Banking Standards Joint Committee

Banking Standards Joint Committee Contents

6 Structural separation in the first instance

The starting point

79. The Chancellor of the Exchequer contended that not only is there a consensus on the need for structural separation, a matter considered in chapter 3, but that there is also a UK consensus that this Commission should not seek to re-open on what form such separation should take:

I do agree we have a consensus that some form of separation is required. That consensus has been created through the Vickers process, but Vickers specifically looked at what form of separation and specifically addressed the question of complete separation. Indeed, it was part of his remit to look at that issue. He looked at it and rejected it. When asked by this Commission, he rejected it again and, indeed, other people who have come before your Commission have rejected it. So I would say to you that there is a consensus about structural reform and that you need to pull apart investment and retail banking in some way, but we also have in this country a consensus of how that is done.[123]

80. There is widespread, but not universal, support for structural separation in some form. However, views in evidence to the Commission about how separation should operate, where a ring-fence should be placed and indeed whether ring-fencing can achieve the desired policy aims, fell well short of consensus.

Contagion, diversification and cost

81. One argument for full separation, i.e. moving some activities completely outside the banking group containing the core activities, is that anything short of this will not be able adequately to insulate the retail bank from the risk of contagion, in the sense that a loss of trust in an investment bank, should it fail, could lead to a run on the retail bank. Paul Volcker highlighted the importance of a shared brand to a bank's reputation:

If the name is on the institution, or parts of the institution, it will be protected; to the extent possible, each part will be protected. If your name is on the door, you are going to protect it. As I understand it, under the philosophy of Vickers and Liikanen that would not happen. I think Britain is going a little bit uphill.[124]

The ICB, in its final report, argued that reputational links between different parts of the banking group could be a feature of ring-fencing that enhances rather than threatens financial stability.[125] An investment bank, in order to secure its own reputation, may be encouraged to provide assistance to a retail bank that bears its name. Martin Taylor explained that:

One sees banks going to enormous lengths to protect subsidiaries or commitments they have made with their brand on. If you look at this with the object that the investment bank is the only dangerous thing, you can make a bad mistake. I don't terribly like the idea. If we enforced a split in the UK, you would have a rather strange ecosystem with very large, very highly correlated retail banks with no earnings diversification from elsewhere, and I don't think that is a particularly good idea.[126]

A number of banks also emphasised the benefits of having a number of diverse functions or operating in a number of territories, broadly suggesting that this would help to insure them against one narrow area of the business falling into difficulties.[127] Sir John Vickers told us:

I do not believe that we are in a simple world of utility and casino, where the utility is totally safe. There are risks in any form of credit extension and they can be correlated risks, when an economy hits trouble. One, in my view, cannot dismiss the possibility that a stand-alone, undiversified sector would get into trouble. If it got into trouble when the rest of the world or the rest of banking was doing okay, one would have lost a great deal by a full split, because there would not be the group resources to mitigate the losses in UK retail.[128]

82. The Chancellor of the Exchequer argued that the additional costs from full structural separation as opposed to a ring-fence would need to be justified by additional benefits:

there is a very considerable cost to the industry in what we are doing. [...] I have taken a judgment [...] that this is a price worth paying and that it is outweighed by the broader economic benefits that greater stability will bring. However, there is a cost to the industry, and exactly the same Members of Parliament who get up and say, "You must screw the banks down" are the same people getting up and saying, "We've got to get the banks to lend and when are we going to do it?" [...] I think that full separation would be an even greater cost and I'd have to justify it. I would be prepared to do so if I thought it was bringing benefits that outweighed that cost.[129]

Sir John Vickers explained that the ICB's rejection of full separation was founded on the belief that ring-fencing could deliver similar benefits at a lower cost.[130] The contention that the costs of ring-fencing would be lower was set out in the ICB's final report:

there are a number of factors which would lead full separation to be more expensive than ring-fencing. This is principally because ring-fencing preserves those diversification benefits which arise from the ability to move excess capital [...] average analyst estimates suggest [diversification benefits] could be as much as £4bn annually but the empirical evidence is mixed.

