With the growth of interest in alternative investments, pressure continues for increased transparency in the operation of hedge funds. As the essence is to minimise opportunities for organisational stakeholders to experience negative surprise, good governance can only be achieved through complete accountability for the factors that drive: i) the search for investment opportunities and ii) the resolution of investment decisions in the face of risk and uncertainty.
Insofar that the risk universe comprises both quantitative and qualitative spaces, hedge fund managers must be able to describe in a unified manner three key factors:
- the element of the whole investment decision-making process that is supported by quantitative model output;
- its boundary conditions, ie the assumptions supporting model construction and the limitations of its output; and
- how the residual uncertainty is resolved.
The management of quantitative and qualitative risk spaces, each of which provides value in the decision-making process, has become disconnected of late; we introduce 'bounded rationality' as an enabling concept for an organisational architecture that allows for their reunification and the role of brand management as securing its delivery.
Words from 'The Fed'
"We should focus more attention on parts of the risk-management process where uncertainty is greatest and materiality of the risks that we can’t readily quantify is highest. This means more attention on the risk factors where the measurement challenges are most complex...
These challenges require using a mix of different analytical tools to help illustrate the range of possible outcomes and the dimensions of uncertainty that apply to the measurement of exposure. The focus should be not on the specific estimates produced for various types of asset price movements or stress events, but the uncertainty that surrounds those estimates and the magnitude of the potential underestimation of losses. Another way to say this is that we probably need to spend as much time discussing the limits of the quantitative outputs of the risk management process as we do on the estimates produced by the models...”
The Federal Reserve Bank of New York
Timothy F Geithner, President and CEO
"Although the importance of the quantitative aspects of risk measurement may be quite apparent the importance of the qualitative aspects may be less so. In practice though these qualitative aspects are no less important to the successful operation of a business as events continue to demonstrate...
Some qualitative factors such as experience and judgment affect what one does with model results. It is important that we not let models make the decisions, that we keep in mind that they are just tools because in many cases it is management experience that helps to limit losses... "
The Federal Reserve Board
Governor Susan Schmidt Bies
How did Quant and Qualitative Issues become Disconnected? – The 'Missing Link'
Financial markets are influenced by human behaviour. Whilst their operation is rationalised through the application of economics, the theory and deployment of quantitative techniques has advanced to such an impressive extent that considerations not yet quantifiable (if ever to become so) have become divided into separate subjects such as behavioural finance. It is, however, all one risk space and whereas quantitative finance restricts its interpretation and description of risk to the market as price volatility, end-investors recognise risk on a broader basis where a more appropriate, fundamental definition of risk might be:
“Known situations of opportunity and danger arising from future events in which something of human value is put at stake and where the outcome is not definite”
after Rosa – Metatheoretical foundations for post-normal Risk, 1998
Qualitative factors in investment decision-making attempt to reconcile this disparity – stakeholder equilibrium is achieved through communicating the presence of uncertainty (aptly described as comprising 'the known unknowns' and 'the unknown unknowns') and being able to account for how it is resolved through the role of organisational culture. The same need for transparency of decision-making and attempting to resolve uncertainty is a principal driver of the interest manifested these days in responsible investment practices.
The Need for a 'Next Generation' Approach
Recent events evidence polarisation within the industry – whilst larger hedge funds are beginning to operate on an industrial scale, smaller funds find it increasingly difficult to contemplate long-term growth prospects. Investors restrict themselves, by default, to dealing with funds they recognise by name or otherwise are familiar with (often the larger funds), even though they might no longer demonstrate the best performance.
Investors find themselves unable to search amongst the smaller funds to the extent that returns are virtually discounted – funds with up to US$100 million AUM account for 63% of funds by number but only 12% of sector assets whereas funds with AUM in excess of US$1 billion account for only 2% of funds by number and 23% of sector assets (IFSL March 2006). This polarisation is likely to increase (and included in the definition of 'small' more vehicles below US$1 billion in AUM) and is a particularly pressing issue given projected increases in allocations by pension funds.
Transparency maximises the number of possible 'docking' opportunities between funds and investors and supports the outreach of the alternatives industry; it underlines the opportunity available for innovative platforms to step in and enable access between institutional investors and funds that are scale or resource constrained.
What about the Cost Downsides?
