England and Wales High Court (Commercial Court) Decisions

IN THE HIGH COURT OF JUSTICE
QUEENS BENCH DIVISION
COMMERCIAL COURT

		24 November 2017

His Honour Judge Waksman QC:

INTRODUCTION

1. This is an application made by the Claimant, The ECU Group PLC ("ECU") for pre-action disclosure ("PAD") pursuant to CPR 31.16 against three banks in the HSBC group: HSBC Bank Plc ("HSBC Bank"), HSBC Private Bank (UK) Ltd ("HSBC Private"), and HSBC Bank USA, NA ("HBUS"). I shall refer to them collectively where appropriate as HSBC. ECU is a currency debt management company, managing multi-currency loan facilities for its clients through its multi-currency debt management program ("MCDMP"). At the material time it controlled about US$1bn of clients debt and 45% of that involved mutual clients of ECU and HSBC Private.
2. As at 2006, ECU had had a long-standing relationship with HSBC Private and part of their dealings involved the placing of "stop-loss" orders by ECU. Such orders were a form of currency damage-limitation for ECU and its clients where foreign exchange rates for a particular pair of currencies rose above an expected level. If that happened then the stop-loss was triggered and there would be an instruction to HSBC Private to buy/sell that currency pair to avoid further losses. If the relevant currencies did not rise above the pre-determined ceiling, then no action would be taken on the order. Such orders were typically placed a few times each year. If a stop-loss order was triggered it would in fact be executed not by HSBC Private but either HSBC Bank in London, or HBUS in New York. As ECU had no direct relationship with the latter, it was not able to see how any particular stop-loss order had been executed.
3. By a letter dated 7 September 2005 the relationship between ECU and HSBC Private was formalised and in particular (according to ECU) HSBC Private was obliged to achieve best execution on the orders and, along with its agents, it was at all times to "safeguard the best interests of" ECU. Where there were mutual clients of ECU and HSBC then ECU had to place its stop-loss orders with HSBC.

