full details of AIG Life’s brochure. It is necessary to reproduce only the following passages:

“Inside the Premier Access Bond your investment buys units in a range of funds, all of which invest in money market instruments to give you a competitive alternative to bank or building society deposits…

The Standard Variable Rate Fund (“the Standard Fund”) is a unit-linked cash fund investing in the short term, high quality financial money market instruments. It aims to generate growth rates that reflect general trends in short-term interest rates whilst providing a very high degree of safety by holding a diversified portfolio of very high quality assets…

A typical fund portfolio would be invested in banks such as Barclays, Citibank, Halifax Bank of Scotland, Lloyds TSB and UBS..

As all the assets purchased are very high quality and short term in nature, the fund shall be considered to be very cautious, with a very high degree of capital protection…

The Enhanced Variable Rate Fund (“the Enhanced Fund”) is similar to the Standard Fund but is aimed at achieving a slightly higher return by investing in a wider range of assets within the money markets. The fund offers a high degree of safety by holding the highest quality assets commensurate with its enhanced yield…

The fund should achieve a higher yield than the Standard Variable Rate Fund because it has access to:

(i)

a wider range of companies…

(ii)

a wider range of instruments issued by the companies it identifies…

(iii)

assets with slightly longer periods until maturity…

Although the fund carries slightly more risks than the Standard Fund it should still be considered to be a cautious fund. Although the criteria are clearly wider than those of the Standard Fund, the fund places high importance on the preservation of capital…”

The judge rightly observed:

“7…the essential distinction between the two variable rate funds lay in the classes of asset held in those funds, the credit rating of those assets, and their duration to maturity. The EVRF was slightly more adventurous than the SVRF in the quality of the assets it purchased, and it contained securities with a longer term to maturity than the SVRF. It was therefore less liquid than the SVRF in two senses. First, the proportion of assets maturing in less than 6 months was very much lower, so fewer of the assets were due to mature as cash in the short term. Second, the cash value of the longer-dated assets, if sold before maturity, would depend on the health of the secondary market in which they could be sold. The discount to acquisition cost on an early sale of such assets could fluctuate.”

The price risk associated with such an early sale was addressed in AIG Life’s literature, although the warning was couched in terms of “costs” rather than losses. Thus the brochure stated (on its sixth page under the heading of “Adjustments to unit prices”):

“If large numbers of Bonds are encashed at the same time, the fund(s) may incur costs in selling assets to meet these encashments and these costs would normally be reflected in the unit price(s). Alternatively AIG Life may defer encashments for up to three months if it considers that this would be more beneficial for Bondholders generally. This would only happen in very exceptional circumstances.”

In the key features document, under the heading “Risk Factors”, the following appeared:

“● The value of your investment can go down as well as up…

● If large numbers of Bonds are encashed at the same time, the funds may incur costs in selling assets prior to their intended maturity date to meet these encashments, and these costs may cause a fall in unit price and therefore the return on your Bond. Alternatively, AIG Life may defer encashments for up to three months if it considers that this would be more beneficial to Bondholders generally. This would only happen in very exceptional circumstances.”

Thus an investment in the EVRF was not the same as a deposit, or even akin to a deposit (although an investment in the SVRF was more akin to a deposit in that the brochure suggested that the essential investments in it were deposits in the major listed banks). The only risk the creditor of a deposit takes (Footnote: ²) is with the creditworthiness of his bank. Unless the bank fails, the creditor is entitled to the return of his deposit, with interest. However, the investor in one of the PAB funds was not entitled to the return of his investment, only to its value at the time of request, which could fluctuate on a daily basis. Value depended on the underlying assets held within the fund. It was only an aliquot share of those assets to which an investor was ultimately entitled. Those assets could vary between cash and short-dated securities to longer dated securities and complicated derivative products, including as things turned out products which fell foul of the “sub-prime” credit crunch which overtook markets in the autumn of 2008. The investment lacked the transparency of a deposit, because the underlying holdings were not marked to market on a daily basis or even independently valued, but instead were valued by an actuary of AIG Life. It was the internal actuary’s valuation which determined the value at which an investor could redeem his investment at any time. As the judge observed: “There was in fact no promise to pay any particular sum in response to a request for withdrawal.” Moreover, whereas AIG Life published a factsheet each month for each fund containing fund information for financial advisers, which included pie charts illustrating the underlying assets by asset rating, sector, and length of maturity, these were sent to HSBC but were not copied by it to the individual financial advisers in its employ, nor were they sent to investors.

There was expert evidence from actuaries at trial to the effect that, as of September 2005, there was nothing to suggest any risk to capital in the EVRF, at any rate looking forward over a 3-12 month period and without the benefit of hindsight. On the other hand, Mr Muhammad (“Icki”) Iqbal, Mr Rubenstein’s actuary expert, also said that “AIG Life’s investment policy was not without risk and this risk was downplayed”; that its product literature minimised the fact that the higher yield of the EVRF entailed higher risk, and that Mr Marsden should have had a discussion about fund selection with Mr Rubenstein.

There was also expert evidence from two financial services experts, Dr Jane Thompson for Mr Rubenstein and Mr Marcus Egerton for HSBC. The judge preferred Dr Thompson’s view that a bondholder in the EVRF not only took the risk that AIG Life might fail but also a risk that the value of his units could fall as well as rise, depending on market conditions; and that, in general cash is a lower risk than cash derivatives. The judge observed (at para 91):

“even in September 2005 Mr Rubenstein’s capital would have been safer in a cash deposit with HSBC than in the EVRF. The spread of assets held by the EVRF was wider than Treasury bills and municipal bonds. It included money market instruments or “commercial paper” linked to securitised loans in the U.S. sub-prime mortgage market and the buy-to-let mortgage market. Whilst in 2005 the degree of risk attaching to these instruments was thought to be very low, and whilst it is doubtful that HSBC had more than a general notion of the kind of bonds which were included in the portfolio of the EVRF, the bank would have realised that the underlying assets were not confined to Treasury bills and bonds of that character. The first part of the advice given to Mr Rubenstein was that an investment was the same as cash deposited. That was not correct. The structure of the investment was materially different and the risk was commensurately greater.” (Footnote: ³)

The judge therefore concluded (“although not without some hesitation”) that Mr Marsden was negligent in recommending the EVRF as being suitable for Mr Rubenstein. He gave two reasons for that judgment (at para 94):

“The first is that it was wrong of Mr Marsden to suggest that the EVRF was the same as a cash deposit. The second is that Mr Marsden made no attempt to consider the other funds in the PAB as possible alternatives. As to the first of these criticisms, I recognise that Mr Marsden was responding to a question from Mr Rubenstein about risk and that it might have been right to say in September 2005 that an investment in the EVRF was very safe and only marginally more risky than a cash deposit. But it was not the same as a cash deposit. The differences were not adequately explained and they should have been. As to the second criticism, Mr Marsden admitted that he did not investigate the other funds and what they had to offer. He put forward the EVRF because of the interest rate. It was his duty to examine the alternatives within the PAB and he should have done so. If he had looked at the alternatives and had given proper weight to Mr Rubenstein’s attitude to risk, he would in my view have concluded that the SVRF was more suitable than the EVRF, even in the circumstances prevailing in September 2005.”

HSBC disputes that conclusion on this appeal.

The Rubinsteins tried continuously to find a new property, but failed in their endeavours. In February 2007 they very nearly succeeded in buying something, but the seller pulled out at the last moment, so that the £130,000 that Mr Rubenstein had withdrawn from his bond in order to pay the deposit was returned into the bond. (Footnote: ⁴)

That remained the position until September 2008. On the weekend immediately before Lehman Brothers’ failure (which occurred on Monday 15 September) and at a time when AIG Life’s parent, AIG, was considered as another possible victim of the market difficulties (because it had insured many of the sub-prime credit derivatives) (Footnote: ⁵), Mr Rubenstein decided to withdraw his investment. At first he was told that his application was being processed (save for £100,000 which apparently he was required to maintain invested: on the latest valuation available that equated to £1,194,076.26). But late on 15 September AIG Life e-mailed him as follows:

“…we have witnessed withdrawal requests far in excess of normal levels from the Enhanced Fund…

Given the volatility of the bond markets, we are unable to put a precise value on the fund assets. Therefore, we feel that it is in the best interests of policyholders to temporarily suspend withdrawals from the Enhanced Fund (including the Notice Funds) until the conditions within the financial markets have stabilised.”

On 22 September 2008 HSBC forwarded another announcement from AIG Life to Mr Rubenstein. This said that the EVRF would be closed on 15 December 2008, at the end of a three month moratorium. Prior to that investors would receive 50% of the fund value (representing cash and liquid assets in hand) by means of a switch of the corresponding holdings and assets into the SVRF. The investors could then liquidate their new holding in the SVRF or not as they wished. As for the remaining 50%, investors could choose either to take whatever value there was as of 15 December or to leave that 50% invested until 1 July 2012 (the “maturity date”). The first option was called the “exit plan” and the second option was called the “maturity plan”. The maturity plan involved transferring the remaining investment from the EVRF into a “protected recovery fund” (or PRF). AIG Life promised to underpin the PRF with its guarantee that the investor would suffer no loss of capital value from the value of the remaining investment as of the transfer date of 14 December 2008 (but there was no guarantee about any income).

The AIG Life announcement contained this passage:

“Those who wish to leave the Enhanced Fund on 15 December 2008 will be able to do so and will receive their share of the sale value of the fund’s assets. In order to meet the requests we will need to sell some of the fund’s assets before they mature. The prices we are likely to achieve in current markets are poor. This will mean that policyholders leaving the fund on 15 December are very likely to receive less than the current value of their Enhanced Fund holding.”

No such similar disaster befell the SVRF, which on the contrary became the medium through which the initial 50% was provided to bondholders in the EVRF.

