Select Committee on Treasury Written Evidence


Memorandum from the Investment Management Association (IMA)

EXECUTIVE SUMMARY

  1.  The IMA's members, managing over £3 trillion of assets and undertaking no proprietary trading, characteristically take a long-term investment view and support measures that enhance the long-term stability and depth of financial markets.

  2.  The difficulties identified with Northern Rock were caused by a loss of confidence in wholesale markets that meant a credit crisis in one part of the global economy impacted liquidity even at the level of inter-bank lending.

  3.  IMA raises concerns about the operation of the tripartite system after the initial protection of depositors.

  4.  The need to preserve confidence requires further work on the conflicts of interest to which credit rating agencies (CRAs) may be subject.

  5.  Care must be taken to resist an over-reliance upon CRAs by regulators and investors for determining capital treatment and eligibility for inclusion in portfolios.

  6.  The complexity of modern derivative products is such that it is hard to look through to see the true economic drivers, from where gains and associated risks actually derive. Institutional investors commonly take great care that there is sufficient dispersal of risk, that is to say that different risks are not correlated and that there is no over-concentration in any one risk. Simply, the market response over the summer revealed that participants felt they could no longer (if they ever could) ensure that risk could be judged efficiently and priced appropriately in several key markets.

  7.  Transparency alone is not enough, the limitations of risk models in dealing effectively with the unexpected should be a factor that is kept in the forefront of regulatory focus upon capital strength and resilience to shocks. There are growing concerns that current models focus overly upon normal distributions of risk and are not just weak but fail in what are often perceived as "abnormal" or "unexpected" markets.

  8.  There is a role for regulators to ask afresh whether some of the market structures serve investors well. Even when liquidity dried up, it is worth remembering that there were still customers in the market. However these customers had difficulty in accessing any form of market support, or indeed each other. This does not tend to indicate a robust market structure from the investor perspective, featuring as it does a block on the operation of the market imposed by the principal intermediaries. Whilst these intermediaries acted separately, in reality the freeze on the market revealed the very substantial concentration that exists within the credit markets and the effective marginalisation of the outer circle of investors.

  9.  In conclusion, IMA would encourage regulators and legislators:

    9.1  to encourage market-led improvement in transparency, disclosure and conflict management at CRAs;

    9.2  to address afresh the risks involved in over-reliance upon ratings (being based upon default probability) for prudent risk-management and capital adequacy assessments, especially as regards resilience and sensitivity to a default;

    9.3  to work with industry to consider the adequacy of valuation techniques for today's highly complex derivative products, especially in what are seen as highly improbable circumstances.

INTRODUCTION

  10.  The IMA welcomes the opportunity provided by the Treasury Select Committee to give evidence concerning financial stability and transparency.

  11.  The IMA represents the UK-based investment management industry. IMA members include independent fund managers, asset management arms of banks, life insurers and occupational pension scheme managers and are responsible for the management of over £3 trillion of assets. These assets are invested on behalf of clients globally. Our members undertake no proprietary trading and their entire investment business is conducted on behalf of others.

  12.  The Committee has welcomed written evidence under three principal topics:

    12.1  Northern Rock;

    12.2  deposit protection; and

    12.3  the overall functioning of financial markets.

  13.  The comments below are principally restricted to the third point, the overall functioning of the financial markets. We have focused our comments upon issues which we consider relate both to the sub-points identified by the Select Committee and to the particular circumstances which led to the difficulties most obviously demonstrated with Northern Rock.

  14.  As regards the particular matter of Northern Rock, from experience, IMA is confident that the authorities will ensure there is a robust review of practices and that FSA will address any shortcomings in supervision. We would however caution against any hasty revision of the tripartite-nature of the system for safeguarding financial stability.

  15.  IMA has no criticism of the guarantee provided to depositors as at the date of announcement, despite the risk that future behaviour of banks may be conditioned by a belief that Government will bail depositors out above and beyond any then-existing deposit protection scheme. However, we do have a real concern as to the wisdom of then extending protection to new depositors. Intervention under the tripartite system should be restricted narrowly to issues of financial stability, including stopping contagion. In the present case, such intervention could be seen as now prioritising the continuance of business as a going concern in order to facilitate a sale. With hindsight this may turn out to be successful, but it would be a real concern if all it did was postpone a failure. It must not be forgotten that many other members of the public are directly and indirectly invested in Northern Rock. Passive investment managers, including FTSE tracking ISAs for example, are almost bound to remain invested at this time. Beyond protection of the then existing depositors, there is no need to protect a bank from the judgement of the market merely because it is a bank.

