Address by Gareth Murphy, Director of Markets Supervision, to the InvestoRegulation Conference, London

Hedge Fund Regulation

Introduction

I would like to thank Mark Berman and InvestoRegulation for the invitation to give the keynote speech this morning. It is good to be back in London where I have spent most of my career and I am delighted to have the opportunity to discuss the issue of hedge fund (HF) regulation.

My thoughts, this morning stem from my experiences working in the hedge fund sector, investment banking and central banking and regulation.

Some of the language I use may seem strange to the non-economists in attendance but I hope to persuade you that we must think about the raison d’être for regulation using the insights of some famous economists like Akerlof, Stiglitz, Kahnemann and Tversky and others.

I will argue that the ‘integratedness’ of global financial markets and the current processes for developing financial regulations within regional blocs pose systemic risks for the future. And the lessons of the recent financial crisis should prompt us all to take a step back and consider how we develop financial regulation.

I will spend the next 20 minutes (or so) looking at (i) how regulation comes about, (ii) regulation in the HF sector and (iii) conclude by looking at how best we might develop regulation in the future.

MiFID as an example of the process for developing EU regulation

As a starting point, let’s consider how financial regulation is developed in Europe at present.

By way of an example, the above slide sketches the process by which the (first) MiFID was developed. There is logic to this process: starting with the high-level principles and progressively become more granular.

But like a busy kitchen, there are many chefs tossing in their favourite ingredients – politicians, civil servants, central bankers, regulators, industry lobby groups, investors, consumer groups, academics, journalists and the odd economist. However, the end-product can be sometimes difficult to digest especially when it is one of many such courses.

I will say more on this later.

Genesis of regulation

But let’s take a step back from this and think about how regulation comes about.

Of the various drivers of regulation, four spring to mind (and are set out in the above slide)....

‘Something must be done’

We are all familiar with situations where a catastrophic or tragic event occurs, there is a public outcry, grinding and gnashing of teeth, political ructions and a hasty legislative change. One real-life example which springs to mind is the Dangerous Dogs Act of 1991 when dog muzzles where introduced for a range of species following a spate of incidents (in this country) where young children were attacked. Tragic though these events were, the legislation has been criticised for being conceived in haste in reaction to a public outcry and (and not allowing for the physiology of some dog species).

'Towards a single market’

Closer to home, the creation of the European Coal and Steel Community in the 1950s was the precursor to a lengthy process to create a single market for production and distribution of goods and services in Europe culminating the Maastricht Treaty in 1992.¹    Legislation and regulations in this instance were motivated by a desire to foster peace through economic integration and build an economic bloc which would grow through increased cross-border trade and the creation of a single continental market place.

‘Towards a single rule-book’

In the face of the emerging single market, individual countries have struggled to resist the impulse to use their respective regulatory regimes to protect local interests. In European financial services, this has led to the need to develop a 'single rule book' to promote consistency, reduce barriers to entry and eliminate opportunities for regulatory arbitrage.²

My point in mentioning these first three regulatory drivers is that, whilst they may have merit in their own right and in their own time, in a globally integrated marketplace we can no longer ignore their unintended consequences by leaving spill-over effects to be swept under the carpet. As an aside, Stiglitz (2010) notes that while the integration of financial markets – whether at a European or global level – was expected to bring greater financial stability, there are reasons to believe that autarky (in this sense of each country for its own) may be more financially stable. This is a strong and provocative conclusion and food for thought which I will come back to later.

This leads me to the fourth driver for regulation – which I will argue has attained greater importance as a result of the financial crisis – this is the need to tackle the sources of what economists term 'market inefficiencies' or 'market failure'. The financial crisis has shown that the consequences of market failure can be extreme costly and they can last for generations, so financial authorities and market participants have a specific responsibility to think about the ways in which financial markets do not perform efficiently. The underlying sources of market inefficiency have been well-studied since the 1970s.