In the event Mr Rubenstein opted to take his second 50% in cash and so to go down the route of the exit plan. HSBC wished at one stage to take issue with that decision, to say that it was a failure to mitigate on Mr Rubenstein’s part, but the point was abandoned. The maturity plan option would have prevented the Rubinsteins using more than 50% of their funds to buy a new home, so the option did not have much appeal to them. Because the choice between the exit plan and the maturity plan did not ultimately matter, the judge did not go into their respective details. As a result it is not possible to say how AIG Life calculated the value of Mr Rubenstein’s investment as of 3 October 2008, the operative date for the transfer into SVRF of the first 50%, at £1,293,614.10, providing a 50% transfer of £646,807.05. That was very close to the valuation on the basis of which Mr Rubenstein had applied on 15 September 2008 to withdraw his investment. It is difficult to see how that might be, but it does not matter for present purposes.

On 27 October 2008 Mr Rubenstein withdrew the first half of his investment from the SVRF. Under the exit plan, Mr Rubenstein received only £469,472.84 for his second 50%. The judge commented that this represented a capital loss of £179,530.17. But the pleaded case was for the net loss after tax suffered by comparison with the total return which would have been earned had the money been placed in deposit account over the same period as the EVRF investment: and that figure was said to be £186,612.87.

Mr Rubenstein’s first complaint of mis-selling was made in a telephone call to Mr Marsden immediately after reading AIG Life’s announcement about the closure of its fund. He complained that he had wanted protection of his capital whereas his capital was now under threat.

It is now common ground that the statutory and regulatory regime applicable involved the application of the COB rules made by the FSA under the general rule making power conferred by section 138 of FSMA. (Footnote: ⁶) At trial HSBC alleged that the investment was an “execution only” transaction, in which case the COB rules would not have applied. The judge found however that investment advice had been given, and there is no appeal from that.

The judge set out large sections of COB in his judgment, but it will be sufficient to refer only to those rules which he found had been breached by HSBC through Mr Marsden. The letter “R” stands for “Rule” (as distinct from other parts of COB which are either guidance (“G”) or evidential), breach of which gives rise to an action for damages for breach of statutory duty pursuant to section 150 of FSMA. Section 150 (1) provides as follows:

“(1)

A contravention by an authorised person of a rule is actionable at the suit of a private person who suffers loss as a result of the contravention, subject to the defences and other incidents applying to actions for breach of statutory duty.”

It is said that a section 150 claim is subject to identical principles relating to causation, foreseeability and/or remoteness of damage as may apply in contract or tort, and the judge generally made no distinction between any of Mr Rubenstein’s three causes of action for these purposes, or for the purposes of his finding of negligence. However, whereas the underlying principles may be the same, they may operate in different ways, seeing that the purpose of a statutory rule may be more focussed than the general law of tort or contract is likely to be. Lord Hoffmann referred to this possibility in SAAMCO at 212D, where he said –

“How is the scope of the duty determined? In the case of a statutory duty, the question is answered by deducing the purpose of the duty from the language and context of the statute: Gorris v. Scott (1874) L.R. 9 Ex. 125. In the case of tort, it will similarly depend upon the purpose of the rule imposing the duty. Most of the judgments in the Caparo case are occupied in examining the Companies Act 1985 to ascertain the purpose of the auditor’s duty to take care that the statutory accounts comply with the Act. In the case of an implied contractual duty, the nature and extent of the liability is defined by the term which the law implies.”

It would seem therefore that, at any rate in the context where the COB rules apply to investment advice provided to a private person, the applicable principles in contract and/or tort will be guided by the focus and purpose of the statutory provisions.

In this connection sections 2 and 5 of FSMA (under Part 1, dealing with the FSA as regulator) are relevant:

The COB rules begin with a rule of “purposive interpretation”, to be found in –

“GEN 2.2.1 R

Every provision in the Handbook must be interpreted in the light of its purpose.

The purpose of any provision in the Handbook is to be gathered first and foremost from the text of the provision in question and its context among other relevant provisions. The guidance given on the purpose of a provision is intended as an explanation to assist readers of the Handbook. As such, guidance may assist the reader in assessing the purpose of the provision, but it should not be taken as a complete or definitive explanation of a provision’s purpose.”

As for the COB rules which the judge found to have been breached, they are as follows:

“COB 2 – Rules which apply to all forms of conducting designated investment business

The judge found a breach of this rule. Both financial services experts were agreed that the rule was breached. The judge preferred the reason given by Dr Thompson, namely “because the advice that an investment in the EVRF was the same as cash was misleading” (at para 96). There was no cross-appeal by HSBC against this finding (although, as will be explained below, Mr Cogley on behalf of the bank sought permission, during the hearing, when the possible consequences of not appealing this finding of breach were canvassed, to amend his appeal to introduce a challenge to the finding). Mr Cogley at times submitted that no case in misrepresentation had been brought at trial by Mr Rubenstein. Whatever a case in misrepresentation may or may not have added, in circumstances where a case was brought for negligent advice and for breach of this rule concerning misleading communication, it may be doubted whether Mr Rubenstein was, in practical terms, lacking anything in his armoury.

The next rules breached were under COB 5.2 – Know your customer, as follows:

“COB 5.2.4 G (Purpose)

Principle 9 (Customers: relationships of trust) requires a firm to take reasonable care to ensure the suitability of its advice and discretionary decisions. To comply with this, a firm should obtain sufficient information about its private customer to enable it to meet its responsibility to give suitable advice…

COB 5.2.5 R (Requirement to know your customer)

Before a firm gives a personal recommendation concerning a designated investment to a private customer, or acts as an investment manager for a private customer, it must take reasonable steps to ensure that it is in possession of sufficient personal and financial information about that customer relevant to the services that the firm has agreed to provide.

COB 5.2.9 R (Record keeping: personal and financial circumstances)

(1)

Unless (2) applies, a firm must make and retain a record of a private customer’s personal and financial circumstances that it has obtained in satisfying COB 5.2.5 R…

(2)

A firm need not retain the record where…the private customer does not proceed with the recommendation or any part of it.”

COB 5.2.12 R (Statement of demands and needs)

(1)

Unless either COB 5.2.13 or COB 5.2.14 applies, a firm must provide the client with a statement of his demands and needs if:

(a)

it makes a personal recommendation of a life policy to a client; or

(b)

it arranges (whether through issuing a direct offer financial promotion or otherwise) for the client to enter into a life policy..,”

The judge found the three above stated rules to have been breached. Once it was accepted that Mr Marsden had been giving advice, that was inevitable: for he took the view that he had never given any advice and for that reason had never sought to know his customer in the way required by COB. The bank’s case, however, was that breach of these, procedural, rules, as not the cause of any loss.

The next rule is under COB 5.3 – Suitability, which the judge regarded as “the rule that really matters”. Thus –

“COB 5.3.5 R (Requirement for suitability generally)

(1)

A firm must take reasonable steps to ensure that, if in the course of designated investment business:

(a)

it makes any personal recommendation to a private customer to:

(i)

buy, sell, subscribe for or underwrite a designated investment…

the advice on investments or transaction is suitable for the client.

(2)

If the recommendation or transaction in (1) relates to a packaged product:

(a)

it must, subject to COB 5.3.8 G – COB 5.3.10 R, be the most suitable from the range of packaged products, on which advice on investments is given to the client as determined by COB 5.1.7 R; and

(b)

if there is no packaged product in the firm’s relevant range of packaged products which is suitable for the client, no recommendation must be made.

(3)

In making the recommendation or effecting the transaction in (1), the firm must have regard to:

(a)

the facts disclosed by the client; and

(b)

other relevant facts about the client of which the firm is, or reasonably should be, aware.”

As to this rule, which he found to have been breached by the bank, the judge said this:

“97.

However, the rule that really matters, as I have already said, is COB 5.3.5(2). An investment in the EVRF was a packaged product because it included life cover. It was therefore required to be “the most suitable” from the range of packaged products on which the bank held itself out as giving advice (ie those on the panels approved by the Product Research Team). Mr Egerton [HSBC’s expert] accepted that the SVRF was more suitable for Mr Rubenstein in terms of risk but was not prepared to say whether it was more suitable overall or the most suitable. Even without the benefit of hindsight, I think it is possible to say that the SVRF would have been a more suitable home for Mr Rubenstein’s money than the EVRF. That conclusion is enough to establish that there was also a breach of COB 5.3.5(2) R.”

The judge said nothing about sub-rule (1). Nor did he say anything expressly about sub-rule (2)(b). Logically, however, he must either have thought that the SVRF was suitable, as well as being more suitable than the EVRF, or have considered that no packaged product was suitable. In either event, there was a breach. It also follows to my mind that, again in either event, the recommendation was not suitable under sub-rule (1), of which a recommendation under sub-rule (2) is merely a species (see the opening words of sub-rule (2)). However, I am content to proceed on the basis that there has been a breach of only sub-rule (2). This was the only finding of breach about which HSBC was given permission to appeal by the judge. It may be noted that whereas sub-rule (2) is framed in terms of a requirement “to be the most suitable”, sub-rule (1) is framed in terms of “reasonable steps to ensure” suitability. The distinction was not addressed by the parties. The judge had, however, found in any event that the bank was negligent in recommending the EVRF (at para 94, cited above). It seems to me that, in part for this reason as well, it follows that there was probably a breach of sub-rule (1).

A further procedural breach found by the judge (see para 95 of the judgment) was as follows:

Finally, there was a further substantive breach found by the judge by reference to COB 5.4.3 R:

“COB 5.4 – Customers’ understanding of risk

…

COB 5.4.3 R (Requirement for risk warnings)

A firm must not:

The judge referred to this breach at para 103 in the context of finding that Mr Rubenstein relied on the advice he had received. The judge said:

“I also find that there was a breach by Mr Marsden of COB 5.4.3 R because he did not adequately explain to Mr Rubenstein that, in theory at least, he could get back less than the capital he put in, even if Mr Marsden was right to think that the prospect of that happening was negligible.”

