Risk in investment is getting into the news rather more in the present decade. The obvious stimulus to this attention is that investment markets are producing poor returns (risk is seen as dull in bull markets), but we can cite three new factors at work in addition to this.
o markets are going through a particular patch of more volatile conditions;
o the measurement of risk is evolving and an increasingly sophisticated view of what is ‘risk’ is being applied (you could characterise this as the new field of risk budgeting);
o a new set of legalistic concerns about risk are emerging following the Unilever - Merrill Lynch case.
I’m not sure much has really changed underneath as the key principles in risk management are largely inviolable. They can be summarised in three key points:
o successful investment involves risk taking so risk management will always be a key success factor in investment management;
o risk is multi-faceted and means different things to different people, in particular clients define their risk tolerances in different ways in different markets;
o risk management introduces particular challenges to global portfolios and investment groups where consistency is critical.
How strong are the global factors? Of course it all comes back to what we mean by ‘global’. The most obvious sense is an external view – considering the extent to which local market share is dominated by firms that operate globally. As we look at the top managers by assets under management, we can find a number who have substantial market share in multiple markets with service infrastructure in support. But none have top ten positions in all the leading markets.
In asset management though, global characteristics show up more in what the firms ‘manufacture’ and more particularly how they produce their performance – you can see this as an internal view of globalisation. In pursuit of competitive advantage in managing global portfolios, some firms have built ambitious plans for the global connectivity of their investment management processes:
o to apply a common investment philosophy across the world
o to do research the same sort of way
o to have portfolios spring up from a global view of industries in which country factors are limited
o and to adopt common risk management disciplines that are aligned across all their products.
Risk management has come in for particular attention in this list as it represents the foundation of many of the investment themes that these groups are pursuing.
Most focus on risk management has been in the analysis of portfolios. Various software packages have been developed that use past patterns of stock prices to characterise the likely future risk characteristics of a portfolio going forward. While some of these methods have been developed in-house, the dominant risk measurement approaches are provided by external providers. The market leader in this area is the BARRA organisation and it is hard not to think of BARRA as something of an industry standard such is its penetration.
Much recent attention has focused on the accuracy of current methods. BARRA has been singled out for particular attention. Much of the criticism of BARRA is based on the recent experiences in which predicted tracking errors have appeared to be too low and there are concerns that the measure is a biased estimator. The periods of trending markets when momentum style effects are in evidence have been the ones when the estimates have been least reliable.
My view is that these estimates do vary in their accuracy over time but that this is not too surprising. The critical issue is learning to adapt your risk thinking to fit the different phases of the market. To criticise the risk measurement software for this sort of failing is statistically naïve. It is better to recognise the considerable extent to which patterns in portfolio performance go through changing phases and apply more skill to interpreting the risk measurement tools. That is not to say that research into risk and improved model building should not continue to improve. But, the principal challenge lies in the interpretation of the risk metrics.
Many global firms are engaged in building more sophisticated risk tools for themselves. Much of the enormous increase in the IT spending of investment groups has been in risk systems of various sorts. The challenge has a number of dimensions:
o enhancing portfolio construction
o placing a check on existing risk measurement tools
o covering other broader interpretations of risk, including the risk of non-compliance with contractual terms.
In addition to the technology involved, these systems bring various cultural challenges. Investment managers that previously could exist with a more vague idea of risk in their portfolios, are forced to confront much harder data concerning risk. All of this must be absorbed into new style portfolio construction processes. It has been a major challenge and has presented a real competitive edge to those that have adapted best.
While the influence of risk measurement was apparent in the Unilever-Merrill Lynch case, the discussions there most highlight the importance of more dialogue in risk management issues. The critical question for every manager to consider is what is the client’s goal? This is not just the overall performance target – that should be reasonably easy to specify. What both parties find difficult to express is the appropriate amount of risk in the manager’s approach and an acceptable distribution of the outcome. Much of the reaction by investment groups to the Unilever case has focused on the legal aspects. Managers have been keen to gauge whether their client contracts are OK. It is an understandable line to take but really only scratches the surface of the issue.
The industry needs a better frame of reference for risk. This applies equally to managers and funds and we see below the best practice points that should be present in manager-fund relationships.
1. Form a better mutual understanding of risk
Funds need to form a clearer view of what they see as risk following dialogue with managers. We think the concept of the risk budget is critical to this understanding: that funds should express how much risk they expect managers to take and give some guidance as to what they consider an appropriate pay-off for that risk in terms of additional return.
2. Express fund risk preferences in the manager contracts
While it is tempting to retreat from discussing risk in contracts, I think this is short-sighted. In practice, most agreements will focus on aspirations for risk rather than specific contractual obligations and with appropriate drafting managers need not fear undue liability risks. While risk has been handled lightly in most contracts to date, I favour much more detailed attempts to describe expectations in future versions.
3. Establish better dialogue with managers on risk
Central to the risk dialogue is a quantitative framework for considering risk. The key tool forthis is the risk budget. But numerical methods have their limitations because risk is complex and fast-changing. This needs regular discussion and debate between manager and client to provide qualitative context.
The major challenge in risk management is moving from ad-hoc and unstructured views of risk to a more coherent and consistent framework. While investment groups have invested substantially in the field, they remain vulnerable in two areas:
o integration of the quantitative methods with the qualitative and intuitive views of risk. I would view any situation where one of these two is dominant as suspect
o better client communication on risk issues, with a dialogue that helps reduce surprises.
I expect the major efforts in the next few years to be applied to these two issues. Success in these areas of risk management will be critical to successful fund management.
Roger Urwin is global head of investment consulting at Watson Wyatt