Risk Management in Risky Times
As the dust begins to clear from the fallout following the detonation of the bombshell that has gone off in the financial markets this autumn, many investment managers will now be looking at themselves to see how best to operate in a slimmed down, potentially more volatile environment. Whether this is driven internally or forcibly by the regulatory authorities, it seems clear that Risk Management will once again be taking a front seat in everyone’s list of priorities for 2009. If we learn just one thing from the crisis – it is that firms on both the buy and sell sides need to focus on risk management in general, and on those less operationally mature areas such as OTC derivatives in particular.
However, the slew of banking defaults and near defaults, and the huge losses made in the sub- prime market are not the sole driver in this trend. For several years now the fund management industry has been moving away from its relative return, beta- seeking roots towards a new paradigm. Alpha generation has been the watchword on the streets for several years now – driven in part by the need for positive differentiation in a crowded rising market, but also by the increasing sophistication and risk-awareness of the investors themselves. Investment products servicing more specific investor needs, such as capital preservation, time horizon management, and downside risk reduction have appeared and all require much more precise and proactive risk management. It is this outcome oriented investment, as people plan for early retirement, or to put their kids through fee-paying schools, that is driving the importance of risk management as much as the recent crisis.
The concepts and the practices of risk management in an investment manager are not new, but are implemented in a number of different ways and with varying degrees of success. In general terms however there are a number of issues that will need to be addressed when looking to improve internal practices – some that are more general in nature and some that are focused on OTC derivatives in particular;
Risk Management as a function is not always viewed as top priority, especially as it not seen as a revenue stream in and of itself. Risk Man- agers do not always have Board representation and sometimes the governance structures (if they exist) cannot by their nature react fast enough to events or have the power to implement suggested solutions.
Risk is often managed well at the micro, fund level but not at the enterprise- wide level. In order to achieve a holistic view of a firm’s diverse risks presents some difficult, but not insurmountable issues. These would include standardisation of risk measures and of the data required to calculate those measures – your front, middle and back offices should be using the same reference data for the risk analytics to make any sense. In the world of OTCs there are new XML standards (FpML) for messaging that need to built into the operating model and potentially new systems and services that automate current confirmation and processing workflows that require integration.
The main tool used by risk managers, Value at Risk or VaR, is based on centuries old mathematics. Stemming back to Gauss in the 18th century, the normal distribution that under- pins VaR calculations appears to work well in stable times, but does not seem to be able to cope with some of the large “six sigma” events that have occurred in the past (1987’s Black Monday Long Term Capital Management and the Russian Bond Default, as well as the current credit crunch). Indeed David Einhorn, of the Hedge Fund Greenlight Capital has been quoted as saying that VaR is “an airbag that works all the time, except when you have a car accident.”, and by Barry Schacter (who runs the risk management website Gloria-Mundi) that it is “a number invented by purveyors of panaceas for pecuniary peril intended to mislead senior management and regulators into false confidence that market risk is adequately understood and controlled.”
The OTC market in particular is about to go through a sea-change in its structure as a result of the credit crunch. For investment managers whose product range and trading strategies require the use of these instruments, this will mean that they must react to new regulations and processes. It seems likely that there will be more regulatory oversight on collateral management (see White Paper, “Buy, Build or Bounce”), and a shift in industry focus from the confirmation arena towards post- trade lifecycle management and central clearing
Organisational and Technological Issues
Essentially good risk management is about building awareness throughout an investment organisation, learning to apply quantitative techniques without being ruled by them, and developing a sound sense of judgement. In order to do this investment firms must be organised in such a way as to facilitate these things. In practice therefore that means that:
Knowledge of risks and risk management techniques must pervade throughout the organisation all the way up to Board level
Management of risk must be enterprise-wide and not silo-based
Governance and management information structures are in place
There is an agreed set of parameters for measuring risk
Data is standardised for use by all areas of the firm
The first starting point for any firm should therefore be the publication of their Risk Management Policy. Recent studies (c.f. Deloitte 2007, E&Y 2005) have shown that investors are increasingly asking for stated policies and that the regulatory bodies (including the FSA) consider it to be best practice. The stated policy can be a powerful tool for communicating and focussing efforts internally towards the implementation of a fully articulated risk management framework and many external investment consultants require this as part of their manager assessments. Secondly firms should revisit their governance structures (an example of which is below), and ensure that there is sufficient and accurate management data available to be able to set the parameters for and then manage the risk levels of the organisation. Data is clearly key here. Those measuring the risks, whether in the front, middle or back offices, should be doing so using the same data used in a standardised way – something that obviously has major technological implications. The implementation of a risk management system or service may go some way to achieving this, but should not be seen as a cure-all without continued organisational focus on the overall risk culture.
Technical and Structural Issues
In the longer terms firms will also need to start addressing some of the issues detailed above. There are many different flavours of VaR – the most widespread being historical and parametric. However, as explained earlier these are based on the assumption that the markets have a normal probability distribution. Traditional theory has it that the capital markets are an equilibrium system, where prices tend to converge back towards stability. Recently however academics are increasingly seeing them as complex, adaptive (evolutionary) systems that are far more prone to periods of chaos than was at first thought. As far back as the 1960s Benoit Mandelbrot (he of the eponymous Mandelbrot Set) argued that markets behaved more in line with his fractal theory than with the normal distribution, and more recently Nassim Taleb’s book on “The Black Swan: The Impact of the Highly Improbable” argues that large-impact, hard-to-predict, and rare events beyond the realm of normal expectations are far more common than is the case if you see the world through the normal curve.
I am not advocating that we ditch VaR forthwith however. VaR is extremely useful; not only as it is in fact a pretty good approximate measure in more stable times, but also as it gives us our only currently-available universal “language” of risk. But investment firms should be looking beyond VaR to see what other techniques are available. As Peter Cotton, the CEO of Julius Finance (a firm that provides innovative new risk and pricing models for complex credit derivatives) said to me “...it is very difficult to create a coherent mathematical picture of what the market is saying – you cannot use copulas to fit the whole picture...”. There are therefore firms, such as Julius Finance, that are capable of providing risk models based on new parameters. Indeed many firms are also using EVT (Extreme Value Theory) and ETL (Expected Tail Loss) that have come from research on the physics of tsunamis (which apparently have much “fatter tails” than you get with the normal curve) – as well as Stress Testing their models to see the effects of extreme market movements. All these new techniques however come with their own peculiar set of problems and have not yet become common currency throughout the industry.
From a structural point of view many firms have already begun to ensure that they have robust and scalable operating models that cover all asset classes, including OTC derivatives. This is being helped by the mushrooming of the number of service providers offering them assistance in these areas. In the last few months alone there have been announcements from the likes of State Street, Northern Trust, SmartStream, DST Inter- national and Checkfree about their new offerings for monitoring OTC products post trade. IT budgets may be tightening across the industry, but I suspect that many boards will see this as a necessary, and a risk- mitigating expense.
In the coming downturn, and with conditions as uncertain and volatile as they are at present, Risk Management will become a top priority for the investment management community in 2009 and beyond. Driven both by an internal need to more closely control risk in risky times, by the coming regulatory backlash, and by the ever increasing sophistication of investor’s requirements all managers will need to reassess their risk management structures, processes and systems—a situation that is unlikely to be transitory as we enter a new era in financial services, and an era where we will likely to be referring to ourselves in the future as Risk Managers rather than as Fund Managers.