Thus, in summary, HSBC was responsible, on the judge’s findings, for a variety of serious breaches of COB, both procedural and substantive. Substantively speaking, it failed (a) in the information it provided, (b) in the advice it tendered, and (c) in the recommendation it made. It was negligently responsible, in breach of statutory duty, and in contract and (subject to the judge’s findings as to the absence of foreseeable loss) in tort: for making a recommendation without taking reasonable steps to ensure that Mr Rubenstein understood the nature of the risks involved (COB 5.4.3R); even more seriously, for communicating its advice and information regarding it in a misleading way (COB 2.1.3R); and for making a recommendation which was not the most suitable from its range of packaged products. Procedurally, it failed to meet its obligations to “know your customer” (COB 5.2.5R), to retain proper records (COB 5.2.9), to provide Mr Rubenstein with a statement of his demands and needs (COB 5.2.12R), and to provide him with a suitability letter about its recommendation (COB 5.3.14). The judge may be right in thinking that the procedural breaches only become important if they help to lead to an unsuitable recommendation: nevertheless the relevant rules are designed to assist the adviser not to fail substantively, and their breach makes such substantive failure the more likely. It may not be easy to give suitable advice: but it is harder to do so if one goes about it in the wrong way.

There is no cross-appeal against the judge’s findings about the important substantive breaches of COB 2.1.3R and COB 5.4.3R. When the absence of any such cross-appeal was brought home to HSBC by means of the skeleton arguments served on behalf of Mr Rubenstein, Mr Cogley, on behalf of HSBC, but only at an advanced stage of the hearing, applied, if necessary, to amend his respondent’s notice to include a complaint about the judge’s findings of breach in respect of these two rules. Up to that point, his only complaint was about the finding of breach of COB 5.3.5R. He had submitted that all other breaches were parasitic or ancillary on that breach, so that, if he succeeded in demonstrating that the judge had been wrong to find a breach of COB 5.3.5R, all other breaches, even though made good, would be of no effect.

However, in my judgment, the application for this late amendment could not be granted without causing unfairness to Mr Rubenstein. It would widen the argument and necessitate fresh investigations of the evidence in the case for which the parties had not prepared. Moreover, it would be unacceptable to give HSBC permission to amend in circumstances where Mr Cogley candidly told the court that HSBC had deliberately chosen not to challenge the judge’s findings of breach in respect of these two rules because the bank accepted that it had been at fault. Thus Mr Cogley told us that the bank had been at fault in eliding the investment in the EVRF with a cash deposit, and in failing to take proper steps to ensure that Mr Rubenstein understood the risks of his investment. I would therefore refuse permission to amend.

Despite finding that Mr Marsden had been negligent, in breach of common law and statutory duties and in breach of contract (see at para 34 above), the judge went on to deny any relief to Mr Rubenstein other than nominal damages in contract.

This was not because Mr Rubenstein had not relied on Mr Marsden’s advice and recommendation to invest in the EVRF: he had. Thus I refer to the judge’s judgment at para 103: “he relied on Mr Marsden’s response as providing reassurance that the EVRF met that requirement [of posing the minimum possible risk to his capital]”; and at para 108: “He relied on Mr Marsden’s advice and was willing to invest his money in whatever Mr Marsden recommended. So I accept that, but for the negligent advice, Mr Rubenstein would not have invested in the EVRF”; and at para 117: “The misdescribing of the risk attaching to an investment in the EVRF as being the same as cash deposited in one of HSBC’s accounts was undoubtedly instrumental in persuading Mr Rubenstein to invest in the EVRF…Similarly the breach of COB 5.3.5(2)R by recommending the EVRF, when it was not the most suitable investment, influenced Mr Rubenstein to decide to invest in that product rather than in one with less risk”. It seems to me to follow from those findings, and from the findings that there had been both a negligent failure to explain the risks inherent in the investment and a positive misleading of Mr Rubenstein in telling him that the investment was the same as cash, that, if Mr Rubenstein had explained the matter properly and not misleadingly, Mr Rubenstein would have been most unlikely to have accepted the recommendation. As the judge said, Mr Rubenstein would otherwise have invested in something with less risk. For, as the judge also said (at para 89): “No risk to capital (which I interpret as meaning “the minimum possible risk”) was to be the sine qua non of the investment”. Therefore, but for Mr Marsden’s negligent advice and breaches of statutory duty, Mr Rubenstein would not have invested in the EVRF, but (as the judge went on to find subsequently) would have been safely invested elsewhere.

However, because the judge considered that Mr Rubenstein’s loss had been unforeseeable and too remote, and had been caused not by Mr Marsden’s negligent recommendation but by what Mr Cogley called “market hysteria”, or what the judge called the “extraordinary and unprecedented financial turmoil which surrounded the collapse of Lehman Brothers” (at para 117), Mr Rubenstein failed to achieve any remedy.

How did the judge express these conclusions? I will set out his reasoning, which is to be found in the section of his judgment headed “What is the recoverable loss?” He immediately opened this section, at para 104, as follows: “This question raises issues as to the scope of duty, causation, foreseeability and remoteness.” He then discussed SAAMCO and other cases for the purpose of dealing with Mr Cogley’s submission that Mr Marsden’s duty was like that of the valuer in SAAMCO, namely to take responsibility only for the consequences of his information being incorrect. However, the judge rejected that submission (at para 106) and placed the present case in Lord Hoffmann’s second category where the defendant provides advice, not mere information. This categorisation is no longer disputed.

It will be recalled that in SAAMCO at 214E/F Lord Hoffmann said this:

“The principle thus stated distinguishes between a duty to provide information for the purpose of enabling someone else to decide on a course of action and a duty to advise someone as to what course of action he should take. If the duty is to advise whether or not a course of action should be taken, the adviser must take reasonable care to consider all the potential consequences of that course of action. If he is negligent, he will therefore be responsible for all the foreseeable loss which is a consequence of that course of action having been taken. If his duty is only to supply information, he must take reasonable care to ensure that the information is correct and, if he is negligent, will be responsible for all the foreseeable consequences of the information being wrong.”

The judge continued:

“107.

Mr Cogley’s second submission was that the negligent advice did not cause the loss. He submitted that the loss was caused by market hysteria which followed the collapse of Lehman Brothers. Mr Egerton put it rather more broadly. He said that no loss was caused by investing in the EVRF. All of the loss had resulted from the banking crisis created by the widespread defaults in the US sub-prime mortgage market…

He then found that Mr Rubenstein would not have invested in the EVRF but for Mr Marsden’s negligent advice (at para 108, see above). On this appeal, Mr Cogley has submitted that this was a mere finding of causation in fact, absent considerations of remoteness, or causation in law. The judge made this distinction in the following passage:

“109.

In Mr Virgo’s submission this was enough to establish the chain of causation. The structure of the EVRF rendered it vulnerable to a loss of investor confidence and to a high demand for the withdrawal of funds, because the underlying assets provided insufficient liquidity. The same problem would not have occurred, or was far less likely to occur, with a true cash deposit. The higher risk inherent in the constitution of the EVRF compared with a bank deposit (or even the SVRF) was the root cause of Mr Rubenstein’s loss rather than market events themselves. I am not persuaded that this is correct. The experts were all agreed that the constitution of the EVRF was very different from a cash deposit but, save for Dr Thompson, they were also agreed that in September 2005 the risk inherent in the EVRF was only marginally or slightly higher than that of a conventional deposit. Insofar as Dr Thompson suggested that the risk was significantly higher, I think her evidence was coloured by hindsight and I do not accept it. The damage which eventuated, namely, the closure of the fund and a substantial loss of investors’ original capital, was triggered by subsequent events. If those were not events of a kind which were foreseeable when the investment was made, I do not think that it can be said that the structure of the product truly caused the loss.”

Subject to the important new factor of the stress now placed at this appeal by HSBC on the short-term period of the proposed investment, that paragraph in its way contains the essence of Mr Rubenstein’s argument and of the judge’s preference for the bank’s submissions.

The judge then went on (at para 113) to identify what had gone wrong. It was not so much the danger of a demand for withdrawals, which could be said to be foreseeable and to have been foreseen in AIG Life’s reference to the possibility of a moratorium on withdrawals. As the judge remarked, “Mr Rubenstein did not understand the moratorium to pose any risk to his capital. He thought that the moratorium was designed to give AIG Life a breathing space at times of high demand for withdrawals, so that within the period of suspension arrangements could be made for an orderly pay out to investors, but with no risk to their capital. It is fairly obvious that that was indeed the intention.” However –

“What in fact happened when the moratorium was imposed between September and December 2008 was that demand for withdrawals was so great that even a fire sale of assets over a 3 month period was going to be insufficient to repay investors without loss to their capital even greater than was achieved by repaying half of each investment and closing the fund. The secondary market for the assets in the EVRF had collapsed. Mr Rubenstein did not appreciate that this could happen: but neither did HSBC.”

Thus what had caused Mr Rubenstein to suffer a loss might be said to be the very thing which he had wished to avoid: the risk of loss to his capital. There were two risks which might be said to be potentially involved. One was the risk of default of the institution to which Mr Rubenstein entrusted his money. I shall call that the “default risk”. Mr Rubenstein was in a sense alive to that risk (but still relying on Mr Marsden to recommend a safe home for his money in that respect). The other risk, however, was a risk that arose from market movements (which I can call the “market risk”). Mr Rubenstein had no idea that he was exposed to that risk at all.

The judge then went on to put flesh on his description of what he regarded as the unforeseeable events which had caused the loss, but in doing so I consider that he confused the two risks which he had properly analysed. He said:

“115.

The suspension of the EVRF was triggered by a volume of requests for withdrawal of funds between 15 and 18 September 2008 which was greater than that which AIG Life had received in any previous 3 month period. The run on the fund was triggered by a well-founded rumour in the U.S. financial markets that AIG was going to go bankrupt. It is now known that it might well have done if it had not received support from the US Federal Reserve. Investors in the PAB, ignorant of the fact that the assets of AIG Life were held separately from those of its American parent, rushed to cash in their investment. The idea that one of the world’s largest insurance companies might go bankrupt was unthinkable in September 2005, just as it was unthinkable then that one of the UK’s major clearing banks might find itself unable to repay depositors. But that is what would have happened if the UK Treasury and the Bank of England had not stepped in to assist the Royal Bank of Scotland in the autumn of 2008. As Mr Cogley put it, the concept of a run on AIG was “so remote that no financial adviser would have been required to point it out as posing a risk to capital”. In my judgment he was right. What happened to the EVRF on 15 September 2008 and the days following was wholly outside the contemplation of the bank or any competent financial adviser in September 2005.” (My emphasis.)