In addition, there are other synergies which ring-fencing would preserve but which full separation would not. There may be a valuable benefit to some customers of being able to purchase from a single banking group a range of services which would straddle the divide between retail banking and wholesale/investment banking. Within banking groups, there need not be restrictions on the sharing of information and it may be possible to preserve a greater degree of operational synergies than under full separation.[131]

Martin Taylor referred to how the ICB made a deliberate attempt to preserve these operational synergies where they did not pose a threat:

Where we had a choice, we tended to take the line that would not put extra cost or inconvenience on the banks. For example, we allowed the ring-fenced bank and the non-ring-fenced bank to share treasury functions and IT functions. This saves these banks a lot of money.[132]

Culture and standards

83. Paul Volcker considered that the cultural advantages of structural separation would best be secured through full separation.[133] The ICB did not consider standards and culture as part of their remit, and it should not be expected that their solution would necessarily be that which best addresses the wider problems in banking standards. Nevertheless, Sir John Vickers argued that they did examine cultural issues:

We certainly gave thought to questions about culture. "Culture" was the word we used more than "standards", and it was before some of the events of this year. Our view was that, given the questions that we had been set, it was not for us directly to seek to regulate cultural standards, which is a difficult thing in any case, but that the issues that we were looking at, both on structure and on loss absorbency and on the competition and consumer side, all had a very clear bearing on questions of culture and, as we might now say, of standards.

I do not believe that the recommendations that we made on the questions that were put to us would have been materially different if standards had been explicitly among, let us say, the issues that we were to have regard to.[134]

Permeability

84. One of the arguments for full structural separation compared with a ring-fence is that full separation would entail fewer rules and therefore less monitoring and enforcement, because the two entities would be separately owned and would have no more incentive to create interdependencies than any other two banks. Paul Volcker's main doubt about the effectiveness of a ring-fence was that they "tend to be permeable over time".[135] He added:

If you really want to separate some operations very clearly and decisively, you put them in different organisations. In my experience, you do not put two functions in the same organisation and say that they cannot talk to each other or interact.[136]

85. Sir John Vickers and Martin Taylor both argued that full separation would not be much simpler than a ring-fence, concentrating on the fact that determining the location of the split would be just as complex for full separation as for a ring-fence.[137] In the words of the latter:

[An] error that people are prone to make is that somehow splitting is simple, and a ring-fence is complicated. In fact, if you are going to split, you have to go through all the complexity that we have gone through with ring-fencing and decide exactly where the split should come. You would have just as much regulatory complexity, and of course you would also risk putting it in the wrong place.[138]

86. Lord Turner pointed out that Glass Steagall, which is thought of as a full-separation approach, nevertheless eroded away over time.[139] However, Paul Volcker argued that the failure of Glass-Steagall was not because full separation was ineffective, but because the separation became less full, saying "the restrictions between the 'commercial bank' and the 'investment bank' in Glass-Steagall broke down over time. I have a little fear that that might happen in Vickers too."[140]

Market-driven separation

87. It is possible that even if not mandated, some banks may decide to pursue full separation of their retail and wholesale activities voluntarily. A robust ring-fence would retain some synergies between the two banks, but would remove many of the benefits that banks obtain from conducting retail and wholesale activity side-by-side. In such circumstances, shareholders might consider themselves better served by spinning off one of the banks and allowing it to compete without the costs and constraints of the ring-fence. António Horta-Osório said:

if the synergies that customers perceive in having an integrated approach are more than offset by the internal costs that you impose on the organisation, I think that shareholders in the future will say that the synergies do not compensate for the costs, and that banks should spin off their investment banks or their retail banks. They will separate, but it will be a market force separation.[141]

Professor Kay went further, arguing that if a ring-fence really did result in effective separation, then banks should want to split themselves up:

