Bank Mellat v HM Treasury

The English Court welcomes litigants from all parts of the world. Gratifyingly, it enjoys a much prized reputation for fairness. There is no “home ground” advantage; it matters not whether the litigants are domestic, foreign, governmental or private. All are treated the same.

For all litigants, the procedure in this Court is governed by the lex fori – English law. That is the norm internationally, as a matter of the conflict of laws. Disclosure and the inspection of documents form a part of the law of procedure governed by the lex fori. On occasions, a tension can arise between the English law requirement for the inspection of documents and the provisions of foreign law in the home country of the litigant.

Where such a tension arises, it is for the Court to balance the conflicting considerations: the constraints of foreign law on the one hand, and the need for the documents in question to ensure a fair disposal of the action in this jurisdiction, on the other. That balance is struck by a Judge sitting at first instance, making discretionary, case management decisions. As is well-established, this Court will only interfere if the Judge has erred in law or principle or has (in effect) reached a wholly untenable factual conclusion.

This is an appeal by the Iranian Appellant (“the Bank”) against the judgment and order of Cockerill J, made on 19 September 2018 (“the judgment” and the “order” respectively), on the application of the Respondent (“HMT”) under CPR Part 31.19(5), requiring the production of documents in unredacted form but subject to various confidentiality provisions. As will be seen, it is common ground that production of the Iranian documents covered by the order and with which this appeal is solely concerned, would constitute a breach of Iranian law.

The background concerns a claim by the Bank for damages under the Human Rights Act 1998 (“the HRA 1998”) against HMT, in respect of loss it allegedly suffered as the result of a “financial restriction”, the Financial Restrictions Iran Order 2009 (“the 2009 Order”), made by HMT in October 2009 and held unlawful by the Supreme Court, by a majority, in 2013: Bank Mellat v HM Treasury (No. 2) [2013] UKSC 38; [2013] UKSC 39; [2014] AC 700.

In overview, the issue at the trial, fixed for June 2019, is whether the Bank can prove that the losses it allegedly suffered were caused by the 2009 Order. The pleaded quantum of the Bank’s claim has been some US$4 billion; a recent amendment suggests a reduction to some US$1.7 billion, plus interest. In very broad terms, the Bank’s claim divides into two parts. The first part is now put at some US$31.7 million plus, relating to specific transactions (letters of credit (“L/Cs”), bank guarantees and penalties) said to have failed in consequence of the 2009 Order. The second – and far larger part – does not turn on existing, specific transactions and includes as its most sizeable component a claim for over US$800 million relating to the Bank’s share of the Iranian foreign currency L/C market; the allegation is that the Bank’s existing share of that market in October 2009 was approximately 25%; but for the 2009 Order, that share would have been maintained and increased to about 35%.

The litigation is massive and complex. Though loosely described as a quantum trial, major issues of causation are live. For example, the trial is intended to determine challenges as to the lawfulness of further United Kingdom (“UK”) financial restrictions orders made in 2011 and 2012 (“the follow-on measures”). The knock-on effect on causation in respect of the Bank’s claim is readily apparent. So too, questions of causation will arise with regard to measures taken elsewhere (including by the European Union (“EU”), the United Arab Emirates (“UAE”) and South Korea), allegedly attributable to HMT encouragement (“the copycat orders”). Further, insofar as it is claimed that the Bank lost business due to the 2009 Order, we are told that there will be an issue as to the extent to which other international measures (not challenged in these proceedings, such as US sanctions) had the effect of reducing Iranian trade more generally. Still further, there may be “cut-off points” as to the second part of the Bank’s claim (not linked to specific transactions), turning on the meaning to be given to “possessions” within Article 1 Protocol 1 to the European Convention on Human Rights (“A1P1” and the “ECHR”, respectively).

As we understand it, some 33,000 documents have been disclosed by the Bank since 2016. Of those documents, around one third (roughly 12,500 documents) are said to contain confidential banking data in relation to identifiable customers, said to be protected under Iranian, Turkish and South Korean law – the Bank having overseas branches in Turkey and South Korea.

In a nutshell, the documents in question had hitherto been produced in redacted form.

It is unnecessary to trace the history of the parties’ respective positions. Suffice to say that by the time the matter reached this Court, the Bank’s case was that customer identities were to be redacted and the needs of the trial could be catered for by way of ciphers representing those identities. For its part, HMT contended for the production of unredacted documents but only to members of a “confidentiality ring” (“the confidentiality club”) and in ciphered form, with a master list of cipher codes being available to members of the confidentiality club – and not for use in open court.

Cockerill J ruled that the risk of prosecution under Turkish law was such that no master list should be provided to HMT in respect of documents in Turkey. The Turkish aspect of the case is now academic as the Bank has recently indicated that it will not be pursuing its claim in respect of losses allegedly suffered by its Turkish branch.

With regard to the South Korean documents, the Judge ruled in favour of HMT and the Bank has not appealed.

As already noted, this appeal is solely concerned with the Iranian documents where the Judge ruled in favour of HMT – the parties essentially dividing on the production of a master list accessible to members of the Confidentiality Club.

It is to be underlined that the Bank resists the order (made by Cockerill J) root and branch. The Bank has not sought to engage with the constitution of the Confidentiality Club, for example by seeking to confine its membership. Nor has the Bank adopted any intermediate position on the documents covered by the order; there is no fall-back contention that even if some documents must be produced unredacted others must not. The Bank’s position is all or nothing.

After recounting the history of the proceedings, Cockerill J helpfully outlined their scope as follows (at [12]):

“A number of interesting legal issues arise, including questions as to causation, whether the interference in the Bank’s enjoyment and control of its possessions has been caused by the 2009 order, the extent to which it can prove its loss, whether an award of damages is necessary to afford just satisfaction to the Bank and, if so, in what sum. In this respect it is relevant to note that there has been a recent amendment to the defence in April 2018 to plead that by reason of the Bank’s own conduct and/or its ownership and/or control by the Government of Iran no order for damages is appropriate. Alternatively, there should be a significant reduction in any damages awarded.”

