Thursday, June 23, 2011

Hector Sants asks for a new bank supervision model, the FDR Framework provides one

In all the commotion over whether or not Greece will default, Hector Sants', the future chief executive of the UK's Prudential Regulation Authority, speech on how the PRA would prevent future financial crisis was not given adequate attention.

This speech is very important because Mr. Sants asks for help in rethinking the bank supervision model.

Allow me to offer the FDR Framework's bank supervision model.  Under this framework, bank supervision harnesses the market to protect the safety and soundness of the financial system.

As regular readers know, under the FDR Framework, it is the responsibility of the regulators to ensure that all market participants (including the regulators) have access to all the useful, relevant data in an appropriate, timely manner (and yes, using 21st century information technology this is easily done).  In the case of financial institutions, this useful, relevant data is their current asset and liability-level data.

Making this data available to all market participants dramatically improves the lot of a bank supervisor.

First, it brings market discipline to financial institutions.

Previously, since the regulators had a monopoly on all the useful, relevant data, it was impossible for market participants to exert market discipline by changing the pricing and amount of their exposure based on the riskiness of the financial institution.  As a result, the stability of the financial markets was dependent on the regulators properly analyzing this data and taking the appropriate steps to discipline the financial institutions.

Under the FDR Framework, market participants, including competitors, have an incentive to use this data because they know that under the practice of caveat emptor (buyer beware) they are responsible for any losses that result from their exposure to any financial institution.

As a result, the stability of the financial markets is now dependent on market participants adjusting the pricing and amount of their exposure to exert market discipline.  To the extent that a bonehead management team misses the signal, it is up to the supervisor to point out the obvious:  the market thinks you are taking on too much risk, either cut back your risk or raise more equity.

Second, it brings many more eyes looking at the data to see if something is wrong.  Instead of substituting the supervisor's judgement for the market's, supervisors can now turn to the market and its participants.  After all, who would be better at analyzing the current asset and liability-level data for RBS, bank regulators or firms like Barclays, HSBC and JP Morgan?

His speech raises a number of issue that the FDR Framework's bank supervision model addresses.
... The purpose of this event is to lay out our views of the supervisory regime required to deliver the government’s vision of prudential regulation for banks. 
To facilitate both this conference and subsequent discussions, we published today a document which outlines our initial thinking on this issue. It is not a formal consultation document, but we do invite you to engage with us in dialogue which will help us as we proceed to the detailed design and implementation stage. 
Thanks for the invitation to provide some feedback on your initial thinking and introduce you to bank supervision under the FDR Framework.
... It is vital that we take this opportunity to go to back to first principles. We need both to recognise why we supervise and have a clear, coherent and transparent methodology for that supervision. 
... Turning first, therefore, to the purpose of the PRA. That purpose should rightly be determined by Parliament, and the government’s intentions are clear. The purpose is:
‘to contribute to the promotion of the stability of the UK financial system. It will have a single objective – to promote the safety and soundness of regulated firms – and will meet this objective primarily by seeking to minimise any adverse effects of firm failure on the UK financial system and by ensuring that firms carry on their business in a way that avoids adverse effects on the system.’ 
This purpose is fundamentally different from that of previous regulatory regimes and will lead to a fundamentally different model of supervision to that which was in use before the financial crisis. 
As discussed above, the disclosure of all the useful, relevant data fundamentally changes the model of supervision because it allows the supervisors to harness the market to promote stability, safety and soundness.
These words I believe are straightforward, but I would like to highlight four points: 
• Firstly, they make it clear that the purpose of the PRA is to focus on the stability of the system overall, albeit through the mechanism of the supervision of individual firms. This will require close coordination with the new Financial Policy Committee (FPC), whose role it will be to manage the risks in the system as a whole. The PRA will be a key contributor to the information and analysis on which the FPC will base its judgements, recognising the determination of the intervention tools with respect to system-wide risks rests with the committee, not the PRA.  
• Secondly, in the light of this objective, there has been much debate as to whether a regulator whose purpose is to promote financial stability could do so by solely minimising the consequence of firm failures, or whether such a regulator has an obligation to minimise the risk of individual firms failing in the first place
In a world without uncertainty, it might well be a justifiable theoretical position to state that the regulator should not be obliged to seek to reduce the risk of firm failure. However, in practice the inherent uncertainty and complexity of both firms and the overall system mean that this is not a realistic objective. The PRA will thus always seek to reduce the risk of individual firm failure, but will be giving particular focus to ensuring that if failure occurs, it does so in an orderly manner. 
Furthermore, it needs to be recognised that international regulatory standards are explicit with regard to major firms and the need to maintain a baseline of supervisory oversight, with the intention of reducing the probability of failure. 
The FDR Framework minimises both the consequences of firms failing and the risk of an individual firm failing in the first place.