Correcting 'market inefficiencies'

In economics, we use the word 'friction' to describe a phenomenon which creates market inefficiency or contributes to a market failure. As a regulator, the four main frictions which I encounter most often are³ :

1. Principal-Agent problems:

These occur where there is a principal who delegates responsibility to an agent to act on his/her behalf, for example, shareholders expect a CEO to maximise the share price not so siphon off profits into executive expense accounts.

2. Incentive misalignment problems:

These occur where different people in the same financial arena are motivated by different incentives. My favourite example from my past-life is the tension between traders who manage risk and salespeople who get paid for getting traders to show good prices to clients. If a trade goes wrong it is invariably the trader’s problem for showing an ‘overly-competitive’ price, not the saleperson who has pocketed his commission.

3. Information asymmetry issues:

This is one of the best studied frictions in financial markets. Akerlof's paper on 'The Market for Lemons' highlighted how markets for used cars can become dysfunctional when you cannot tell a quality car from a dodgy car - or 'lemon' - the famous result being that in such circumstances only the 'lemons' trade.

4. Agent behaviour issues

This fourth friction deals with our cognitive response to complex situations, put simply, how do we process information and how do we deal with uncertainty ⁴ ? There is currently a burgeoning literature of 'behavioural finance' looking at the many phenomena which exemplify this friction at work. You will be familiar with some of the issues here such as: ⁵

• the Prisoners’ Dilemma which is the famous Nash game which considers whether you should betray your accomplice and gamble that he will not betray you knowing that he will be considering just the same;
• traders’ loss aversion where winning trades are cashed in and losing positions are let run (and possibly fester);
• consumers’ inconsistent approach time: “if you would prefer $90 now to $100 in a year’s time does that mean that you would prefer $90 in nine year’s time to $100 in 10 years time?”.
• Lastly, it is worth mentioning that legal uncertainty can be a significant source of market inefficiency.⁶

The bail-out of Long-Term Capital Management (LTCM) is a useful example to view through the prism of economic frictions. LTCM had lots of trades which other banks had mimicked, the market could not establish the size of the positions, during the Summer of 1998 investment banks' self-interest lead them to try and liquidating their positions thereby driving the marks wider, thereby generating larger losses for the fund and for the banks. Finally the Federal Reserve Bank of New York (FRBNY) stepped in, brought all the key banks with exposures to the fund together, realigned incentives by asking each bank to provide capital to the fund (thereby diluting the partners) and dealt with the principal-agency problem by appointing a six-person management team representing the banks.

This slide maps the various familiar types of regulatory intervention with the four economic frictions. Without discussing all the regulatory interventions noted here, I will note two.

Principal-agent problems are tackled through corporate governance standards and fitness and probity assessments. Much has been written ⁷ and much work has been done to strengthen corporate governance ⁸ and to require that persons occupying controlled functions meet fitness and probity standards.⁹

Many of the above economic frictions have been at the heart of the crisis in the banking industry but the one I would highlight is the herd-like behavioural weaknesses which lulled banks to underestimate liquidity and solvency risks through weak origination standards and excessive maturity transformation. The upshot of this is the Basel III protocol – and the fourth iteration of the EU Capital Requirements Directive (CRD IV).

From a European perspective, I have (in this slide) sought to align the different financial markets directives with the various frictions. As we can see (from this non-exhaustive list), there are many directives (some of which have gone through multiple revisions and iterations.)¹⁰

When thinking about the corpus of financial regulations in an increasingly integrated global financial market place, it seems to me that in order not to be overwhelmed by the tsunami of directives, regulations, technical standards and guidance, we need to fix our frame of reference on the underlying reasons for regulating, that is, the economic frictions which are the sources of market inefficiency.