I have thought, and in some ways I continue to think, that the effectiveness of ring-fencing would be demonstrated by whether Barclays wanted to split itself up. If the ring-fence were really effective, they would have little reason to want to maintain that structure. In that world, the interest that certainly the previous management of Barclays would have in the retail side of the activities would probably be rather small.[142]

European law

88. One difficulty that the imposition of full structural separation might pose is that it may conflict with European law. The final ICB report noted that:

full separation would give rise to legal obstacles which are not applicable to ring-fencing because European law places particular constraints on the degree to which ownership of companies can be controlled. Member states can object to the change of ownership of a bank only on certain grounds, and it is far from clear that these would enable the authorities to prevent the acquisition of a UK-incorporated retail bank by a European universal or wholesale/investment bank. [...] while it might be possible to secure changes to the relevant EU law, there seems little reason to pursue this difficult and uncertain course given that the merits of the economic arguments do not clearly favour full separation."[143]

89. Martin Taylor referred to "the deep difficulties under European law of mandating full separation" as a factor in the ICB's deliberations when he gave oral evidence.[144] The limitations on change of ownership are set out in Article 19(1) of the Banking Consolidation Directive (2006/48/EC), which was incorporated into that Directive by the Acquisitions Directive (2007/44/EC), and which was transposed into national law by sections 185 and 186 Financial Services and Markets Act 2000 (FSMA). The provisions were intended to ensure that acquisitions were blocked only on strictly prudential grounds, rather than discriminatory grounds. A ban on the acquisition of a retail bank by an investment bank, because it is an investment bank, could be held to breach the obligation on the FSA under section 185 of FSMA to object to an acquisition only on the limited set of prudential criteria set out in section 186 FSMA, which do not include the criterion that an acquirer of a retail bank is not an investment bank.[145] In considering the impact of these current restrictions, it needs to be borne in mind that, if the European Commission puts forward a legislative proposal based on the work of the Liikanen Group, as expected, then this might provide an appropriate vehicle within which to negotiate the necessary changes to EU law if full separation were to be pursued.

International context

90. The ICB remit included a requirement to have regard to the impact of their recommendations on "the competitiveness of the UK financial and professional services sectors".[146] Bill Winters told the Treasury Select Committee in May 2011 that "we spent some time thinking about whether either large universal banks or parts of universal banks were likely to re-domicile".[147] He noted that the ICB made recommendations based on "what we thought the right structure was for the banking industry in the UK, considering financial stability, competition costs, and service to society, not focused primarily on whether banks would re-domicile".[148] In evidence to us, Martin Taylor voiced scepticism over the suggestion that banks would be driven to relocate by the current set of proposals:

I do not believe that we will have wholesale moving of banks' head offices, which is what we were worried about two or three years ago, simply because pretty much the entire world is going in the same direction. I work in Switzerland and spend half my time there. I remember in 2009-10, all the people in the City were saying they wanted to move to Switzerland, and all the Swiss banks wanted to move to London. Each of them was ignorant about the changes taking place on the other side.[149]

Michael Cohrs noted that the UK has a number of distinct advantages as a location for banks which would remain regardless of the UK's approach to financial regulation:

generally speaking, it is really hard—just as it is really hard to separate a bank—for a bank to move its jurisdiction. That is before you get into the cultural issues, which, for a bank, should be very important. So I am dismissive of bankers when they tell me—I used to be one of them—"If you don't give us a good regime, we will go elsewhere." It is rubbish... You know the most important reason? Greenwich Mean Time.... Two, language is critically important. The world has adopted our language; that is very important. Three, our legal system is very clear. It works. People want to litigate in this country. That is a big asset that we have. We should make this into an industry, as a country. We are probably not charging enough for people to come here and use our courts [...] Finally, London is a pretty neat place to live. These people make a lot of money. They want to spend their money in a pleasant place, and London is a very pleasant place.[150]

