Dahabshiil Transfer Services Ltd. v Barclays Bank Plc

The basic issue on these applications for interim injunctions, which I heard on 15 and 16 October 2013, is whether the defendant Barclays Bank Plc (“Barclays”) should until trial or further order be restrained from terminating the supply of banking services to the claimants, or (which in substance amounts to the same thing) should be ordered to continue to supply them with such services. Each of the claimants carries on a “money service business”, defined in the Money Laundering Regulations 2007 (SI 2007 No. 2157, Reg. 2(1)) as:

“an undertaking which by way of business operates a currency exchange office, transmits money (or any representations of monetary value) by any means or cashes cheques which are made payable to customers.”

Following an internal review in late 2012 and early 2013, Barclays decided that it wished to reduce its exposure to this sector. The implementation of that decision forms the background to the present applications.

There is no dispute that Barclays is contractually entitled to terminate its provision of banking services to each of the claimants. Like any other private business, Barclays is entitled to choose its customers. Although heavily regulated in the public interest, banks are under no public law duty to make their services available to particular categories of customer. The injunctions which the claimants seek do not depend on any actual or threatened breach of contract by Barclays. Instead, the claimants contend that by giving them notice of its intention to withdraw banking services from their businesses Barclays has acted (or is threatening to act) unlawfully, because (put shortly) Barclays is alleged to be in a dominant position in the market for the provision of banking services to money service businesses, either generally or in relation to the particular sector in which the relevant claimant operates, and by ceasing to provide such services without objective justification Barclays would be abusing its dominant position contrary to Article 102 of the Treaty on the Functioning of the European Union (“TFEU”) and the Chapter II prohibition in the Competition Act 1998. That is the sole basis on which the claimants say they are entitled to interfere with the contractual freedom which Barclays would otherwise undoubtedly enjoy to terminate their banking relationship.

Although I heard the three applications together, and they raise a number of common or related questions of law and fact, I should stress at the outset that one of the claims is entirely separate from the other two.

Dahabshiil Transfer Services Limited (“Dahabshiil”) has for many years carried on an international money transfer and remittance business. The main focus of its activities is the Horn of Africa and, in particular, Somalia where it has over 280 payment outlets. Dahabshiil issued its claim form against Barclays on 24 September 2013, seeking a declaration, damages and injunctive relief. Its case was fully pleaded in the attached particulars of claim, settled by leading and junior counsel (Mr Alan Maclean QC and Ms Sarah Love). On the same date Dahabshiil issued an application for injunctive relief, returnable on 30 September (the last day of the Long Vacation). Dahabshiil’s solicitors are Shepherd and Wedderburn LLP.

The other two claimants, Harada Limited (“Harada”) and Berkeley Credit and Guarantee Limited (“BCG”), are commercially related to each other and (as I understand it) under common ultimate management and control, although they are incorporated in different jurisdictions (Harada in the Republic of Ireland, BCG in the British Virgin Islands) and they carry on separate businesses. Harada is a medium sized provider of bureau de change facilities to members of the public, operating from six locations in Central London and Oxford. BCG’s main business is pawnbroking, which it carries on at some 12 shops in and around London. Pawnbroking is not a money service business, but some of BCG’s shops also offer bureau de change services, although these form a relatively small part of the company’s overall business. Both companies have their UK office at 85 Cromwell Road, London, SW7.

On 23 September 2013 Harada and BCG, through their London solicitors Cubism Law, issued an application notice in the Chancery Division seeking an interim injunction requiring Barclays to reinstate the provision of banking services to Harada, and to continue providing services to BCG, in each case pending trial or further order. The application was again made returnable on 30 September. No claim form had yet been issued, but the application notice said that the order was required “to prevent [Barclays] from abusing a dominant position in a market or markets for the provision of banking services to money service businesses and to prevent the Claimants from suffering irreparable loss pending trial”. The application was supported by a witness statement of Jennifer Pollock, who is the legal officer for both companies with responsibility for their legal and compliance issues.

On 30 September the two applications came before Warren J, who ordered an expedited hearing to be listed on 15 October for half a day, with a timetable for evidence in the meantime. Barclays agreed to reactivate Harada’s account, and to continue to provide all three claimants with the same banking services as before, until close of business on 16 October.

By way of evidence, Dahabshiil principally relies on two witness statements of its chief executive officer, Mr Abdirashid Mohamed Duale, and an expert report on definition of the relevant market by Dr Iestyn Williams, a professional economist and partner at RBB Economics LLP, a consultancy specialising in competition economics analysis. Further evidence in reply was filed on behalf of Dahabshiil by Mr Adam Matan, who is a director of the Anti-Tribalism Movement, a registered non-profit organisation based in London and Mogadishu that provides educational and other services to the Somali community; by Mr Abdul Rahman Awl, the vice chairman and a non-executive director of Dahabshil Bank International; and by Mr Ajay Chowdhary, who is a strategic adviser for Dahabshiil.

The evidence on behalf of Harada and BCG comprises, apart from the statement of Ms Pollock which I have already mentioned, three statements by Mr Costas Haji-Georghiou, who describes himself as the director of BCG.

Barclays relies on the same evidence in answer to both applications. It is contained in: (a) two witness statements of Mr Andrew Reid, who is a managing director at Barclays and global head of its Non Bank Financial Institutions (“NBFI”) team; (b) a statement by Ms Barbara Ann Wastle, who is the Head of Financial Crime for EMEA at Barclays; (c) an expert report on the money laundering and regulatory risks involved in the provision of banking services to money service businesses by Mr Brian Dilley, a partner in KPMG LLP who is the global head of anti-money laundering (“AML”) services and a Fellow of the Institute of Chartered Accountants: and (d) an expert report by Mr Alan Overd, a vice president in the European competition practice of Charles River Associates, which answers Dr Williams’ evidence on a provisional basis and expresses some views on whether Barclays’ behaviour constitutes an abuse of a dominant position.

The material contained in the second exhibit to Mr Reid’s first statement, which relates to the internal review conducted by Barclays, is confidential, and its disclosure is confined to a confidentiality ring agreed between Barclays and Dahabshiil. For the purposes of this judgment it is unnecessary for me to refer in any detail to the contents of this exhibit, and I will therefore confine my references to general summaries of matters contained in it, or to specific matters which were mentioned in open court without objection from Barclays.

