- Tracking Error is a flawed metric. Its role as the denominator in Information Ratio inhibits Information Ratio’s capacity to retroactively assess the success of a strategy.
- A productive and helpful way to discern differences in anticipated performance patterns is to disaggregate a strategy’s performance into periods when it outperforms its benchmark and periods when it underperforms.
- When disaggregating a strategy’s performance into periods of outperformance and underperformance, investors should focus on the relative magnitude of excess returns in each environment and the frequency with which a strategy outperforms its benchmark.
Myriad metrics exist to evaluate the historical performance of an actively managed investment strategy. Many of these metrics are risk-adjusted and provide unique insights into a strategy’s past performance. Since no metric is perfect, each of these retrospective measures has its own strengths and weaknesses. Some of the widely adopted performance metrics that we use include:
• Sharpe Ratio
• Jensen’s Alpha (“Alpha”)
• Information Ratio
The Sharpe Ratio
The Sharpe Ratio measures the excess return a strategy generates over the risk-free rate per unit of standard deviation, a commonly used proxy for the total risk of an investment. What makes the Sharpe Ratio effective is the ease with which it facilitates comparisons between investments with different risk profiles by scaling a strategy’s excess return over the risk-free rate by the amount of risk taken by the strategy. The primary drawback of the Sharpe Ratio is that the risk-free rate is not a relevant benchmark for most actively managed strategies; instead, investors are mostly concerned with whether an actively managed strategy generates a return greater than its relevant benchmark. If a strategy fails in that endeavor, many would consider it to have failed its primary objective even if it generated a strong Sharpe Ratio.
Additionally, the Sharpe Ratio fails to distinguish between risk drivers. For example, a strategy could be penalized for having a high standard deviation even if it is driven by a few periods of strong outperformance. To partially remedy this issue, the Sortino Ratio scales excess return over the risk-free rate by downside deviation, which measures the standard deviation of returns in periods when the strategy’s return is negative or fails to meet a predefined threshold. Unfortunately, this adjustment has the same conditional volatility flaw that we address later in our discussion of Information Ratio. Furthermore, in regard to the Sortino Ratio, a manager can still generate outperformance in periods when its return is negative (or fails to meet the predetermined threshold).
Jensen’s Alpha (“Alpha”)
Another common metric is Jensen’s Alpha, which measures the excess return a strategy generated after adjusting for the amount of systematic risk taken by the strategy. This metric is most commonly based on a strategy’s Beta to its benchmark. While Alpha can be calculated using any asset-pricing model, the most commonly used methodology is the Capital Asset Pricing Model (“CAPM”). The CAPM states that a manager’s expected return equals the risk-free rate plus the beta of the security multiplied by the equity market risk premium, which is equal to the market’s return over the risk-free rate. The remainder after the adjustment is Alpha, which equals the difference between the strategy’s actual return and its expected return. If a strategy generates positive Alpha, it means that it generated a positive excess return after adjusting for systematic risk. While this metric is widely used, we question whether Beta and the market risk premium are the best measures of systematic risk and the price of risk, respectively. In addition, Alpha fails to account for its own variability and says nothing about a strategy’s ability to consistently generate Alpha, making it difficult to determine if the positive result was due to luck or skill.
The information contained herein is confidential and the dissemination or distribution to any other person without the prior approval of Fiducient Advisors is strictly prohibited. Information has been obtained from sources believed to be reliable, though not independently verified. Any forecasts are hypothetical and represent future expectations and not actual return volatilities and correlations will differ from forecasts. This report does not represent a specific investment recommendation. The opinions and analysis expressed herein are based on Fiducient Advisor research and professional experience and are expressed as of the date of this report. Please consult with your advisor, attorney and accountant, as appropriate, regarding specific advice. Past performance does not indicate future performance and there is risk of loss.