Thus, in this stage of his analysis the judge is suggesting that the default risk (or rather fear of the default risk) was what caused the loss, rather than the movement of markets, which was his earlier finding (at para 113: “The secondary market for the assets in the EVRF had collapsed”). In fact, neither AIG nor AIG Life suffered any default (nor would it seem that the investors in the PAB were exposed to AIG default in any event. Their investment depended on the value of the assets within their fund). Thus, significantly, investors in the SVRF suffered no loss at all. We are not told whether there was a similar run on withdrawals from the SVRF, but I would infer that there was, albeit it appears that it remained sufficiently liquid to prevent the need for any moratorium. (Footnote: ⁷) Nevertheless, what caused the loss for investors in the EVRF was the unfavourable movement (“collapse”) of markets in the assets held by that fund. At an earlier stage of his judgment, the judge had said that the SVRF was a more suitable investment for Mr Rubenstein than the EVRF and thereby implied that Mr Marsden should have recommended either it or nothing (among packaged products). At the stage of his judgment which I am now considering, however, the SVRF has dropped out of sight.

Para 115 contained the judge’s ultimately critical reasoning, for he immediately went on to his conclusions, as follows:

So, at the end of this passage, the judge possibly glossed both or either the run on AIG and/or the collapse of markets as his “extraordinary and unprecedented financial turmoil”. In para 113 he had emphasised the collapse of markets. In para 115 he had emphasised the run on AIG. Which ultimately did he have in mind? It looks to me as though it was fear of the default risk. In fact, however, it was the collapse of markets generally, rather than fear of a default by AIG, which seems to me to have been critical.

The judge then proceeded to another section of his judgment, headed “The calculation of the loss”, in case he should be wrong about liability. He reasoned:

“124.

The measure of damages is the sum which will place Mr Rubenstein in the position he would have been in if the contract with the bank had not been broken. It follows that Mr Rubenstein’s loss should be calculated as if HSBC had succeeded in recommending the most suitable investment, using that investment as a comparator.”

Various scenarios were canvassed. One was a comparison with what would have happened if Mr Rubenstein had invested in the SVRF. That gave rise to a loss figure of £122,853.44, net of tax. Another, put forward by Dr Thompson, was based on investing 10% of the capital in the HSBC Premier account and the remaining 90%, at three months’ notice, in Investec’s High 5 account. That produced a loss figure of £132,346.15. Mr Egerton’s scenario was to place the full amount in the HSBC Premier account. That produced a loss of £92,492.42.

Mr Cogley had submitted that Mr Rubenstein was unable to prove any loss on the ground that nowhere in his evidence had he stated what he would have done with his money if he had not followed the recommendation to put it in the EVRF. The judge rejected this approach (at para 126). He said:

“It was not for Mr Rubenstein to say what the most suitable investment would have been. The tenor of his evidence was that he would have invested in whatever Mr Marsden advised him to do. The experts have now given their views as to the alternatives. That evidence is sufficient, in my judgment, to mount a claim for damages, and to invite the court to conclude what the appropriate comparator should be. However, there is one caveat. Since the particulars of claim contained the positive allegation that the most suitable investment was one or more deposit accounts, and it has never been part of Mr Rubenstein’s pleaded case that the most suitable investment was the SVRF, I do not think it would have [been] open to me to award damages on the basis of the SVRF calculation even if I had held that the SVRF was the most suitable investment rather than a cash deposit.”

In the event, the judge’s reasonable solution, which is not challenged on appeal, was to select the HSBC Premier account for 10% of the capital (on the basis that this gave instant access for that portion of the capital) and to direct a calculation based on the best (in terms of rate) of the 3 month notice deposit accounts of the UK clearing banks as of 22 September 2005.

It appears therefore that the judge might have decided, but was not called upon to do so, that the SVRF was not only a more suitable investment but the most suitable investment. However, in the end his choice was between a deposit account or accounts solution and the SVRF, and he chose the former on the ground that that had been pleaded, while the latter had not been.

It is not clear why the SVRF resurfaced as an important element in the (obiter) calculation of Mr Rubenstein’s loss, together with the ultimate solution of an investment in one or more deposit accounts, whereas these satisfactory investments were ignored for the purposes of considering whether Mr Rubenstein’s loss was unforeseeable and too remote. There is something rather counter-intuitive about on the one hand finding that there was negligence and significant breach of the COB rules in giving the wrong advice and yet concluding that any loss was too remote to be recovered in law, while on the other hand being in a position to find how much loss Mr Rubenstein had suffered by comparison with what would have happened if care had been taken to give him the right advice. It is this puzzling feature of the case that has led Mr Cogley to submit that the judge was wrong in the first place to find that there had been any mis-selling, and Mr Palmer QC to submit on behalf of Mr Rubenstein that the judge’s legal conclusions as to remoteness were mistaken.

It therefore makes sense to begin with Mr Cogley’s cross-appeal against the finding of negligence and breach of COB 5.3.5R.

Mr Cogley made no distinction between liability in negligence, contract or under the COB rule. He submitted that the essence of the matter was that the judge had been wrong to rule that Mr Marsden had failed to recommend the most suitable investment. If, as he submitted, the EVRF had been suitable and indeed the most suitable investment, then nothing else mattered, even if the bank had failed to explain matters properly or had misled Mr Rubenstein. It may have got the mechanism wrong, but it got the product right.

In developing this submission Mr Cogley highlighted three aspects of Mr Rubenstein’s “brief” to the bank which were at the core of his argument. First, Mr Rubenstein wanted a better rate of return than was available from a deposit account, having rejected the rates on offer which he had investigated; second, he only wanted a short-term investment, for at most six months to a year; and third, he was prepared to accept some risk, because the judge glossed his requirement of “no risk to capital” as “minimum risk” (in fact “the minimum possible risk”, see at para 89). In this context, Mr Cogley went on to stress the experts’ view, as found by the judge, that in September 2005, and looking forward over the period of up to a year, there was nothing to suggest any risk to Mr Rubenstein’s capital (at para 63). In effect Mr Cogley was repeating the submission he made to the judge, recorded at para 88 of his judgment and set out immediately below, save that he now stressed the period of up to 12 months as a matter of foresight rather than hindsight:

“Mr Cogley submitted that, if what was said amounted to advice, the advice was correct at the time it was given, and that, with hindsight, it held good for the period of time that it was anticipated the investment would be held.”

In my judgment, however, the judge was correct to reject that submission, and for essentially the reasons which he gave. I would seek to express the matter in my own words as follows.

First, Mr Rubenstein stressed the importance of his capital to him and his family: it was the money they had to invest in a home, for which they were looking while they rented in the meantime. Therefore, they could not afford to risk any loss of capital. In the words of his email to Mr Marsden on 24 August 2005: “We can’t afford to accept any risk in the investment of the principal sum”. In the words of the judge, this was the “sine qua non of the investment”. Secondly, while the judge chose to gloss that requirement as “the minimum possible risk”, that was merely reflective of the thought in the judge’s mind that any investment involved some at least theoretical risk about the solvency of the counter-party. Moreover, in stressing “minimum possible risk”, the judge had it in mind that the risk involved in the EVRF was measurably greater than in the SVRF or in a deposit account (see his para 91, cited at [35] above). Thirdly, Mr Rubenstein had not rejected the concept of a deposit: a fortiori, he was not a “rate chaser”, which had been Mr Marsden’s evidence about the view which he had formed of him, but evidence and a view which the judge rejected. Mr Rubenstein, in the opinion of the judge, had not rejected the idea of a deposit. Indeed, a deposit had been Mr Rubenstein’s first intuition: his approach to the bank was generated by the thought that an investment of over £1 million might give rise to a better deposit rate than the rates which were quoted for up to £1 million. The experts were in any event to give evidence, which the judge accepted, that a prudent adviser would have pointed out to Mr Rubenstein that he only needed instant access for 10% of his capital, and could allow the bulk of it to be placed on 3 months’ deposit, thereby in any event obtaining a better rate. Fourthly, and in this connection, it emerged in the evidence that Mr Marsden never gave a thought to giving any other advice than to invest in the EVRF. And why, it might be asked, should he, if he regarded Mr Rubenstein as a rate chaser? Why not go for the maximum rate obtainable in an investment which even Mr Marsden regarded as a cash deposit?

Fifthly, Mr Marsden did not explain to Mr Rubenstein, and did not even understand for himself, that the structure of the EVRF was not akin to a cash deposit. There was no obligation on the recipient of the investment to return the capital received, but only its current value in terms of the assets invested in by the fund. Therefore, the investor was taking the risk of market movements. However, Mr Marsden’s advice, in support of his recommendation, was that the investment in the EVRF was “the same as cash deposited in one of our accounts” (his email of 24 August 2005). That was wrong and misleading advice. Mr Marsden went on to address the “fall as well as rise” warning as dealing only with the risk of default, not with the risk of market movement. That was also wrong and misleading, but was of a piece with the previous error.

It is not clear on the judge’s findings what Mr Rubenstein’s reaction to being properly advised would have been, but the (rebuttable but not rebutted) presumption must be that he would not have proceeded. Indeed, if Mr Marsden had properly understood what was involved in the EVRF, properly understood that he was advising Mr Rubenstein and not merely acting as an execution only conduit, properly understood that the COB rules applied to the transaction and that it was his duty to know his customer, to record the relevant information and to record his advice in a suitability letter to his client, it seems unlikely that he would have recommended the EVRF to Mr Rubenstein. He in fact recommended it because he regarded the EVRF as the same as cash in a deposit account. It is true that he gave evidence that, even if he had known everything then that he had come to know because of the litigation, he would still have recommended the EVRF. However, the judge said nothing to suggest he accepted that evidence, and he did not accept other aspects of Mr Marsden’s evidence. It seems unlikely to me.