Thus it was that the two applications came on for hearing before me on 15 October. Dahabshiil was represented by Mr Maclean QC and Ms Love, while Harada and BCG were represented by Mr Timothy Mousley QC and Miss Sukwinder Dhadda. Barclays was represented by Mr Mark Brealey QC and Ms Sarah Ford, instructed by Simmons & Simmons LLP. The half day time estimate proved to be seriously inadequate, and the hearing in fact lasted for one and a half days, at the end of which I reserved judgment. During the hearing, Barclays agreed to continue to provide banking services to all three claimants on the existing basis until at least one day after I gave judgment.

By an apparent oversight, which I confess I found surprising, Harada and BCG had still neglected to issue a claim form by the date of the hearing. I required this omission to be rectified before the close of business on 17 October. This was duly done, and on 18 October I received a brief Note from Barclays’ legal team commenting on Harada’s and BCG’s claim form and particulars of claim, which in turn prompted a Note in response from Dahabshiil on 21 October.

I have already quoted the definition of money service business (“MSB”) in the Money Laundering Regulations 2007. Under the 2007 Regulations, the Commissioners for Her Majesty’s Revenue and Customs (“HMRC”) are designated as the supervisory authority for MSBs and are required to maintain a register of them. HMRC’s AML annual report to the Treasury for 2010/11 states that the number of MSBs regulated by HMRC as at 31 March 2011 was 3,633. Before undertaking its internal review in late 2012, Barclays provided corporate banking services to 414 MSBs, or approximately 11% of the total number registered with HMRC in March 2011.

For internal purposes, Barclays categorises MSBs as undertakings which carry out any one of the following six types of activity:

It is common ground that the MSB sector is a high risk one, and is subject to regulation in many jurisdictions including the UK, the USA and the EU. Two particular risks which are widely recognised within the sector by regulators and law enforcement agencies are the risk of money laundering and the risk of terrorist financing. A global financial institution such as Barclays is subject to intensive and increasing regulation in these areas, and this impacts on the compliance costs of providing banking services to MSBs, particularly the smaller ones.

The relevant regulatory regimes are outlined by Ms Wastle in her evidence, and she explains that Barclays has classified all MSBs as high risk in accordance with its risk scoring criteria. According to her evidence, the two main areas of the MSB sector that are of particular concern to law enforcement agencies are money transmitters and bureaux de change. Moreover, failure by a financial institution to comply with its regulatory obligations can prove very expensive, as well as generating bad publicity. As Ms Wastle records, in 2012 the US Senate published a report which was heavily critical of HSBC’s money laundering controls, and this culminated in a settlement of cases brought against HSBC by the US authorities for no less than $1.9 billion. She goes on to say:

“In the UK, from 2012 to the present date, the FCA has fined five firms for having inadequate controls to prevent money laundering, the largest of which was an £8.75 million penalty (after early settlement discount) imposed on Coutts and Company. In addition to taking action for breaches of its own rules, the FCA can impose civil penalties (pursuant to Money Laundering Regulation 42). The Financial Services Authority (being the predecessor of the FCA) utilised this power to fine RBS £5.6 million for sanctions systems and controls weaknesses in August 2010.”

After its settlement with the US authorities, HSBC withdrew from offering any banking services to MSBs. This led in turn to an increase in the number of MSBs wishing to bank with Barclays, and a corresponding increase in the potential risks faced by Barclays.

It was against this background that Barclays began its review of the MSB sector in late 2012. The decision to undertake the review was approved by senior management. All the relevant departments of Barclays participated in it, although the relationship managers for individual MSB customers were not consulted or informed about the review until it had been concluded. Three options were considered: first, to maintain the current model; secondly, a strategic tightening of the criteria and risk framework surrounding Barclays’ exposure to the sector; and thirdly, to exit from the sector altogether. The second option was the one chosen.

In carrying out the review, Barclays recognised that the risks from providing services to money remittance MSBs were higher than those from the provision of services to other types of MSB. The main reasons for this perceived higher level of risk were:

As a result of this analysis, says Mr Reid, Barclays identified the type of remitters which it felt represented a lesser degree of risk. These were, in general, those which operated on a sufficiently large scale, and had a large enough turnover, to allow them to implement and monitor a robust system of controls. Barclays was willing to retain and strengthen its relationship with such customers, particularly if they were themselves quoted on an exchange or owned by a financial institution. Conversely, the customers with the highest perceived potential for risk would be removed.

The proposed new criteria were set out in a draft “Minimum Standards” document, which was prepared in March 2013 and finalised following further discussions on 1 May 2013. One of the minimum requirements at the draft stage was that the customer should yield actual or potential annual revenue to Barclays of at least £100,000. This criterion was removed from the final draft, on the basis that it was not a financial crime factor. Nevertheless, it remained a consideration to which the review team still had regard, and the view was taken that it would not be commercially viable for Barclays to continue to provide services to any customer representing less than £100,000 in annual revenue.

Another minimum requirement was that the customer should have a specified level of net tangible assets. This hurdle was deliberately set high, and eliminated many of the smaller MSBs and most remitters. The minimum level of net tangible assets for MSBs carrying out money remittance services was set at £10 million. According to Mr Reid, the purpose of this “was to ensure that Barclays could have comfort that the MSBs involved would be in a position to resource a strong AML process”. The minimum level of net tangible assets for other MSBs was £500,000 in the client entity or group parent, or £200,000 for existing clients.

The Minimum Standards document also set out an approval process, and made it clear that the MSB Approval Committee retained an “overriding discretion” to approve or reject any prospective or existing MSB clients, regardless of whether the minimum standards had been achieved.

The result of the review was that Barclays decided to continue to provide corporate banking services to 156 MSBs (out of the total of 414), including 19 money remitters (out of a total of 165). Thus the total number of MSB customers was reduced by about 62%, but the reduction in money remitter customers was approximately 88%. Mr Reid’s evidence is that:

“Barclays has not targeted particular MSBs. It has taken a risk based approach and acted in accordance with defined eligibility criteria and by reference to regulatory exposure and the costs of providing services bearing in mind the increased cost incurred as a result of that regulatory exposure.”

It was only when these decisions had been made that the customers’ relationship bankers were informed of the decision to withdraw banking services from them, and notices of termination were sent out in early May 2013 giving them approximately 2 months’ notice until close of business on or around 10 July 2013. This was substantially longer than the minimum one month’s notice in writing required in most of the relevant customer agreements. A number of extensions were then granted to allow time for discussions between the customers concerned and alternative banking providers, but with only very limited exceptions a deadline of 30 September 2013 was then set beyond which no further banking services would be provided to companies which did not meet the new criteria. One of the exceptions was the extension granted to the present claimants, initially to 16 October 2013.