Conclusions

91. Sir Mervyn King drew the Commission's attention to the fact that he had long supported structural separation:

I have made no secret of the fact, and I have spoken about it for five years, that I have always felt that total separation was the right way ultimately to go. I have been joined in arguing this by some distinguished company, but it has been a lonely and difficult furrow to plough. I am glad that many more people are now coming on board the idea that a move to some kind of serious separation is the right thing to do. Even the Financial Times has now advocated a move in this direction.[151]

However, he went on to say why he strongly supported introducing a ring-fence "now":

I really do not want the last five years of effort to go to waste through the whole issue being kicked into the long grass by not implementing Vickers. We appointed the Vickers Commission in the UK, and I think this is the best-qualified group of people to serve on such a Commission in my lifetime in the UK. These are very impressive individuals, and they have thought about it.[152]

This viewed was echoed more recently by a fellow member of the Financial Policy Committee, Michael Cohrs:

I am not completely pessimistic. I think that Vickers is a step in the right direction. To me, however, it is only a step in the journey because I think a modern Glass-Steagall will ultimately see total separation. [...] I think that we are on a journey, and Vickers is a good path for us to follow.

92. The Chancellor of the Exchequer emphasised in his evidence how a change of approach at this stage would cause significant delay and be hard to justify:

[W]e have reached this point, we have got agreement, we are fundamentally going to change the structure of British banking. We have got that consensus. Let's get on and implement it and legislate for it, instead of getting to the top of the snakes and ladders board and then going all the way down the big snake that takes you to the bottom again.[153]

93. Whatever their views on arguments for and against full separation, which are finely balanced, the majority of witnesses told the Commission that the partial structural separation of the ring-fence would probably bring significant benefits for public policy and for banking. The Commission therefore welcomes the Government's action to bring forward legislation to implement a ring-fence.

94. The ICB's proposals should be the starting point for proposals for legislation for implementation of structural separation. However, that does not mean that they should be the final destination. The current proposals may not be sufficient. In addition to concerns about proprietary trading, the case that a ring-fence will in practice be able to achieve the necessary level of separation remains unproven. The ring-fence may also be tested and eroded over time. The Commission considers it essential that steps are taken to reinforce the ring-fence, and makes specific recommendations to this effect in chapter 9.

95. There is evidence to suggest that proprietary trading, which under the current proposals could still take place within the non-ring-fenced part of banking groups, is an activity which is incompatible with maintaining the required integrity of customer-facing banking and which could have harmful cultural effects if permitted to continue. This was the primary concern of Paul Volcker in suggesting the prohibition of such activity in US banks.

96. The Commission has not considered fully the ramifications and practical issues of supplementing the proposed UK ring-fence with something akin to the Volcker rule. The Commission intends to take further evidence on this in the New Year. The Bill which the Government will shortly introduce provides the appropriate vehicle for establishing the future structural form of the UK banking industry.

97. The Commission will consider further the implications of introducing a prohibition on groups containing a ring-fenced bank from engaging in proprietary trading and, in particular, the contribution such a prohibition could make to the changes needed to banking culture and standards. The Commission expects to report in good time in order that legislative effect to any recommendations can be given as the Bill progresses.

98. Measures to tighten the regulation of UK banks beyond international norms should be assessed for their potential to cause an unwelcome shift of activity abroad. However, concerns about relocation of banks may be over-stated. They should not be allowed to dominate the decision on the measures necessary to remove the implicit guarantee and ensure the banking system serves the UK economy. We will address this in our final Report.

123 Q 1066 Back

124 Q 53 Back

125 Independent Commission on Banking, Final Report, September 2011, p 63 Back

126 Q 369 Back

127 Q 860 [Peter Sands]; Ev w30 [Barclays] Back

128 Q 760 Back

129 Q 1066 Back

130 Q 762 Back

131 Independent Commission on Banking, Final Report, September 2011, p 64 Back

132 Q 395 Back

133 Qq 55 and 62 Back

134 Q 746 Back