Sixthly, however, it is suggested that Mr Marsden’s recommendation would have been the same, and would have properly and prudently been the same, because the investment world at that time was regarded as so benign. It is submitted that there was in fact no risk, at any rate over the time-scale of one year, as hindsight showed. In the circumstances prevailing, it is said that hindsight is the proof of foresight. It is said that Mr Marsden owed a duty to recommend the EVRF, because otherwise Mr Rubenstein would have been deprived of one of the benefits which he sought, which was a higher rate of interest. It is said that in an exercise of professional judgment, Mr Marsden should be given the benefit of any doubt, or afforded the margin available to those called upon to exercise their judgment in such circumstances.

However, although the experts may have agreed that the outlook was fair and seemed without risk, at any rate over the short-term, Mr Rubenstein was not looking for a product which subjected him to market risk which could be discounted as a matter of judgment on a short-term basis (which might be said, more or less, of many investments, such as gilt edged securities): he was looking for an investment without risk. But Mr Marsden was not listening to his customer, or at any rate not doing so with care: because he did not recognise him as a customer whom he was advising and had to know, because he regarded him as a rate chaser, and because he did not himself understand the product he was recommending. In such circumstances, it does not seem to me acceptable to accord Mr Marsden the benefit of a professional margin. The question is not one of risk in the abstract, but of suitability for the customer. The judge was right to conclude that “the structure of the investment was materially different [from cash deposited] and the risk was commensurately greater” (at para 91); and that Mr Marsden was negligently wrong to suggest that the EVRF was the same as a cash deposit and to make no attempt to consider the other funds in the PAB as an alternative (at para 94): to which I would add a particular reference to the SVRF and to straightforward cash deposits. It might be possible to say that both a deposit with HSBC and an investment in the EVRF were without risk (either in the abstract or compared with other possible investments), but the question is whether the deposit and the investment were equally suitable for Mr Rubenstein. The very fact that the EVRF yielded materially more than the SVRF or a cash deposit demonstrates how the market priced the difference in risk. That is the outstanding objective evidence of the situation at the time.

Seventhly, I come back to Mr Cogley’s leitmotiv of the short-term indication of Mr Rubenstein’s investment. This is more appropriately considered under the subsequent heading of scope of duty and remoteness, where I will revisit it. However, in the present context, it is sufficient to say (a) that although Mr Rubenstein suggested that a new home would probably be found within the year, the indication given was nevertheless ultimately an indefinite one, namely until a new home was bought; and (b) that it was one of the vices of Mr Marsden’s advice that, because he represented the investment as equivalent to a cash deposit, he said that “it is unlikely that you will need further advice” (see his email of 24 August 2005 at [14] above). For both those reasons, Mr Marsden could not afford to render his advice on the basis that the current outlook for market conditions looked set fair. He ought to have realised, therefore, that once Mr Rubenstein made his investment, he intended to stick with it until he found a new home, and that could be for an uncertain period. Of course, Mr Marsden made no such calculations at all, nor would one have expected him to do, where he regarded the investment as equivalent to a cash deposit.

Eighthly, I return to the important matter of the SVRF. Whether or not that was the most suitable investment, it was clearly a more suitable investment for Mr Rubenstein (perhaps not for a “rate chaser”), for, although its structure was still dependent, at any rate in part, on market valuations of the assets within it, those assets seem to have been made up in very large part by deposits with major banks. It was that cash liquidity which meant that it survived unscathed, while the EVRF failed. However, Mr Marsden did not even consider it. He ought to have done, and, if he had been listening to his customer and recognised that he was advising him, it is difficult to understand how he would not, acting properly, have recommended that fund to him in preference to the EVRF, even if its rate was less than the EVRF. To recommend the EVRF in preference to the SVRF was deliberately to make the decision for Mr Rubenstein that he was prepared to accept more risk than in the SVRF. At any rate, he would have had to have had a proper discussion with Mr Rubenstein about the differences between the two funds.

Finally, it is difficult to understand how an adviser who did not realise that he was an adviser, and failed to prepare himself for his advice in the way that COB required him to do, and failed to understand what he was advising about, and misled his customer in the advice he gave, would serendipitously have chosen for him the most suitable of products. I agree with the judge that Mr Marsden was negligent and in breach of substantive COB rules in advising Mr Rubenstein.

Mr Cogley again repeated the submissions which, in this respect, had succeeded before the judge. He emphasised the expected timescale for the investment of up to one year, and submitted therefore that the events of September 2008 were not only unforeseeable (or even “unthinkable”), and outside the reasonable contemplation of the parties, but that it was entirely illegitimate to take into account anything that happened outside that timescale. As to foreseeability, he relied on the judge’s findings, which I have cited and analysed above.

On the other hand, Mr Palmer, on behalf of Mr Rubenstein submitted that market movements are eminently foreseeable, and are inherent in the concept of markets. Sometimes such movements are unpredictable and even extreme. That is merely a difference in the extent of a kind or type of loss which is foreseeable, and will not prevent the consequential loss from being recoverable; nor will such a loss be regarded as too remote. Whether in terms of the test in tort, of foreseeability, or in terms of the test in contract, of the reasonable contemplation of the parties, the loss of capital by market movements was in law to be regarded as caused by Mr Marsden’s advice and recommendation, as it was in fact, and was within the scope of HSBC’s duty and responsibility. This was particularly so in the context of investment advice by a bank to a “private person” pursuant to the statutory duties set up by the COB rules. As for the timescale of Mr Rubenstein’s investment, it may have been short-term, but it was essentially indefinite, for it depended on the Rubinsteins’ ability to find a new home. If Mr Marsden had wanted to put a time cap on his responsibility, he should have made that clear, especially in the context of telling Mr Rubenstein that his investment was the same as a cash deposit and that he would not need further advice.

A relatively small number of authorities have been cited to the court to establish and illustrate the applicable principles.

In Hughes v. Lord Advocate [1963] AC 837 at 845 Lord Reid said:

“No doubt it was not to be expected that the injuries would be as serious as those which the appellant in fact sustained. But a defender is liable, although the damage may be a good deal greater in extent than was foreseeable. He can only escape liability if the damage can be regarded as differing in kind from what was foreseeable.”

In that case, a known source of danger, a paraffin lamp, acted in an unpredictable way, by causing an explosion rather than a fire, and the boy plaintiff was badly injured. The House of Lords took the doctrine, that an unforeseeable extent of injury of a foreseeable type of injury did not render the injury too remote, and applied it in the same way to an unforeseeable mechanism of injury, provided the source of injury was foreseeable. As Lord Reid said at 847, contrasting a case where “the accident may have been due to the intrusion of some new and unforeseeable cause like the falling of a ceiling”, nevertheless –

“This accident was caused by a known source of danger, but caused in a way which could not be foreseen, and, in my judgment, that affords no defence.”

The Heron II [1969] 1 AC 350 is a case concerning damages in contract, for loss of market by reason of delay in the carriage of goods. It is of course a leading authority, but it is also a case between merchants, rather than between a bank and its private customer, a consumer, to whom the bank owed statutory duties. In it, in a passage of which only the first part was quoted by the judge, Lord Reid said at 385-386:

“I am satisfied that the court did not intend that every type of damage which was reasonably foreseeable by the parties when the contract was made should either be considered as arising naturally, i.e., in the normal course of things, or be supposed to have been in the contemplation of the parties. Indeed the decision makes it clear that a type of damage which was plainly foreseeable as a real possibility but which would only occur in a small minority of cases cannot be regarded as arising in the usual course of things or be supposed to have been in the contemplation of the parties: the parties are not supposed to contemplate as grounds for the recovery of damage any type of loss or damage which on the knowledge available to the defendant would appear to him as only likely to occur in a small minority of cases.

In cases like Hadley v. Baxendale [(1854) 9 Exch 341] or the present case it is not enough that in fact the plaintiff’s loss was directly caused by the defendant’s breach of contract. It clearly was so caused in both. The crucial question is whether, on the information available to the defendant when the contract was made, he should, or the reasonable man in his position would, have realised that such loss was sufficiently likely to result from the breach of contract to make it proper to hold that the loss flowed naturally from the breach or that loss of that kind would have been in his contemplation.

The modern rule of tort is quite different and it imposes a much wider liability. The defendant is liable for any type of damage which is reasonably foreseeable as liable to happen even in the most unusual case, unless the risk is so small that a reasonable man would in the whole circumstances feel justified in neglecting it. And there is good reason for the difference. In contract, if one party wishes to protect himself against a risk which to the other party would appear unusual, he can direct the other party’s attention to it before the contract is made…But in tort there is no opportunity for the injured party to protect himself in that way…”

The modern law of damages in contract has been revisited in The Achilleas [2009] 1 AC 61. That was also a dispute between merchants. The question was whether a charterer who redelivered the chartered vessel to her owner late, in breach of contract, should be liable for only a conventional rate of damages representing any increase in the market rate for the vessel over and above the charter rate for the period of the overrun, or should be liable for the full extent of the owner’s undoubted loss where, due to the late redelivery, the owner had lost a new fixture at the higher market rate. It was held that the owner was limited to the former. Lord Hoffmann reasoned the relevant principles as follows:

“21 It is generally accepted that a contracting party will be liable for damages for losses which are unforeseeably large, if loss of that type or kind fell within one or other of the rules in Hadley v Baxendale: see, for example, Staughton J in Transworld Oil Ltd v North Bay Shipping Corpn (The Rio Claro) [1987] 2 Lloyd’s Rep 173, 175 and Jackson v Royal Bank of Scotland plc [2005] 1 WLR 377. That is generally an inclusive principle: if losses of that type are foreseeable, damages will include compensation for those losses, however large. But the South Australia and Mulvenna cases show that it may also be an exclusive principle and that a party may not be liable for foreseeable losses because they are not of a type or kind for which he can be treated as having assumed responsibility.

22 What is the basis for deciding whether loss is of the same type or a different type? It is not a question of Platonist metaphysics. The distinction must rest upon some principle of the law of contract. In my opinion, the only rational basis for the distinction is that it reflects what would reasonably have been regarded by the contracting party as significant for the purposes of the risk he was undertaking. In Victoria Laundry (Windsor) Ltd v Newman Industries Ltd [1949] 2 KB 528, where the plaintiffs claimed for the loss of the profits from their laundry business because of late delivery of a boiler, the Court of Appeal did not regard “loss of profits from the laundry business” as a single type of loss. They distinguished, at p 543, losses from the “particularly lucrative dyeing contracts” as a different type of loss which would only be recoverable if the defendant had sufficient knowledge of them to make it reasonable to attribute to him acceptance of liability for such losses. The vendor of the boilers would have regarded the profits on these contracts as a different and higher form of risk than the general risk of loss of profits by the laundry.