Dahabshiil is a company registered in England and Wales. It is one of the largest African MSBs, and forms part of the Dahabshiil network which includes companies in Dubai and Somalia. The network has over 180 UK branches, operated by independent agents, and more than 400 payout locations across the Horn of Africa, including 286 in Somalia. Dahabshiil has a healthy balance sheet and an annual turnover for the year ended 31 December 2011 in excess of £3 million. The business was originally founded in 1970 by Mohamed Saeed Duale, and over the next 18 years it grew to become the leading remittance broker in the Horn of Africa. In 1988, the business collapsed as a result of the civil war in Somalia, but the business was re-established in London in 1994. Since then the business has steadily grown, and it now processes the majority of all remittances to Somalia from the UK. Its customer base includes individuals (especially migrant workers) and also corporate entities, international aid organisations and charities (including the United Nations Development Programme, the World Health Organisation, Oxfam, Save the Children UK and Action Aid International).

There is no formal banking system in Somalia, so the only way in which money can be transferred to individuals in Somalia is by using the services of MSBs such as Dahabshiil. Each year Dahabshiil transfers approximately £177 million on behalf of individual customers to Somalia, and approximately £48 million on behalf of corporate entities. There are three main ways in which individual transfers are effected. First, money may be deposited by customers at branches across the UK (either in cash or by electronic payment) and then transferred via Dahabshiil’s Barclays account to a local bank in Dubai and onwards to Somalia. This method accounts for approximately £103 million. Secondly, cash may be deposited with agents in London, collected by Moneycorp, and then transferred by Moneycorp (via its own bank account) to a local bank in Dubai and onwards to Somalia. This accounts for approximately £74 million. Thirdly, customers may transfer money at Barclays’ branches across the UK directly into Dahabshiil’s Barclays account, which is then transferred to a local bank in Dubai and onwards to Somalia. This method is predominantly used by corporate customers, and accounts for approximately £48 million.

Dahabshiil has a compliance and AML programme in place to ensure compliance with the relevant legislation and regulations relating to money laundering and terrorist financing. The programme is operated by a dedicated staff unit led by a senior board member. Apart from a purpose-built AML compliance system, which meets all relevant international regulatory standards, the company also uses sophisticated software called AML Compass to monitor lists of suspect individuals and organisations issued by the UN, the EU, HM Treasury, Canada and the US Office for Foreign Asset Control. Dahabshiil is also subject to regular AML audits by HMRC, as well as annual or six monthly AML audits by Barclays. Barclays has provided banking services for Dahabshiil’s money transfer activities for 15 years, but has never raised any concerns or made any complaint about the company’s policies and procedures. On the contrary, Barclays has often acknowledged that the company’s policies and procedures were satisfactory.

On 8 May 2013, without any prior warning, Barclays wrote to Dahabshiil giving notice of the withdrawal of banking facilities with effect from 10 July. The letter began as follows:

“Barclays takes a risk-based approach to its business and continually reviews the areas in which it operates. The Corporate and Investment Banking divisions of Barclays have recently undertaken a review of money services business clients and, as a result, we have amended our acceptance and eligibility criteria across the sector.

Having reviewed your business against the amended eligibility criteria, I regret that I must inform you that the Corporate and Investment Banking divisions of Barclays will no longer be able to provide services to you. As at the close of business on 10 July 2013 (the “Termination Date”), we will close your company’s accounts and terminate any services provided to you. This does not include services provided by the other divisions of Barclays, including Barclaycard and Personal Banking.

We therefore request that you make alternative banking arrangements as soon as possible …”

The letter was signed by Dahabshiil’s Relationship Director, Mr Paul Barraclough, and on 14 May Dahabshiil wrote to him in courteous terms asking for the decision to be reconsidered and outlining a number of reasons in support of the request. Further correspondence then ensued, both through and independently of the parties’ respective lawyers, but Barclays declined either to change its mind or to explain the nature of the assessment which had been carried out and details of the criteria which had been applied. It was only when Mr Reid’s evidence was filed in answer to the present applications that Dahabshiil learnt about the procedure and reasoning on Barclays’ part which I have summarised above.

In view of the clear social and humanitarian benefits associated with an efficient and properly regulated system for the transmission of money to Somalia, it is not surprising that Dahabshiil’s cause, and that of other money remitters from whom Barclays has withdrawn its services, have been taken up by other interested parties and been the subject of political lobbying as well as widespread reporting in the press. Thus, for example, on 2 September 2013 the Overseas Development Institute (“the ODI”) wrote to Mr Antony Jenkins, the group chief executive of Barclays, asking Barclays to reconsider its decision to close accounts with money transfer companies in Somalia. The letter referred to recent research carried out by the ODI on a major cash transfer programme introduced in response to the Horn of Africa famine, and continued:

“Apart from underlining the effectiveness of cash transfers relative to food aid, we found that the risks of money laundering can be substantially mitigated through careful monitoring and engagement with Hawala agents.

As I know you are aware, remittances play a vital role in Somalia. The $1.3bn transferred annually – some US $500m of it through the UK – far exceeds international aid. With acute malnutrition rates among displaced people in South Somalia ranging from 12 per cent in Mogadishu to 19 per cent in Kismayo, any loss of remittances will pose significant food security risks. It will also undermine the efforts of families to keep their children in school. Moreover, remittances are supporting a fragile economic recovery: some 80 per cent of all new business ventures in Somalia are funded by remittances.”

The letter acknowledged the “enormous challenges associated with aid and banking operations in Somalia”, and the “chilling effect across the banking sector” occasioned by the $1.9 billion fine imposed on HSBC, but concluded:

“There will be no winners from the closure of Barclays’ Somali accounts. Desperately poor and vulnerable people will lose a vital source of finance. The international community’s efforts to support recovery and respond to humanitarian emergencies will be compromised. And Barclays will suffer the reputational damage that will come with closure of a vital lifeline.”

In addition to pressure of this kind, the evidence indicates that HM Treasury has taken an interest in the matter, and a round table meeting was organised to attempt to find a solution to the problems faced by money remitters such as Dahabshiil. Unfortunately, however, government-sponsored discussions of the problem have not yet borne fruit, and nobody suggested that they were likely to do so in the near future.