Taking stock of the Hedge Fund sector

As this is a conference about HF regulation, let me be a little bit more concrete. But first, let's consider the economic function of the HF sector. In this regard, four particular functions come to mind:

1. HFs increase the set of investment opportunities which investors may access. This is evidenced by the wide array of HF strategies being implemented at present. Looking at a recent HFR Global Review (see chart 2), I count 28 different types of single-manager strategy. In the economics literature, we say that this makes markets more complete. This allows asset managers to be more diversified ¹¹. This is exemplified by the emerging growth of tail-funds which offer investors the opportunity to hedge serious financial crises.
2. The HF sector plays a valuable role as part of the market information network where prices adjust to market news - without HFs, I believe that prices would take longer to adjust.
3. By virtue of the incentive and lock-up arrangements, many HFs can take contrarian views which challenge conventionally held market ideas and inefficient herd behaviour.
4. HFs can provide liquidity to the banking and insurance sectors and can be a very useful sounding board for pricing risk.¹²

In presenting the case that the HF sector does not need to be regulated, many commentators would advance the following arguments:

1. that no tax dollar has been spent bailing out a hedge funds
2. that HF leverage has generally been lower than that of broker/dealers, investment bank and even commercial banks and there are studies to back up these claims, see for example Ang, Gorovyy and van Inwegen (2011)
3. that the incentives in the HF sector are better aligned than in banking for the following reasons:

• HF managers get paid for positive performance and for assets under management,
• Poor performance leads to redemptions
• Substantial redemptions lead to fund closures
• Investors demand that HF principals have 'skin in the game' by requiring that they are co-invested in their own funds

Thinking further about incentive alignment:

  4. The HF habitat - to use the language of Darwin - is one characterised by high attrition rates (see Xu et al (2011) where weak performers face extinction. This is evidenced in the above chart which shows how new HFs have come and gone over the last decade. Based on HFR data, HF attrition has run at 6% on average over the last 15 years and peaked at 15% in 2008.¹³

Other arguments advanced against greater HF regulation include:

  5. the fact that risk management culture and discipline within HFs is generally strong, the measurement and monitoring of risk is seen to be an area of comparative advantage and there are numerous vendors of off-the-shelf risk management suites which obviate the need for substantial up-front in-house development.

  6. acting in their own self-interest, prime brokers do play a quasi-prudential role in ensuring that weak or distressed HFs are not allowed to impose credit losses on their counterparts.¹⁴

These arguments do have merit in their own right.

But they are not the whole story. The economic frictions I outlined earlier explain why.

First, let’s think about the issue of information asymmetry.

1. HF investment strategies are proprietary by nature. Investors will have very limited knowledge of trading strategies. This means that HF returns and track records are especially important. Owing to the risk of redemptions, HF managers seek to avoid 'down months' and this has been noted in the empirical work of Bollen and Pool (2009) and Cumming and Dai (2010)¹⁵.
2. The diversification benefits, referred to earlier, have encouraged the interest of a wider and less sophisticated array of investors. As a result, investor protection is a greater concern as the impact of a failing HF is likely to extend beyond that of high-net worth individuals to pension funds, insurance and other long-term investment portfolios.

In sum, investors need the protection of an independent administration function to ensure the integrity of monthly performance returns.

Investor co-ordination problems and legal uncertainties

1. Whilst investors may have to live with a veil of secrecy about the details of trading strategies, Lehman's bankruptcy has shown that they must understand what I call ‘the plumbing of the HF’. Due diligence must be conducted on the interactions between the prime broker, custodian and administrators.
2. In the past, many HFs have not prioritised attention to these 'plumbing matters' relative to performance management and capital-raising (or perhaps because of these issues). Having consulted with the HF industry in the past, many HFs did not have much leverage over the prime brokers when it came to negotiating terms around custodial and rehypothecation arrangements. (I recall how in late 2008, some HFs were severely punished by investors for not being able to answer key questions about how their assets were being held at the UK subsidiaries of US prime brokers. Concern about the insolvency regime created strong incentives for HFs to move prime brokers.)