THE TRADES

4. In January 2006, ECU placed three substantial stop-loss orders. On 5 January 2006 there was an order to buy €167,313,926.23 and sell a corresponding amount of Canadian Dollars ("CAD") should the spot inter-bank exchange rate for this pair of currencies rise above CAD$1.4071. On 6 January 2006 ECU placed an order to buy the same amount of euros and sell a corresponding amount of US dollars should the exchange rate for that pair rise above $1.2176. Finally, on 31 January 2006 ECU placed an order to buy CAD$242,350,919.02 and sell a corresponding amount of Japanese Yen should the exchange rate for the pair exceed JPY103.06. In this particular case, while the order was placed at 5:15pm GMT it was only valid from 7.45pm that day. I shall refer to these three trades collectively as "the Trades".
5. In the case of each of the Trades, they were triggered very shortly after they were placed. For the 5 January Trade, when placed at 5:10pm, the relevant exchange rate was 1.4025 and falling. But within minutes of this order being placed, the market reversed and climbed steadily above the trigger level by 5:41pm at which point the order was activated. Similarly, for the 6 January Trade, it was placed at 1:42pm when the relevant rate was 1.2119. But within minutes, the market climbed sharply above the trigger level by 1.52pm at which point this order was activated. Finally, at the time when the 31 January trade was placed at 5:15pm, the relevant rate was 102.09 and falling. Within minutes of the order being placed the market reversed and climbed steadily to 102.75 at 7.45pm when the order became valid. Within 5 minutes the exchange rate had climbed sharply above the trigger level of 103.06 when that order also was activated. ECU's case as set out in its Skeleton Argument is that this was an unusual sequence of events and the triggering of those stop-loss orders so quickly, or indeed at all, was not expected. It meant that for some reason, the market had moved significantly against ECU in an unexpected direction and very shortly after the trades were placed. There are, according to ECU, only two explanations for this. It could be an innocent, although unusual, volatility in the marketplace. For this to happen on each occasion of the Trades would be extremely unusual. The other explanation is that the traders at HSBC had manipulated the market by their own buying and selling activities in respect of the currencies so as artificially to push the relevant rates up causing the stop-loss orders to trigger. Provided they had secured their own positions in anticipation, the result would be very large profits for HSBC at the expense of ECU and its clients. This practice, which was well known, is called "front-running" because it involves manipulating the market in the knowledge of the particular stop-loss orders placed. The traders would therefore be running ahead of or in front of the orders and their trigger points. Another factor which ECU says indicates front-running is that the relevant rate dropped once the orders had been activated, as opposed to continuing to rise, accelerated by the buy/sell orders which occurred when the stop-loss orders had been activated. But if this was all being manipulated by traders at HSBC, then the rate would not necessarily rise.
6. All of this would have been at ECU's expense because absent the front-running, the rates would not have risen through the relevant ceilings and the stop-loss orders would not have been activated because there would have been no "loss" to stop. ECU calculates that the total losses as a result of the activation of the Trades was £6.17m. This would in fact have been sustained by ECU's clients, but as ECU would earn a 20% fee on profitable transactions it argues that it has suffered a loss as principal of £1.234m. It also says that it could expect to receive authority from the affected clients to bring representative proceedings on their behalf if indeed they are commenced.
7. Because ECU found these Trades surprising it expressed its concern to HSBC Private in particular in a long letter of 1 February 2006. ECU wanted HSBC Private to tell them of the extent to which HSBC Private had benefited from buying ahead of the stop-loss orders. Following what was said to have been a full investigation of the matter, Mr Whiting of HSBC Private wrote to ECU on 9 March 2006. That letter contained a detailed rebuttal of any suggestion of front-running or other wrongdoing and stated that full and extensive enquiries into the Trades had been made with no evidence of wrongdoing. The currency movements were simply due to general market activity.
8. ECU's evidence is that although it was not entirely satisfied by this response it felt that was not in a position to take the allegation further and it did not do so. It makes the point that this was in the pre-2008 crash period where assurances given by banks were more likely to be taken at face value than later. Indeed there was no independent way of ECU checking what had happened before and when the Trades were activated because all of the relevant information and documents were internal to HSBC.
9. There the matter rested until July 2016. The following summary is taken from paragraph 5 of ECU's Skeleton Argument. I did not understand it to be controversial. On 19 July 2016 the US Department of Justice issued a formal indictment against two of HSBC's most senior former FX traders, Mark Johnson and Stuart Scott ("the Indictment"). This related to alleged front-running committed in 2011 at the expense of an HSBC client called Cairn Energy. As with ECU, the client had complained to HSBC at the time but was assured by HSBC after its full investigation that there had been that no wrongdoing. It was only years later that this conclusion was challenged. HSBC's flawed internal investigation was widely reported in international media and there were further consequences. For example, on 29 September 2017 the US Federal Reserve issued a Cease and Desist Order highlighting HSBC's failure to identify FX trading malpractice and requiring it to amend its practices within 90 days. It was also fined $175m for FX abuse. On 23 October 2017 Mr Johnson was convicted of conspiracy and fraud. Mr Scott now faces extradition to the US. All of this came on the back of an earlier fine of £216m imposed on HSBC by the FCA in late 2014 as a result of misconduct which included front-running and abuse of stop-loss orders and then in the same year an agreement by HSBC with the Commodity, Futures and Trading Commission to pay $275m in New York for similar conduct.
10. The events of 2016 caused ECU to review its position in relation to the Trades. There was now evidence, albeit concerning different trades in different years, which bore all the hallmarks of the Trades which had given rise to suspicions at ECU at the time but which HSBC had assured it were innocent.
11. It still remains the case, however, that ECU does not know how the Trades were effected nor what trading was going on at HSBC in the currency pairs between the time when the stop-loss orders were placed and when they were activated. Documents showing this, ECU says, are likely to give the answer to the question as to whether there was front-running or not. If so, there would be a real impetus for HSBC to settle the matter without, or without protracted, litigation. Equally, if the documents show that indeed there was nothing untoward in respect of at least these Trades, ECU would be bound to consider whether any claim was likely to be viable.

THE APPLICATION

12. That is the brief background to ECU's present application for pre-action disclosure.
13. I will refer to the individual classes of document in detail below but broadly speaking, in order to provide information as to what happened prior to the activation of the Trades, ECU has sought data relating to the Trades themselves, documents evidencing communications between the traders concerned, and documents relating to the investigation which took place in early 2006 and which would themselves shed light on what had happened at the time of the Trades. Some preliminary information about such documentation has already been helpfully provided by HSBC as well as indicating what classes of document it knows now are available and could be produced albeit at some cost, what further classes of document are potentially available and finally those documents which it says are not available.
14. HSBC resists this application in its entirety. It is appropriate here to deal with the relevant law before articulating the grounds of that resistance.
15. I should add that both sides adduced evidence: Mr Alexopoulos, the solicitor for ECU, put in three witness statements and Mr Gadhia, HSBC's solicitor put in two. It is not necessary to refer to much of that evidence. Where I do, it is in context, below.