IMA REVIEW OF THE LIQUIDITY CRISIS

  16.  IMA's board has commissioned an internal review of the impact of the summer's liquidity crisis and the lessons to be learned amongst asset managers. That review is currently under way. However early indications are that our members' investment styles positioned them to continue to manage assets under their care largely unaffected and in contrast to firms that may have taken a more short-term approach to enhancing yield. This memorandum addresses what might be seen as a cause of the crisis and how that may lead in the lines of inquiry into the wider operations of the financial markets.

THE INTEREST OF IMA FIRMS IN CLEAN, STABLE MARKETS

  17.  The members of IMA are concerned with the management of assets for occupational pension schemes, local government pension schemes, other institutional funds, life funds, individual portfolios, and charities as well as assets in UCITS and other pooled investments, such as unit trusts and OEICs. Individual consumers are, therefore, the underlying beneficiaries of most of the assets managed by IMA members. Characteristically our members take a long term investment view and support measures that enhance the long-term stability and depth of financial markets. Put simply, one day's gains are of little value if they come at the cost of returns over a five year outlook. Nevertheless, individual consumers can be disadvantaged by short-term market disruption eg if it is on that day that a person's pension is valued for buying an annuity.

THE NEED TO PRESERVE CONFIDENCE

  18.  Although Northern Rock is understandably the focus of this enquiry, due to the impact upon its depositors and because this has been the first very public test of the tripartite agreement, at its heart IMA considers that the recent crisis could be described as deriving from a loss of confidence in the wholesale market. Regulators and government are rightly spending time and money seeking to preserve and improve consumers' confidence in the markets. But the first regulatory objective for the FSA in the Financial Services and Markets Act 2000 is the market confidence objective; and that is described as "maintaining confidence in the financial system". The recent crisis exhibited failures in three critical areas for the wholesale markets: confidence, concentration and correlation. These are explored below. Although the impact may have been keenly felt on a wide public basis through retail reliance on, and exposure to, the banks, in fact the damage to the operation of financial markets, and therefore the wider economy, was already done.

  19.  As the description produced by the Bank of England in the MPC minutes of 5-6 September 2007 and comments by SEC's Commissioner Nazareth both suggest, the current crisis is one that evolved from a revaluation of credit risk in one sector of the financial market into a wider liquidity crisis.

  20.  The importance of looking at the underlying causes is twofold; first, it may assist in determining whether regulatory, government and market responses to such events could be improved; and secondly, it may assist in determining whether such events could be better predicted and what role regulators, government and markets might play in providing such early warnings.

  21.  A question to ask is whether the liquidity crisis was expected or unexpected and whether or not there was a great deal that regulators at various levels within the financial system internationally could have done to plan for, or prevent, the developments that have taken place since July. If the precise form that the crisis took was not foreseen, still there are legitimate questions to be asked as to precisely how unexpected was some form of problem resulting from lending practices in parts of the US real estate market and the way in which these were packaged and sold as financial products. Whilst the Northern Rock crisis hit the UK in late August 2007, the US sub-prime problems had been publicly reported since early spring. In this respect, the risk of a credit crunch could be said to be a "known unknown" as opposed to an "unknown unknown", which would imply a truly unexpected development within the US sub-prime market.

  22.  In a speech in April 2005, Alan Greenspan spoke of the widespread adoption of more advanced credit scoring models and stated [our emphasis applied] "where once more-marginal applicants would simply have been denied credit, lenders are now able to quite efficiently judge the risk posed by individual applicants and price that risk appropriately. These improvements have led to rapid growth in sub-prime mortgage lending; indeed, today sub-prime mortgages account for roughly 10% of the number of all mortgage outstanding, up from just 1 or 2% in the early nineties". The argumentation was not wrong, but what may have been in doubt was the assertion that the credit scoring models did actually allow lenders to "quite efficiently judge the risk posed by applicants ... and price that risk appropriately". Whatever the position may have been in the US sub-prime market, belief in the general assertion that risk can be judged efficiently and priced appropriately underpins confidence across financial markets, regardless of product or strategy.

