Risk Management: A Dynamic Process
IN PRINT ARCHIVE CIR Winter 1999
|Risk Management: A Dynamic Process|
|"Measurement in itself is not the solution...rather, it is the systematic process of identifying, monitoring and reviewing risk"|
|Shawn Menard, CFA|
From a plan sponsor's perspective, risk deals mainly with whether the plan will be able to meet its liability commitments or maintain its surplus status. Factors which can influence this desired outcome include market, exposure, currency and liquidity risk. How can a plan sponsor correctly analyze and manage these risks? The key is timely and accurate knowledge of the plan's investments at all times.
Purpose of Risk Management
Traditionally, institutional investors have allocated the lion's share of their analytical and technology budgets towards making money, with little or no resources devoted to minimizing the potential of losing money. Risk management systems have too often been viewed as a costly control function with little or no benefit to the bottom line.1
Pension plans can manage risk in several ways. The simplest method is to develop guidelines to identify specific risks which could jeopardize the plan's objectives, then institute proper supervision of investment policies. According to risk consultant Christopher Culp of CP Partners in Chicago, "A proper risk management program should complement, not replace the investment strategy put in place by the plan sponsor, but should be able to measure, monitor and manage the risk of an investment portfolio. Risk measurement involves quantifying the risks faced by investors. The measurement system should examine all risks that the plan is currently exposed to relative to the risks of a target benchmark. Although risk resides at the individual security level, it is important to look at summary risk measures." 2
Until recently, limited tools were available to a plan sponsor who wanted to aggregate market risk across all portfolios and asset classes. Sponsors need answers to questions such as:
* What do you do if your equity portfolio's beta exceeds its risk threshold or the duration on the fixed income portfolio is longer than that of the benchmark?
*Which portfolio is riskier?
* How do you compare risk or aggregate risk across portfolios?Standard deviation has some usefulness, but since the measurement is historical in nature and only shows the past behaviour of the portfolio, it does not warn whether a fund has suddenly changed its risk profile. Unless risk can be measured it cannot be managed!
That's why certain plans looked towards VaR methodologies to see if VaR could be used to manage their pension fund's risk. Why Use VaR?
VaR is gaining increased acceptance as a measure of plan-wide risk. VaR is an attractive measure for plan sponsors for many reasons. It is an easy number to understand and explain to the board of trustees. It is also useful for monitoring and controlling risk within the fund. Lastly, VaR allows risks across different types of securities to be aggregated and represented using a common denominator.
Everyone can understand risk when it is explained in dollar terms. A portfolio that has a $1 million VaR with a 95% confidence interval would expect to lose no more than $1 million, only 5% of the time. VaR is an exhaustive measure of market risk and one that can be aggregated at all levels, by manager to the total fund level. You can't simply add up risk measures such as beta, duration and standard deviation to get a complete measure of risk. VaR allows you to do just that.
VaR is also useful for monitoring and controlling risk within the fund. As a monitoring tool, VaR draws attention to large swings in risk caused by the purchase of risky financial instruments and/or excessive concentration in securities, sectors or countries. The use of this tool helps the plan sponsor evaluate the effectiveness of managers or investment strategies by providing a consistent methodology for estimating risk over time, relative to the overall market.
Sponsors can also use VaR to develop risk budgets for managers. Managers are assigned a dollar amount of risk, and must manage their portfolios within this prescribed amount. A red flag goes up when a security is purchased that places the fund above its stated risk threshold. The benefits of this approach are twofold: the sponsor no longer needs to approve every transaction, and managers are able to be evaluated on a risk-adjusted basis.
The sponsor must ensure that the benefits of implementing a VaR system do not exceed the cost. In order to decide which model is most appropriate, it is important to understand not only the intricacy of the models, but how the plan intends to use the information provided by a VaR system. The complexity of the needs will often dictate the model, and hence the cost of a VaR system.
Another factor affecting implementation is how often the VaR is calculated. The purpose of VaR is to identify risk, so the more frequently the tool is used, the better equipped a manager will be to manage risk when it arises.
Despite the benefits VaR provides, it is not the 'be-all and end-all' for risk worried plan sponsors. VaR is simply one tool in the risk manager's toolbox. VaR will not prevent, or even forewarn a plan sponsor of market crashes or sudden changes in interest rates. It does, however, allow the plan sponsor to establish a particular risk target and monitor whether the fund has exceeded the threshold or desired risk level. It also ensures that the risks the sponsor believes and wants the plan to be exposed to are the same as the actual risks of the plan.
Detailed exposure reporting is key to any risk management program. As plan sponsors continue the trend of sub-contracting the investment management function to external advisors, there is a risk that the individual holdings within the externally managed accounts will overlap, thus exposing the fund to unwanted concentration in particular issues, sectors and markets. For this reason, it is vitally important to look through a plan's individual holdings and aggregate exposures to different markets, sectors, and issues to ensure that the plan is achieving the full benefits of diversification.
Although the process of managing managers can vary from one sponsor to another, an effective review of investment managers generally requires access to technology, fundamental style characteristics and a listing of portfolio holdings. Portfolio information, linked to a database of analytical characteristics and technology provides a consistent means of analyzing this information on a regular basis.
Alternative Risk Measures
In their simplest form, alternative risk measures can offer insight into which investment advisor has demonstrated superior risk adjusted performance relative to a benchmark. By using more complicated risk measures one can examine downside risk volatility, in order to capture the probability of achieving negative performance. "You can't really separate risk analysis from performance analysis. If a manager is making a large return, obviously he or she is taking on some kind of risk to gain those returns," according to Desmond MacIntyre, director of risk management for General Motors Investment Management Corporation.3
Among the measures available to help ascertain plan risk, plan sponsors consider downside, variance and beta to be the most important. Downside risk captures the uncertainty of not achieving one's goals; variance captures the risk of not achieving the average return; and beta captures the risk of being in the market.4 While the greater number of ways used to measure risk the better, no risk management program should omit the consideration of these three key measures. Additional measures, such as information ratios and Sortino ratios, will also allow the sponsor to examine the value added by investment strategies, on a risk-adjusted basis.
2. Christopher L. Culp and Andrea M.P. Neves, "Risk Management by Securities Settlement Agents," Journal of Applied Corporate Finance, vol. 10, no. 3 (fall 1997): p. 62-63.
3. Paul G. Barr, "Guidelines to Improve Information Control," http://www.pionline.com, October 1997.
4. Author Unknown, "How Risk Analysis was Accomplished," http://www.pionline.com, March 1995.
Shawn Menard is director of business development for Russell / Mellon Analytical Services.