The immediate priority for Dahabshiil, in the light of Barclays’ withdrawal, was to attempt to find replacement banking services. Without such services, Dahabshiil’s business of remitting money to Somalia would be forced to close, or at any rate would be unable to operate in anything like its present form. In Mr Duale’s words, “the impact of Barclays’ decision would be life-threatening for the Dahabshiil business in the UK”. On 19 September 2013, Dahabshiil succeeded in opening three bank accounts with Abu Dhabi Islamic Bank, one of which is a “safeguarding account” which will enable Dahabshiil to retain its authorisation by the FCA. The second is a general client account that will be used for receipt of corporate client funds, while the third is an expenses account. As matters stand, these accounts will only permit Dahabshiil to undertake corporate transactions. Abu Dhabi Islamic Bank will not permit any transactions for individuals, nor will it provide facilities for the cash transactions which currently account for “the significant majority of Dahabshiil’s business”.

According to Mr Duale, Abu Dhabi Islamic Bank is a new entrant to both the UK and the MSB market, and it needs time to better understand the UK MSB market and its regulatory framework before it can reconsider whether it would be willing to provide facilities for individual customer transactions. Meanwhile, Dahabshiil’s ability to offer money transfer services for individuals is limited to those cash transactions for which Moneycorp currently provides a collection and transfer service. This accounts for approximately 32% of Dahabshiil’s total business. Furthermore, Moneycorp has apparently informed Dahabshiil that it will not collect money from Dahabshiil’s agents across the country, but will only deal with cash which it will collect from one of Dahabshiil’s London locations. Nor will it deal at all with any of Dahabshiil’s customers’ card transactions. Mr Duale also says he understands that Moneycorp is currently operating at full capacity, and would not be in a position to provide any further collection and transfer business. Dahabshiil’s existing contractual relationship with Moneycorp is structured as a framework agreement, under which Moneycorp has the right to refuse to enter into any contract without giving any reason or being liable for any resulting loss.

I have already described the relationship between Harada and BCG, and the basic nature of their respective businesses, in paragraph 5 above. Harada has provided bureau de change services since 1997. As a MSB, it is registered with, and regularly inspected by, HMRC. Harada has had a bank account with Barclays since about December 2011. BCG has provided pawnbroking services since 1988, and has also had bank accounts with Barclays since around that time. It is a member of the National Pawnbrokers’ Association, and is also regulated by, and holds licences from, the Office of Fair Trading and (in respect of its bureau de change business) HMRC.

According to the uncontradicted evidence filed on their behalf, both companies have been model customers of Barclays, and in the case of BCG for some 25 years. There are no outstanding loans, fees or charges for any of the bank accounts operated by either Harada or BCG. Harada is a more recent customer of Barclays, but in January 2013 Barclays carried out a regular on-site audit of Harada’s operations, as it does for all its MSB customers. The audit was very detailed and extensive, and was followed by a due diligence exercise which continued until March 2013. Compliance with the exercise cost the companies around £20,000. It was concluded to the entire satisfaction of Barclays. Harada was also the subject of an annual inspection by HMRC in May 2013, which again raised no adverse issues.

Despite this apparently exemplary history, on 9 May 2013 Barclays served a notice of termination on Harada. The letter was in similar form to that sent to Dahabshiil. Although addressed to the directors of Harada, the letter informed them that banking services would also be withdrawn from “the companies listed in the attached schedule”, which included BCG. This was apparently not noticed by Harada, and no similar letter was sent by Barclays to the directors of BCG.

The notice of termination came as a complete surprise to Harada, and also caused it great concern as Harada’s only bank account is with Barclays. As in the case of Dahabshiil, correspondence ensued in which Harada tried in vain to obtain details of the eligibility criteria which Barclays had applied and to address any concerns which Barclays might have.

On 13 August 2013 Barclays closed the bank accounts of both Harada and BCG. So far as BCG was concerned, this action was completely unexpected, because no separate notice of termination had been served on BCG, and the previous discussions with Barclays had been about Harada’s account. BCG complained vigorously to Barclays, and on 16 August Barclays agreed to reinstate BCG’s account for a further month until close of business on 16 September. BCG was informed that no further extensions would be granted. In the light of further correspondence, however, there was a further agreed extension to 30 September, by when Harada and BCG had issued the present application.

Although Harada was left without any banking facilities on 13 August, it was able to continue trading on a temporary basis by making use of BCG’s account. On 1 October 2013, Barclays agreed to reopen Harada’s bank account pending the hearing of this application.

BCG employs around 40 members of staff. It holds long leases on about 12 commercial properties, and has many suppliers and thousands of customers. According to Mr Haji-Georghiou, a closure of its banking facility would mean the collapse of the business. BCG has applied for alternative banking facilities to ten other UK banks, but each of them has refused to open a bank account for BCG, even though BCG was not asking for any loans or overdraft facility, and requires no more than a standard business account.

The criteria which Barclays applied in deciding to close the accounts of Harada and BCG were revealed for the first time in Mr Reid’s first statement. His evidence (which has not been challenged) is that each company’s net tangible assets are well below the £200,000 threshold which was applicable to it. In addition, neither company meets the criterion of generating annual income for Barclays of at least £100,000.

Neither company has adduced any detailed evidence of its financial position, or exhibited any accounts or financial statements. Mr Haji-Georghiou says that Harada’s “annual revenues” are approximately £2.6 million, while those of BCG are approximately £2.5 million. Harada’s services are mainly provided to individual tourists exchanging foreign currencies for sterling. The average value of these transactions is about £200, although I was informed (in response to a question from the Bench) that the maximum limit for a single transaction is £20,000.

Section 18 of the Competition Act 1998 provides as follows:

“(1)

Subject to section 19, any conduct on the part of one or more undertakings which amounts to the abuse of a dominant position in a market is prohibited if it may affect trade within the United Kingdom.

(2)

Conduct may, in particular, constitute such an abuse if it consists in –

(a)

directly or indirectly imposing unfair purchase or selling prices or other unfair trading conditions;

(b)

limiting production, markets or technical development to the prejudice of consumers;

(c)

applying dissimilar conditions to equivalent transactions with other trading parties, thereby placing them at a competitive disadvantage;

(d)

…

(3)

In this section –

“dominant position” means a dominant position within the United Kingdom; and “the United Kingdom” means the United Kingdom or any part of it.