23 If, therefore, one considers what these parties, contracting against the background of market expectations found by the arbitrators, would reasonably have considered the extent of the liability they were undertaking, I think it is clear that they would have considered losses arising from the loss of the following fixture a type or kind of loss for which the charterer was not assuming liability. Such a risk would be completely unquantifiable, because although the parties would regard it as likely that the owners would at some time during the currency of the charter enter into a forward fixture, they would have no idea when that would be done or what its length or other terms would be.”

Lord Hoffmann also said this about questions of remoteness:

“25 The owners submit that the question of whether the damage is too remote is a question of fact on which the arbitrators have found in their favour. It is true that the question of whether the damage was foreseeable is a question of fact: see Monarch Steamship Co Ltd v Karshamns Oliefabriker (A/B) [1949] AC 196. But the question of whether a given type of loss is one for which a party assumed contractual responsibility involves the interpretation of the contract as a whole against its commercial background, and this, like all questions of interpretation, is a question of law.”

These, then, together with the well-known authority of SAAMCO [1997] AC 191 on the importance of scope of duty in contract and tort, and the significance of statutory duties in this regard (see at [45] above), are the general principles which must be applied. Lord Hoffmann there said (at 212E):

“The scope of the duty, in the sense of the consequences for which the valuer is responsible, is that which the law regards as best giving effect to the express obligations assumed by the valuer: neither cutting them down so that the lender obtains less than he was reasonably entitled to expect, nor extending them so as to impose on the valuer a liability greater than he could reasonably have thought he was undertaking.”

Much reference was made by Mr Cogley to Lord Hoffmann’s example of the mountaineer in his speech in SAAMCO (at 213). The importance of that example is that it illustrates the significance of the scope of a defendant’s duty for the purpose of questions of causation and remoteness. A mountaineer is told by his doctor that his knee is fit for a mountain climb, but the doctor is negligent, ie the knee is not fit. If the mountaineer had been told that his knee was not fit, he would not have gone climbing. On the climb the mountaineer suffers an accident which, however, had nothing to do with the knee. Is the doctor liable for the consequences of the mountaineer’s injury? No, suggested Lord Hofmann, even though the injury would not have happened but for what Lord Hoffmann called both “information” and “advice”. The reason is: “The injury has not been caused by the doctor’s bad advice because it would have occurred even if the advice had been correct” (at 213F). The importance of the example, as it seems to me, is that it illustrates the manner in which we think naturally of causation and responsibility. Ex hypothesi, the injury was caused by something else entirely, for which the doctor had no responsibility. Although the mountaineer would not have gone on the mountain unless he had been given the all clear from the doctor, we would not select the doctor’s negligent advice as the cause of the mountaineer’s injury unless the injury had been contributed to in some material way by the unfitness of the knee. This is despite the fact that an accident on the mountains – whether it is due to something entirely fortuitous such as an avalanche, or is the result of some faulty piece of equipment, or of the weather, always unpredictable but inherently so – is always, more or less but readily, foreseeable. However, the doctor is responsible for the mountaineer’s knee, but not for the weather, the equipment, or sheer bad luck.

But what does the mountaineer’s example teach us in the present case? An investment adviser, with his statutory duties of various kinds, owed to a consumer as a result of the latter’s statutory status as a private person, who as adviser recommends a particular investment, which he must take care to be suitable for his client and, if a packaged investment, to be the “most suitable” on the adviser’s menu, may well be responsible if some flaw in the investment turns out materially to contribute to some investment loss. The doctor did not advise, let alone recommend, his patient to go mountaineering: he merely told him that his knee was in good shape. Mr Marsden, however, not only advised Mr Rubenstein on the investment of his capital, he recommended a particular investment. He, so to speak, put him in it. If such an investment goes wrong, there will nearly always be other causes (bad management, bad markets, fraud, political change etc): but it will be an exercise in legal judgment to decide whether some change in markets is so extraneous to the validity of the investment advice as to absolve the adviser for failing to carry out his duty or duties on the basis that the result was not within the scope of those duties.

Ultimately, Mr Cogley’s reliance on the mountaineer was by reference to his leitmotiv about the timescale of the investment. A doctor who advises his patient about tomorrow’s climb, he submits, is taking no responsibility, for next year’s climbing season. Two things may be said about that, however. First, the submission assumes that the adviser is at least prima facie responsible in the first place, ie what has gone wrong is connected with the knee’s lack of fitness. Secondly, if one assumes that the advice may be consistent with a rather static medical condition, it does not follow that the doctor’s responsibility ceases with the next day’s climb. That may depend.

Pursuant to these principles set out in decisions of the highest authority, a number of cases have been cited to us as illustrations to guide the decision in this appeal. I would refer to the following.

In Bates v. Barrow Ltd [1995] 1 Lloyd’s Rep 680 the plaintiff insured sued their insurance brokers for negligence in the placing of their insurance. Their insurer, Kansa, had denied liability on the basis of its own unlawful conduct in carrying out insurance business without being authorised in the UK. Kansa was an entirely respectable and solvent firm. Gatehouse J held that the brokers were in breach of duty to their insured in placing the insurance with an unauthorised company. He also held that the insured’s loss flowed directly from and was caused by the brokers’ breach. However, he held that the loss was too remote (see question 2(c)(ii) at 684 lhc). He directed himself by reference to The Heron II and the question, which he recorded as “most difficult to assess”, whether “the ordinary competent broker would have regarded the danger of the [illegality] point being taken as a serious possibility” (at 691 rhc). He concluded that at the time in question and in the light of the ethos of the insurance market at that period, and in the case of a respectable and solvent company, “it was virtually unthinkable that an insurer or reinsurer would take the point”. Mr Cogley relied on that authority. In sum, the brokers were taken to be aware of the risk of illegality, and their breach of duty in not guarding against it caused the insured’s loss, but nevertheless the loss was too remote and irrecoverable. I am not much assisted by that decision. It seems to have turned on the view that Kansa’s decision to rely on its own illegality was so unethical, ie against the ethos of the market, as to amount in effect to an extraneous cause.

In Brown v. KMR Services Ltd [1995] 2 Lloyd’s Rep 513 the court of appeal considered a claim by a Lloyd’s name against his members’ agents for exposing him to membership in syndicates which reinsured catastrophe excess of loss without warning him of the high risk nature of his participation. At trial, the judge, Gatehouse J, found that no one had anticipated the size and frequency of the various disasters that occurred between 1987 and 1990 (at 515 rhc). Nevertheless, he found that the plaintiff could recover for breach of the agents’ duty on the basis that if he had been warned he would have limited his exposure. On appeal, this court upheld his judgment in this respect (although Stuart-Smith LJ dissented, he joined with the majority on this point). Stuart-Smith LJ said (at 542 rhc): “the fact that the scale or amount of the losses were not foreseeable does not make them too remote”. Hobhouse LJ observed (at 557 lhc): “If it was the duty of the defendants to protect the plaintiff from losses of the kind which he subsequently suffers, how can it be just or appropriate to say that, because those losses are larger than either party anticipated, the plaintiff must bear those losses not the defendants?” Mr Palmer relied on this authority.

In Supershield Ltd v. Siemens Building Technologies FE Ltd [2010] EWCA 7, [2010] 1 Lloyd’s Rep 349 the dispute was between a subcontractor (Supershield), who had installed a sprinkler system which malfunctioned causing a flood, and the contractor which had employed it (Siemens). The building occupiers settled with Siemens, but its Part 20 claim went to court. At trial, Supershield was held liable. On appeal, Supershield debated inter alia whether Siemens’ settlement with the occupiers had been reasonable, despite strong arguments available to Siemens on issues of causation and remoteness. That was because a protection system for which Supershield was not responsible (among others for which it was) had not functioned as it should. That led Toulson LJ to comment (at [40]):

“If those responsible [for multiple protection systems] fail to do [as they ought] and the unlikely happens, it should be no answer for one of them to say that the occurrence was unlikely, when it was that party’s responsibility to see that it did not occur.”

In the course of reaching that conclusion, Toulson LJ considered the recent decisions of the House of Lords in SAAMCO and The Achilleas. He commented:

“43.

Hadley v Baxendale remains a standard rule but it has been rationalised on the basis that reflects the expectation to be imputed to the parties in the ordinary case, ie that a contract breaker should ordinarily be liable to the other party for damage resulting from the breach if, but only if, at the time of making the contract a reasonable person in his shoes would have had damage of that kind in mind as not unlikely to result from a breach. However, South Australia and Transfield Shipping are authority that there may be cases where the court, on examining the contract and the commercial background, decides that the standard approach would not reflect the expectation or intention reasonably to be imputed to the parties. In those two instances the effect was exclusionary; the contract breaker was held not to be liable for loss which resulted from its breach although some loss of the kind was not unlikely. But logically the same principle may have an inclusionary effect. If, on the proper analysis of the contract against its commercial background, the loss was within the scope of the duty, it cannot be regarded as too remote, even if it would not have occurred in ordinary circumstances.”

In Camerata Property Inc v. Crédit Suise Securities (Europe) Ltd (No 2) [2012] EWHC 7 (Comm), [2012] PNLR 15 the claimant, an investment vehicle of a financially astute high net worth individual, was advised in 2007 by the defendant to invest in a note, due to mature in 2009, issued by a Lehman Brothers subsidiary. The note was a bet on the dollar weakening against the euro. If Lehman Brothers had survived, the note would have paid off handsomely in 2009: as it was, the claimant lost everything because of the issuer’s default. The claimant alleged that in 2008 the defendant should have advised it to sell the note because of the doubtful solvency of Lehman Brothers. In the first trial, Andrew Smith J held that there had been no negligence, but that in any event, even if advised to sell, the claimant would have retained the note. In a second action the claimant now claimed that it had been negligently advised to buy the note in the first place, also saying that there had been a breach of section 150 of the FSMA, inter alia because of the recommendation of an unsuitable investment. The claim was struck out on the basis that the findings at the first trial made the claim impossible (and it was also said that the second claim was an impermissible collateral attack on the first judgment). Flaux J went on, however, in a collateral aside, to opine that the claim would have been bound to fail for another reason, namely that even if the defendant had been at fault, the 2008 collapse of Lehman Brothers had been unforeseeable in 2007. The submission is set out at [68] as follows:

“even if Camerata could establish its general wrong advice case and even if it could show that it would not have invested in the Note had it been given the right advice, the claim for damages would still fail because the actual cause of the loss was issuer default as a consequence of the collapse of Lehman Brothers, which was wholly unexpected and unforeseeable.”