THE LAW

16. CPR rule 31.16 states as follows:

"(1) This rule applies where an application is made to the court under any Act for disclosure before proceedings have started.

(2) The application must be supported by evidence.

(3) The court may make an order under this rule only where-

(a) the respondent is likely to be a party to subsequent proceedings;

(b) the applicant is also likely to be a party to those proceedings;

(c) if proceedings had started, the respondent's duty by way of standard disclosure set out in rule 31.6, would extend to the documents or classes of documents of which the applicant seeks disclosure; and

(d) disclosure before proceedings have started is desirable order to:

(i) dispose fairly of the anticipated proceedings;

(ii) assist the dispute to be resolved without proceedings; or

(iii) save costs."

17. The leading case is Black v Sumitomo [2002] 1 WLR 1562. The following propositions may be derived from the judgment of Rix LJ:

(1) The requirements in sub-paragraph (3) (a) and (b) are simply about the likely parties to any claim, not its underlying merits and "likely" in this context means "may well"; see paragraphs 71 and 72;

(2) Requirement (c) will raise the question of the clarity of the issues which would arise once the litigation has started, without which such clarity it will be difficult to say if the documents now sought would fall within standard disclosure; see paragraph 76;

(3) Requirement (d) with its three possible variants constitutes both a jurisdictional threshold and also a set of factors which are required to be considered in more detail when the question of discretion is dealt with; see paragraphs 81 and 82;

(4) The jurisdictional threshold is not intended to be a high one and the real question is likely to be the exercise of discretion which will not be much assisted by the simple fact that the jurisdictional threshold is met; see paragraph 73; if it were otherwise, that would tend to suggest that orders would be made much more frequently under this provision than they are; see paragraph 85;

(5) The discretion itself is not confined and will depend on all the facts of the case; important considerations will include the nature of the injury or loss complained; the clarity and identification of the issues raised by the complaint; the nature of the documents requested; the relevance of any protocol or pre-action enquiries and the opportunity which the complainant has to make his case without PAD; see paragraph 88;

(6) In addition, if there is considerable doubt as to whether the usual disclosure staged would ever be reached, the court can take this into account as affecting discretion; see paragraph 77. This must be a reference to practical or legal obstacles which the putative claim may face;

(7) At paragraph 92 Rix LJ stated "unless there is some real evidence of dishonesty or abuse which only early disclosure can properly reveal and which may, in the absence of such disclosure, escape the probing eye of the litigation process and thus possibly all detection, I think that the court should be slow to allow a merely prospective litigant to conduct a review of the documents of another party, replacing focused allegation by a roving inquisition". This observation was made in the context of Rix LJ having found that the complaint in that case was factually and legally "speculative in the extreme" see paragraph 91. Context is important because it is otherwise hard to see why it must be shown that in the absence of early disclosure the evidence would (later) escape the eye of the legal process;

(8) The more focused the complaint, and the more limited the disclosure sought in that connection, the easier it is for the court to exercise its discretion in favour of PAD, even where the complaint might seem somewhat speculative or the request argued to be mere fishing. The court might be entitled to take the view that transparency was what the interest of justice and proportionality most required. But the more diffuse the allegations and the wider the disclosure sought more sceptical the court is entitled to be about the merit of the exercise; see paragraph 95.