In terms of performance,
"There is an increasing awareness of the crucial role that governance plays in achieving the aims of funds due to a growing body of research that makes the link between good governance and good investment performance..."
Watson Wyatt Global Pension Assets Study 2007
The point of adding a transparency layer is to be able to properly engage with stakeholders and inter alia increase AUM through improved marketing (a related issue of transparency exists in respect of value added/the overall cost of operations/the remuneration of the fund manager – we discuss this separately below).
Given that in excess of 50% of funds that fail, do so for non-market risk reasons (Understanding and Mitigating Operational Risks in Hedge Fund Investment – Capco, March 2003), we further argue that enhanced transparency is an important asset in the management of operational risk.
Accounting for the Investment Decision
The 'Classical' Approach and its Limitations
The traditional or 'fully rational' concept of decision-making (unbounded rationality) seeks to 'optimise' – to compute demonstrably, through the exercise of rationality, one solution to a problem that is superior to all other possible solutions. In order to do so, we need both to define a utility function and then to compute/solve for it.
With complex non-linear, time dependant problems such as making investment decisions the limitations of the fully rational approach become obvious: in the real world we are faced with uncertainty and with risk and uncertainty and 'real-world' limitations of resources.
The task of determining one unique investment solution can be said to be self-defeating to the point of implausibility as it potentially dictates an ever increasing requirement of resources – as we sort through, identify and discard sub-optimal solutions, those remaining become increasingly difficult to differentiate between.
In solving for utility functions, notwithstanding the sophistication of the mathematics available to the investment industry continuously being improved, the fully rational model of decision-making does not provide well for 'resolving' uncertainty whether arising from the limitations of non-linear modelling techniques or from a more fundamental structural unknowability. As a result, two approaches to the application/use of quantitative modelling techniques are usually employed:
- trading 100% on a signal where the quant model indicates one 'optimum' solution (referred to as 'program trading' herein); or
- trading on a combination of signal (for one or more solutions – refer below) plus judgement.
A single solution can be output by a model as an optimum result where it is constrained by set targets. Whilst all quant models contain assumptions and simplifications in their construction, programme trading contains all of its market assumptions/simplifications within the model so as to produce one 'satisfying' solution.
Non-program traders have two choices: build a model that gives either an answer that you use as a prompt for an investment decision or build a more complex model but one that can only solve for a set of solutions and then choose from amongst the set. In addition to the mathematical limitations of models, how can we account for judgement calls included in the construction or interpretation of model output?
An Indicative Survey of Status
We conducted a 'flash' online survey of hedge fund managers of how some aspects of investment selection were being dealt with – 26 validated responses were collected; AUM was not requested. We found the results illustrative and it is planned to undertake further research as a result. The text of this article should be read with the results below in mind.
The use of quant model output
1. Methodologies ranged from 35% of respondents utilising 100% quant with the remaining 65% clustered in the 25-50% range.
1. Do your investors understand how you make decisions and the uncertainties you have to deal with?
1. Does your investment strategy contain a default action if you have conflicting investment options?
2. In addition to quantitative output and your experience, do the fund's brand and reputation management issues influence your investment decisions?
3. Do you wish you could spend less time on administration and more time on investment search and selection?
4. Do you as a manager wish you had the resources (time or money) to keep developing your quantitative model?
Using a Bounded Rationality Architecture – Recognising Uncertainty
The application of the concept of bounded rationality (refer Herbert Simon - Nobel laureate 1978) specifically addresses decision-making in the face of natural, real world limitations of information gathering/processing, cognitive capability, time (and/or cost equivalent of resource).
The critical aspect of bounded rationality is that compared to the classical model it starts from the other end of the decision-making scale by emphasising the existence of uncertainty and in effect abandoning the search for an optimum solution.
Bounded rationality is not 'irrationality' but rather emphasises the role of accounting for qualitative factors through the recognition of the application by decision-makers of a heuristic 'toolbox' (acquired experience/rules of thumb) and of organisational/cultural norms (influences from their decision-making environment) in resolving equilibrium amongst stakeholders.
Bounded rationality is a decision-making architecture and whilst in effect it abandons the search for perfect maximisation, we would emphasise here that any attempts to account for the qualitative factors present in the decision-making process are of little (if any) consequence without being placed in the context of identifying and working together with the application (and limitations) of the rational, quantitative model.