As to this submission, Flaux J said (at [100]) that it was unnecessary to decide it, but he would deal with it briefly. He did so as follows (at [102]):

“It seems to me that, in principle, this argument is correct. Were it not for the unforeseeable bankruptcy of Lehman Brothers, Credit Suisse would have had a complete answer to Camerata’s case that, but for the wrong advice, it would not have entered into the transaction. That answer would have been that the transaction had in fact been profitable, so that any negligence or breach of contract had not in fact caused any loss. In other words, the only reason why Camerata has suffered any loss at all, as opposed to making a substantial profit, is because of the collapse of Lehman Brothers, which was unforeseeable.”

Mr Cogley submits that these brief obiter remarks are directly applicable to the present case: but I doubt that they are. If I understand this decision correctly, in its second claim Camerata was thrust back (unsuccessful as it was) on complaining about the investment in the Lehman Brothers note on grounds entirely distinct from the creditworthiness of the issuer and connected rather with the suitability of the investment in terms of its logic and risk. However, there was nothing wrong with the way in which the investment turned out in those respects – indeed the investment was brilliantly successful. What went wrong was something wholly different, namely Lehman Brothers’ issuer default. In the present case, however, there was no similar issuer default by AIG Life: what went wrong was the investment itself, and for the very reason that it was structured in a way which exposed Mr Rubenstein to the very risk (of loss to his capital by reason of market movements) which he had wanted to avoid by investing, as he was led to believe he had, in a safe investment equivalent to a cash deposit. Contrary to Mr Cogley’s submission, I am not assisted by Camerata.

In my judgment, these facts and the principles and authorities discussed above are ultimately to be analysed in favour of Mr Rubenstein’s appeal, and the only matter which has caused me any anxiety in that conclusion has been the timescale point which for understandable reasons Mr Cogley has now put in the forefront of his response.

In his careful judgment, to which I would otherwise pay tribute, the judge considered matters first under the banner of liability in negligence and contract, and only subsequently did he turn to questions of statutory duty. This may have reflected the way in which submissions were deployed before him: but I do not think that that is an entirely satisfactory way to the correct result especially so far as issues of remoteness are concerned. As Lord Hoffmann pointed out in SAAMCO in the passage cited above at [45], in a case of statutory duty the question as to scope of duty is to be answered by reference to the statute itself, and in such a context the position in negligence and contract will fall in behind the statutorily discerned purpose. If, however, the position in tort or contract, absent the context of statutory duty, might lead to a separate result, as it might, there seems to me to be no profit in considering that position first in a case where breach of statutory duty has been established. To do so increases the risk of error.

In the present case, therefore, it seems reasonably clear that the statutory purpose of the COB regime pursuant to FSMA is to afford a measure of carefully balanced consumer protection to the “private person”. That purpose is elucidated not only by the content of the COB rules themselves, but also by section 2 of FSMA, which speaks of “the protection of consumers”, ie “securing the appropriate degree of protection for consumers” (section 2(2)(c) and section 5(1)) as among the regulatory objectives. The rules to be created by the regulatory authority are to be informed by a proper regard for “the differing degrees of risk involved in different kinds of investment…the need that consumers may have for advice and accurate information…the general principle that consumers should take responsibility for their decisions” (see section 5(2)). In the present case it is not suggested on this appeal (although it was at trial) that Mr Rubenstein is seeking to avoid responsibility for his decisions. These basic principles and purposes are reflected in the imposition under the COB rules of onerous duties (albeit in a well conducted operation these should not be difficult to achieve and they are couched for the most part in terms of “reasonable care”) designed to ensure that the investment adviser understands his client and his client understands risk. Of course, much investment business is conducted with investors who are familiar, even expert, in investment markets. But in the present context of Mr Rubenstein and HSBC we are dealing with a consumer on the one side and an expert on the other. Unfortunately, the judge’s findings establish that Mr Marsden understood neither the client he was advising, nor the product he was recommending. He did not even understand that he was advising, as distinct from merely executing his client’s instructions. He failed therefore to undertake the standard statutory procedures designed to assist the parties to a satisfactory transaction. He misled his client, by omission and commission, into thinking that he had invested in something which was the same as cash. This is not, to my mind, a promising context in which to find that a loss suffered as a result of following a recommendation to enter into an unsuitable investment, when that loss came about because of the very factor which made the investment unsuitable (namely its inherent susceptibility to risk from market movements) was too remote to be recovered from the defaulting advising bank. For reasons discussed above, this is wholly unlike the case of the mountaineer’s knee.

Three arguments are raised by HSBC as to why nevertheless the judge was right to conclude that the loss was too remote.

The first is that, as found by the judge, the loss was ultimately caused by the “extraordinary and unprecedented financial turmoil which surrounded the collapse of Lehman Brothers” (at para 117). To the extent that the judge was saying that this was unforeseeable (see his para 116), what was unforeseeable about it? Was it the insolvency of Lehman Brothers? Was it the run on AIG’s PAB funds which accompanied that insolvency (fear of the default risk)? Or was it the collapse of market values of the securities in which the EVRF, but not, it must be remembered, the SVRF, was invested (the market risk)? The insolvency of Lehman Brothers may have been unforeseeable, but Mr Rubenstein was not invested in Lehman Brothers. (If he had been, because his adviser had recommended him to be so invested, but unsuitably so, would the unforeseeability of Lehman Brothers’ collapse have saved the negligent adviser? It is unnecessary to consider that case, which may well depend on the particular features of the breach of duty.) The extent of the run on AIG may have been unforeseeable, but it was not the run which ultimately caused Mr Rubenstein’s loss: it has to be restated that the SVRF survived any apprehension about AIG’s solvency. In any event, a run on AIG Life was both foreseeable and foreseen, for its brochure discussed the need, in such circumstances of high demand for withdrawals, for a three months’ moratorium. Ultimately, however, it was the collapse in the value of the market securities in which the EVRF (but not the SVRF) was invested which in my judgment, on the findings of the judge as I would analyse them and on my understanding of the situation, caused the loss. If it were otherwise, the SVRF investors would have been caused a similar loss, but they were not. But such a loss was both foreseeable and foreseen, because AIG Life’s brochure referred, albeit rather obliquely, to the “costs” which might accompany “selling assets prior to their intended maturity date” (see at [30] above). Those losses or “costs” may have been unforeseeably high, but that is the nature of markets at a time of stress, and in any event that merely represents an unforeseeable extent of loss of a kind or type which is foreseeable (see Brown v. KMR Services). And in truth, although the Lehman Brothers collapse was both a symptom and a contributory cause of market turmoil, the underlying causes of that turmoil went infinitely beyond Lehman Brothers’ difficulties. It stretched to a failure of confidence in marketable securities in which there had previously been greater confidence. And what is new about that?

It seems to me that in the relevant passages of his judgment, the judge was implicitly selecting, for the purpose of giving effect to the law on remoteness, one out of a number of possible causal factors as the essential cause of Mr Rubenstein’s loss. This, as Lord Hoffmann remarked in The Achilleas, is ultimately an exercise in legal assessment: this court will pay regard and respect to the judge’s choice, but it is not a factual finding such as should deter an appeal court from acting upon its own understanding and analysis of the primary findings of the trial court. In my judgment, the judge was ultimately selecting as the cause of loss an “unthinkable” run on AIG (see his [115]). However, to my mind this was not the right selection. Against the background of the facts found and of the origin of the transaction, and the scope of HSBC’s duties, what connected the erroneous advice and the loss was the combination of putting Mr Rubenstein into a fund which was subject to market losses while at the same time misleading him by telling him that his investment was the same as a cash deposit, when it was not. Therefore, the correct selection of the cause of Mr Rubenstein’s loss was the loss in value of the assets in which the EVRF (but not the SVRF) was invested. Therefore, unlike the case of the mountaineer’s knee, advice and the loss were not disconnected by an unforeseeable event beyond the scope of the bank’s duty. It was the bank’s duty to protect Mr Rubenstein from exposure to market forces when he made clear that he wanted an investment which was without any risk (and when the bank told him that his investment was the same as a cash deposit). It is wrong in such a context to say that when the risk from exposure to market forces arises, the bank is free of responsibility because the incidence of market loss was unexpected.

The second matter to which Mr Cogley has drawn attention has been the judge’s finding (among others) that at the time of investment in September 2005, the EVRF would have been regarded as without risk. So it might, but then nearly all the greatest losses come out of a cloudless sky. In some circumstances, which were not those which obtained between Mr Rubenstein and the bank, that might have closed the door on any recovery by the investor. However, as discussed above when I was setting out the judge’s findings, that did not prevent HSBC from being in breach of its statutory duties in recommending an investment in the EVRF, when a proper understanding of Mr Rubenstein’s desires and needs would have led to a recommendation of either the SVRF or a combination of bank deposit accounts. If that was not already obvious when Mr Rubenstein made it clear that he could afford no risk to his capital, it would have become still more obvious had Mr Marsden performed his statutory duties pursuant to COB. In either event, that is to say, whether Mr Rubenstein had invested in the SVRF or in a number of bank deposit accounts, he would have suffered no loss: not merely because he would not have followed the wrong advice (and thus avoided the EVRF), but because it was the bank’s duty to take care to guide him to the right advice which reflected a suitable response to his needs.