The Judge then went on to deal with the topic of sampling (which does not arise at least directly on this appeal) before turning to the question of redactions. After noting the parties’ positions (as they then were), Cockerill J observed (at [47]) that the “relevant legal principles were not very much in issue”:

“It is agreed that obligations of confidentiality arising under some other legal system do not provide an automatic entitlement on a party litigating in this forum to withhold documents from disclosure and it is a matter for the court’s discretion whether production should be ordered.”

After citing authority (to which I shall return), the Judge (at [50]) noted the HMT submission that it was relevant to the Court’s discretion “how real any risk of prosecution in the foreign state is found to be”. Ultimately (at [51]), it was agreed between the parties that:

“….the question is a discretionary one. The issue….is to what extent the fact that risk is invoked by the claimant affects the principle and to what extent the interest of the litigant trumps those of the party claiming that it is not legitimate to comply.”

Coming to her conclusions, the Judge (at [79]) did not accept that there was “exactly the same presumption in favour of disclosure” where, as here, the disclosure of confidential information was said to be contrary to foreign law, as where the issue was “simply ‘business confidential’ information”. After some reference to authority, the Judge observed (at [82]) that each case needed to be evaluated on its own facts

The Judge acknowledged that the expert evidence on Iranian law came only from the Bank, in the event through a Dr Kakhki (discussed further below). HMT had not been able to put forward evidence on its own behalf. Nonetheless, it was for the Judge to evaluate that evidence (at [84]), rather than simply take it at face value. The Bank was advancing an extremely large claim and seeking the Court’s “sanction to provide less than full disclosure by reason of redactions”. In particular, the Judge was required to consider what the evidence was telling her as to risk.

The Judge continued as follows, in a passage much debated on the appeal:

“85.

I accept Dr Kakhi’s expertise which was not in issue. I also accept the evidence indicates that there would be a breach of the law in Iran in providing the documents unredacted. The real question is as to the risks of sanction. In this area, while I

see and I have read a few times now what Dr Kakhi says, which is that there are such risks from breach which he describes as even ‘real’ and ‘probable’ and ‘likely’, I am not entirely happy with his evidence. In particular it seems to me that there is a failure to address head on the critical question; that is, what is the risk of sanction in the face of the breach being as a result of compliance with an order of a competent court elsewhere.

86.

Despite Mr McLaren’s submissions [i.e., Mr McLaren QC, then representing the Bank], I do not read the report’s allusion to deliberate/knowing wrongdoing as engaging with that question. It engages rather with the question of compliance voluntarily. That appears to be the basis on which the questions as to confidentiality in ciphering were posed to him [i.e., Dr Kakhi] for the purposes of preparing his second report…. He was not asked and he does not appear to be telling me what the situation would be if the court imposed a confidentiality ring. Similarly, in his earlier report where there is a section….dealing with the question of the impact of this court’s order, it seems to me that he is not dealing with it in relation to the question of risk of sanction in the face of compliance with a court order, but only in relation to the question of enforcement.

87.

The question of risk is…plainly in play on the authorities. I would have expected it to be actively and specifically engaged with on that basis…..It is…still more surprising in that (i) Dr

Kakhi does say that legal compulsion provides an excuse and

(ii)

that question of legal compulsion providing an excuse (when it is an order of the Iranian court) provides the very obvious jumping off point for engaging with the concept of what qualifies as legal compulsion. And yet he does not do so. He does deal…with the effect of the English court’s order; but there he does so only as regards the separate question of enforcement, without dealing at all with the question of the Iranian Courts’ attitude to complying with such an order. Nor does he specifically deal with the question of legal compulsion and comity. The result is that in performing the weighing exercise I lack evidence on this key point from the Bank.”

The inference which the Judge drew (at [88]) from Dr Kakhki’s evidence, particularly given that “it is conceded that compulsion may be taken into account” was that the answer to the question of risk “if grappled with head on, is not one which is entirely coherent with the answers of ‘real risk’ and ‘probability’” which Dr Kakhki had given in relation to voluntary disclosure. The Judge (at [89]) was not “entirely happy” with Dr Kakhki’s evidence in other respects; his evidence as to the bases for claiming illegality appeared to her to have shifted between his two reports, without explanation.

Other materials had been adduced as to risk but the Judge was not persuaded (at [90]) that she could extrapolate from them to infer “such a real risk” were she to make the order. The fact that HMT could not obtain an expert involved an “apples and oranges” comparison; such an expert would be assisting resistance to a claim brought by the Bank in which the Government of Iran through its substantial shareholding had a financial interest; by contrast, compliance by the Bank with a Court order would take place in the context of the Bank doing its best to prosecute “the same claim in the indirect interests of the Government”. Similarly (at [91]), media items as to detentions in Iran were not valid comparisons at all. On the contrary:

“92.

…it seems to me fair to conclude that the position of the Iranian Government vis a vis the Bank does suggest that it would be well placed to offer support in any legal challenge and that, given that non-compliance with the order could only harm the chances of success of the Bank in this important claim which it brings, the Government support would be likely to render the chances of sanction very much less than they would be even in a normal case of compelled production…..

94.

I therefore conclude that the evidence demonstrates the production would be a breach of Iranian law but the factors which I have considered indicate that the risk of prosecution and sanction are not as serious as Dr Kakhi’s indications in relation to what might be called voluntary disclosure. I do not accept that Dr Kakhi’s evidence, which does on reading appear to address voluntary production, should be read as extending without nuance to the facts of this case. I conclude that there is no real evidence evaluating any risk as a result of complying with the order. I do, however, accept his evidence as to extraterritoriality.”

Turning (at [97] and following) to the need for the documents in an unredacted form, the Judge underlined that though not impressed with the evidence of risk, she would not make an order “unless…persuaded that the material contained in the redactions had a relevance”. In this regard, the Judge’s view was that the material would not just be relevant “but may have a probative influence on some issues”. The Judge did not propose to go through all the heads of argument and (at [98]):

“…while not emulating Flaux J [who had conducted an earlier hearing] and being able to think of a ‘multitude of reasons’ off the top of my head, I can even looking at the question fairly critically, see the following.”

First (at [99]), “factors such as reasons for a transaction not completing may well be customer related”. Insolvency was a “prime example”. The Judge disagreed with the Bank’s contention that this was de minimis. Similar issues arose “as regards ascertaining which transactions were affected by which measures: original, Follow-on or Copycat, although to a lesser extent…”.