(4)

The prohibition imposed by subsection (1) is referred to in this Act as “the Chapter II prohibition”. ”

Similar provision is made by Article 102 TFEU, which it is unnecessary for me to set out.

It is common ground that Barclays is an “undertaking” within the meaning of section 18(1) of the 1998 Act.

In a case which Mr Maclean QC drew to my attention when making his submissions in reply, Jobserve Limited v Network Multimedia Television Limited [2001] EWCA Civ 2021, Mummery LJ (with whom Longmore LJ and Harrison J agreed) said at paragraph [12]:

“In general, abuse of a dominant position is a complex question of mixed fact and law, which should be determined at trial on the basis of tested oral and documentary evidence and rival submissions, rather than in the summary setting of an application for an interim injunction.”

The context in which that observation was made was an unsuccessful appeal from the grant of an interim injunction, restraining Jobserve Limited from doing various acts alleged by the claimant, Network Multimedia Television Limited, to be an abuse of a dominant position in the relevant market. It was common ground on the appeal that there was a serious question to be tried on the issue of Jobserve’s dominance in the relevant market: see paragraph [4]. The general nature of the alleged contravention of section 18 was Jobserve’s “refusal to deal with existing customers who also deal with Network as a new competitor in the market already dominated by Jobserve”: see paragraph [9]. As Mr Maclean QC pointed out, this shows that a refusal to deal with existing customers is, in appropriate circumstances, at least arguably capable of amounting to a contravention of section 18.

There was no disagreement between the parties about the relevant principles which the court should apply in deciding whether or not to grant an interim injunction. They are the familiar American Cyanamid principles: see American Cyanamid Co (No. 1) v Ethicon Limited [1975] AC 396 at 407-408 per Lord Diplock. Mr Maclean also reminded me of the illuminating discussion by Lord Hoffmann, delivering the opinion of the Privy Council, in National Commercial Bank Jamaica Limited v Olint Corpn Ltd (Practice Note) [2009] UKPC 16, [2009] 1 WLR 1405, at paragraphs [16] to [20]. In particular, Lord Hoffmann said at [16]:

“The purpose of such an injunction is to improve the chances of the court being able to do justice after a determination of the merits at the trial. At the interlocutory stage, the court must therefore assess whether granting or withholding an injunction is more likely to produce a just result.”

He went on to say at [17]:

“The basic principle is that the court should take whichever course seems likely to cause the least irremediable prejudice to one party or the other.”

Lord Hoffmann then explained that this underlying principle is the same, whether the injunction is prohibitory or mandatory in form:

“19.

… In both cases, the underlying principle is the same, namely, that the court should take whichever course seems likely to cause the least irremediable prejudice to one party or the other … What is true is that the features which ordinarily justify describing an injunction as mandatory are often more likely to cause irremediable prejudice than in cases in which a defendant is merely prevented from taking or continuing with some course of action: see Films Rover International Ltd v Cannon Film Sales Ltd [1987] 1 WLR 670, 680. But this is no more than a generalisation. What is required in each case is to examine what on the particular facts of the case the consequences of granting or withholding of the injunction is likely to be. If it appears that the injunction is likely to cause irremediable prejudice to the defendant, a court may be reluctant to grant it unless satisfied that the chances that it will turn out to have been wrongly granted are low; that is to say, that the court will feel, as Megarry J said in Shepherd Homes Ltd v Sandham [1971] Ch 340,351, “a high degree of assurance that at the trial it will appear that the injunction was rightly granted”.

20.

For these reasons, arguments over whether the injunction should be classified as prohibitive or mandatory are barren: see Films Rover [1987] 1 WLR 670, 680. What matters is what the practical consequences of the actual injunction are likely to be.”

It should be noted that the requirement for “a high degree of assurance” that the injunction was rightly granted applies to the position of the party enjoined, which in the present context would be Barclays. It is not a hurdle which has to be overcome by the applicant in every case where a mandatory injunction is sought. It comes into play only where the court is satisfied that the grant of the injunction is “likely to cause irremediable prejudice to the defendant”.

The first question which I have to consider is whether there is a serious issue to be tried as to whether Barclays has a dominant position in the market or markets upon which the claimants rely. Unless this question is answered in their favour, the applications for interim relief cannot get off the ground.

Before proceeding further, I need to sound a note of caution about terminology. There is considerable scope for confusion in the differing senses attributed to the term MSB by the parties, both in their pleadings and in their evidence. Thus, for example, Dahabshiil’s pleaded case is that Barclays has a dominant position in “the MSB Banking Market” in the UK, but as paragraph 1 of the particulars of claim makes clear, “MSB” is here being used in the limited sense of “companies that provide a mechanism to transfer money from one country to another”, or in other words international money remittance businesses. Such businesses form only one sub-category of MSBs as defined in the 2007 Regulations. On the other hand, the belatedly pleaded case of Harada and BCG is that Barclays has a dominant position in “the MSB Banking Market” in the UK, where “MSB” is defined by reference to the 2007 Regulations. In other words, the allegation is of dominance in the entire market for the provision of banking services to MSBs in the UK. It is not alleged that there is a smaller relevant market consisting of the supply of services to bureaux de change. For its part, Barclays uses the term “MSB” in the evidence of Mr Reid to mean undertakings which carry out any of the six types of activity referred to in paragraph 15 above. It is not entirely clear whether these six types of activity, as categorised by Barclays, coincide with the definition of MSB in the 2007 Regulations, although there must on any view be a very considerable degree of overlap.

Dahabshiil’s case: the provision of banking services to money remitters in the UK

Dahabshiil’s submissions on this part of the case are briefly as follows. There are two aspects to the definition of a relevant market in competition law: the relevant product market, and the relevant geographic market. The determination of each is a context-specific exercise that depends on the facts of the individual case. The critical issue, for defining a relevant product market, is the extent to which the goods or services in question are interchangeable: see, for example, Case 6/72, Continental Can, [1973] ECR 215 at paragraph [32]. On this issue, Dahabshiil relies on the expert evidence of Dr Williams, who in his report applies the so-called hypothetical monopolist or “SSNIP” (“small but significant non-transitory increase in price”) test to both demand-side and supply-side substitution in the UK money remittance market.

Dr Williams’ conclusions, on the basis of the evidence currently available to him, are as follows:

“43.