Furthermore, though absolute numbers might not be attributable to the qualitative component of a judgement call, its proportion of the overall decision can be determined as well as its relative value within a distribution of comparable decisions.
When to Stop Searching – A 'Stopping Function'
How far should a fund manager contemplate improving on an investment choice?
If, in the first place, we have generated results which meet minimum criteria and suffice as an investment choice, the application of judgement by the manager should yield a result somewhere between no worse than the minimum criteria judged to meet investors' expectations and somewhere improved upon that (in the direction of an optimum solution).
We would suggest the construction of 'stopping functions' where through a self-assessment/ confidence flag methodology (comprising inter alia the effects of on-going experience/ learning) decision-makers identify their limitations of adding to the decision's value.
How are Qualitative Factors Accounted for?
Identity & Brand Management
An organisation attempts to communicate its identity (that is a description of how it responds to risk or the opportunities and dangers of wealth-creating activities at large – importantly, in respect of both quantitative and qualitative issues) through its brand, ie the association of sounds and images with an idea, expectation or experience. Brand describes capabilities, values and aspirations. A brand is interpreted by each audience within the context of its cultural environment – comparable to a projection in geometry where varying the 'angle of projection' yields different 2D perspectives of a 3D reality.
Organisational Culture & Normative Cues
Recognition of normative cues is important - investment choice is modified according both to the investor's and the fund manager's environments. Examples of this can include i) expectations from the end-investor about return (absolute and with reference to peer groups) ii) a mandate for responsible investing and iii) from the fund as an enterprise - its investment style, corporate values, etc. Normative cues are descriptive of decision-makers' worldviews; in order to attempt to optimise marketing opportunities (as part of stakeholder engagement at large) a fund should attempt to engage with an investor's worldview - this concept derives from philosophy and refers to the framework of ideas and beliefs through which an individual interprets the world and interacts in it.
It is important to note that worldviews are unlikely to correspond in all respects and in practise engagement is often sought at the level of a number of 'sub-beliefs'.
In respect of hedge funds, it is critical for a fund as an organisation to demonstrate an organisational culture in its own right (in addition to that of the manager) that (includes and) guides the fund manager.
Reputation, coveted in financial markets, can be considered to comprise the coherence over time between an organisation’s brand and each audience’s experience of whether the expectations created are met or not. The brand should avoid establishing an identity that is unsustainable either as an act of self-legitimation or in response to false or unrealistic demands. A brand not perceived as honest is recognised often as representing more than is true or held as a belief and damages an organisation’s reputation through creating undeliverable stakeholder expectations. It is equally incumbent upon stakeholders to take responsibility for acting with regard for the values, capabilities and limitations that define an organisation’s identity.
A recognised organisational identity and culture, in conjunction with an effective real-time public relations and communication programme, minimises opportunities for the experience by stakeholders of ‘negative surprise’ and the realisation of conflict. Conflict, particularly where manifested beyond an individual in the form of negative crowd formation, predisposes damage to an organisation’s reputation and may result in a decline in the customer base, revenue reductions, greater than usual levels of uncertainty and extraordinary cost all of which in turn may affect the ability of an organisation to continue its present or future activities.
Investors want and need to know a fund. Brand management is a holistic activity that maintains the integrity of an organisation's identity/culture with regard to both internal and external stakeholders. How is a brand's function managed across an organisation? If the fund manager is focused on managing market risk, who can implement and manage the condition of this 'cultural glue' across the decision-making architecture?
Much work has been done on the subject of organisational governance in the last few years that clearly distinguishes the (nonetheless inseparable) roles of the governance of economic performance and governance of the corporate architecture (International Federation of Accountants 2004 - Enterprise Governance: Getting the Balance Right). We would commensurately suggest there is a strong argument for hedge funds to employ a C-suite function to include/supervise brand management as a critical part of the enterprise architecture/operational management (eg refer to Harvard Business Review May 2006 - Second in Command: the often misunderstood role of the Chief Operating Officer).
An Accounting for Value Added
Through the application of transparency, investors can start to assess more properly whether a fund manager is adding value beyond a random investment selection – this is already an issue and will prove more so as alternatives evolve further into the mainstream.
"…while allocations to alternative assets are attractive because of the combination of strong diversification and access to skill they may lack merit in many cases going forward given higher pricing and cost structures …
The structure of incentives is poorly aligned with funds' requirements and needs to change..."