Secondly (at [100]), “alternative routes or replacement transactions” would be client specific. The Judge was not persuaded by the Bank’s argument that this was overstated. Though ciphering and specific transaction numbers (“UTNs”) had been suggested as the solution to this concern, the Judge regarded this as a dubious assumption. The Judge (at [101]) accepted the submission of Mr Foxton QC (for HMT) that information could be missed if fields were redacted, all the more so where it was sought to extrapolate from a sample.

Thirdly (at [102]), whether a client was “a long-term customer” could be relevant to show how much business had been lost.

The Judge was satisfied (at [103]) the (unredacted) material would be relevant “on numerous fronts”. Primarily (but not exclusively), this related to the first part of the

The Judge’s overall conclusion was as follows: “106. When I balance these various factors, the one against the other, I am satisfied that the appropriate answer is that an order along the lines sought by HMT should be made.

Pausing there, the Bank’s skeleton argument on appeal (pre-dating the involvement of Mr Young QC who did not appear below) appeared to criticise the Judge for producing an ex tempore judgment. That criticism was misconceived and, very properly, Mr Young expressly disassociated himself from it. This impressive judgment was delivered after overnight reflection so that, in any event, it was not, strictly, ex tempore. More importantly, however, Judges are not to be criticised for ex tempore judgments, which have the beneficial result that the outcome is known promptly after the hearing. If there are proper grounds for differing from the judgment under appeal, they do not arise from its delivery “ex tempore”.

In respect of the Iranian (and South Korean) documents, the order provided for the establishment of a “Confidentiality Ring” (i.e., the confidentiality club) and a cipher code for each document disclosed which included redacted information. A “Master List” of the cipher codes was to be maintained by the Bank, which would itself be a confidentiality club document. Schedule 1 to the order dealt, inter alia, with the membership of the confidentiality club. Thus, with the exception of the Court and individuals permitted by the Bank, the Master List was only to be made available to confidentiality club members. “Confidentiality Ring” Members were strictly limited to:

“…Ministers, employees or contractors of HM Government, the Government Legal Department, the Defendant’s expert accountants and counsel instructed in these proceedings.”

That is, it may be observed, a broad membership – but, as already highlighted, the Bank has not sought to advance any proposals or submissions for narrowing the membership. At all events, a “Confidentiality Ring Register” is to be maintained by HMT’s solicitors. A copy of or detail from the Confidentiality Ring Register is to be provided to the Bank “…if agreed by the parties or so ordered by the court”.

The principal issues on the appeal fall conveniently under the following headings:

For the Bank, Mr Young QC submitted that the question was whether the Iranian documents should be produced (1) with customers represented by a cipher system or (2) with customers represented by a cipher system plus a master list – the master list being accessible to the wide membership of the confidentiality club. The background was to be kept in mind: the 2009 Order had been imposed to exclude the Bank from London based international business; the Supreme Court had held the 2009 Order to be unlawful; the Bank sought to vindicate its rights and could only do so before this Court.

The Judge had conducted a balancing exercise but not a legally valid balancing exercise. As to risk, she had misunderstood the evidence and drawn her own inferences straying outside the permissible bounds of the uncontradicted expert evidence. With regard to the question of need, the criterion was the need for the documents in unredacted form (i.e., with a master list) for a fair trial but the Judge had made no mention of that criterion. The Judge’s balancing exercise had thus failed to reflect the components on both Issues I and II.

The relevant risk (Issue I), was the risk of prosecution – not the risk of sanction if prosecuted and convicted. The Judge was obliged to follow the uncontradicted expert evidence of Iranian law and practice but she had, wrongly, imposed her own thinking. At [84] and following, the Judge had simply gone wrong. Her assessment of risk had no foundation in the evidence. Dr Kakhki had “head-on” addressed the treatment in Iranian law and practice of production being made in consequence of compliance with an English court order. Under Iranian law, such production remained a criminal offence, violating the Sharia tenets of privacy in the Iranian constitution. The Judge’s criticisms of the expert had been unfair and unfounded. Unlike the position in some of the authorities, there would be here a real crime with a real risk of prosecution. On the advice received by the Bank, an application to the Iranian court for an order permitting production of the documents in unredacted form would be pointless. Though the English Court had jurisdiction to do so, it would not lightly order a party to do something which constituted a criminal offence abroad.

As to need (Issue II), the HMT submissions did not survive scrutiny; they could all be addressed by production of the documents with a cipher code and in unredacted form. The suggested needs were speculative, “Micawberish lite” and hypothetical. The Bank would in any event not have information relating to the underlying transactions. Further and in any event, the risk was on the Bank, which bore the burden of proof of making good its claim.

On the facts of this case, the exercise of the Court’s discretion (Issue III) permitted only one right decision. The order was contrary to comity. The Bank should not be put in a position where it could only have its rights fairly determined by committing a crime in Iran. With regret but on instructions, Mr Young said in terms that “if push came to shove”, the Bank would not comply with the order. Accordingly, the order as it stood would generate a contempt of court. In this last regard it is also right to record that, through Mr Young, Mr Zaiwalla of the Bank’s solicitors, told the Court that his firm would strongly advise the Bank to comply with the Court’s order.

For HMT, Mr Foxton QC underlined, as to Issue III, that the Judge had made a case management decision after a balancing exercise conducted in the exercise of her discretion; this Court should not readily interfere. The fact that the burden of proof rested on the Bank was neither here nor there; as Mr Foxton put it, HMT was entitled to attack as well as defend.

As to Issue I, the focus was on the actual risk of prosecution. That was a question fundamentally different from the content or interpretation of foreign law. On that question, Dr Kakhki did not have relevant expertise. No evidence had been produced by the Bank of any Iranian prosecution of any person who had given disclosure pursuant to an order of the foreign court. The Judge was aware of the real issue – the actual risk of prosecution. She had not fallen into error. It was further to be remembered that many of the Bank’s customers could be expected not to be Iranian – by way of examples, suppliers under a performance bond or buyers under a L/C. In any event, even if there was a real risk of prosecution, that factor fed into the discretionary balancing exercise; it did not dictate the permissible outcome of the exercise.