It is the third matter on which Mr Cogley has relied (in fact his primary submission) which has all along given me greatest reason to consider that the judge might ultimately have come to the right answer, albeit on a somewhat different analysis. That is Mr Cogley’s leitmotiv concerning the short-term nature of Mr Rubenstein’s investment. The submission is that a loss two years outside of the period about which Mr Rubenstein spoke is simply beyond the scope of the bank’s duties of care and foresight. That is a powerful submission. On balance, however, I have not been persuaded by it.

Thus, in one sense the time for investment was undefined and uncertain: it was until the Rubinsteins had bought a new home. That was thought to be likely to occur within a year, but the possibility obviously existed that that timescale would be exceeded. The achievement did not lie within their hands, as the example of their failed attempt to purchase in 2007 demonstrated. Moreover, Mr Marsden had himself told Mr Rubenstein that no further advice would be needed, since “once the account is open it is effectively an instant access account”; also that “We view this investment as the same as cash deposited in one of our accounts”. With instant access to a cash deposit account, why should Mr Rubenstein be concerned about changing financial weather, and why should HSBC be free of responsibility once a year was up? Moreover, it follows from the fact that Mr Rubenstein was misled as to the nature of his investment, that he did not understand that he was exposed to a risk that he did not want. That risk, of market movement in the value of an investment of a type Mr Rubenstein did not even realise he was committed to, was exactly the risk which caused his loss. Finally, the whole purpose of COB was to protect the consumer from a failure to understand risk. If Mr Marsden had done his duty, for instance by warning Mr Rubenstein that, because his investment was not like cash, its safety depended on the financial weather, then Mr Rubenstein would have either been on the qui vive for more advice, which he had been told he would not need, or, as was still more likely, he would have queried the investment, and that would have led to reformulated advice, or he would not have proceeded with the recommended investment.

The question remains: if the scope of the bank’s duty is not set by Mr Rubenstein’s own timescale of up to one year, then what is it set by? Three years, ten years, twenty years? It is a good question, with some reminiscence of a similar question posed by Lord Hoffmann with respect to the length of the follow-on fixture in The Achilleas. Nevertheless, I consider that the question is answered by the factors mentioned above: and in any event, a period of three years is, in terms of a “cash” deposit, not significantly different from an indefinite period of about a year.

Ultimately, the question of remoteness (at any rate in a contractual setting, which Lord Reid in The Heron II suggested was the more restricted one, because a claimant could stipulate contractually for his own protection) is a matter of the reasonable contemplation of the parties. In the context of statutory protection for the consumer, it seems to me that a bank must reasonably contemplate that, if it misleads its client as to the nature of its recommended investment, and thereby puts its client into an investment which is unsuitable for him, when it could just as easily have recommended something more suitable which would have avoided the loss in question, then it may well be liable for that loss. Lord Reid contemplated, but he was thinking in the context of merchants, that a claimant could stipulate for his own protection. However, what may be true of merchants is not likely to be true of consumers. In effect the obligation of explaining matters properly to its clients is put by statute on the advising expert. In such circumstances, if HSBC is to be protected by some relevant, albeit indefinite, time limit for its advice, then perhaps the obligation of making that limitation clear rests on the recommending expert, not on the misled consumer.

Where the obligation of a defendant is not merely to avoid injuring his claimant but to protect him from the very kind of misfortune which has come about, it is not helpful to make fine distinctions between foreseeable events which are unusual, most unusual, or of negligible account (cf Lord Reid in The Heron II). Whether the test of remoteness is expressed in the classic terms found in the leading authorities, or has to reflect that sense of balance (an exercise in judgment) to which Lord Hoffmann referred in SAAMCO at 212E (see [101] above), or has to take account of the manner in which the scope of duty may extend responsibility for even unusual events (see Supershield, cited at [108]-[109] above), in my judgment it should not be said that the loss which Mr Rubenstein has suffered by reason of HSBC’s breach is to be regarded as too remote.

For all these reasons, I consider that the judge came to the wrong conclusion on questions of remoteness, and I would allow the appeal on this issue.

Although I have reached these conclusions without taking into account two decisions of the FOS (Financial Ombudsman Service) and the FSA (Financial Services Authority) respectively, which have been relied on by Mr Palmer not as authorities, but as representative of his submissions, I mention them here as supportive of the judge’s and my views on the current subject.

In February 2012, that is to say after the judge’s decision had been published, the FOS had to consider the case of Mr and Mrs V (the FOS decisions are anonymised), who had invested in the EVRF on the advice of their bank. The investment was the major part of the proceeds of sale of a business. The argument between the customers and their bank was very similar to that in this case, save that the investment was in early 2008, and thus closer to the debacle of September 2008, and new COBS rules had replaced the earlier COB rules. Therefore, it is necessary to be cautious. Nevertheless, having found breach of statutory duty, the principal ombudsman had to ask himself whether it was “fair to require Bank D to pay all of the losses incurred?” He was pressed with the decision of the judge in this case and with the submission that the losses were unforeseeable, but he commented that investment is inherently unforeseeable and that extreme market conditions, including the irrational behaviour of other investors, are an established risk of all investments. He concluded that Bank D, having exposed the complainants to the risk of capital loss which it should not have done, the complainants were in an investment that they should not have been in and were entitled to compensation for their losses.

In November 2011 the FSA published a Final Notice to which the FOS referred in the above decision regarding Mr and Mrs V. This Final Notice criticised Coutts & Co for its failure to comply with COB rules in relation to recommendations to its clients to invest in the EVRF. The period in question was December 2003 to September 2008. Apparently Coutts & Co had also referred to the EVRF as a “cash product”. The FSA observed:

“In order to generate an enhanced return, the Fund exposed customers to greater level of capital and liquidity risk than that typically associated with a traditional bank or building society account.”

It is of some comfort to observe that the FOS and FSA, with all their experience in this field, have for their part come to these conclusions.

HSBC had a point to the effect that there could be no liability at all in contract, because the contract between Mr Rubenstein and HSBC only came into existence after the recommendation to invest in the EVRF had been made.

Since there would, on the basis discussed above, be liability for breach of statutory duty and in negligence, I fail to see the significance of this point. Perhaps it was pursued as seriously as it was for costs purposes: on the basis that, if loss was too remote, then there was no liability at all in statutory duty or tort, and HSBC was anxious to be able to say that there was not even a nominal liability for breach of contract (which does not depend on the proving of material loss).

Be that as it may, in the circumstances I consider that the question of breach of contract can be treated briefly. The judge may be right to look at the matter in terms of collateral contract. In my judgment, however, a simpler way of looking at the matter is that when the advice was given in August, it was on the basis that it would be paid for in accordance with the bank’s scheme for fees (see [12] above). If the FEEPAY contract document only came forward and was signed at the time of investment, the investment advice was always given and received on the basis that a fee would be paid. I see no difficulty in seeing the giving and taking of advice for a fee as a matter of contract.

One final issue for which Mr Rubenstein also received permission to appeal arose out of the circumstance, described by the judge as a “footnote to the argument on quantum”, that on 7 April 2011 ALICO or AIG Life paid Mr Rubenstein a sum of £7,195.23 on an ex gratia basis. The explanation for the payment was that ALICO had recently made a recovery in respect of three assets that had been held within the EVRF. A decision was made to distribute this recovery, pro rata, to the policyholders in the fund when it closed. Although it was considered that investors had no contractual entitlement to the money, the money was distributed within the spirit of ALICO’s “Treating Customers Fairly” policy. The judge had to decide whether this sum should be treated as a credit against any damages due to Mr Rubenstein, as a matter of avoided loss, or whether it was analogous to res inter alios acta, ie something divorced from the contractual relationship which had given rise to the claim in damages.

The judge considered British Westinghouse Electric and Manufacturing Co Ltd v. Underground Electric Railway Co of London Ltd (No 2) [1912] AC 673 and McGregor on Damages, 18^th ed, at paras 7-137 to 7-168 and concluded that the payment was made as a continuation of the original transaction and not as a matter entirely collateral to it. He distinguished Needler Financial Services v. Tauber [2002] Lloyd’s Rep PN 32, where Sir Andrew Morritt V-C held that a demutualisation payment did not go in reduction of a pension mis-selling claim. The judge held (at [131]):

“It arose directly in consequence of the investment in the bond and although it was not a payment to which Mr Rubenstein was legally entitled, it was paid to him out of moral obligation. Despite the gap in time since December 2008, I regard the payment as a continuation of the transaction whereby the defendant caused Mr Rubenstein to invest in the PAB.”

Despite Mr Palmer’s submissions to the contrary, I agree with the judge on this point. Although expressed as an ex gratia payment, it was not made out of pure benevolence: see an analogous line of cases discussed in Pirelli v. Gaca [2008] EWCA Civ 994 and Pope v. Energem Mining (IOM) Ltd [2011] EWCA Civ 1043. As for Needler, after discussing the jurisprudence Sir Andrew Morritt said:

“[24] In my view the authorities to which I have referred establish two relevant propositions. First, the relevant question is whether the negligence which caused the loss also caused the profit in the sense that the latter was part of a continuous transaction of which the former was the inception. Second, that question is primarily one of fact.”

In the present case, the payment represented the fund in which Mr Rubenstein was invested and the assets to which he was entitled by way of aliquot share. It was only because of a delay in recovery on some of those assets that the payment was not rolled into the money which Mr Rubenstein received under the “exit plan”. It may be that under the terms of the investment and its settlement the investors were not entitled contractually to the belated recovery, but there might always have been an argument that AIG Life or ALICO were liable to account for it. As a matter of what AIG understandably regarded as its moral obligation, the recovery was due to its erstwhile investors. It might be asked, who had a better claim to it? The payment was not made by a stranger to the transaction, or out of pure benevolence. The judge viewed the payment as part of a continuous transaction, and that was a finding to which he was entitled, and with which I agree. I would dismiss Mr Rubenstein’s appeal on this point.

In sum, I would allow Mr Rubenstein’s appeal on the remoteness issues, but dismiss it on the ex gratia issue. Mr Rubenstein will be entitled to recover damages in accordance with the judge’s obiter findings as to quantum, giving credit for the ex gratia payment of £7,195.23.

I agree.

I also agree.