Watson Wyatt Global Pension Assets Study 2007
Is the Investment Industry Alone?
Whilst the investment industry recognises the existence of uncertainty, it has not 'embraced' it into the investment decision-making process at large and does not appear to have utilised the experience of other professions. The progress of quantitative techniques has to a large degree held the attention of the industry possibly to the detriment of attempting to accounting for qualitative issues.
It is worth noting two other professions in particular – the Military and Medicine.
The Institute for Strategic Studies, Washington DC commented on the US Department of Defence quadrennial report 2006 that of two fundamental imperatives identified by the Pentagon, one was the reform of strategic decision-making to recognise the role of bounded rationality and the need for a decision-support structure that formalises drawing upon multi-disciplinary experience in attempting to make informed command decisions.
Whilst the military has focused to a large degree on supporting command decision-making, the medical profession has made notable progress in communicating risk to patients as part of the delivery of informed choice. Particular attention has been drawn to translating the meaning of statistics (such as disease survival rates) applicable to a population through to what they mean at the level of an individual.
If progress can be achieved in respect of complex medical issues there appears to be room to move against the 'defence' raised by the finance industry in that amongst lay people there is little understanding of the complexities of finance.
It is further worth noting that within a number of complex medical issues it has been evidenced that i) where there exists a common understanding that uncertainty exists in respect of treatment options and that ii) the treatment plan is a dynamic process subject to re-evaluation/revision then prospects for recovery are improved.
Natural Language Forms
Percentage probabilities at the level of an individual who experiences a 0/100 result are difficult to grasp. The use of natural language forms in for example medicine has been explored and could be utilised to a greater degree by the investment industry – a simple example could be to describe, for example, portfolio future price probability distributions rather than as a 'functional curve' to an investor in the form of 100 squares (like a 'stacked boxes' histogram); it's relatively simple for the investor to visualise (subject to limitations of the quant analysis) that their future pension value can be any one of those 100 squares. This methodology facilitates the development of user interfaces demonstrating the effects of news, portfolio change, stress-testing, etc – it's a strong methodology for engagement.
Increased transparency leads to funds engaging more closely with investors. Bounded rationality provides a compelling illustration as to why successful investment is not dependant on a single outcome but requires revisiting and reconfirmation over time. With enhanced opportunities for informed choice, potential for the distribution of hedge funds based on improved stakeholder communication from both regulatory and marketing perspectives should be increased.
There exists a primary need for hedge fund managers to account for:
- the construction, application and the limitations of the quantitative models used in making investment decisions; and
- the qualitative factors involved in resolving uncertainty.
Many funds are resource constrained – in our survey, 73% of managers wanted to spend more time on investment search and selection. As such, an opportunity exists for service providers/platforms to meet on a scale effective basis a number of 'next generation' needs of the hedge fund industry:
Provision must be included for brand management in order to account for and help communicate the investment decision-making process and how uncertainty is resolved in a fully unified, transparent manner comprising both quantitative and qualitative factors.
In our survey, only 19% of respondents said that brand currently influenced their investment decision-making indicating substantial potential for development.
Brand management unifies issues critical to the support/delivery of equilibrium of stakeholder relationships, both internal and external.
We recognise the difficulty of describing complex probabilistic modelling techniques. However, greater effort must be made to describe the assumptions and simplifications that go into model building and the usefulness of the output; communication with non-experts can be challenging - we encourage reference be made to other industries, particularly the use of 'natural language' forms in communicating complex risk propositions to stakeholders.
Risk comprises opportunity and danger. A clear exposition of the opportunities being sought by the fund manager and the dangers they are likely to encounter is a necessary context to the accounting of their search.
In our survey, 69% of managers thought their investors did not understand the uncertainties of the investment process.
Investment is a dynamic process and a strong corollary to accounting for the decision-making process is to provide support to the fund manager in the on-going maintenance and development of modelling/risk management efforts. In our survey, 69% of managers wished they had the resources to keep developing their models.
Fifth Capital combines banking industry expertise with academic support to be at the leading edge of considerations relating to the foundations and organisational architectures that enable the pursuit of Sustainable Wealth Creation. Its Magus division focuses on providing services to the Alternative Investment sector.