As to Issue II, the Bank’s case was driven by its (alleged) loss of customers. Mr Foxton submitted that the dispute could not fairly be tried without knowing who they were, together with when and why business had been lost. Further questions included whether transactions had really not completed or whether other routes had been used and whether the Bank had been involved in an attempt to evade lawful sanctions, before or after 2009. Without a master list, ciphers would not be satisfactory and would be vulnerable to minor variations, front companies, group companies and so on. HMT was entitled to test the very sizeable claim brought against it. Insofar as extrapolation from a sample was involved, the need for customer identities was heightened. As the Judge had observed, citing an observation of Flaux J (as he then was, at an early Case Management Conference (“CMC”) in this case), there were “a multitude of reasons” why customer identities needed to be provided to ensure a fair disposal of the trial.

We were most grateful to both Mr Young and Mr Foxton, together with their respective teams, for their excellent submissions.

I come next to the expert evidence of Iranian law, given by Dr Mohammad M.

Dr Kakhki’s evidence was uncontradicted. HMT was unable to instruct an expert on

Dr Kakhki’s first report was dated 29 September 2017 (“the first report”). He explained the Sharia law nature of the system, a matter of importance as the Iranian Constitution – with its Sharia foundations – protects the right to privacy of Iranian citizens. Dr Kakhki drew attention to criminal penalties contained in the Iranian Criminal Code for professionals and officials who disclose confidential information, where not permitted by law. Against this backdrop, the Bank owed a duty of confidentiality to its customers under the Iranian Constitution. According to Dr Kakhki, the location of the customer (whether within or outside Iran) would make no difference to the duty owed by the Bank. That duty could be overridden under Iranian law but only “…by a domestic court as Iran does not recognise and enforce foreign court orders within the Iranian jurisdiction automatically”.

Dr Kakhki was asked this question:

“Does it make any difference…that Bank Mellat is required to give disclosure and inspection of such documents by an Order of the English Court in litigation which Bank Mellat has brought in England?”

He gave the following extended answer:

“44.

As explained above, the order of a foreign court is not capable of being enforced automatically, and would have to be taken through the domestic judicial process to be first recognised, and then enforced where appropriate….

45.

In principle, there is no bar on the ‘recognition’ of foreign court orders/judgments in Iran but their ‘enforceability’ is contingent on their approval at the discretion of the domestic court through the issuance of an ‘enforcement order’.

…..

47.

According to Article 1295 of the Civil Code, a foreign judgment/document will only be considered as legal if the following conditions are met:

…

 Their content is not in contradiction with public policy or good morals in Iran.

49.

….privacy and confidentiality are key principles of Sharia law, and therefore Iranian law. In my opinion, an order issued by a foreign court for disclosure of client information and/or documents is likely to be found contrary to public order/ morality, per Article 1295, due to the emphasis on the preservation of individual privacy within Islamic doctrine. In my view, a disclosure order, such as that of Flaux J, would not be directly enforceable by the courts in Iran; and as noted, additional scrutiny by the Iranian courts in the context of Iranian principles and policies would likely find such disclosure incompatible with preservation of good morals and public policy and not recognise the order as valid under Article 1295.”

Turning to the question of sanction, Dr Kakhki stated that if the Bank disclosed customer information to HMT without the requisite Iranian court order, “…they will be held liable for criminal and civil law penalties…”. The parties affected, as well as the Public Prosecutor, could initiate court proceedings. The risk of criminal prosecution applied equally to the Bank’s branches, subsidiaries, officers and employees outside Iran.

The question of an English Court order was raised again, as follows:

“Does it make any difference to this question that Bank Mellat is required to give disclosure and inspection of such documents by an Order of the English Court in litigation which Bank Mellat has brought in England?” Dr Kakhki answered in these terms:

“61.

The applicable provisions relating to disclosure are part of Iranian law, and are regarded as public policy by a sovereign state. In the absence of a bi-lateral treaty between the two countries any disclosure based on an English court order would not negate the overriding duty of the bank to preserve confidentiality under domestic Iranian law. Therefore, in my opinion, if Bank Mellat disclosed information based on an English court order they would be held responsible for the breach of their duty.”

The first report concluded with the following paragraph:

“64.

To summarise…Bank Mellat is bound across its branches and subsidiaries to act in accordance with Iranian law, which protects customer privacy in accordance with Sharia principles.

…..these provisions are enshrined within the Iranian Constitution…. Therefore Bank Mellat has a duty to protect the confidentiality of its customers’ details which extends to redacting identifiable information. Even if the court order for disclosure of the information were taken through the court system in Iran it is likely that this information would similarly be ordered to be redacted/concealed considering the fundamental importance and emphasis within Sharia on the privacy, dignity and respect of the individual and their associated rights. I would like to reiterate that the banking system in Iran is fully Sharia compliant and it is…highly likely that an order for full disclosure would be found to be against public order in a country where there is no separation between religion and the state and all laws and regulations are based on fundamental principle of Sharia including the right to privacy.”

Dr Kakhki’s second report was dated 16 May 2018 (“the supplementary report”). By this stage, the question of the confidentiality club had emerged. In that regard, he was asked this question:

“Does Bank Mellat have a right or duty under Iranian law not to provide customer information to HMT through a confidentiality ring, in the manner proposed by HMT in its application? Please identify the relevant right or duty.” He answered as follows:

“5.

…the liability explained in my First Report would still apply despite disclosure being limited to a confidentiality ring. As mentioned previously, Bank Mellat has both a right not to reveal customer information and a duty not to disclose confidential information, and would be at risk of criminal investigation and liability were it to do so. It is not difficult to find examples of arrest and prosecution of Iranian bank employees for privacy law offences. For example, one account from 26^th July 2016 details the arrest of 4 employees for disseminating information regarding over 200,000 bank customers (including names and addresses) without authorisation….”

The source of the example appears to be a BBC website.

Reiterating what had been said in the first report, Dr Kakhki said that the Bank’s duties under Iranian law and the consequences of non-compliance would continue to apply “even in the event that an order for disclosure of confidential information is made by the English Court”. Additional complications arose from the confidentiality club in circumstances where the proposed disclosure of confidential information would be by electronic transmission to servers located outside Iran.

