Panel on Regulatory Approach

Written evidence submitted by Andy Haldane (SE002)

At my hearing before your Panel of the Parliamentary Commission on 21 January, I was asked for a series of follow-up written responses to questions on remuneration, the auditor/regulator relationship and market- making and the Volcker rule.

National discretion over remuneration

The Commission asked how much national discretion existed in setting rules for remuneration. The European Capital Requirements Directive (CRD3) contains provisions relating to bankers’ remuneration, including rules on bonus deferral and payment in a non-cash medium, such as shares. In line with our legal obligations as an EU member state, these provisions have been transposed (with some additions) into UK law within the FSA’s Remuneration Code. Any further changes prescribed by CRD4, which is currently being agreed, will also need to be transposed.

In transposing Directives, member states have a degree of discretion in two respects:

 Under the principle of proportionality, the Remuneration Code is applied proportionately in that firms with higher levels of systemic risk are subject to the full scope of the Code, while lower-risk firms are subject to more limited scope.

 Unless a rule is "maximum harmonising", member states may apply a stricter version than that prescribed in a Directive.

At the hearing, I discussed the desirability of tightening the existing Remuneration Code in three respects: (a) reducing the scope for basing pay on non-risk adjusted metrics; (b) extending the minimum deferral/clawback periods beyond 3 years; and (c) paying bonuses in debt-like instruments. These suggestions are in line with the ideas outlined in the Bank’s November 2012 Financial Stability Report, overseen by the Financial Policy Committee.

Under CRD4, the rules in respect of (a) and (b) will remain "minimum harmonising", which allows scope for national discretion to apply a tougher line. Work in respect of (a) is under way. In respect of (b), as I mentioned at the hearing, there is national discretion to set the minimum deferral period at a higher level – say 5-10 years rather than 3. That would better align risk-taking horizons with the duration of a typical risk cycle. It would shape incentives in ways which encouraged long-termism, creating a more "partnership- type" model. Given the national discretion, this is a proposal the Commission may wish to consider carefully.

Clearly, any move to impose a tougher line would need to be implemented on a proportionate basis and be consistent with the aims of the Directive. In this case, as the policy is clearly risk-focussed and could be targeted at those senior individuals within a bank with direct responsibility for managing that risk, I believe it would satisfy those criteria.

In respect of (c), although the rule itself will also be minimum harmonising, as part of the CRD4 process binding technical standards will be developed that will determine which instruments can be used. This is a matter of some concern to the Bank and we are working with the FSA and HM Treasury.

At the hearing, Mr Garnier asked if these rules could be worked around if a new employer offered to buy out the outstanding deferred pay. Prior to the Code, it was common for buy-outs to comprise an upfront lump sum payment, which effectively cancelled out the deferral. But under guidance issued by the FSA, the new employer must now pay out the deferred awards under the same timetable that was in place under the previous employer. This guidance would increase in importance if deferral periods were to be lengthened. At present, this guidance is only applicable to UK-regulated firms.

The auditor-regulator relationship

At the hearing, I was asked by Lord McFall whether auditors should have a statutory "duty of care" to regulators. I take this to mean whether statute should provide for regulators to have formal powers over auditors, backed up by powers of enforcement (not whether the scope of the "duty of care" for auditors established under tort law should be expanded by statute so as to include regulators).

Section 340 of the Financial Services and Markets Act 2000 (as amended by the forthcoming Financial Services Act) already provides the PRA with rule-making powers over auditors of PRA-authorised firms. The PRA has not yet decided the nature of the rules, if any, it may wish to impose on auditors. The Commission itself may have views on that. This is a complicated issue, since auditors are already subject to regulation by the Financial Reporting Council and various industry bodies and the PRA needs to be mindful of creating regulatory overlap or duplication.

Designated market makers and the Volcker Rule

A Designated Market Maker (DMM) is an intermediary who has been contracted by a trading venue to stand ready to trade a financial security or contract against their own inventory on a continuous basis, thereby providing liquidity to the market. Being a DDM comes with obligations including ensuring trade and price continuity. In return for fulfilling their obligations, DMMs are typically rewarded - for example, through lower trading fees.

DMMs have traditionally been an important component of trading venues worldwide. Almost all the stock exchanges of the major industrialised countries feature DMMs. DMMs are also present in the largest trading venues for standardised derivatives contracts such as futures and options. But over recent years, DMMs have faced significant challenges. The fragmentation of equity trading as a result of regulatory initiatives (MiFID I in Europe and Reg NMS in the US) has also fragmented trading activity, meaning that DMMs’ clientele (and revenue) in traditional trading venues has substantially decreased.

Computerised trading has also allowed many market participants to take on some of the activities of market makers in an unofficial capacity, further eroding DMM revenues. As a result, DMM volume has decreased to the point that they appear to be slowly fading away in some of the world’s largest stock markets. Indeed, some exchanges have responded to this by diluting DMM obligations so that they may have a better chance of competing against unofficial market-makers.

This dilution of the role of DMMs can be a financial stability concern if it results in gaps arising in market liquidity during periods of market stress, which may serve to amplify this stress. An example of this came during the US Flash Crash of 6 May 2010. There have been several mini-Flash Crashes since then. [1] As a by-product, this new model of non-designated market-making has also resulted in a blurring of the distinction between market-making and proprietary trading. This in turn complicates the task of implementing something like the Volcker rule.

Some hope for progress lies in the revisions to the Markets in Financial Instruments Directive (MiFID II). The text is still being negotiated, but Article 51.1a obliges regulated markets to provide sufficient incentives to investment firms to act as DMMs. In practice, this means that exchanges are likely to extend the benefits they provide to DMMs. MiFID also places new restrictions on traders who engage in algorithmic trading in order to execute market making strategies. In particular, Article 17.3 of MiFID obliges these traders to register with the exchange and execute their market making strategy on a continuous basis.

This could have relevance to the question of implementing a Volcker rule. For products traded solely on exchanges, the act of becoming a DMM could help in differentiating between market-making and proprietary trading, albeit not perfectly. In particular, should a firm not be a DMM for certain activities, this would strongly suggest its trading activities were proprietary and so, under a Volcker rule, should be separated off from the main group. On the other hand, should a firm be designated, that burden of proof is lowered, although of course designation would not by itself prevent a firm from assuming proprietary risks. In effect, the designation decision could act as an initial screen of a firm’s trading intentions. Some additional tests, similar to those outlined in the CFTC Volcker proposal, may then be needed for DMM firms to ensure proprietary risks were not being run. Such an approach would have the additional effect of providing incentives for firms to become a DMM in the first place, with resulting benefits to market liquidity and financial stability. This, however, is not a complete solution and would not work for purely bilateral (OTC) markets, where DMMs do (can) not exist.

I hope these responses are useful in the deliberations of the Commission.

8 February 2013