As discussed above, market participants who are at risk of loss from their exposure to an individual firm have an incentive to exert discipline on the firm to keep it from failing.  It is a fundamental principle of finance that as investment risk increases, so too does the rate of return required by investors.

As the return to hold a financial institution's debt and equity increases relative to its peers, it is a easily understood signal that the institution is becoming riskier.  Either management will voluntarily react by lowering the financial institution's risk profile or supervisors will need to step in.

Despite this, some financial institutions with truly bonehead management will fail.  Fortunately, this is not a source of financial instability as investors have had a chance to analyze the useful, relevant data and adjust their exposure to levels where they can afford the resulting losses.
• Thirdly, the obligation of baseline supervision should not detract from a key strand of the PRA’s approach, namely to ensure that the role of regulators is to complement and promote the disciplines of the marketplace, not to substitute for them. The ultimate responsibility for managing a firm prudently rests with its management, board and shareholders.
By definition, bank supervision under the FDR Framework complements and promotes the disciplines of the marketplace and does not substitute for them (with the exception for bonehead management).
• Finally, the consequence of this approach undoubtedly means that the PRA needs to recognise that if firms fail, society will rightly ask whether that was a regulatory failure. 
Conversely, society needs to recognise that the PRA should not be held accountable for all failures. Failure should be seen as a necessary element of a healthy, innovative system. 
Furthermore, as I have just said, we need to consistently remind ourselves that the primary obligation to ensure a firm is prudently managed within its statutory obligations lies with the management, and in particular the board, not with the regulator. 
Under the FDR Framework, failure is seen as a necessary element of a healthy, innovative system.
Moving now to how we do it. A supervisor effectively has three buckets of tools: 
• rules and regulations, primarily in the form of capital, liquidity, leverage and governance standards; 
• resolution plans; and 
• supervisory oversight of management actions and strategies, including recovery plans. 
The effectiveness of each of these tools is amplified when combined with market discipline under the FDR Framework.
There is a view that says if there is sufficient capital and liquidity buffers, then the probability of failure could be all but eradicated, which would render the concept of supervisory interventions over and above compliance with those standards irrelevant. 
Theoretical studies can demonstrate what that number might be, but the reality is, for a number of practical reasons, such a level of capital is not going to be achieved.
True, we are not going to set capital equal to 100% of assets.
There is also a view at the other end of the spectrum which says, if we could be absolutely sure we could resolve all firms without any material cost to the system, then we need neither capital nor supervisory oversight. This perfect world of resolution is not likely to be achieved and indeed at present we seem some years off being able to offer even reasonable certainty about being able to resolve a complex cross-border group.
This is a very important point.  Simply put, we are a long ways from being able to resolve a Goldman Sachs.
Given these realities, the PRA supervisory approach will be to deploy all three tools.... Furthermore and critically, this supervisory approach, to be effective, will need to be based on judgement and a forward-looking assessment of risk. 
One of the strengths of the FDR Framework is that it harnesses the market to make a forward-looking assessment of risk.  Naturally, the market will have different views on this forward-looking assessment.

Bank supervision under the FDR Framework benefits from being able to examine these different views and their implications.
In designing our supervisory approach, as well as going back to first principles, it is vital to ensure we have drawn on all lessons of the past. 
Bank failures occurred under the past supervisory regimes of both the Bank of England and the FSA. In designing the PRA’s approach, we have taken into account the lessons from both periods of supervision. 
Let me briefly summarise those principal supervisory lessons. 
The central regulatory failing in the period of FSA supervision was the inadequate standards for capital and liquidity. These failings have been addressed in the short term by the FSA’s interim capital and liquidity regimes, while the long-term solution is the responsibility of the Basel Committee. I believe that if we had had effective capital, liquidity and leverage standards in place in 2005, then history would have been significantly different.
This blog has spent considerable time debunking the myth of capital and liquidity standards.  Higher standards would not have prevented the banks from holding toxic securities.  In fact, had higher standards been in place, it is entirely possible that the credit crisis would have been even worse.  For example, Merrill Lynch might have chosen to own more toxic securities to enhance its return on equity had there been a higher capital ratio.

Clearly, the conclusion from this supervisory lesson is not the need for higher capital and liquidity standards.  The conclusion is the need for ensuring that market participants have access to all the useful, relevant information in an appropriate, timely manner.