Dr Kakhki went on to express further concern as to whether any of the (proposed) disclosure material might be regarded by the Iranian authorities as “secret data”. The relevant legislation was vaguely worded:

“9.

…it is not difficult or unlikely that the Iranian authorities would define ‘state security’ or ‘national interest’ so as to cover many types of information, including the customer information being considered in the present case. It is….probable that the Iranian authorities would conclude that the disclosure of banking customer information on this scale could potentially lead to widespread distrust of the domestic banking system of which customers would previously have trusted the secrecy and privacy, which could undermine the national interest. These considerations would make prosecution and a significant penalty more likely in my view.”

In relation to the confidentiality club, Dr Kakhki’s conclusions were these: “21. ….revelation to a confidentiality ring as proposed would not eliminate liability and is unlikely to reduce the punishment if investigated. Disclosure of confidential customer information by Bank Mellat to a confidentiality ring in the manner proposed would carry a very real risk of prosecution and potential persecution of the offenders by the Iranian authorities, in addition to the risks to the Bank’s reputation and future business. ….. In summary, under Articles 604 and 648 of the Iranian Penal Code offenders face three months to one year’s imprisonment in addition to payment of a fine and compensation for damages caused. Moreover, as well as payment of damages to affected customers or third parties, the Bank would be likely to lose its licence to operate…..

Finally, the use of ciphers representing customers – without a master list giving HMT access to customer identities – would meet the requirements of Iranian confidentiality law.

There was no significant dispute before us as to the legal framework, covering two broad areas which it is convenient to outline here. The first concerns the approach of this Court to evidence of foreign law. The second goes to the position of this Court where it is admitted or established that an order for production and inspection of documents will involve the party subject to the order in a breach of foreign criminal law.

(1) Evidence of foreign law: The propositions which follow are well-established and emerge clearly from the judgment of Scott LJ, in A/S Tallina Laevauhisus v Estonian State S.S. Line (1947) 80 Ll. L. Rep. 99, at pp. 107-108, together with Dicey, Morris and Collins On The Conflict of Laws (15^th ed.), at paras. 9-015 – 9-016:

(2) Production, inspection and contravening foreign criminal law: For present purposes, we are concerned with a species of confidentiality, namely, where a party to litigation in England asserts a right or a duty to withhold inspection of documents because a failure to do so would give rise to a contravention of foreign criminal law.

In general terms, the right to inspect documents, under CPR Part 31.14 and following, is not unqualified: National Crime Agency v Abacha [2016] EWCA Civ 760; [2016] 1 WLR 4375, at [30]. In that case, I went on to say (at [31]):

“….while disclosure and inspection cannot be refused by reason of the confidentiality of the documents in question alone, confidentiality (where it is asserted) is a relevant factor to be taken into account by the court in determining whether or not to order inspection. The court’s task is to strike a just balance between the competing interests involved – those of the party asserting an entitlement to inspect the documents and those of the party claiming confidentiality in the documents. In striking that balance in the exercise of its discretion, the court may properly have regard to the question of whether inspection of the documents is necessary for disposing fairly of the proceedings in question…..”

Coming closer to the present context, in Mackinnon v Donaldson, Lufkin and Jenrette [1986] 1 Ch 482, the Court discharged an ex parte order and subpoena requiring an American bank, not a party to the action, to produce books and other papers held at its head office in New York. In necessarily obiter observations, Hoffmann J (as he then was) said this (at pp. 494-495):

“…I am not concerned with the discovery required by RSC Ord. 24 from ordinary parties to English litigation who happen to be foreigners. If you join the game you must play according to the local rules. This applies not only to plaintiffs but also to defendants who give notice of intention to defend…..Of course a party may be excused from having to produce a document on the grounds that this would violate the law of the place where the document is kept…..But, in principle, there is no reason why he should not have to produce all discoverable documents wherever they are.”

In Ventouris v Mountain [1991] 1 WLR 607, Bingham LJ (as he then was) underlined (at p.622) that under the (old) RSC Ord. 24 regime, production and inspection were not automatic once relevance and the absence of entitlement to privilege were established. He added this (ibid):

“While the court’s ultimate concern must always be to ensure the fair disposal of the cause or matter, it need not be unmindful of other legitimate concerns nor is it powerless to control the terms upon which production and inspection may be ordered. I would not wish it thought that because, as I conclude, production and inspection may be ordered therefore they must at once be ordered unconditionally.”

Brannigan v Davison [1997] AC 238, a decision of the Privy Council on appeal from the Court of Appeal of New Zealand, concerned a commission of inquiry summoning witnesses to give evidence in circumstances where their testimony was likely to render them criminally liable under foreign law. While the issue related to the privilege against self-incrimination, rather than the inspection of documents, the considerations articulated by Lord Nicholls, giving the judgment of the Board, are nonetheless of interest for the present debate. The “chief strand of reasoning” discernible in the common law privilege against self-incrimination was (at p.249) the “…undesirability of the state compelling a person to convict himself out of his own mouth. There is an instinctive recoil from the use of coercive power to this end.” However, where prosecution under a foreign law was involved, the privilege (if applicable) “…would have the effect of according primacy to foreign law in all cases”. Lord Nicholls continued (ibid):

“Another country’s decision on what conduct does or does not attract criminal or penal sanctions would rebound on the domestic court. The foreign law would override the domestic court’s ability to conduct its proceedings in accordance with its own procedures and law. If an answer would tend to expose the witness to a real risk of prosecution under a foreign law then, whatever the nature of the activity proscribed by the foreign law, the witness would have an absolute right to refuse to answer the question, however important that answer might be for the purposes of the domestic court’s litigation.”

The “opposite extreme” (at p.251) involved the proposition that the prospect of prosecution under foreign law was neither here nor there; the witness would always be required to answer a relevant question in the domestic proceedings, regardless of the likely practical consequences for the witness under foreign law. This, Lord Nicholls stated, “would be a harsh attitude”. He went on:

“It would be a reproach to any legal system. One would expect that a trial judge would have a measure of discretion….”

