| Performance & Risk Systems: Lessons Learnt from Implementing such Solutions in Asset Managers |
|
Background
This article has been written by David King, who runs the consultancy function at FusionExperience, who have been very active in the performance and risk arenas over the past 2 years. David provides some insight into the lessons learnt from such projects.
|
What Do We Mean By Performance & Risk?
As a consultancy company advising our customers on products and services available in the market, it is sometimes quite difficult to distinguish between the solutions which have been traditionally aimed at the needs of the performance or risk manager and the solutions aimed at the fund manager or quant analyst.
The market for risk systems is littered with a large number of suppliers, many of whom can be very small in terms of company size and numbers of client implementations. Such tools tend to be proprietary and aimed at individual users’ need to manipulate portfolio, benchmark and fundamentals data to satisfy specific requirements at a point in time. Risk systems tend to provide ex-ante risk modelling algorithms which allow future forecasting of portfolio characteristics based on certain event types. Similarly, this can be applied to back testing of the resilience of portfolios were such events to occur in the past.
Performance systems are by their very nature backward looking in terms of investment returns. Resilience of past returns can be measured in the form of ex-post risk measures through statistical analysis of the volatility of returns and expressed in the form of statistics such as Sharpe ratio. Where we have seen confusion is in areas such as UCITS 3 fund reporting, which requires the ex-post analysis of portfolio Value at Risk (VaR). Such analysis is currently only available through risk type systems.
Organisational changes within asset management organisations have tended to try to align the risk and performance management functions under a single management structure. This makes sense in terms of ensuring a consistency of methodology, but is often a problem for the fund managers who see ex-ante risk modelling as their own tool to assist in the construction of portfolios.
Instruments, Products, Fund Types…
The landscape becomes more complex when we introduce new instruments or products and new fund types. One aspect that we at FusionExperience always look at in the selection of a supplier, is their heritage – where have they come from?
One of the issues around fixed income performance attribution, for example, is that the majority of players in the traditional performance attribution market come from an equity background. It is a major step for such suppliers to move into the fixed income market. One might argue that it may be an impossible step to move into the fixed income OTC market; the skill of supplier selection is making intelligent bets and subsequently backing them up with risk mitigation strategies around implementation.
At FusionExperience, our product and supplier scoring mechanisms tend to focus not only on the product but the heritage of the supplier, and the number of degrees by which they have had to adjust their core focus to provide the services that our client is now asking for. The size and agility of the supplier concerned is also a key factor. In most cases big is not better; but on the other hand small is not necessarily beautiful….
Project Scope
Any performance project is heavily reliant on data. Implementation of performance systems can be made simpler by a significant factor if the supplier of the system also provides data. However, be careful what this means. There are certain suppliers that offer a data aggregation service; that means that they receive market data from multiple sources and validate it prior to providing it to you in a single format. Other vendors may offer benchmark and other data but charge you a premium for collecting it from the various benchmark providers without providing a validation/cleansing service.
It is also important to establish, up front, whether you sign just one master contract with the supplier of the performance system in order to receive market data or whether you must enter into separate agreements with the market data vendors too. In short, be aware of the service you are receiving and be clear on any premium that you are paying in price terms and how that elates to improved service.
The performance team itself will have a number of differing requirements ranging from measurement of returns on a buy & hold basis, full daily time weighted stock level returns, attribution based on equity style models (such as Brinson, Karnovsky-Singer), fixed income attribution (such as the GRAP methodology), global GIPS compliance under CFA Institute standards, specialist drill down into performance effects and downstream dissemination of performance data.
From a downstream perspective, the performance system needs to deliver return information to many parts of the business including marketing & client services, fund managers and front office, client reporting and risk management. How such information is provided is important to determine.
Components Of A Performance Application Strategy
Based on the overall scope of the requirement to provide information on performance returns, there are a number of components to consider: |
|
|
|
|
Conclusion
The world of performance and risk measurement & attribution is clearly complex and rapidly changing. Like any problem, it is important to decompose it to the basic components and then to build a sensible implementation approach from there. |
|
|