18. There was a certain amount of debate before me, and in the cases, as to whether it was also necessary for the applicant to show (a) that without the disclosure it could not properly plead a case at all and/or (b) that even without the disclosure, the case had at least a real prospect of success, or had reached some other merits threshold. In fact, of course, there is a degree of tension or inconsistency between those requirements if both had to be made out, as noted in paragraph 28 of the judgment of Underhill LJ in Smith v Energy Secretary [2014] 1 WLR 2283. Since the view of Rix LJ at paragraph 68 of his judgment was that the provision was addressed to situations where disclosure (a) would help the applicant who could plead a cause of action to improve it or (b) was necessary as a vital step in deciding whether to litigate at all or (c) necessary to provide a vital ingredient in the pleading, it seems to me that questions of underlying merit of the claim should be dealt with in the context of discretion. As Underhill LJ put it in Smith "If there were a jurisdictional requirement of a minimum level of arguability the question would necessarily arise of how the height of the threshold is to be described…It is inherently better that questions about the likelihood of the applicant being able in due course to establish a viable claim are considered as part of a flexible exercise of the court's discretion in the context of the particular case."; see paragraph 24.
19. The potential width and focus of the classes of documents sought is a further matter for discretion. See paragraph 72 of the judgment of Rix LJ. I can see that in an extreme case, where the documents sought were hopelessly wide, that might even militate against the jurisdictional thresholds being achieved. In any event, as Morison J put it in Snowstar v Graig [2003] EWHC 1367 "… Every action for pre-action disclosure should be crafted with great care, so that it is properly limited to what is strictly necessary.". In the case before him where the disclosure sought was "wide and woolly" he did not regard it as a satisfactory suggestion that any flaws in the application notice could be dealt with after judgment. I accept that where the categories of disclosure sought are extremely wide or unclear, the Court is unlikely to be prepared to rescue the application by, in effect, re-drafting them. However, in my judgment that does not mean that the Court has no power to adjust the categories of disclosure sought so as to deal with any particular problems, whether in terms of scope or availability, which may become apparent in the course of the application. The extent to which the Court would, as part of its discretion, consider it appropriate to vary the disclosure sought obviously depends on the facts and circumstances of each particular application. See the observations of Marcus Smith J at paragraph 34 of his judgment in Attheraces v Ladbrokes [2017] EWHC 431.
20. Beyond those points of principle, I do not consider it helpful or necessary to examine all of the PAD cases put before me, many of which turned on their own particular facts, especially where the role of discretion, as compared to achieving jurisdiction, is so great. For the same reason, I do not think it assists much by adding a gloss to the provision by suggesting that it only applies "exceptionally" or some such. Obviously it is not the norm for potential claimants either to need or secure PAD and to that extent such applications are by their nature unusual. Beyond that, it is for the court to decide whether the jurisdictional thresholds are met and if so how the discretion should be exercised and, as with any such exercise, with regard to the overriding objective and in particular proportionality.
21. In the light of a particular issue of limitation which has been raised here, it is necessary to make some observations about that.
22. Section 32 (1) of the Limitation Act 1980 ("the Act") provides as follows:

"…. Where in the case of any action for which a period of limitation is prescribed by this Act either-

(a) the action is based upon the fraud of the defendant; or

(b) any fact relevant to the plaintiff's right of action has been deliberately concealed from him by the defendant;

the period of limitation shall not begin to run until the plaintiff has discovered the fraud, concealment or mistake (as the case may be) or could with reasonable diligence have discovered it."

23. As to this, it is well-established that the "fact" concealed must be one without which the claim or cause of action would be incomplete. It is not enough that the relevant fact is something which if known, would simply improve the claimant's case as a matter of evidence or otherwise. See paragraphs 33 and 36 of the judgment of the Chancellor in Arcadia v Visa [2015] EWCA 883, referring back to the decision of the Court of Appeal in Johnson v Chief Constable of Surrey 23 November 1992. So, for example, if the missing fact related only to the ability of the claimant to rebut an exemption to liability raised in the defence and which the defendant would have to prove further down the line, that would not be material for the purposes of s32.
24. Second, there is no particular rule of law or practice, in my judgment, as to how a court dealing with an application for PAD should deal with any question of limitation as opposed to any other "merits" point. Obviously, if it is submitted and the court finds that the claim is hopelessly time-barred with no real prospect of overcoming it, that would be a powerful if not conclusive reason not to order PAD since the entire exercise would be a waste of time. But if that is not the submission or finding, then it is simply a matter to be weighed in the exercise of discretion.
25. In this context, I was referred by Mr Valentin to the case of Lampert v Lloyds [2012] EWHC 2312 where Mann J dismissed an application for PAD as being totally without merit. He observed this at paragraph 28:

"There would be a case for saying that the limitation clock started running then, if not earlier. I do not of course rule on that but I think that in a case like this, if there were a claim with a prospect of a limitation defence and the need to run a concealment riposte, the proper course would be to start the proceedings and to deal with disclosure matters at the proper disclosure stage. On balance I think that the limitation point just taken by itself may not be a sufficiently clear bar to these proceedings at this stage of the reasoning to demonstrate that there can be no sustainable cause of action. I do not consider that in a case like this it would be right to embark on the sort of consideration that would be necessary to determine that. However the nature of the material and the history of this matter probably means that pre-trial disclosure would not be appropriate."

26. However I can derive no particular rule from those observations (for example that where there is an issue as to whether s32 applies PAD should not be ordered), nor do I think that Mann J was intending to lay one down. He was simply dealing with the exigencies of the (very unmeritorious) case before him.
27. In this case, while Mr Valentin submits that the limitation difficulties facing ECU should be a strong or even compelling factor militating against PAD as a matter of discretion, he does not submit that ECU's position is hopeless on the point.