Morris v Banque Arabe et Internationale d’Investissement SA [2001] IL Pr. 37 concerned a dispute arising out of the BCCI saga. The claimant liquidators commenced proceedings against the defendant French bank, seeking (inter alia) the disclosure and inspection of documents held in France. The defendant bank resisted inspection on the basis that under a French “Blocking Statute”, the production of such documents for use as evidence in foreign legal proceedings was prohibited. The bank argued that the claimants should instead seek the evidence via a letter or request under the 1970 Hague Convention on Obtaining Evidence Abroad.

Neuberger J (as he then was) held (at [50]) that the Court had jurisdiction to order inspection of the documents. Under the CPR, the Court had a discretion whether or not to order a person resident and domiciled in another country to do something which would be a breach of the criminal law of that country (at [59] and following). On that footing, putting aside considerations relating to the Hague Convention (which ultimately did not dissuade him), he had little hesitation in exercising his discretion to do so; the absence of the documents in question (at [68]) would “….very substantially interfere with the liquidators’ ability to pursue the case and would clearly hamper the Court’s ability to try the case fairly”. Although (at [71]), by affording inspection, the bank would be committing an offence under the Blocking Statute:

“and could, at least in theory, suffer the imposition of a penalty, it appears to me that this risk, on the evidence I have heard, is little more, and indeed is probably no more, than purely hypothetical.”

Neuberger J (at [74]) added these observations:

“…in connection with litigation of this sort, involving a substantial sum of money, alleged wrongdoing and in the context of a massive and notorious international financial scandal….[it]…would be highly unusual if the French criminal authorities were to prosecute a party to an action such as this in England, in circumstances where he was required to comply with an order of the Court for production of documents for the purposes of that action. The enforcement of a law such as the Blocking Statute in a case such as this would not correspond with generally accepted notions of comity.”

In Secretary of State for Health v Servier Laboratories Ltd [2013] EWCA Civ 1234; [2014] 1 WLR 4383, proceedings for breach of EU competition law were brought in England. French defendants were ordered to provide further information and specific disclosure; compliance potentially exposed those defendants to criminal prosecution in France under the Blocking Statute. Their appeals to this Court were dismissed. For present purposes, it suffices to refer to the following passage from the judgment of Beatson LJ (at [117]):

“Whether or not compliance with the orders of the English court…is illegal under French law, the English court has jurisdiction to make them as part of the ordinary process of disclosure in civil proceedings because such matters are governed by English law as the lex fori. In the exercise of its jurisdiction, it is legitimate for the court to take account of the real risk of prosecution. On the information available to Henderson and Roth JJ when they made their orders, it cannot be said that their exercise of discretion was flawed in law. First, there is no evidence of any prosecutions under the French blocking statute….apart from that in Christopher X (unreported) 12 December 2007. That was a case in which…the facts were exceptional, involving as they did the use of deception by a French lawyer without the protection of a court order.”

The question of compliance with foreign legal obligations is specifically dealt with in Matthews and Malek, Disclosure (5^th ed.), at para. 8.26, as follows:

“The court may take into account, in deciding whether to order disclosure, the fact that compliance with the order would or might entail a breach of foreign law……. It will…need to be shown that the foreign law contains no exception for legal proceedings, and that it is not just a text, or an empty vessel, but is regularly enforced, so that the threat to the party is real.

Even so, the court has a discretion and, on the basis that English litigation is to be played according to English and not foreign rules, it will rarely be persuaded not to make a disclosure order on this ground. More often than not where foreign law is raised as an objection, any threat of a sanction abroad against the disclosing party is found to be more illusory than real.”

Pulling the threads together for present purposes:

(1) Nature of the Issue: As already foreshadowed, this Issue goes to the actual risk of prosecution of the Bank and/or its employees in Iran. It is important to highlight what this entails. First, it is concerned with the risk of prosecution, rather than the risk of subsequent sanction, if prosecuted and convicted. Secondly, however, the question focuses on the actual risk of prosecution – not on whether the conduct in question discloses a breach of Iranian criminal law, without more.

(2) The Judge’s conclusion: The Judge’s conclusion on this Issue appears from [94] of the judgment: although production of the Iranian documents unredacted would constitute a breach of Iranian law, the risk of prosecution (and sanction) was “not as serious” as suggested by Dr Kakhki in relation to “voluntary disclosure”.

(3) Overview: For my part, I think that the Judge addressed the right ultimate question – the actual risk of prosecution – and came to the right answer. En route, however, there appears to have been a mis-step. My reasons follow.

(4) Dr Kakhki’s evidence: Dr Kakhki’s expertise as to matters of Iranian law was undisputed before the Judge and before us. Properly assessed, Dr Kakhki’s evidence yields the following conclusions:

(5) The judgment: I turn next to the judgment. First, I am satisfied that the Judge addressed the right ultimate question – namely, that going to the actual (or real) risk of prosecution in Iran, as clearly emerges from both [50] and [82] of the judgment – and gave her answer at [94] of the judgment.

Secondly, I would reject the criticism of the Judge that she had misunderstood the relevant risk. Thus, the Judge made express reference to the risk of prosecution (at

Thirdly, I am unable to accept the Bank’s submission that the Judge strayed outside her remit and impermissibly imposed her own thinking in the face of Dr Kakhki’s uncontradicted expert evidence. To my mind, Dr Kakhi’s evidence, uncontradicted though it was, did not preclude the Judge from reaching the conclusion to which she came:

Fourthly, I part company with the Judge in one respect only. At [85] – [87] and [94] of the judgment, the Judge appears to have concluded that Dr Kakhki did not consider the risk to the Bank flowing from a breach of Iranian law consequent upon compliance with an order of a competent Court elsewhere. With respect, this conclusion appears to conflate two separate questions. As to the actual risk of prosecution faced by the Bank, for the reasons already given, I think Dr Kakhki’s evidence was of very limited assistance and the Judge was entitled to reach the conclusion she did. However, also as already explained, I think that Dr Kakhki did indeed address the question of whether the Bank’s breach of Iranian law would be excused by reason of the breach being a consequence of an English Court order. On that question, his evidence was clear: the fact that production of the Iranian documents unredacted was attributable to compliance with a foreign Court order would be neither here nor there in assessing whether or not the conduct in question gave rise to a breach of Iranian law.

