Style analysis examines the manager’s risk exposures in relation to an appropriate benchmark and the changes in those exposures over time.
The risk exposures allow for the classification of managers by style for selection purposes and to perform returns-based style analysis (RBSA) and holdings-based style analysis (HBSA).
Monitoring the RBSA and HBSA output over time can help to detect style drift, whereby the manager’s actions are deviating from the manager’s stated style.
Returns-based and holdings-based style analyses provide a means to determine the risks and sources of return for a particular strategy. To be useful, style analysis must be:
- Meaningful: The risks reported must represent the important sources of performance return and risk.
- Accurate: The reported values must reflect the manager’s actual risk exposures.
- Consistent: The methodology must allow for comparison over time and across multiple managers.
- Timely: The report must be available in a timely manner so that it is useful for making informed investment decisions.
Style analysis works best for publicly traded investments with frequent pricing data. For less liquid investments, style analysis can still be used to generate questions in the due diligence process.
RBSA estimates the portfolio’s sensitivities to security market indexes for a set of key risk factors. One complication of RBSA is that the risk factors are estimated rather than using predetermined style categories. However, the approach is top-down in nature and little additional data is needed to perform the analysis so the computational approach is relatively easy. RBSA can determine the key risk factors and return drivers for basic and complex strategies. Also, RBSA uses objective data and allows for comparability between managers and through time. Finally, RBSA can be performed on a timely basis.
RBSA lacks precision since it essentially assumes that there is a static portfolio for the period. As a result, it makes it difficult to ascertain the impact of multiperiod investment decisions for a given period, and could alter the breakdown of the sources of value added. The portfolio may also contain illiquid securities, so stale prices could understate risk exposure. When performance is computed based on appraisals of ongoing projects and the internal rate of return of cash distributions, the short-term volatility may be understated. A manager’s true return standard deviation is best gauged over a longer time period. Finally, reporting timeliness will decrease in the presence of illiquid or nontraded securities due to the longer time required for pricing.
HBSA looks at the actual securities included in the portfolio at one time. That allows one to estimate the current risk exposures using a more security-specific (bottom-up) approach. Many of the advantages are the same as for RBSA. Overall, HBSA is most appropriate for equity-based strategies.
A key drawback of HBSA is the increased computational requirement as complexity increases and transparency decreases. HBSA uses a point in time analysis format that may not be useful in projecting into the future or if the portfolio has high turnover. Similar to RBSA, the presence of illiquid and nontraded securities results in stale pricing and that could understate risk exposure as well as decrease reporting timeliness.
Capture Ratios and Drawdowns
Capture ratios determine how suitable a manager is with respect to the investor’s risk tolerance and time horizon.
The upside capture ratio (UC) looks at capture when the benchmark has a positive return. Based on the benchmark return, UC that is higher (lower) than 100% is indicative of out- (under-) performance.
Conversely, the downside capture ratio (DC) looks at capture when the benchmark has a negative return.
The capture ratio (CR) = UC ratio / DC ratio. The CR is a measure of return asymmetry, > 1 = positive asymmetry (convex shape), and < 1 = negative asymmetry (concave shape).
In examining positive asymmetry (convex shape), the question is whether the investment strategy is naturally convex or if the convexity occurs because of manager skill. With the former, consider an example of a hedging strategy that involves buying a series of out-of-the-money puts prior to severe market downturns. The result is positive asymmetry with many small losses (low DC ratio) due to the puts expiring worthless with far fewer large gains in this scenario (high UC ratio). With the latter, consider an example of a long-only equity strategy requiring active management to minimize losses and maximize gains. That requires manager skill, but may not result in consistent positive asymmetry.
When betas are increasing (decreasing), momentum-driven strategies should have higher (lower) UC than value-driven strategies. A low-beta (high-beta) strategy will have lower (higher) UC and DC. Therefore, CRs can be used to confirm the investment strategy.
Drawdown is the total peak-to-trough loss for a specified time period; maximum drawdown is the largest peak-to-trough loss during that time period. Large drawdowns are not appropriate for investors approaching the end of their investment horizon. Drawdown duration is the total time from when the drawdown begins to when the total drawdown recovers to zero (the latter achieved with offsetting gains).
A low-beta strategy may have lower absolute returns but the lower risk in the form of lower drawdowns may result in better risk-adjusted returns.
Drawdowns are useful for identifying poor or poorly executed investment strategies, weak internal controls, and operational problems. Significant or extended drawdowns could cause a manager to utilize self-preservationist tactics that could harm the investors.
There is a fine line between risk management versus self-preservation. For example, it may be prudent to immediately sell assets with unrealized losses in a market downturn, because those losses may become worse later due to a fundamental change in the asset. It might be risky to sell assets in a market downturn should there be a subsequent reversal. The question is whether the manager acted properly in accordance with the IPS, acted for self-preservation reasons, or had a sudden overreaction.
In applying the concept of drawdown to the IPS, those investors with shorter time horizons and lower risk tolerance with less time to recover from losses should invest with managers with smaller and less extended drawdowns.