  23.  This need for confidence was even more necessary in the specific context of the recent US sub-prime market. As Federal Reserve Board Governor Frederic Mishkin has noted, a challenge arose from the distance between a loan originator in the United States and the ultimate holder of the debt (increasingly to be found outside the United States).[30] The most problematic part of the distance was not geographical, but generated by the way in which loans were securitised and re-packaged, which diminished the incentives on originators to protect the ultimate holders of the debt from potential risk.

  24.  Having introduced the connection between confidence and assessment and pricing of risk, we explore this under three headings:

    24.1    the role of credit rating agencies ("CRAs"): "conflicts and confidence";

    24.2    the transparency and complexity of many financial products: "correlation, concentration and confidence"; and

    24.3    the Blind Watchmaker model of market development.

THE ROLE OF CRAS: "CONFLICTS AND CONFIDENCE"

  25.  Commissioner Nazareth in her speech on 19 October 2007[31] indicates one of the phases in the collapse: "Investors stopped buying CLOs [collateralised loan obligations] in July, however, because of fears that they—like the asset-backed securities tied to sub-prime—were not the safe bet that their investment grade ratings would imply".

  26.  The roles and responsibilities of CRAs have been in discussion for many years. In February 2005, IMA submitted a response to CESR's consultation paper on technical advice to the European Commission on possible measures concerning CRAs. Then, as now, the importance of CRAs grows as corporate debt and complex instruments displace equities in most portfolios. Recently CRAs have been relied upon to describe or circumscribe asset allocation in clients' mandates or funds' definitions.

  27.  IMA remains of the view that more competition in the ratings process will encourage a higher level of analytical input and thereby improve the quality of ratings overall. IMA members have a considerable interest in transparent and robust practices at CRAs. IMA does not advocate regulation in this area. Forms of registration and regulation raise barriers to entry when arguably the role of the regulator should be to encourage more competition. There are areas however which must remain under continuous consideration.

  28.  Firstly, with respect to conflicts of interest, as happens in UK-regulated firms, the IMA believes that CRAs should put in place policies and procedures to identify, manage and, where incapable of being managed out, disclose conflicts inherent in their individual business models. Group-wide links to the issuer's client base and the provision of other services to those for whom they provide credit ratings are all necessary disclosures.

  29.  Secondly, IMA strongly supports efforts to enhance the transparency of the credit rating process and the publication of CRAs' methodologies. The temptation to which CRAs must be subject is to establish a credible rating so that a bond can be issued rather than to challenge aggressively the veracity of the information with which they have been provided. Some of that information is non-public and therefore is not available to an investor and is unverifiable. Classically, a fund manager being presented with such information would challenge it as part of his due diligence. Publication of methodologies reassures markets that CRAs are designing, applying and monitoring those methodologies with independence and consistency. All these factors again emphasise the importance of conflict management within the CRAs.

  30.  Accordingly, IMA believes regulators and the legislature should encourage market-led improvement in transparency, disclosure and conflict management at CRAs.

  31.  Having noted that conflict management at CRAs and the transparency of their methodology remain areas that must be kept under review, risks also arise from perceptions that CRAs do something more than they actually do. Whilst frequently important opinion providers, there will always be a range of opinions amongst practitioners in the wider market about any particular borrower or bond issue. CRAs would say that their ratings are restricted to the likelihood of there being a default. In this sense a "AAA" rating might be seen by some as a "safe bet" but likelihood of default is not the same as a measure of the recovery level that might be expected were there to be a default and the sensitivity of these recovery levels to external market conditions. In seeking to ensure that risk can be judged efficiently and priced appropriately, it may be that a lesson from the recent crisis is that a rating addresses only a part of what a market ought to know. CRAs have an interest in promoting their services; regulators, market participants and investors more widely need to be aware of the real limitations to this.

  32.  Though IMA does not advocate registration or regulation of CRAs, it is important that CESR, the US SEC and IOSCO remain alert given the role of CRAs in the market. However CRAs are merely one opinion in the market. The more weight, or reliance, is put on an opinion, the more difficult it is for a credit analyst to adapt to changing circumstances, so slowing down opinion forming. Forms of registration and regulation may not only raise barriers to entry but regulation may have the unexpected effect of slowing down opinion formation. Additionally, a moral hazard can arise where there is regulatory oversight of CRAs because investors may then be misled as to the role, quality and nature of a rating.

  33.  But even without regulation in the EU, CRAs play an important role in determining how international regulators treat some assets for capital purposes. Regulators must remain alert to providing favourable capital treatment based overly upon the rating of an investment. For example, the standardised approach to credit risk under the Capital Requirements Directive is reliant upon the credit assessment of a recognised credit rating agency. A firm may benefit from a capital requirement of 20% for exposure to a high rated entity; whereas a similar exposure to a lesser rated entity would warrant a 150% capital charge. It is worth noting also that whilst the largest credit institutions will rely upon their own internal models for assessing credit risk, these models are not immune from reliance on ratings, for instance in respect of collateral.

  34.  Accordingly, IMA would encourage regulators and legislators to address afresh the risks involved in over-reliance upon ratings (being based upon default probability) for prudent risk-management and capital adequacy assessments, especially as regards resilience and sensitivity to a default.

THE TRANSPARENCY AND COMPLEXITY OF MANY FINANCIAL PRODUCTS: "CORRELATION, CONCENTRATION AND CONFIDENCE"

  35.  Although only anecdotal, some IMA members have commented that once liquidity concerns arose, investors started to exit a wide range of investments and sought safe investments. The range of asset-backed securities (ABSs), collateralised loan obligations (CLOs) and equities included many that with hindsight ought not to have been significantly affected by the market turmoil. This cannot merely be dismissed as characteristic of any moment of panic, but rather it reveals two other critical characteristics of the modern financial markets. First, the drive to maximise earnings leads to every available asset being put to work such that when liquidity dried up in one part of the market, the effect was felt in uncorrelated assets. For instance, even equities had to be sold by some investors to raise cash to meet their obligations, including with respect to regulatory capital. Secondly, it revealed that the interposition of conduits and structured investment vehicles (SIVs) between ultimate investment and investor, driven as it commonly is by regulatory treatment of issuing banks' balance sheets, may have exacerbated investors' lack of confidence as it impeded their understanding of what exactly were the risks to which their money was exposed.

  36.  The complexity of modern derivative products is such that it is hard to look through to see the true economic drivers, from where gains and associated risks actually derive. Institutional investors commonly take great care that there is sufficient dispersal of risk, that is to say that different risks are not correlated and that there is no over-concentration in any one risk. Simply, the market response over the summer revealed that participants felt they could no longer (if they ever could) ensure that risk could be judged efficiently and priced appropriately in many ABSs, CLOs and asset-backed commercial paper issued by SIVs. Re-investment was re-directed as the need for liquidity and capital security over-rode the search for short-term yield and ultimately that led to banks being unwilling to lend to one another except in the very-short term.

  37.  Whilst markets will re-evaluate and re-price these complex products, it may be instructive for regulators to learn what it was that investors in these products thought they were relying upon—credit ratings, their own due diligence, the brand name of the issuing bank or a mere promise of returns.

  38.  As an aside, it is in this context that IMA has recently called upon HM Treasury and the FSA to re-think their proposals for introducing a recognised covered bond regime into UK regulation. We fear that the FSA and HMT have not sufficiently recognised the risk to quality that the current proposal could have on funds and we are concerned with the lack of prior engagement with investors. HMT identifies a risk of this regime as being to the depositors of the issuing bank. IMA agrees but considers there could be wider investor risk as well under the current proposals.

  39.  Returning to the wider concern as to the limitations upon assessing risk in complex derivative products, this does not merely depend upon transparency. There are growing concerns that current models focus overly upon normal distributions of risk and are not just weak, but fail, in what are often perceived as "abnormal" or "unexpected" markets. This is a very deep issue which goes to the heart of modelling techniques that have been in existence for years. It relates not only to the models themselves but also raises questions as to the use to which they are put by senior managers who are either not fully aware of the technicalities of the models on which they are dependent, or who are incentivised (by their own managers and ultimately by shareholders) to seek return enhancement ahead of risk reduction.

  40.  In her speech, Commissioner Nazareth went on to say [again our emphasis supplied]:

    "As these hedge funds sold their equity positions, it did more than drive down equity prices. Instead, in early August it changed market dynamics in a way that some fund managers thought was statistically impossible—the prices of the securities the quant funds owned tended to decline, and the prices of those they sold short tended to increase. Regardless of whatever statistical improbability may have existed, these events occurred, and the hedge funds performed very poorly as a result".