THE ISSUES ON THE APPLICATION

28. In summary, HSBC resists the application on the following grounds:

(1) The jurisdictional threshold is not met;

(2) Even if it is, the court should not, in its discretion, make any order because

(a) the underlying claim is speculative; and/or

(b) ECU can now in fact plead a case without sight of the documents sought; and/or

(c) as is common ground, the claim is time-barred without recourse to s32, which ECU relies, upon, alleging deliberate concealment by HSBC, but it is unlikely that ECU will be able to make that out; and/or

(d) the classes of documents sought are too wide and insufficiently focused and/or the exercise of identifying them will be too costly and/or otherwise disproportionate.

ANALYSIS: JURISDICTION

29. It is plain here that any proceedings brought will be by ECU as against HSBC entities. It is also plain that the documents sought by way of PAD would fall within standard disclosure. HSBC has submitted that the jurisdictional threshold is not met because the claim itself is too speculative. As noted above, questions as to the merits or otherwise of the intended claim should properly be considered as part of discretion.
30. But in any event, I do not accept that the claim is so speculative or devoid of particularity. ECU's essential point is that if there was front-running in relation to the Trades, then this would be a clear breach of contract as far as HSBC Private is concerned, and there would also be claims in tort against HSBC Bank and HBUS for inducement to HSBC Private's breach of contract. There could also be claims in conspiracy if more than one trader was involved and claims for breach of ECU's confidential information that is to say the terms of the stop-loss orders. I accept that there could be relief by way of damages or an account of profits. The fact that such claims would be viable if there was front-running is given additional support by the events of 2016. Given those events, and the earlier events of 2014, regrettably it cannot be said that it is inconceivable that HSBC could be liable.
31. Otherwise, it is obvious that the parties to this application would indeed be the parties to any subsequent claim.
32. Further, ECU have satisfied me that there is in principle a real prospect that if PAD was ordered, it would be likely to shed real and direct light on whether there was front-running here or not. That is because the documents are likely to show whether the traders involved at the material times were indeed engaged in a course of buying/selling which pushed up the price of the relevant currency shortly after the stop-loss orders had been placed and passed on to HSBC Bank or HBUS. If front-running is revealed by the documents, then there must also be in principle a real prospect that any claim would be settled without litigation or at least at an early stage even with the prospect of a separate argument on limitation. Alternatively there would be a real prospect of narrowing the issues between the parties going forward. All of this would lead to the prospect of a real saving on costs. Conversely, if the documents produced revealed no front-running but simply trading in volatile market conditions, then ECU would be forced to consider very carefully whether any further action could be justified. At this stage, I am quite satisfied that the modest jurisdictional threshold set by CPR31.16 has been reached and I therefore turn to discretion.

ANALYSIS: DISCRETION (1) - MERITS AND OTHER MATTERS

Overall merits

33. At various points, HSBC has submitted that the putative claim (leaving aside any question of limitation) is speculative (I have dealt with that to some extent in the context of jurisdiction above) and that it can in fact be pleaded out with relative ease now so that the documents are not required. Both propositions cannot be right.
34. In terms of legal merits, for the reasons given above I do not regard this claim as speculative. On the facts there is a very clear distinction between this claim and that made in Sumitomo not least because the foundation for the claim here is a direct and long established contractual relationship between ECU and HSBC Private. It is not made by a "stranger" who is seeking to recover general trading losses or lost opportunities to make profits as against one player in the market.
35. The putative claim is backed up, now, by evidence of precisely the sort of conduct it seeks to allege against HSBC albeit at different times and in relation to a different client. There can be arguments about how strong any inferential case may be (apparently, for example, Mr Johnson was not at HSBC in 2006) but at least it shows that the intended claim is not fanciful. Nor has it been dreamt up now. ECU did have its suspicions back in 2006 and communicated them to HSBC but was told there was nothing in them.
36. I cannot see anything in respect of the general merits of the putative claim which would count seriously against the exercise of discretion here.

Limitation

37. However it is submitted strongly by HSBC that there are formidable difficulties for ECU on the question of limitation. This all turns on the application or otherwise of s32.
38. ECU argues that if there was indeed front-running then the letter of 9 March 2006 was manifestly false since it stated the opposite. ECU goes on to argue that the letter itself would constitute the concealment, or part of it, in the sense of suggesting a different version of events which would involve no liability on the part of HSBC at all.
39. Further, ECU says that if there was front-running, HSBC, or some material part of its operations involving its traders, must have known about it at the time, and thus any concealment was deliberate. It must be remembered that while this allegation is serious (as is the underlying allegation of front-running) HSBC has already been found to have failed to disclose front-running in a different investigation where the client was assured that nothing was amiss. I appreciate that it might be said that any such concealment here was not deliberate but looking at HSBC's recent track record as stated above, that cannot be assumed if indeed there was front-running in respect of the Trades.
40. However, HSBC next argues that if the relevant concealed fact was the existence of front-running, then ECU had already discovered it back in 2006. It had observed the spikes in the trading and had found them to be so suspicious that they required a full explanation from HSBC. Indeed, they referred specifically to the possibility of front-running. I see all of that but in truth ECU had not discovered the fact of front-running - it had its suspicions but felt it could not push the matter further after receipt of the letter of 9 March. It had, in effect, been "put off the scent"; for an analogous case see JD Wetherspoon v Van Den Berg [2007] EWHC 207. And at the end of the day, the question is why the spikes occurred - which ECU did not then know.
41. HSBC retorts that it is not as simple as that because in fact ECU was not satisfied or wholly satisfied with the letter of 9 March, and still had reservations as the later correspondence showed. In particular it stated to HSBC that the position it took was "lamentable" although it is not clear that this meant that ECU did not believe the response. In any event, it seems to me again to be perfectly arguable to say that there is a real issue as to whether the suspicions held by ECU and largely dissipated by the letter could really amount to knowledge of front-running as denied in the letter. Moreover, it cannot be (nor is it) said that on any view, ECU could with reasonable diligence have discovered the truth (if that is what it was) at the time, since all the relevant information was within the exclusive possession of HSBC.
42. A further point taken against ECU is that the fact of front-running was not itself a fact without which any claim would be incomplete. Therefore it is not material for the purpose of s32. Rather it was merely a fact which went to strengthening ECU's case against HSBC or improving its evidence which is not enough. However, it is important not to conflate two quite separate questions:

(1) Why is PAD appropriate now and to what extent could ECU still mount a case against HSBC without it? ("the PAD Question"); and

(2) The correct characterisation of the fact said to be concealed for the purpose of s32 ("the Limitation Question").

43. The PAD Question involves in part a comparison between the position of ECU now, either with or without the documents; it therefore would include the knowledge which it now has from the 2016 and earlier actions against HSBC. The Limitation Question, on the other hand, invites a comparison between ECU's position back in 2006 and either with or without the knowledge of front-running. The fact of front-running was obviously an essential element of its claim. And I have already held that certainly for present purposes, it is well arguable that ECU did not otherwise "know" that fact. Therefore, it seems clear to me (or it is at least well arguable) that the fact which ECU says had been deliberately concealed was material.
44. Accordingly, I am far from persuaded that the strength of HSBC's limitation argument is such that in truth any claim issued would never even get to the disclosure stage because it would be struck out on limitation grounds or that there were very serious obstacles here for ECU to surmount. I accept that the limitation issue is one which must be weighed in the balance against ECU and be accorded significant weight. But it is far from an overwhelming or compelling factor in my view.
45. To the extent that the PAD would itself be necessary so that ECU could show, or show clearly, that there was deliberate concealment, that is really no different from the overarching point which is that the PAD will (if it yields the fact of front-running) make it much easier for ECU to make the underlying claim. ECU does not say that without PAD it could not plead a case at all, but that is not the sole criterion - see above.
46. Finally, as noted above, I do not accept that there is any rule to the effect that where there is this kind of dispute on limitation in the context of an application for PAD the proper approach is to dismiss it and let the proceedings take their course.

Other matters

47. While ECU may well be able to plead out an inferential case against HSBC at present, that is not fatal to the application. Direct evidence of front-running will have a real bearing on the ability to dispose of the case without proceedings or to narrow the issues and to save costs in any event. It is in my judgment wholly unrealistic to suppose otherwise. Therefore, at the discretion stage, it seems to me that the requirements of subparagraph (d) (i) (ii) and (iii) are made out in fact and not just as a matter of abstract principle.
48. I also do not consider that any uncertainties about how the loss claim would be pursued has any real impact on discretion. The same is true for any confidentiality concerns there may be surrounding the documents. That can be dealt with by the use of a confidentiality ring in the usual way or redactions. I do not consider any steps here to be unduly burdensome.
49. HSBC points to the fact that in ECU's financial statements for the year ended 31 December 2016, it is said that ECU undertook a fresh investigation of the Trades and had decided to initiate proceedings for disclosure, ie this application. It is then said that a substantial claim will be made against HSBC in due course. This litigation was said to be one purpose of a new bond issue of up to £4.5m. HSBC says therefore, that PAD does not matter because ECU is going to start a claim anyway. I do not accept that so much can be read into this part of the statement. What was said was that a claim was "anticipated" and in any event as I understand it no formal decision has yet been taken. I do not think much turns on this.
50. A final general point is the fact that this is both a serious and a very substantial claim. That has a bearing on the question of the proportionality of the documents required and one which favours ECU. This especially since there is no-one other than HSBC that could provide the documents.

ANALYSIS: DISCRETION (2): THE DOCUMENTS

51. Any exercise of discretion must itself take into account the nature of the disclosure sought before a conclusion is expressed. Accordingly, I deal with the various categories here and by reference to the documents which are available, potentially available or stated to be unavailable.
52. Before doing so, I should say that the PAD sought is designed to achieve one simple aim which is to find out what happened in the short periods between the placing and activation of the Trades so as to see whether or not there was front-running. That is, itself, a perfectly focussed objective.

Available documents

Bloomberg messages

53. These are sought in the following parts of the draft order:

"1.2 (d) Bloomberg messages, from five minutes before the relevant order was placed until 10 minutes after it was executed, to and from:

i. G10 Spot FX traders responsible for HSBC's London and New York Spot EUR/USD, USD/CAD and, if operated at that time, EUR/CAD trading and order books on 5^th January 2006 (and specifically limited to a truncated time zone of between 17:05 and 17:51 - a period of just 46 minutes);

ii. G10 Spot FX traders responsible for HSBC's London and New York Spot EUR/USD trading and order books on 6^th January 2006 (and specifically limited to a truncated time zone of between 13:37 and 14:02 - a period of just 25 minutes);

iii. G10 Spot FX traders responsible for HSBC's London and New York Spot USD/CAD, USD/JPY and, if operated at that time, CAD/JPY trading and order books on 31^st January 2006 (and specifically limited to a truncated time zone of between 17:10 and 20:00 - a period of just 2 hours and 50 minutes);

iv. G10 Spot FX traders and G10 Spot FX managers specifically authorised to trade "back books" and/or "proprietary trading books" on 5^th, 6^th and 31^st January 2006 (expressly limited to the times between five minutes before the placing of ECU's Stop-Loss orders and ten minutes following the reported execution of the Relevant Trades as cited in a) to c) above).

54. It cannot be seriously in dispute that FX traders would typically use Bloomberg messages and if they were engaged in front-running here, this would be likely to show up in such messages. The recent allegations against HSBC show this (see for example 1/1/106). The cost of retrieval is relatively modest, perhaps as low as just a few pounds. The time period for all of the Trades is a total of only 4 hours and 1 minute. HSBC has already helpfully identified six traders who dealt in the relevant currencies; in addition, there may be others who dealt with the Traders themselves; it is not suggested that the work to obtain these messages cannot be done. ECU's estimate of costs is about £4,500. ECU will have to bear the costs of this exercise in the first instance. I do not regard this category as overbroad or disproportionate.

Emails

55. These fall within the following categories in the draft order:

"1.3 (d) Emails to and from any of the following: Mr Steve Whiting; Mr Andrew Brown; Mr Ben Welsh; the relevant HSBC personnel working within the Compliance Departments of London and New York offices that were involved in investigating the Complaint and the Trades and/or conducting their independent review of "the findings" (referred to in the Respondents' March 2006 Letter).

1.4 All emails to and from Mr Alan Ramsay from 2^nd February 2006 to 27^th April 2006."

56. HSBC has said that it has no access now to underlying confirmations and market data pertaining to the actual Trades which had been the subject of category 1.2 of the draft order (apart possibly from the 6 January Trade). The next best thing, according to ECU, would be emails to and from those involved in the investigation which apparently led to the letter of 9 March 2006 concerning the Trades. That is because it is to be expected that any investigation would require all the relevant information about the execution of the Trades and surrounding trades. It is likely that such information would be contained in or attached to the emails sought. ECU has now limited the period applicable to paragraph 1.3 (d) to 1 February to 15 March 2006, there is a limited number of custodians, being those involved in the investigation, and keywords. This category seems to me to be pertinent and proportionate.
57. The same applies to category 1.4 which relates to Mr Ramsay, HSBC's Head of Global Compliance at the time. An internal ECU email dated 27 March 2006 referred to a conversation with Mr Ramsay who at that point was reported to have said that HSBC's investigation was still continuing and that he had called for the file. I accept that there is no direct evidence from Mr Ramsay himself that this was the case but on the other hand there is no evidence to contradict it from HSBC. He was clearly a relevant individual and on the face of it emails to or from him are likely to contain information about the Trades. The fact that the date range here goes beyond the date of the letter is irrelevant since it seems to be the case that investigations were still ongoing.
58. As to costs of the email search, ECU says that because it has reduced the scope of that search the £40,000 estimated by Mr Gadhia on behalf of HSBC will be much reduced, to about £12,500. Again, ECU will bear those costs in the first instance.