Had this been in place in 2005, there would have been no such thing as opaque, toxic securities.  While there would have been securities holding sub-prime mortgages, the market could have more accurately priced these because it could have evaluated the true risk of the underlying collateral.  Similarly, the market would have exerted discipline on Merrill Lynch when it saw that its risk was increasing from an accumulation of high risk securities.
However, in addition to the issue of over-reliance on the then capital and liquidity standards, we also need to recognise that the FSA’s supervisory approach rested on another presumption which has proved to be false; namely that senior management judgement and market discipline should not be questioned by supervisors.
Clearly, supervision should not rely on the judgement of senior management about their own firm.

The presumption that supervisors should not question market discipline has not been proven false.  As I discussed above, since the regulators had a monopoly on all the useful, relevant data, the markets were not able to provide market discipline.

Under the FDR Framework, the markets will, for the first time, be able to exert market discipline on financial institutions.  In addition, supervisors are encouraged to ask for senior management's judgement when it concerns one of their competitors who they have an exposure to.
... The PRA will take a different approach. It will, as has already been said in the Treasury’s Consultation Document, build on the post-crisis approach adopted by the FSA, which focuses on outcomes and thus can be termed ‘judgement-based’. 
Central to this supervisory model is the presumption that regulators cannot rely on the judgement of the management of the firms they supervise, and must take their own view formed from their own analysis about the significant issues which affect the safety and soundness of the firm. Furthermore, where that judgement differs from the firm’s management, the regulator must act. 
The firm failures in the period of supervision by the Bank of England throw light on how these regulatory judgements should be reached. In particular, they demonstrate the importance of basing those interventions on thorough analysis, particularly on the firm’s business model, capital and funding strategy. 
Along with the importance of ‘close intensive engagement’ with both the management of the firm being supervised and informed third parties, such as auditors and market participants, they also demonstrate the need for effective global coordination, particularly where the UK is not the home supervisor. 
It bears repeating that under the FDR Framework, supervisors will have many more informed third parties to talk with in analyzing any financial institution.  Many of these third parties will be acting on their analysis of the financial institution.
.... In order to make effective judgements, the PRA must equip itself with: 
• high quality, experienced supervisors who are willing to make difficult judgements and command the respect of the firms they supervise;
• high quality analysis of the critical risks in relation to a firm’s soundness, notably: its business model, particularly in relation to risk and profitability, its
• capital model and its funding model;
• high quality analysis of the effectiveness of a firm’s governance model and the competency of its key executives;
• a clear understanding of the firm’s culture and the implications of that culture on its risk profile;
• a clear understanding of its recovery and resolution capability;
• a clear understanding of the consequence of the failure of that institution on the wider economic system; and
• a clear understanding of informed third parties’ views of the risks that a firm is running.  
... The effectiveness of the PRA will rest heavily on its ability to deliver a judgement-based model. This raises the question of how will we ensure we equip the PRA to make the best possible decisions. 
A key factor in this goal will be ensuring we have individuals with the optimal experience and technical ability.
A feature of the FDR Framework is that it makes it much easier to find qualified supervisors.
... The challenge of delivering high-quality judgments will, however, not be entirely solved by the greater involvement of the senior executive team. The reality is, in order to make a credible and effective judgement about a firm, indepth analysis is required, along with the perspective that comes with continuity of oversight and understanding of that firm’s business model, management and culture. 
That skill set is one which requires a supervisor who is dedicated to an individual firm. The requirement for dedicated individual supervisors for each major firm will inevitably mean that more than a handful of senior executives are needed.   
The new model will not have solved the problem that such supervisors have overall compensation levels which are multiples less than the comparable role in a major bank or professional services firm. The economic reality is that good staff are hard to attract and retain.
By harnessing the market, the PRA effectively "hires" the needed senior executives without having to put them on its own staff.  As I discussed above, each competitor has an incentive to analyze its peers. This analysis will be overseen by exactly the type of individual that the PRA would want to hire as a senior executive (and this is before we get to all the credit and equity market analysts).

By providing the market with all the useful, relevant data, the PRA is not hiring a few senior executives, it is effectively hiring senior executives at every financial institution.  What the PRA needs from its bank supervisors is the ability to talk with and understand the analysis and conclusions of these executives.
The other challenge I would like to highlight is technology and data. If the envisaged small high-quality teams of supervisors are to make the best possible judgements, they will need a greater level of technology support and better quality data than historically supervisors have had. Achieving this goal will be a non trivial exercise. 
Mr. Sants, I would be happy to help you with this non trivial exercise.  You can get my number from the Bank of England's Andrew Haldane.

No comments: