Library

Introduction

For years, investors have universally accepted that risk and return are inherently two sides to the same coin. You cannot have return without risk. However, while investment returns are obvious; investment risk is much more subtle. If you ask three investors to define risk, you will likely receive three different answers. One investor might define risk as significant variations in a portfolio’s value, another that a portfolio is risky when it performs poorly relative to a specific benchmark, while another might say that risk is the potential to lose money. Despite whatever definition of risk an investor considers important, how do you measure and evaluate that risk? Appropriately defining risk, as well as identifying appropriate risk measures, are critical first steps to understanding a portfolio’s risk exposure.

Over the following pages, we will explore several common measures of risk that investors have developed to both calculate and evaluate risk in an investment portfolio. We will also explore risk-adjusted return measures that attempt to showcase the relationship between risk and return in order to derive an approximation of manager skill. Despite some practitioners that use these measures definitively, we will demonstrate how these measures focus on only a very narrow view of what risk is and how they may fail to accurately describe how risky a portfolio can be. No single measure can effectively address every investment risk. It is important for investors to clearly understand what these measures mean and what their potential drawbacks are. In doing so, investors may be able to better navigate through the undeniable hazards that dot the investing landscape.

Risk Measure: Standard Deviation

Volatility is Not Necessarily a Proxy For Risk of Capital Loss

Many investors are concerned about their risk of loss. Traditionally, they may associate higher standard deviation with an increased risk of loss. In turn, they expect investments with higher standard deviation to be more adversely affected in market downturns. However, standard deviation statistically reflects the absolute volatility of an investment’s returns, giving equal weight to observations both below and above the average. As such, large deviations on the upside, while increasing standard deviation, may not be meaningful to an investor concerned with capital loss.

Furthermore, historical data illustrates that the relationship between high standard deviation managers and bear market performance isn’t always evident. The table to the right analyzes the performance of U.S. large-cap equity mutual funds over the past six bear markets. Specifically, the table illustrates the historical relationship between a fund’s standard deviation relative to the benchmark in the five year period before the start of each bear market and its return relative to the benchmark during the bear market. Should the traditionally implied relationship between standard deviation and performance hold, one would expect those funds with higher standard deviations relative to the benchmark to underperform in the subsequent bear market and conversely those funds with lower standard deviations relative to the benchmark to provide comparatively better bear market protection. However, over the six bear markets shown, on average 45% of the funds failed to exhibit this relationship.

As such, for those funds, past volatility (as represented by standard deviation) was not an accurate predictor of excessive losses during adverse market environments. For investors concerned with the risk of capital loss, standard deviation failed as a measure of risk when risk mattered most.

Does Standard Deviation Predict Relative Bear Market Protection?

This analysis examines the bear market performance of current U.S. large cap equity funds in the Morningstar database relative to the performance of the S&P 500 Index (the benchmark). “No” implies that either a fund achieved (1) above benchmark bear market returns and above benchmark standard deviation or (2) below benchmark bear market returns and below benchmark standard deviation. “Yes” implies that either a fund achieved (1) above benchmark bear market returns and below benchmark standard deviation or (2) below benchmark bear market returns and above benchmark standard deviation. Standard deviation is defined as the five year annualized standard deviation computed using monthly returns for the period prior to the start of the bear market. A bear market is defined as a period of negative returns in the S&P 500 Index for which it takes at least four quarters to recover the loss. Analysis: Manning & Napier. Data source: Morningstar.

Standard Deviation is a Backward Looking Measure

As with many common risk measures, standard deviation is based on historical observations and thus provides information on an investment’s historical volatility profile. Unfortunately, a portfolio’s historical volatility does not seem to provide a meaningful indication of future volatility. For example, as the chart below shows, the standard deviation of the Morningstar Large Growth, Value, and Blend categories all fell meaningfully from 2003 to 2007 implying a reduction in risk. However, this decline in standard deviation preceded one of the largest market declines in history (i.e., the S&P 500 Index declined 54.89% from 10/09/2007 to 03/09/2009). Likewise, the standard deviation of the Large Growth and Value category averages have varied significantly over time. For example, the five year standard deviation for the Large Growth category has been as high as 23.33% and as low as 9.71% indicating a high degree of inconsistency. While historical information regarding an investment’s volatility may provide investors with a useful initial framework, simple standard deviation measurements seem to lack the continuity needed for an in-depth risk evaluation.

Market Returns May Not be Normally Distributed

Standard deviation is often applied alongside an assumption of normal distribution where observations in a data series are distributed more or less symmetrically around a mean. A normal distribution allows investors to use several rules of thumb to describe returns. However, based on historical return observations, often times the actual distribution of returns will not conform to what a normal distribution will predict. Looking at daily returns for the S&P 500 Index over the past 42 years (i.e., 10,957 observations), a normal distribution would assume that a daily loss of greater than 4.00% would have occurred approximately once, where in actuality, the S&P 500 Index has experienced a daily loss of greater than 4.00%, 39 times. The chart below plots the S&P 500 Index’s actual daily returns alongside what a normal distribution would predict. In contrast to a normal distribution, the actual returns demonstrate more frequent extreme outliers at the tails of the distribution as well as many more observations centered near the mean. Each investment portfolio will differ in terms of how closely their returns align with the assumptions of a normal distribution. Investors using standard deviation should approach the measure with caution as it may often fail to accurately describe a portfolio’s historical returns.

Risk-Adjusted Return Measure: Sharpe Ratio

The Sharpe ratio is a risk-adjusted return measure that is often used concurrently with standard deviation. In fact, standard deviation is part of the calculation of the Sharpe ratio. Specifically, the Sharpe ratio measures a portfolio’s excess return per unit of risk where risk is measured by standard deviation and excess return is a portfolio’s return above a risk free rate. The Sharpe ratio is typically used as a comparison tool, where the Sharpe ratio of two or more portfolios are compared against each other. In this context, a higher Sharpe ratio is favorable (i.e., greater excess return per unit of risk) and considered an indication of management skill.

Since the Sharpe ratio uses standard deviation in its calculation, it suffers from all of the same drawbacks: it defines risk solely as volatility, it relies on historical return observations to predict future returns, and it assumes that return data is normally distributed. Likewise, since the Sharpe ratio is a backward looking measure, it may not be predictive of future manager skill, a potential concern for investors using the Sharpe ratio as the sole criteria for manager selection.

For example, using the historical U.S. large cap mutual fund returns from the Morningstar Direct database, the analysis to the right attempts to determine if a manager with a high Sharpe ratio in one decade is likely to have a high Sharpe ratio in the next decade. The analysis to the right tests three sets of observations, the 1970s to 1980s, the 1980s to 1990s, and lastly the 1990s to 2000s, and indicates that historical Sharpe ratios may often fail to predict which managers are skillful in the future. Nearly 70% of managers with top quartile Sharpe ratios over the period from 1990 to 1999 had below median Sharp ratios over the subsequent decade. Thus, investors should use the Sharpe ratio with caution, and know that its predictive nature seems to be inherently limited.

Risk Measure: Beta

Beta is a Backward Looking Measure

The most traditional way to calculate a portfolio’s Beta is returns-based, which uses the portfolio’s actual historical returns. Like standard deviation, historical returns-based risk measures may not provide insight in regards to a portfolio’s volatility relative to the market going forward. For example, as the chart to the right illustrates, while rolling three year Betas of U.S. mid cap, U.S. small cap, and international equities relative to the S&P 500 Index have diverged over time, the Betas of all three asset classes converged in the most recent bear market and the subsequent rally. As such, the asset classes’ historical Betas provided little insight into their behavior over the past two years. For the three year period ending 12/31/2006, U.S. mid cap stocks had a Beta of 1.31 compared to a Beta of 1.77 for U.S. small cap stocks. However, for the three year period ending 03/31/2009, the two asset classes had equivalent Betas of 1.16. Thus, making investment decisions solely on the basis of historical Beta calculations could have led investors to draw very different conclusions about the relative risk profile (i.e., exposure to market volatility) of mid and small cap stocks.

The limitations of historical data are especially an issue for returns-based Beta calculations for portfolios that are actively adjusted over time. In contrast, a holdings-based Beta is calculated by taking the weighted average of the Betas of the individual securities currently held within the portfolio. While a holdings-based Beta still relies on the historical returns of the actual portfolio’s holdings, it may be a better indication of the portfolio’s current sensitivity to market movements, as it takes the current portfolio’s holdings into account. Analyzing the historical holdings-based Beta of a portfolio allows investors to compare the portfolio’s exposure to market risk relative to its history and whether the portfolio has taken on risk at the right time.

Beta is a Relative Risk Measure

One of the central features of Beta that is important for investors to understand is how Beta defines risk. Beta assumes that the only risk faced by the portfolio is the volatility of returns relative to the market portfolio, usually a broad market index such as the S&P 500 Index. This risk is often referred to as market risk. In this context, Beta is a relative risk measure as it considers volatility relative to a specific benchmark. Beta doesn’t generally consider absolute risk such as the risk of the portfolio losing money on an absolute basis.

For example, while a portfolio with a Beta of 0.80 would have been considered less risky than the market, a Beta of this level implies the portfolio still would have lost approximately 44.00% (i.e., 80% of the market’s return) in the most recent bear market. In addition, it is well known that an investment portfolio faces risk beyond just market risk. Two examples are inflation risk (a portfolio’s returns being adversely impacted by high inflation) and interest rate risk (the general decline in fixed income prices as interest rates rise). In fact, studies have shown that market risk may be only a small percentage of the actual risk faced by a portfolio, thus resulting in Beta being a potentially incomplete measure of risk.

Benchmark Selection is Very Important

The choice of benchmark may have meaningful implications for a portfolio’s Beta calculation and any analysis where that Beta measure is used. One of the most notable impacts of benchmark selection is found when utilizing style benchmarks (e.g., growth and value). As the various benchmark providers incorporate different measures to classify securities as growth or value, various growth and value benchmarks may have different compositions and return streams. As such, a portfolio’s Beta versus one style benchmark may meaningfully differ from its Beta versus a different style benchmark. The table to the right illustrates the average Betas of mutual funds classified as either Large, Mid, or Small Growth in Morningstar versus various growth style benchmarks from 01/01/2002 - 12/31/2011. The information presented in the table clearly illustrates the benchmark dependence of Beta calculations. For example, the Beta of the Small Growth category calculated with respect to the Russell 2000 Growth is 0.92 while its Beta calculated with respect to the S&P SmallCap 600 Growth is 1.02. Depending on the benchmark used, investors may draw different conclusions about the level of risk in a particular portfolio.

Risk-Adjusted Return Measures: Alpha and Treynor Ratio

Alpha is a risk-adjusted return measure that compares the actual return of a portfolio with the return that should have been earned based on the portfolio’s Beta. It is typically calculated by comparing an investment manager’s return to that of a benchmark (assuming an equal level of risk). In general, the higher the Alpha, the greater the level of excess return a manager earned for a given level of risk. Practitioners typically use this measure to compare the skill of two (or more) managers, in light of the level of risk (i.e., Beta) they took in the management of a portfolio.

The Treynor ratio is conceptually similar to the Sharpe ratio, however it utilizes Beta as the measure of risk while the Sharpe ratio utilizes standard deviation. Specifically, the Treynor ratio measures a portfolio’s excess return per unit of risk where risk is measured by Beta and excess return is a portfolio’s return above a risk free rate. When utilizing the Treynor ratio as a comparison tool, a higher Treynor ratio is considered favorable and an indication of management skill. Investors believe a Treynor ratio is best used for completely diversified portfolios.

Since Alphas and Treynor ratios use Beta within their definition of risk, they suffer from all the same drawbacks as Beta discussed previously. They are both relative risk measures, suggesting that these calculations may not provide insight in regards to the absolute level of portfolio returns. Secondly, as illustrated in the discussion above, the choice of benchmark may significantly influence a portfolio’s Beta, with that impact likely being reflected in the portfolio’s Alpha and Treynor ratio. For example, higher Beta managers are theoretically expected to earn higher returns thus increasing the difficulty of producing Alpha at any given level of actual returns. Lastly, being backward looking measures, Alphas and Treynor ratios over one time period may provide little predictive value over the next time period. As the table to the right illustrates, over the previous four decades, there has been little persistence in top quartile Alpha and Treynor ratio U.S. large cap equity managers (as defined by Morningstar) from one decade to the next. Nearly 70% of managers with top quartile Alphas and Treynor ratios over the period from 1990 to 1999 were below median over the subsequent decade. It is important for investors to realize that various market environments may have a meaningful impact on both risk-adjusted return measures.

Risk Measure: Tracking Error

The illustration below provides the five year rolling annualized returns of two portfolios, as well as the benchmark for the periods ending 12/31/1994 to 12/31/2011. As the return patterns show, portfolio A had a much lower tracking error, as evidenced by its similar return pattern to the benchmark, compared to portfolio B (i.e., 2.16% compared to 10.81%, respectively), whose return pattern exhibited a greater degree of variation compared to the benchmark’s return. Often times, a portfolio with high tracking error is considered more risky than a portfolio with low tracking error, as investors perceive meaningful deviations in return from a benchmark as a potential source of risk. Likewise, within a multiple manager context whereby managers are serving a specified role (e.g., Large Cap Value, Small Cap Growth, etc.), managers with low tracking errors are often times perceived as less risky in the context that they may be less likely to potentially deviate from their specified role. For investors using tracking error as a measure of risk, there are several important issues they must keep in mind.

Tracking Error is a Backward Looking Measure

Tracking error measures historical deviations in performance relative to a benchmark. However, this may not be representative of the tracking error for the portfolio going forward. The chart below provides the five year rolling tracking error for the overall average Morningstar Large Growth, Blend, and Value fund since 12/31/1974. As the chart shows, the tracking error of all three category averages have varied significantly over time, with the five year rolling tracking error for the average Large Growth and Value fund categories deviating by as much as approximately 8.00% since 12/31/1974. An important part of any tracking error evaluation should also include understanding an investment manager’s process in relation to the likelihood that a manager’s tracking error may deviate in the future. For example, a portfolio’s historical tracking error could be a function of the manager’s investment opportunities in the market opposed to relatively tight sector, capitalization, or universe constraints relative to a given Index.

Tracking Error is a Relative Risk Measure

Since tracking error measures deviations in performance relative to a benchmark, it is more often associated with benchmark risk (i.e., the risk of deviating from a specified benchmark) opposed to the risk of capital loss. For investors defining risk as the risk of capital loss, tracking error is unlikely to be an appropriate risk measure to use in evaluating a portfolio. For example, despite a substantially similar rolling tracking error pattern for the 04/01/2000 to 09/30/2002 bear market time period (as shown on the previous page), the annualized performance of the average Morningstar Growth and Value fund varied significantly. Specifically, the average Large Value Fund had an annualized return of -10.00% compared to -28.52% for the average Large Growth Fund and -20.56% for the S&P 500 Index. Likewise, similar to standard deviation, this type of analysis demonstrates how tracking error does not distinguish between deviations in performance on the upside or downside relative to the benchmark.

Morningstar Category Trailing Ten Year Tracking Errors vs. Various Indices

(01/01/2002 – 12/31/2011)

Analysis: Manning & Napier
Data source: Morningstar

Benchmark Selection is Very Important

Similar to Beta, benchmark selection is critical in the calculation of tracking error. Since tracking error measures historical deviations in performance relative to a benchmark, how high or low a portfolio’s tracking error is depends on the selection of the comparative benchmark. While often times identifying the most appropriate benchmark for a portfolio may be difficult, the chart below shows how the choice of a benchmark may affect even a long-term tracking error calculation. When evaluating tracking error as a measure of risk, understanding and determining the appropriateness of the benchmark is important.

Risk-Adjusted Return Measure: Information Ratio

The information ratio is a risk-adjusted return measure that is used to evaluate a portfolio’s excess return in relation to its tracking error. Similar to other risk-adjusted measures, this ratio attempts to quantify how skillful the manager was in generating their excess return. The information ratio measures a portfolio’s excess return per unit of risk where risk is measured by tracking error and excess return is a portfolio’s return above its benchmark. Generally speaking, higher information ratios are considered favorable to lower information ratios.