In the wake of Lehman’s failure – where HF assets were caught up in the general creditor estate of Lehman Brother’s UK subsidiary – prime brokers have sought to create different solutions to the problem of a failure of a UK incorporated prime broker, eg tri-party custodial arrangements, bankruptcy remote custody vehicles and the like. But many of these can only be tested when a firm fails. The lack of certainty regarding the right solution suggests that the state has a role to play in removing legal uncertainty.

The US SIPC trustee and the UK Special Administration Regime are all responses to the need to provide clarity to investors and firms in the event of the failure of an investment firm. The Irish authorities are currently working on introducing special administration powers for investment firms and there is a need for similar regimes in other European countries.

Systemic behaviour

The events of LTCM and Amaranth have shown that individual HFs with substantial outsized positions can cause market turbulence which may undermine investor confidence. It is difficult to argue the case that such market volatility is economically beneficial. Indeed, regulatory authorities are concerned about situations where markets may become disorderly due to the impact and actions of a single player.

We have seen cases where HFs and investment banks can crowd the same trade. A real economy shock to the financial sector can have real economy consequences. For example, the GM and Ford credit downgrades to junk status in 2005 led to substantial dislocation in corporate credit markets.¹⁶

The wide range and dynamic nature of HF strategies poses significant challenges for regulatory authorities and their capacity to monitor markets and take decisive action.
Lastly, let’s think about HFs from a financial stability perspective.

From a macro-financial point of view, HFs manage roughly 1% of global financial ^wealth17 and turn over their portfolios approximately considerably more often than institutional fund managers.

Recent steps by the FSB in the area of (i) data gaps and (ii) shadow banking (and by some national regulators like the FRBNY and the BoE) to map the financial system have been motivated by the need to understand the linkages between different sectors of financial markets.¹⁸

HFs are a dynamic and substantial contributor to market activity. The collection of additional data on holdings, exposures and financial flows is the inevitable consequence of regulators and central banks need to leave no part of the financial landscape uncharted.

In the above slide, I have heroically attempted to compare and contrast hundreds of pages of US and EU HF regulation proposed under the DF and AIFMD regimes.

The key points I would highlight are:

1. that the EU sets prescriptive capital requirements whereas the US has no regulatory capital requirement
2. the EU requires independent administrators whereas US HF managers and/or funds have no requirement to appoint independent 3rd party administrators, (although over recent years there has been a trend towards this mostly driven by institutional investors)¹⁹.

Whilst the US applies client asset requirements which require verification of the assets by accountants, in the EU, in the event that client assets of an EU HF are lost the consequences for the custodian are more severe under the AIFMD’s strict liability rules.

I should also mention that other investment firm regulations (such as MiFID) also matter for HFs.²⁰ The failure of the Lehman’s drew attention to the location of prime broker activity globally and how that was influenced by different rules around the rehypothecation of client assets.²¹

Economic inefficiencies arising from these frictions

To summarise the case for regulatory intervention, I would note three key points:

1. Investors are vulnerable to information asymmetries and the integrity of HF performance reporting is key
2. HFs trading strategies can have systemic consequences for one or more markets from time to time
3. HFs are part of a complex global financial landscape which we must understand better

The challenge is to craft regulations which are focussed, proportionate and which efficiently use public and private sector resources ²²  without unintended consequences or spill-over effects. The differing approaches and priorities of the US and EU are a concern in this regard.

But there is also a substantive point to be made about how we develop and implement regulations in mobile, globally-integrated markets.

My earlier points on how European financial regulation is developed were not aimed at criticising the Lamfalussy process but rather to suggest that the raison d’être for regulation, namely tackling the underlying economic frictions, should bring more discipline and focus to the discussions underlying the process.

As I explained at the outset, we have, in the past, regulated to achieve competition and economic growth. At times, the increasing scale of the financial industry on both sides of the Atlantic and in Asia has seemed like an arms race. But we have seen the adverse consequences when the financial industry outstrips the real economy in terms of size and importance and where economic frictions are not addressed. In this more mature phase of financial sector evolution, I would argue that we need to focus on the so-called economic frictions which may lead to market failure.