Potentially Available Documents

Underlying Trading Data

59. The only trading data that might now still be available concerns the 6 January Trade and that is because it was not executed in the US. The relevant category here is:

"1.2 (b) Documents evidencing the Spot EUR/USD trades executed by HSBC's London and New York Spot G10 FX Trading Desks only on 6th January 2006 within the time zone of 13:37 and 14:02 (a period of just 25 minutes);"

60. It appears to be the case that HSBC has a backup system called OTP which is likely to have the data for this Trade on it although it is not yet been searched. According to Mr Gadhia (see paragraphs 33-35 of his first witness statement) it would require a manual search of the trading data for that particular day. It is possible that for some reason the relevant data will not be found. But that is somewhat speculative at this stage. Nor do I consider that the fact that data for only this one Trade may be available, is a reason not to seek it. The cost of the exercise is put by HSBC at £25-35,000. It is also said that there may be a cost in terms of management time spent by HSBC in relation to this search. That is not a fatal obstacle on the footing that ECU will defray not only the costs incurred by HSBC's solicitors but also a reasonable sum to compensate HSBC for loss of its time.

Global Market and Compliance Documents

61. These are stated in the draft Order as follows:

"1.3 All underlying documents relating to the "detailed enquiries" undertaken in respect of the Complaint and/or the Trades (referred to in the Respondents' March 2006 Letter), from 1^st February 2006 to include:

(a) File notes and/or reports;

(b) Internal communications;

(c) Minutes of meetings held"

1.5 The "findings of the two Foreign Exchange Desks ... [in] London and New York' (referred to in the Respondents' March 2006 Letter); "

62. Again, the reason for the emphasis on these further documents which also relate to the investigation is the lack of documents dealing directly with the Trades themselves. I accept that any documents prepared in connection with the investigation may very well show what actually happened with the Trades - since that was the very purpose of the investigation (as with the emails dealt with above).
63. The difficulty lies in trying to retrieve such documents if they exist. If they do they are likely to be in HSBC's Global Markets and Banking Division Compliance drive which covers compliance issues in this field worldwide. Some searches have been done with a limited number of keywords and by reference to document title. It appears that this has not yielded anything. However, Mr Gadhia accepts at paragraph 56 (b) of his second witness statement that it would be possible to do an electronic search of document content and not merely title. It will have some limitations because some documents will not be able to respond to text searches and others may be password protected. But that is not a reason not to do the search. It also seems to me, as Mr Alexopoulos said, that although this is a very large amount of data there must be a way of refining the search to the narrow period around the investigation.
64. This may still be an extensive exercise but it is far from impossible and again, ECU will pay for it in the first instance. I need also to bear in mind that the searches already undertaken by HSBC have revealed no discrete document which relates to the investigation other than the Compliance Report which is surprising given the thoroughness of the investigation as recounted in the letter of 9 March. I also bear in mind that following a suggestion from Mr Alexopoulos, HSBC did find some further records pertaining to ECU - see paragraph 54 of Mr Gadhia's second witness statement.
65. I therefore consider that a limited search of this kind can and should be undertaken. Subject to any further submissions on the point, I think that the keywords used so far by HSBC (see paragraph 56 (c) of Mr Gadhia's witness statement) should be enough.
66. But otherwise I would not order any further disclosure within this category.

Unavailable documents

67. Subject to some further and more specific confirmation from HSBC of the categories of documents said not now to be available, which ECU requires and which can be dealt with hereafter, it accepts that it cannot have what is not there. Accordingly, it would not be proportionate to order any further disclosure.

ANALYSIS: DISCRETION (3) - CONCLUSION

68. Having taken into account all the various factors, I consider that in my discretion, there clearly should be PAD in respect of the categories positively indicated above. There will no doubt be various drafting points which arise and these can be dealt with following handing-down of this judgment. I am indebted to all Counsel for the excellence of their written and oral submissions.