Since the information ratio uses tracking error as its definition of risk, it suffers from all of the same drawbacks as tracking error discussed previously. For investors defining risk as the risk of capital loss, information ratio does not provide insight in regards to the absolute level of returns for a portfolio. As with tracking error, a portfolio’s information ratio will be highly dependent on the selection of the benchmark. Finally, a portfolio’s current information ratio may not provide insight in regards to a portfolio manager’s ability to maintain a favorable information ratio going forward. As the table to the right illustrates, over the previous four decades, there has been little persistence in top quartile information ratios for U.S. large cap equity managers (as defined by Morningstar) from one decade to the next. Over 80% of managers with top quartile information ratios over the periods from 1990 to 1999 had below median information ratios over the subsequent decade.

Conclusion

All too often, investors rely only on a narrow view of what risk is; devoting their sole risk management focus on measures that are generally misunderstood and misused. As we have demonstrated, some of the most popular risk measures used by investors today perpetuate this narrow view and as a result may fail to accurately describe how risky a portfolio can be. Whether an investor believes risk means volatility, failure to adhere to a benchmark, or the possibility of capital loss, no one risk measure will provide a complete picture of a portfolio’s risk exposure. Included below are several steps investors may wish to consider in order to approach risk from a more holistic perspective.

Set Forth Goals and Risk Priorities

Academic studies have indicated that over 90% of the variation in a portfolio’s return can be attributed to the asset allocation decision. Perhaps the most important step in appropriately managing risk for an investor is to utilize an asset allocation approach that takes into account an investor’s goals and risk priorities. For example, for an investor who’s primary concern is the risk of capital loss, an investment approach that has a meaningful allocation to high-quality, shorter-term fixed income securities would likely help manage that risk. In contrast, for an investor concerned primarily with maintaining pace with a rise in the cost of living (i.e., inflation) over the long term, a portfolio of longer term, growth-oriented securities may better serve to manage their primary risk. Until an investor has clearly defined their goals and risk priorities, it is impossible to effectively manage and evaluate the risk of their portfolio.

Understand the Investment Manager’s Process

As part of any risk management evaluation, it is important for the investor to understand how their investment manager manages risk within their portfolio. For example, does their manager utilize an active investment approach that manages portfolio risk by evaluating the risk/reward trade off of various securities based on current market conditions, or does their manager manage risk by maintaining relatively static sector allocation or market capitalization relative to a broad market benchmark? Evaluating risk measures in the context of the manager’s investment process will help an investor identify the risk measures that may be more relevant, as well as lessen the likelihood of being misled by drawing incorrect conclusions from statistics that may not provide appropriate insight into the portfolio’s risk.

Apply Context to Risk Measures

To achieve a deeper understanding of a portfolio’s risk, investors should apply context to any risk measures or statistics that they use. Context can be applied in many ways including on a historical, environmental, goal-oriented, or current positioning basis. Below are four examples that demonstrate an insightful use of risk management statistics in each of these contexts:

From a historical context…

Evaluating a manager’s historical holdings Beta over time will not only provide insight in regards to whether a manager is taking on more market risk relative to the manager’s history, but also help determine whether the manager historically has taken on more market risk at the correct times.

From an environmental context…

Investors may wish to utilize up/down market capture to help characterize a portfolio’s relative return. Up/down market capture looks at the percentage of the market’s upside or downside return that is captured by a portfolio’s return. For example, if a portfolio had an upside capture of 110%, it has captured 110% of the market’s return in those periods where the market was positive. If that same portfolio had a downside capture of 90%, it has captured 90% of the market’s return in periods when the market was negative.

From an investment goals context…

Investors primarily concerned about the risk of loss may wish to evaluate a portfolio’s bear market performance. This will give the investor an idea of how a portfolio has fared in prior periods dominated by negative market returns. Clearly, these are the periods that mean the most to investor’s concerned with investment loss.

From a current positioning context…

An investor’s primary concern should be the portfolio’s current level of risk (as this is most relevant to its future performance), which is not specifically addressed by historical risk measures. For example, active share is a measure that evaluates a portfolio relative to a benchmark on a holdings basis. High active share (generally 60% or above) indicates that the majority of a portfolio’s holdings are different than the comparative benchmark. As such, it would not be surprising if the returns of a portfolio with high active share were to meaningfully deviate from the comparative benchmark’s returns. Compared to a portfolio’s tracking error (which relies on historical returns), it is more likely that a current holdings comparison like active share will provide greater insight in regards to future return deviations from a benchmark.