I would be concerned that by allowing regulations to be crafted without these underlying economic frictions in mind and on a regional basis, we risk creating misaligned regimes which have unanticipated and undesirable spill-over effects.

Indeed, across the whole panoply of financial regulation, there is an ever greater need to focus on (a) the underlying economic frictions causing market inefficiencies and (b) achieve tighter global co-ordination. Whilst, the FSB has taken on some of this role (working with the BCBS, IAIS and IOSCO), I will leave that to others to consider the precise institutional arrangements – in terms of scope, representation and accountability – which might take such an agenda forward. In some arenas, like bank capital this has been progressed, but this is only the tip of the iceberg.

Thank you.

____________________________________________________________________________________________________

¹ See http://europa.eu/about-eu/eu-history/index_en.htm

² The World Trade Organisation has sought to remove these barriers in a global setting.

³ There are other frictions which I will not dwell on here such as: transactions costs and taxes, barriers to entry, property rights and legal claims.

⁴ See Lo and Mueller (2010) for a taxonomy of uncertainty.

⁵ See for example, Kahnemann and Tversky (1979), Laibson (1992), de Bondt and Thaler (1985), Harford (2011)

⁶ For example, the introduction of limited liability and the creation of certain types of bankruptcy regime.

⁷ Such as Cadbury (1992), Rutteman (1994), Greenbury (1995), Hampel (1998), Turnbull (1999) and Higgs (2003) in the UK.

⁸ Notably the US Sarbanes Oxley Act of 2002 and various directives to strengthen the role of the audit function in the EU: EC 1996 Green Paper (COM 1996/321) on “The Role, Position and Liability of Statutory Auditor in the EU” and ‘the 8th Directive’ 84/253/EC.

⁹ The Central Bank of Ireland has recently introduced a statutory corporate governance code for banks and insurers, assisted the funds industry in the development of its own voluntary code and introduced Fitness and Probity regulations across all regulated financial entities.

¹⁰ I would note in this regard, that the EC is looking to create a harmonised and consistent framework for funds, therefore, as UCITS evolves, one should monitor the read-across for AIFMD, and vice versa.

¹¹ Economists would say that HFs make market more ‘complete’.

¹² For example, HFs can play a valuable role when it comes to pricing structured or out-sized retail and corporate transactions. That said, whilst HFs do provide liquidity to the investment banks, they have limited capacity and such trades can become crowded if the underlying flow, eg retail, continues unabated, a good example being the demand for guaranteed tracker bonds during the bull market of the late 1990s.

¹³ Whilst the market may do a good job in ensuring incentive alignments, especially for the medium and small HFs, it is worth keeping this under review to the supersized HF managers with tens of billions under management. Over time, whereas a billion dollar HF would have been a monster in 1990s, it is tiddler relative to the $30-50bn behemoths of 2012.

¹⁴ That said, when prime brokers raise haircuts in a systematic fashion across one or more lines of business, this can have systemic consequences resulting in fire-sales and trading strategies becoming more correlated from a mark-to-market point of view.

¹⁵ Where there is a discontinuity in the [-0.5%,0] range of the cross-sectional returns distribution (see graph reprinted from Bollen and Pool (2010).

¹⁶ http://www.voxeu.org/index.php?q=""node/2420""

¹⁷ HF $2trn compared with McKinsey estimates for global financial wealth (excluding property investments and derivatives) of $200trn.

¹⁸ See Poszar et al (2010) and Gorton and Metrick (2010).

¹⁹ For example, UBP who are a very large institutional HF investor made a public statement post Madoff that they would only invest in HF with an independent adminstrator

²⁰ For example MiFID and the US Securities Act of 1933

²¹ Namely, the fact that there are strict rehypothecation limits for broker/dealers use of client assets in the US under Regulation T whereas in the UK there are not. See Singh and Aitken (2010)

²² See Sami (2009)