(6) Overall conclusion on Issue I: In my judgment, the Judge addressed the right ultimate question (the actual risk of prosecution under Iranian law) and gave the right ultimate answer (more than a purely hypothetical risk but less serious than that suggested by Dr Kakhki). I am not persuaded that the Judge’s mis-step en route – insofar as she thought Dr Kakhki’s evidence was confined to what she termed “voluntary disclosure” – invalidates her decision. Be that as it may, if it did, I would in any event and emphatically reach the same conclusion as the Judge, for the reasons set out above.

(1) Overview: Issue II goes to the other side of the equation, forming part of the discretionary balancing exercise: namely, as the Judge put it (at [82]), “…the utility of the information sought and the prejudice caused by its absence…”. As the Judge made clear (at [97]), there would be no order for production of the Iranian documents unredacted, unless “…the material contained in the redactions had a relevance”; in her judgment (ibid), there was a “very real prospect” that the material would not just be relevant “but may have a probative influence on some issues”.

One of the Bank’s criticisms may be disposed of at the outset; although the Judge may not have referred expressly to the need for the unredacted materials to ensure a fair trial, to my mind it is clear beyond peradventure that this was the criterion she had in mind. Realistically, I am wholly unpersuaded that, when balancing the risk of ordering unredacted production of the Iranian documents against the utility of their production, she had in mind any other criterion.

Before proceeding further, some idea of the scale of the trial should be underlined. We were told that some 2361 transactions were involved. Disclosure has been given in respect of about 635 of those (156 L/Cs, 21 bank guarantees and 458 penalties). Leaving customer identities to one side, disclosure has thus been given in respect of

Based on experience and instinct, my initial reaction, I readily confess, is that the fair disposal of this trial cries out for production of the Iranian documents unredacted. I am in complete sympathy with Flaux J’s observation that there were a “multitude of reasons” for the unredacted production of the relevant customer identities to the confidentiality club.

Mr Young retorted with force that the need for the documents must be analysed and could not be considered in the abstract; a speculative need and a fishing expedition would not do. That may well be right as far as it goes. But it may not go very far.

First, in assessing the need for the unredacted documents, it is necessary to keep in mind the dynamics of a trial and the developments which may take place as preparation becomes more advanced and even in the trial itself. A decision now to proceed by way of ciphers alone, without a master list, would (in practice) be irremediable should it later transpire that customer identities were needed. Secondly, this will be a trial of very considerable complexity involving a massive claim, even after the amendment to which reference has already been made. Thirdly, that complexity will increase insofar as it is necessary (as appears likely at least in part) to extrapolate from sample transactions. When each transaction is to be taken as representative of (say) ten others, it is all the more necessary to minimise the risks of unfairness, in one direction or the other – and the master list would assist in doing so. An extrapolation exercise, with the trial confined to ciphers and without a master list, strikes me as over-ambitious and fraught with risk. Fourthly, while catering for purely (or essentially) speculative needs would count for little weight in the balancing exercise, I would find it troubling to dismiss the need for unredacted Iranian documents as speculative, on no more than the Bank’s say-so. Moreover, it is important to appreciate that the parties are not here concerned with a fishing expedition in the context of disclosure; the issue goes to the mode of production of documents which already satisfy the disclosure criteria. Fifthly, a healthy dose of realism and practicality is called for. Anonymisation almost invariably makes the trial process more difficult and cumbersome. Here, the claim brought by the Bank alleges loss of custom; as a practical matter, whatever the theoretical attractions of proceeding by way of ciphers only, I would be deeply concerned as to the prospect of proceeding without the production of customer identities. There is an element of unreality in attempting to try a claim for loss of custom without knowing who the customers are or were, together with when and why the business was lost.

Overall, there can be no gainsaying HMT’s right to test the very substantial claim advanced against it. Nor can the matter be resolved by leaving it to the burden of proof resting upon the Bank. Quite apart from the inherent undesirability, as the Judge observed (at [103]), of leaving matters to be resolved by the burden of proof if an alternative solution is available, HMT must be entitled to attack as well as defend – as Mr Foxton tellingly put it.

(2) Particular reasons: As it seems to me, the (non-exhaustive) reasons which follow suffice, individually and cumulatively, to demonstrate the need for production of the Iranian documents unredacted, in the interests of the fair disposal of the trial.

First, as the Judge concluded (at [99]), the reasons for a specific transaction not completing might not be attributable to the 2009 Order but instead be customer related – insolvency comprising an obvious example. Prima facie, that conclusion has obvious force. That the matter might turn out to be de minimis (as the Bank contended before the Judge) or that the information might not be of any use (as Mr Young submitted to us, on the ground that there was no register of insolvent Iranian companies), cannot justify preventing HMT from investigating and pursuing this topic – on, necessarily, no more than the Bank’s assertion that it would be de minimis or pointless. For the same reasons, time will tell whether the information sought by HMT is lacking, given that the Bank’s interest must be expected to lie with the documentation rather than the underlying transactions.

Secondly, as Mr Foxton submitted, HMT must be entitled to explore whether transactions did not complete in consequence of the 2009 Order (as alleged by the Bank) or whether they were in fact completed by other routes, with the Bank obtaining a share of the commission or profits. To my mind, this submission is unanswerable. Plainly, any such instances would reduce the Bank’s claimed loss. It is not at all apparent that any such (obviously relevant) inquiry could be conducted by way of ciphers only and without customer identities. As is plain from the Bank’s disclosure to date (to which we were taken in argument) and as is to be expected, the Bank was actively exploring ways of mitigating the impact of the 2009 Order.

Thirdly, given the complexity of the causation issues (flagged much earlier), it will be of obvious relevance to explore which transactions were affected by the 2009 Order and which by other measures, such as the follow-on measures or copycat orders. So too, it will be relevant to explore the impact of different cut-off points, having regard to A1P1 considerations, flagged above. Once again, the difficulties of conducting such an exercise with ciphers alone – and without a master list – are readily apparent.