¹Morningstar Large Blend (Large Blend) portfolios are fairly representative of the overall U.S. stock market in size, growth rates, and price. Stocks in the top 70% of the capitalization of the U.S. equity market are defined as large-cap. The blend style is assigned to portfolios where neither growth nor value characteristics predominate.

²Morningstar Large Growth (Large Growth) portfolios invest in big U.S. companies that are projected to grow faster than other large-cap stocks. Stocks in the top 70% of the capitalization of the U.S. equity market are defined as large-cap. Growth is defined based on fast growth (high growth rates for earnings, sales, book value, and cash flow) and high valuations (high price ratios and low dividend yields).

³Morningstar Large Value (Large Value) portfolios invest primarily in big U.S. companies that are less expensive or growing more slowly than other large-cap stocks. Stocks in the top 70% of the capitalization of the U.S. equity market are defined as large-cap. Value is defined based on low valuations (low price ratios and high dividend yields) and slow growth (low growth rates for earnings, sales, book value, and cash flow).

⁴The S&P 500 Total Return Index (S&P 500 Index) is an unmanaged, 
capitalization-weighted measure of 500 widely held common stocks listed on the New York Stock Exchange, American Stock Exchange, and the Over-the-Counter market. The Index returns assume daily reinvestment of dividends and do not reflect any fees or expenses. Index data provided by Morningstar.

⁵The Russell Midcap^® Index (Russell Midcap) is an unmanaged index that consists of approximately 800 of the smallest securities based on a combination of their market cap and current index membership. The Russell Midcap is a subset of the Russell 1000^® Index. The Index is completely reconstituted annually to ensure larger stocks do not distort the performance and characteristics of the true mid-cap opportunity set. The Index returns do not reflect any fees or expenses. Index data provided by Morningstar.

⁶The Russell 2000^® Index (Russell 2000) is an unmanaged index that consists of 2,000 U.S. small-capitalization stocks. The Index returns are based on a market capitalization-weighted average of relative price changes of the component stocks plus dividends whose reinvestments are compounded daily. The Index returns do not reflect any fees or expenses. Index returns provided by Bloomberg.

⁷The MSCI EAFE Index (MSCI EAFE) is comprised of 22 MSCI country indices, representing the developed markets outside of North America: Europe, Australasia, and the Far East. The Index returns do not reflect any fees or expenses. The Index is denominated in U.S. dollars. The Index returns assume daily reinvestment of net dividends (which do account for foreign dividend taxation). Index data provided by Morningstar.

⁸The Russell 1000^® Growth Index (Russell 1000 Growth) is an unmanaged, market capitalization-weighted index consisting of those Russell 1000 companies with higher price-to-book ratios and higher forecasted growth values. The Index returns are based on a market capitalization-weighted average of relative price changes of the component stocks plus dividends whose reinvestments are compounded daily. The Index returns do not reflect any fees or expenses. Index data provided by Morningstar.

⁹The S&P 500 Growth Index (S&P 500 Growth) is a market capitalization-weighted index consisting of those S&P 500 companies with higher sales growth, higher earnings growth to price, and higher price momentum. Index returns provided by Morningstar.

¹⁰The MSCI U.S. Large Cap Growth Index (MSCI U.S. Large Cap Growth) represents the growth companies of the MSCI U.S. Large Cap 300 Index. (The MSCI U.S. Large Cap 300 Index represents the universe of large capitalization companies in the U.S. equity market). Index data provided by Morningstar.

¹¹The Wilshire U.S. Large-Cap Growth Index (Wilshire U.S. Large-Cap Growth) is a market cap-weighted index including securities from the Wilshire Large Cap 750 index that meet Wilshires definition of growth. Variables employed for the growth portfolio screenings include sales growth, return on equity, and dividend payout. Index data provided by Morningstar.

¹²The Russell Midcap^® Growth Index (Russell Midcap Growth) is an unmanaged index that consists of approximately 800 of the smallest securities based on a combination of their market cap and current index membership. The Russell Midcap is a subset of the Russell 1000^® Index. The Index is completely reconstituted annually to ensure larger stocks do not distort the performance and characteristics of the true mid-cap opportunity set. The Index returns do not reflect any fees or expenses. Index data provided by Morningstar.