Based on the information provided by Dahabshiil, I think there is unlikely to be a good demand-side argument for broadening the definition of the relevant market beyond the provision of banking services to money remitters.

44.

The case for supply-side substitution is stronger. In particular, many banks that do not currently provide services to money remitters would seem technically capable of providing the range of services demanded. The key issue is whether they would be willing and able to provide such services. Dahabshiil’s experience suggests that they are not.

45.

I believe that the different risks or profile of risks associated with providing banking services to money remitters is relevant in this respect. Specifically, if commencing services to money remitters would involve significant up-front investment in understanding and managing the new risks, then that would preclude supply-side substitution. Similarly, if incumbent providers would hold advantages in understanding and dealing with these risks then that would militate against effective supply-side substitution too.

46.

A complication is that evidence regarding substitution at prevailing prices may be misleading if those prices are not competitive. Normally the concern is that prices are above the competitive level and that evidence of substitution will overstate the breadth of the market. That is a relevant concern in this case.

47.

However, in the specific circumstances of this case I cannot at this stage rule out the alternative possibility that prevailing prices in the provision of banking services to money remitters have been below competitive levels. In that case, the absence of supply-side substitution at prevailing prices could be consistent with a broader market definition. Nevertheless, if this is the case, I would expect Barclays to be able to provide evidence that substantiates this. The evidence I have reviewed does not do so.”

Part of the evidence relied on by Dr Williams in coming to these conclusions was the first statement of Mr Duale. He points out that the money remittance market is characterised by close regulation by a number of authorities and the high risk profile of the business. As a result, money remitters need to have in place comprehensive AML procedures and need to be aware of regulations and legal restrictions in dealing with money and cash transfers worldwide, especially in “problematic regions such as East Africa”. Mr Duale continues:

“77.

Based on Dahabshiil’s experiences (not least in trying to secure an alternative supplier) it seems to me that these specific regulatory requirements and risk exposures inhibit other banks, which have the technical capability, from entering the [money remittance] market and also constrain banks who currently are in the market from expanding their exposure to it further: e.g. Lloyds Banking Group which is active in the [money remittance] market is not willing to take on any new customers.

78.

To my knowledge, there are currently only three providers of banking services to [money remitters]: Barclays, Lloyds Banking Group and RBS Group. It is my belief, again based on our experience over the past months, that if those three banks were to impose a 5-10% price rise for services for [money remitters] it is highly unlikely that this would lead other banks which do not currently provide such services to enter into the [money remitter] banking services market.”

In the above quotation I have replaced Mr Duale’s references to MSBs and the MSB market with references to money remitters, in order to avoid confusion with the wider definition in the 2007 Regulations.

In my judgment the evidence before me is clearly sufficient to raise a triable issue whether the relevant market is, as Dahabshiil contends, the supply of banking services to money remitters in the UK. If that is right, the next question is whether it is seriously arguable that Barclays had a dominant position in that market when it gave notice of withdrawal to Dahabshiil in May 2013. Again, I consider that this question should be answered in Dahabshiil’s favour, for the reasons which follow.

In paragraph 80 of his first statement, Mr Duale said he understood from discussions with the UK Money Transmitters Association (“the UKMTA”) that Barclays was “by some distance the predominant bank” in the UK MSB market, and that based on best estimates it had a market share of 70%. The remainder of the UK market was served by the RBS Group and Lloyds Banking Group, which were respectively estimated to have market shares in the region of 20% and 10%. Mr Duale also understood from his discussions with the UKMTA that Barclays’ high market share had been stable for a number of years, although it would have increased by approximately 5% when HSBC exited the market in 2012.

Counsel for Barclays make the fair point that this evidence is based on discussions with an unspecified source, that no basis for the estimates is given, and there is no statement from the source. However, a degree of further particularity was provided by Mr Duale in his second statement. In paragraph 43 Mr Duale gives evidence that Joanne McDowall, an Associate at Shepherd and Wedderburn LLP, spoke to Dominic Thorncroft of the UK MTA on or around 16 July 2013. Mr Thorncroft did not have the underlying data, but he provided estimates that there were approximately 240 money remittance firms, and that Barclays’ share of the money transfer business was approximately 70% and had been stable since HSBC exited the market. Furthermore, this estimate appears to tally with Mr Reid’s evidence that Barclays provided corporate banking services for 165 money remittance firms: 165 is 69% of 240.

Dahabshiil also places reliance on a March 2010 report by the UK Department for International Development and Developing Markets Associates Limited, entitled “Supply Side Constraints for Remittance Service Providers in the UK” (“the DFID Report”). The DFID Report includes the following statement:

“It is approximated that 70% of MTOs operating in the UK market bank with one bank. A few years ago there was a problem with smaller MTOs not being able to find a bank to open an account with which presented a major barrier to entry. The problem is still significant because not only is one bank dominant but two of the other banks have taken policy decisions not to offer accounts to MTOs and another has set very high barriers to be overcome in order to open an account. Most of the banks maintain that the MTO sector presents a greater risk because of anti-money laundering threat, than other sectors. The consequence of this is that there is presently very little competition in the market and banking costs for MTOs reflect this.”

This passage needs to be read with the definition of “MTO” in the glossary at the end of the report. Unfortunately, the definition appears to be incomplete both in the copy of the report in evidence, and in the online version of it, but it reads as follows:

“Money Transfer Operator (any business that sends money overseas for an individual; including banks, e-money …”

The definition thus appears to be wider than the class of money remitter MSBs registered with HMRC, and at least in some contexts it may include banks. On the other hand, the focus of the extract from the DFID Report which I have quoted was on smaller money remitters which had difficulty in obtaining banking facilities. In that particular context, it seems to me that the class of MTOs in question was likely to be similar to, if not identical with, the class of money remitter MSBs, and this in turn provides further support for the estimates given to Dahabshiil’s solicitors by Mr Thorncroft.

The figure of 70% is clearly a figure based on the number of money remitters, not the value of their business. But for present purposes a high market share of 70% or more, whether in number or value, is generally considered as strong evidence of a dominant position: see generally Bellamy & Child, European Law of Competition, 7^th edition, at paragraphs 10.021 to 10.023 and Robert O’Donoghue and Jorge Padilla, The Law and Economics of Article 102 TFEU, 2^nd edition (2013), at section 4.2.2 (pp143-151). Furthermore, despite the allegation of dominance in Dahabshiil’s particulars of claim and Mr Duale’s evidence, Barclays has chosen to reveal nothing at this stage about its share of the relevant market. I think I am entitled to infer from this reticence, at least on a provisional basis, that Barclays is indeed the “one bank” referred to in the DFID Report, and that in May 2013 Barclays provided banking services to approximately 70% of UK-based money remitter MSBs.

Mr Brealey QC reminded me, in this context, that the 70% by number will include an unspecified number of money remitters who bank with two or more banks, and that the position in terms of value may also be very different. He further suggested that Mr Thorncroft’s reported estimates might be suspect, since the UKMTA is a trade association lobbying on behalf of its members. I bear these points in mind, but at this early stage they seem to me to cut little ice. There is no obvious reason to doubt the general reliability of Mr Thorncroft’s estimates, given in answer to an enquiry from a solicitor, even if the UKMTA is a campaigning organisation. The other points also carry little weight, given Barclays’ refusal, no doubt for tactical reasons, to disclose anything at this stage about its relevant market share, beyond the bare number of money remitters for which it provided banking facilities.

Mr Brealey also submitted that it is wrong in principle to examine the question of dominance as at May 2013, and that the position should instead be examined today (or, in due course, at trial). He submitted that the picture is a changing one, and that even if Barclays held a dominant position in May 2013, it may not do so today after the implementation of Barclays’ new policy and the withdrawal of banking facilities from most of Barclays’ previous money remitter customers. In support of this submission, Mr Brealey relied on the decision of the Court of Appeal in Passmore v Morland [1999] 1 C.M.L.R. 1129, especially at paragraphs [25] and [26] per Chadwick LJ. I found this a surprising proposition, because it seemed to me wrong in principle that a party could (ex hypothesi) abuse a dominant position by unjustifiably reducing its client base, and then rely on the new state of affairs which it had brought about as negating its position of dominance, or at least as sufficient to preclude the grant of injunctive relief until trial. In my view the facts of Passmore v Morland are readily distinguishable from those of the present case, and even if there may be some merit in Mr Brealey’s argument, I am certainly not persuaded that the position is so clear that there is no triable issue on the question of dominance.

Harada’s and BCG’s case: the provision of banking services to the MSB sector as a whole in the UK

As I have already said, the pleaded case of Harada and BCG is that the relevant market is the provision of banking services in the UK to MSBs as defined in the 2007 Regulations. At the moment, the only MSB activities carried on by Harada and BCG are as bureaux de change, but in the past they have carried on some money transfer services as well, and the evidence indicates that they intend to re-enter the international money transfer sector, in particular between the UK and the Czech Republic.

For the purposes of this application, I have no difficulty with the propositions that the MSB sector as a whole is capable of constituting a relevant separate market for the purposes of Article 102, and that there is a triable issue whether it is the relevant market so far as Harada and BCG are concerned. Indeed, the whole thrust of Mr Reid’s evidence is that the MSB sector is a separate one which poses high risks to Barclays and is subject to extensive and exacting regulatory requirements. Consistently with this, it was the MSB sector as a whole which Barclays reviewed and from which it decided to effect a partial withdrawal.

The evidence as to Barclays’ alleged dominance of the MSB sector as a whole, however, is distinctly more problematical. The evidence indicates that well over 3,000 MSBs are registered with HMRC alone, but Mr Reid’s uncontradicted evidence is that Barclays has only ever provided corporate banking services to 414 MSBs. That is plainly far too small a number, taken by itself, to give rise to any inference that Barclays has a dominant position in the sector. Moreover, it is Mr Reid’s evidence that there are a number of other banks, apart from Lloyds and RBS, which provide banking facilities to MSBs, including Bank of America, BNP Paribas, Deutsche Bank, Citibank, Santander and Clydesdale. This provides an unpromising start for the contention that Barclays is, even arguably, in a dominant position over the MSB sector viewed as a whole.

In his oral submissions, Mr Mousley QC attempted to surmount this hurdle by relying on three main areas of evidence.

First, he relied on the passage which I have already cited from the DFID Report. The trouble with this submission, however, is that the passage relates only to money transfer operators, who on any view are only one sub-sector of the MSB sector as a whole. In Mr Reid’s internal classification, money remitters were only one of the six groups into which MSBs were divided by Barclays. Thus even if it is the case, as I have held to be seriously arguable, that Barclays held a dominant position within the money remittance sub-sector, it cannot safely be assumed that Barclays had a similar dominance in the other sub-sectors. At most, a provisional inference to that effect might be drawn. The inference gains a little further strength from Ms Pollock’s evidence that no research is readily available for the bureau de change segment of the MSB market, “but it is accepted throughout the industry for many years that Barclays is virtually the sole provider of services to the bureau de change operators as well”.

Secondly, reliance is placed on the fruitless efforts made by Harada and BCG to obtain alternative banking facilities. I have already referred to Mr Haji-Georghiou’s evidence about this in his first statement: see paragraph 41 above. In his second statement, he takes issue with Mr Reid’s evidence that several banks are prepared to offer services to MSBs. He gives evidence of approaches made on behalf of Harada to over 30 banks, enquiring whether it would be possible to open a business bank account. Not one of the banks thus approached has yet agreed to do so, and in many cases the enquiry was met with a blanket refusal to deal with MSBs as a matter of general policy. So, for example, on 11 October 2013 the London branch of ING Bank wrote to Harada saying:

“We can confirm that we have not in the past and we do not currently, as a matter of practice, offer banking services to money service businesses.”

Similar answers were given by a number of other banks, although it is fair to say that in some instances the refusal had nothing to do with the MSB sector, or left open the possibility that an application might be considered at a later date. In general, however, I accept the evidence that BCG and Harada have made concerted efforts to obtain alternative banking facilities, and that they have not yet succeeded in doing so.

For present purposes, however, the question is whether the failure to find an alternative bank provides any evidence of probative value that Barclays has a dominant position in the MSB sector. In my view, it does not. All it shows is that many banks have now taken a policy decision not to expose themselves to the MSB sector. In view of the high risks and intensive regulations associated with the sector, there are good commercial reasons why banks may decide to adopt this stance. But responses of this nature to enquiries by a new customer cannot ground a reliable inference that in May 2013 Barclays had a dominant position in the UK MSB market.

In reaching this conclusion, I have not overlooked the evidence of a telephone conversation between Ms Pollock and a representative of Metro Bank on 28 May 2013, when she was informed (according to her contact report) that:

“they do not do this any more – he said that their backing bank (Barclays) had stopped this, we were a high risk, ML etc. Due diligence on a foreign shareholder would be too high. He said they only do this for companies with [directors and shareholders] registered in UK.”

I do not read this as any more than a reference to Barclays’ new policy; and the exception for companies with directors and shareholders registered in the UK shows that, so far as Metro Bank was concerned, the refusal was not a blanket one.

Thirdly, in common with Dahabshiil, Harada and BCG rely on the inference to be drawn from Barclays’ refusal (so far) to disclose its true position in the MSB market. I consider that this is a point to which I am entitled to attach some weight, particularly in the light of Mr Reid’s evidence that a consequence of HSBC withdrawing banking services to MSBs was that Barclays “was being asked to bank many more companies in the MSB sector”. This was, indeed, one of the triggers of the review initiated by Barclays in December 2012. If Barclays were able to demonstrate that it had no position of dominance in the MSB market as a whole, I am left wondering why it has not take the opportunity to do so, if necessary in a redacted and confidential exhibit. The onus is, of course, on Harada and BCG to show that there is a triable issue, and at the stage when Mr Reid filed his evidence they had not even pleaded their case. But the requirement of showing a triable issue is a relatively low threshold, and if a party which could in principle disprove it fails to take the opportunity to do so, it runs the risk of an adverse inference being drawn, albeit on a provisional basis.

I have not found this an easy question, but in the end I am narrowly persuaded, on the totality of the evidence as it now stands, that there is a triable issue whether Barclays had a dominant position in the UK MSB market in May 2013. Barclays’ strongest point, in my judgment, is the apparently modest number of MSBs for which it provided banking services when compared with the total number registered with HMRC. But the position in terms of value may turn out to be very different, and in any event the question is ultimately one that requires a multi-factorial evaluation rather than the application of crude market share statistics. I also bear in mind the wise words of Mummery LJ in the Jobserve case, which I have quoted in paragraph 47 above. It is true that they were directed to the issue of abuse, rather than the issue of dominance (in respect of which Jobserve accepted that there was a serious question to be tried); but the issues are in my judgment of a similar nature, and the same approach should generally be adopted.

I can deal with this part of the case much more briefly. It may at first sight seem counter-intuitive that a party in a dominant position in a market can abuse that dominant position by seeking to reduce its participation in it. This point is duly made by Mr Overd in his expert evidence on behalf of Barclays: see paragraphs 26 and 42 of his report. On the other hand, there is at least some authority that a dominant undertaking may commit an abuse where, without justification, it cuts off supplies of goods or services to an existing customer: see Jobserve at paragraph [9], and the opinion of Advocate General Jacobs in Case C-7/97, Oscar Bronner GmbH & Co. AG v Mediaprint Zeitungs-und Zeitschriftenverlag GmbH & Co. AG and others, [1999] 4 C.M.L.R. 112 at paragraph [43] of his Opinion. This is a question of law for determination at trial. I am satisfied that it is not suitable for summary determination.

As to justification, there is a defence to an allegation of abuse if a dominant undertaking can establish that its conduct is objectively justified and proportionate. The burden of establishing the defence lies on the dominant undertaking. The burden will not normally be discharged if there are alternative, non-abusive solutions to the problem. The law is summarised in this way by O’Donoghue and Padilla, loc. cit., at p 283:

“A dominant firm’s conduct may be justified by objective necessity. The issue is whether the conduct in question is indispensable and proportionate to the goal allegedly pursued by the dominant undertaking. This question must be determined on the basis of factors external to the dominant firm. Exclusionary conduct may, for example, be considered objectively necessary for health or safety reasons related to the nature of the product in question. Thus, in refusal to deal cases, capacity limitations or concerns about quality, security, or safety at a facility may justify a refusal to deal.

Such defences will, however, be scrutinised carefully.”

I am satisfied that Barclays’ defence of justification needs to be fully examined at trial. It is not so clear that I can say with confidence that it is bound to succeed. On the contrary, the claimants have drawn attention to a number of points which raise serious questions about Barclays’ proposed solution to the perceived problems in the MSB sector which led to the internal review. For example, the apparently unfettered discretion reserved by Barclays to override the Minimum Standards in individual cases may be difficult to justify on an objective basis, as may the requirement for a minimum level of customer income, even though it had been dropped from the final draft of the Minimum Standards. Again, the requirement for a minimum level of net tangible assets (set at the very high level of £10 million for money remitters) gives rise to a number of obvious questions which can only be properly explored at trial. I emphasise that I am not saying the defence is in my view likely to fail, but merely that at this preliminary stage I cannot safely assume it will succeed.

Again, I can deal with this part of the case briefly. In the light of the evidence, it seems to me all but self-evident that damages would not be an adequate remedy for any of the claimants, and that the balance of convenience favours the grant rather than the refusal of interim relief. There is a far greater danger of irremediable prejudice to the claimants in refusing the grant of injunctions until trial then there would be in granting the injunctions. As matters now stand, the alternative banking arrangements that Dahabshiil has been able to make are far more limited than those previously provided by Barclays, and the arrangements made with Moneycorp may be precarious. As for Harada and BCG, they have been unable to find any bank willing to offer them standard banking facilities, and the refusal of an injunction would on the face of it either force them out of business or compel a merger with a competitor which still has banking facilities. Conversely, the grant of an injunction will require Barclays to do no more than continue providing banking services to established customers with impeccable records. I bear in mind Mr Reid’s evidence that some concerns have been expressed within Barclays about Dahabshiil; but those concerns have never been raised with Dahabshiil, so Dahabshiil has not had an not opportunity to answer them. On the face of it, the continued provision of banking services to the claimants will be profitable to Barclays, albeit probably not at the level set by the review. In the unlikely event that Barclays suffers any loss as a result of the injunction, all three claimants are willing to offer the usual cross-undertaking in damages. In their skeleton argument, counsel for Barclays assert that there is “a very real question mark” over the ability of the claimants to satisfy a cross-undertaking in damages. I heard no oral argument on this question. If it remains a live issue, it can be raised when this judgment is handed down.

For the reasons which I have given, these applications succeed. The parties should consider what further directions it is appropriate to give at this stage, and whether they wish to ask for an expedited trial. My provisional view is that at least a moderate degree of expedition would be appropriate.