Fourthly, given that the Bank’s claim is for “just satisfaction” pursuant to the HRA 1998, HMT must be entitled to probe whether any transactions were linked to attempts to evade lawful sanctions – a matter which would or might impact on any sums payable to the Bank. It is unrealistic to suppose that this exercise could be conducted without customer identities, as demonstrated by the gaps in the sample disclosure to which we were referred in this context. Given, which must be anticipated, determined efforts to circumvent sanctions, a master list will be needed and ciphers alone will not suffice. Further, I am unable to accept Mr Young’s submission in this regard, that without a confidential UK list of “suspect entities”, any such information could not be evaluated; I simply do not see why that should be the case. Still further even if production of such a UK list was a realistic possibility (which I am satisfied it is not), any linkages would require customer identities not ciphers alone.

Fifthly, there is the obvious risk that if the proceedings are confined to ciphers, the use of front companies or group companies, together with minor variations in names would be missed.

Sixthly, although it can be said that the need for customer identities is more readily apparent in respect of the first and smaller part of the Bank’s claim relating to specific transactions (itself now put at US$31.7 million plus) rather than the second larger part of the claim based on loss of market share (put, in respect of L/C’s at over US$800 million), it cannot be gainsaid that the availability of customer identities might demonstrate links between the two. As the Judge observed (at [104]) the fact that no links had been made so far was not a powerful point; the entire complaint was that “this information is needed to …make such links meaningfully”.

It follows that the Judge was entitled to reach the conclusion she did under Issue II; in any event, I agree with that conclusion.

It is now time to consider the balance struck by the Judge between the conclusions as to Risk (Issue I) and Need (Issue II). For my part, I am amply satisfied that the Judge exercised her case management discretion lawfully and appropriately (at [106]) in concluding that an order should be made for production of the Iranian documents, unredacted but safeguarded by the confidentiality club provisions, along the lines sought by HMT. If, however, as Mr Young submitted, it is necessary for this Court to exercise the discretion afresh, then I would exercise it in the same manner as did the Judge, on the fact specific conclusions as to Risk and Need outlined above. My conclusion is not reached lightly; it takes comity very much into account; it neither intends nor entails any disrespect for the relevant principles of Iranian law – and it reiterates a suggestion (hitherto apparently disregarded by the Bank) as to mitigating a possible area of concern. I elaborate in the paragraphs which follow.

First, the order was not made lightly by the Judge and is not lightly upheld in this Court. It must be recognised that the 2009 Order, as with any provision for Orders, was designed to affect the Bank’s ability to do business. The Supreme Court then held that Order to be unlawful. Even if it is too simplistic to say, as Mr Young did, that the (Judge’s) order compels the Bank to commit a crime in Iran, it is apparent that if the Bank wishes to pursue its claim against HMT, it must do so in this Court. It is therefore necessary to be mindful of the Bank’s concern that compliance with the order will involve the commission of a criminal offence in Iran. Comity requires no less. There is no doubt that Cockerill J had the Bank’s predicament well in mind and so do I. It is, however, to be underlined (as already discussed) that an actual risk of prosecution of the Bank is not, ipso facto, determinative of the balancing exercise but is to be taken into account as part of it.

Secondly, in assessing the degree of actual risk of prosecution, it is both appropriate and necessary to take into account the limitations of Dr Kakhki’s evidence (set out above and which need not be repeated here). If necessary to go further, there are attractively rational reasons for taking a guarded view as to the actual risk of prosecution under Iranian law. So:

Thirdly, the English Court order for the production of the Iranian documents already makes provision for the confidentiality club. To repeat, the Iranian documents will not be ventilated, unredacted, in open Court. That is not the issue; the parties divide on the different question as to the provision of the master list to the confidentiality club. If it is the Bank’s view that the safeguarding provisions of the confidentiality club are inadequate, there is no reason why – even now and even though, hitherto, the Bank has declined to engage in this regard - its legitimate concerns as to the breadth of the membership of the confidentiality club could not be further considered. While any such debate cannot be pre-judged, there is apparent scope for an argument that membership of the confidentiality club could be restricted (for example) to those directly involved in the litigation. It will be recollected that the master list (containing the key to the ciphers) is only available to members of the confidentiality club and that in open court ciphers only would be used.

Fourthly, the need for production of the Iranian documents in the manner set out in the order, intends and entails no disrespect for the relevant principles of Iranian law as to customer confidentiality. Instead, the need flows from the procedural requirements of the lex fori, to ensure a fair disposal of the trial. In accordance with English law and international norms, that is, necessarily, a matter for the lex fori. The Judge proceeded in just this fashion and her approach cannot be impugned. The scale of the redaction sought by the Bank must be emphasised. It does not go to an isolated document or even to a small number of documents; it relates to (it would seem) thousands of Iranian documents, notwithstanding the confidentiality club provisions - and the absence of a master list would dramatically impact upon the entire conduct of the trial. There is no reason to suppose that an Iranian Court or the Iranian prosecuting authorities, on the basis of mutual respect, would not appreciate the need for the production of the Iranian documents as contemplated by the order.

Fifthly, having regard to the degree of risk (the actual risk of prosecution in Iran) on the one hand and the scale of redaction sought on the other, and should all overtures to mitigate any legitimate concerns as to the safeguarding provisions of the confidentiality club be rebuffed, then I am in no doubt that the needs of the litigants -

Sixthly, notwithstanding the courtesy with which Mr Young indicated, upon instructions, that (“if push came to shove”) the Bank would not comply with the order (if upheld), any such statement contains an inherent element of threat. It is thus, despite Mr Young’s best efforts, deeply unattractive. This Court does not contemplate that its orders would not be obeyed and is unmoved by this indication on behalf of the Bank. How the matter would be dealt with if the order is disobeyed will be resolved if or when necessary. But, as already emphasised, there is no need for it to come to that.

For the reasons given, I would uphold the order made by Cockerill J and dismiss the appeal.

In the course of argument, the Court expressed its concern as to the case management of this immensely complex trial, with the trial date now rapidly approaching. We were told that no Judge had yet been designated and that the next CMC is scheduled for 9 May. Case management of the Commercial Court proceedings is not for us, but we draw the attention of both the Judge in Charge of the Commercial Court and the President of the Queen’s Bench Division to our concern. We express and imply no criticism but, on the face of it, designation of a Judge to grip these proceedings from now up to and including the trial is overdue and a 9 May date for the next CMC strikes us as worryingly late.

I agree.

I also agree.