  41.  The limitations of models in dealing effectively with the unexpected should be a factor that is kept in the forefront of regulatory focus upon capital strength and resilience to shocks. Benoit Mandelbrot, creator of the eponymous mathematical set, and others have described their unease with any widespread dependence upon the Gaussian distribution (or Bell curve).[32] These are highly technical issues but a real example may illuminate what the debate is about. An approach based on the standard Gaussian distribution would expect the Dow Jones Industrial Average to have moved more than 3.4% during a single day only 58 times over 90 years. Instead, such a wide swing had been seen on 1,001 occasions. Put another way, the under-representation of unlikely events is a growing risk in evermore complex and interlinked financial markets. To repeat Commissioner Nazareth's phrase "Regardless of whatever statistical improbability may have existed, these events occurred".

  42.  Accordingly, IMA would encourage regulators and legislators to work with industry to consider the adequacy of valuation techniques for today's highly complex derivative products, especially in what are seen as highly improbable circumstances.

THE BLIND WATCHMAKER MODEL OF MARKET DEVELOPMENT

  43.  The dominant paradigm of market regulation has been to secure transparent, clean and orderly markets. As such, regulatory intervention has been determinedly agnostic as to the market structure. Legislative intervention in the EU has focused upon fostering competition; as an introductory recital from MiFID states:

    "A coherent and risk-sensitive framework for regulating the main types of order-execution arrangement currently active in the European financial marketplace should be provided for. It is necessary to recognise the emergence of a new generation of organised trading systems alongside regulated markets which should be subjected to obligations designed to preserve the efficient and orderly functioning of financial markets".

  44.  The unspoken statement is that new organised trading systems will emerge. Competition for the business carried by the traditional exchanges, and other trading systems, is seen as good and markets are left to determine which trading systems will survive. This is undoubtedly the right approach in principle. But the operation of trading systems needs to meet standards laid down by regulators, whether operating as an EU regulated market, a multi-lateral trading facility, market maker or other liquidity provider. The liquidity crisis over the summer should cause market participants to consider how different trading systems withstood the market impact and whether some models were more resilient to shock than others. The liquidity squeeze acted as a salutary reminder that the availability of liquidity, and the ability to value assets, go hand in glove. The overriding feature of the recent problems was the banks' apparent loss of confidence in each others' valuation models, possibly also their own, and the complete freeze in credit markets that this engendered. It is worth pointing out that there were customers in the market. However these customers had difficulty in accessing any form of market support, or indeed each other. This does not tend to indicate a robust market structure from the investor perspective, featuring as it does a block on the operation of the market imposed by the principal intermediaries. Whilst these intermediaries acted separately, in reality the freeze on the market revealed the very substantial concentration that exists within the credit markets and the effective marginalisation of the outer circle of investors.

  45.  This may lead to a conclusion that some standards need re-emphasising so as ensure that markets can keep operating in times of crisis. Ultimately, however, if confidence evaporates in any market, then there is little any participant can be expected to do save to look after their own short-term interests. If that means no longer providing liquidity, then the market will stop functioning until participants once again can judge risk efficiently and price it appropriately. This does, nonetheless, indicate very significant weaknesses in the operation and structure of any market driven to such a resort. Indicators of this kind should weigh heavily with regulators when they consider the risk assessment presented by individual credit institutions in respect of their credit operations.

  46.  In conclusion, IMA would encourage regulators and legislators:

    46.1  to encourage market-led improvement in transparency, disclosure and conflict management at CRAs;

    46.2  to address afresh the risks involved in over-reliance upon ratings (being based upon default probability) for prudent risk-management and capital adequacy assessments, especially as regards resilience and sensitivity to a default;

    46.3  to work with industry to consider the adequacy of valuation techniques for today's highly complex derivative products, especially in what are seen as highly improbable circumstances.

November 2007






30   Governor Frederic S Mishkin, Speech to the Money Marketeers of New York University, New York, New York, 10 September 2007. Back

31   SEC Commissioner Annette L Nazareth: Remarks Before The Los Angeles County Bar Association 40th Annual Securities Regulation Seminar 19/10/07. Back

32   The (Mis)behavior of Markets: A Fractal View of Risk, Ruin and Reward, by Benoit Mandelbrot and Richard L Hudson published by Profile Books. Back


 
previous page contents next page

House of Commons home page Parliament home page House of Lords home page search page enquiries index

© Parliamentary copyright 2008
Prepared 1 February 2008