¹³The S&P MidCap 400 Growth Index (S&P MidCap 400 Growth) is a market capitalization-weighted index meant to measure the performance of those companies with growth characteristics within the S&P MidCap 400 Index. Growth factors include sales growth, earnings change to price and momentum. Index data provided by Morningstar.

¹⁴The MSCI U.S. Mid Cap Growth Index (MSCI U.S. Mid Cap Growth) represents the growth companies of the MSCI U.S. Mid Cap 450 Index. (The MSCI U.S. Mid Cap 450 Index represents the universe of medium capitalization companies in the U.S. equity market). Index data provided by Morningstar.

¹⁵The Wilshire U.S. Mid-Cap Growth Index (Wilshire U.S. Mid-Cap Growth) is a market cap-weighted index including securities from the Wilshire Mid Cap 500 Index with growth characteristics as defined by Wilshire. Variables employed for the growth portfolio screenings include sales growth, return on equity, and dividend payout. Index data provided by Morningstar.

¹⁶The Russell 2000^® Growth Index (Russell 2000 Growth) is a market-weighted total return index that measures the performance of companies within the Russell 2000^® Index having higher price-to-book ratios and higher forecasted growth values. The Russell 2000^® Index includes the 2,000 firms from the Russell 3000 Index with the smallest market capitalizations. Index data provided by Morningstar.

¹⁷The S&P SmallCap 600 Growth Index (S&P SmallCap 600 Growth) is a market capitalization-weighted index meant to measure the performance of those companies with growth characteristics within the S&P SmallCap 600 Index. Growth factors include sales growth, earnings change to price and momentum. Index data provided by Morningstar.

¹⁸The MSCI U.S. Small Cap Growth Index (MSCI U.S. Small Cap Growth) represents the growth companies of the MSCI U.S. Small Cap 1750 Index. (The MSCI U.S. Small Cap 1750 Index represents the universe of small capitalization companies in the U.S. equity market). Index data provided by Morningstar.

¹⁹The Wilshire U.S. Small-Cap Growth Index (Wilshire U.S. Small-Cap Growth) is a market cap-weighted index including securities from the Wilshire Small Cap 1750 Index with growth characteristics as defined by Wilshire. Variables employed for the growth portfolio screenings include sales growth, return on equity, and dividend payout. Index data provided by Morningstar.

²⁰The Russell 1000^® Index (Russell 1000) is an unmanaged index that consists of 1,000 large-capitalization U.S. stocks. The Index returns are based on a market capitalization-weighted average of relative price changes of the component stocks plus dividends whose reinvestments are compounded daily. The Index returns do not reflect any fees or expenses. Index data provided by Morningstar.

²¹The Dow Jones Wilshire 5000^® Total Market Index (Dow Jones Wilshire 5000) is an unmanaged index that consists of over 5,000 U.S. equity securities with readily available price data. The Index returns are based on a market capitalization-weighted average of price changes in its components factored against the total market capitalization of those components. The Index returns assume daily reinvestment of dividends, and do not reflect any fees or expenses. Index data provided by Morningstar.

²²The Dow Jones Industrial Average is a price-weighted average of 30 blue-chip U.S. stocks that are generally the leaders in their industry. Dividends are reinvested to reflect the actual performance of the underlying securities. The Index returns do not reflect any fees or expenses. Index data provided by Morningstar.

²³The Russell 1000^® Value Index (Russell 1000 Value) is an unmanaged, market capitalization-weighted index consisting of those Russell 1000 companies with lower price-to-book ratios and lower forecasted growth values. The Index returns are based on a market capitalization-weighted average of relative price changes of the component stocks plus dividends whose reinvestments are compounded daily. The Index returns do not reflect any fees or expenses. Index data provided by Morningstar.

²⁴The S&P 500 Value Index (S&P 500 Value) is a market capitalization-weighted index consisting of those S&P 500 companies with lower book value to price ratios, lower earnings to price ratios, and lower sales to price ratios. Index returns provided by Morningstar.

Unless otherwise noted, all figures are based in USD.

Analysis: Manning & Napier Advisors, LLC (Manning & Napier).

Manning & Napier is governed under the regulations of the U.S. Securities and Exchange Commission (SEC).

Sources: FactSet, Morningstar.