northerntrust.com | Global Low Volatility Anomaly | 1 of 12

Insights on...Global low Volatility anomaly

After the financial crisis of 2008 and the heightened market volatility of 2011, many investors re-evaluated the strategic importance of the biggest source of risk in their portfolios – equities. Some explored alternative asset allocations that shift away from equities but needed to maintain equity exposure to achieve return objectives. Rather than reducing allocations explicitly, investors found viable options in indexed and actively managed, low-volatility equity strategies.

academic ReView The idea that return should be positively related to risk is intuitive and the hallmark outcome of Sharpe’s1 revered capital asset pricing model (CAPM) suggesting that security returns should be a positive linear function of risk. Yet when securities are sorted into volatility quintiles, the lowest volatil-ity quintile of securities significantly outperforms the quintile of the highest-volatility securities over the long-term. This so-called “low-volatility anomaly” has persisted for years across various scenarios. For instance, MSCI World securities in the lowest quintiles of Barra volatility significantly outperformed those in the highest volatility quintile from December 1996 through June 2012 (Chart 1).

Similarly, a comparison of low-volatility stocks in a capitalization-weighted index (MSCI World Index) to a traditional low-volatility index (MSCI World Minimum Volatility Index or WMVI) also reveals the anomaly. From January 1999 through June 2012, available data show that the MSCI WMVI earned excess annualized returns of 1.8%, with a notable reduction in annualized volatility compared to the MSCI World Index (Chart 2, Table 1).

G l o b a l l o w V o l a t i l i t y a n o m a l y : b e n e f i t i n G f r o m a n a c t i V e l y D e s i G n e D a p p r o a c h

The low volatility anomaly suggests that low-volatility, low-beta securities outperform high-volatility

securities over the long-term, producing greater returns over time than the capital asset pricing

model predicts. In this paper, we explore global low-volatility investing strategies and examine

the academic research and empirical evidence supporting the low-vol anomaly. We also discuss

ways to effectively exploit the anomaly in an actively designed equity portfolio and review results

from low-volatility investing in global and U.S. markets.

Our research indicates that the Northern Trust Quality Low-Volatility (QLV) solution, which uses a proprietary quality screen and unique portfolio construction, simultaneously adds alpha and provides a more stable estimate of volatility. Thus, it delivers additional return compared to indexed low-volatility products and more correlation stability, which provides additional protection in stressed markets.

0%

1%

2%

3%

4%

5%

6%

7%

Source: Northern Trust Research, Barra

Low BarraVol

HighBarraVol

2 3 4

Cap

-Wei

ghte

d Re

turn

chaRt 1: annualized RetuRns

by Volatility Quintile

Dec 1996 – June 2012

northerntrust.com | Global Low Volatility Anomaly | 2 of 12

In addition to the historical data alone, substantial academic research confirms that low-beta and low-volatility portfolios outperform the broad market long-term. ■■ Ang, Hodrick, Xing and Zhang2

analyzed a global universe of stocks to show that historically those with recent high idiosyncratic volatility have low future returns. They explore plausible explanations including transaction costs, analyst coverage and trading volume and find that when these variables are controlled, the underperformance of high-volatility stocks persists. They conclude that the spread between extreme quintiles of stocks sorted by idiosyncratic volatility earns an abnormal return of more than 1% per month.

■■ Frazzini and Pedersen3 analyzed leverage and margin constraints of individual investors to provide one explanation for the low-volatility anomaly. CAPM states that all investors should invest in the portfolio with the maximum Sharpe ratio (the market portfolio) and deleverage/leverage this portfolio to suit their risk tolerance level. However, many investors have leverage constraints and invest directly in risky securities to try to achieve greater returns. Frazzini and Pedersen contend that leverage-constrained investors bid up the price of riskier securities, resulting in lower realized returns than CAPM predicted.

$147

Jan

– 99

Jan –

02

Jan –

03

Jan –

05

Jan –

00

Jan –

01

Jan –

04

Jan –

06

Jan –

09

Jan –

10

Jan –

12

Jan –

07

Jan –

08

Jan –

11

Source: Bloomberg

MSCI World Index MSCI World Minimum Volatility Index

$60

$80

$100

$120

$140

$160

$180

$200

$

$186

chaRt 2: outpeRfoRmance of low Volatility (Jan 1999 – Jun 2012)

table 1: aVeRaGe annual RetuRns

Jan 1999 – Jun 2012

msci world index

msci world minimum Volatility

index

Annualized Return 2.9% 4.7%

Annual Volatility 16.6% 11.4%

Return/Risk 0.2 0.4

Source: Bloomberg

northerntrust.com | Global Low Volatility Anomaly | 3 of 12

■■ Baker, Bradley and Wurgler4 assert that the low-volatility anomaly arises partly from investor psychology and the behavior of large institutional managers. In line with Mitton and Vorkink5, they argue that volatile individual stocks have positive return skew – i.e., unlimited upside but a known downside limited to the stock price. Investors, therefore, view buying a low-priced volatile stock as if they were buying a lottery ticket, with volatility serving as a proxy for the potential payoff. As Barberis and Huang6 show, volatile stocks tend to be overvalued because of this lottery preference and, hence, perform poorly relative to low-volatility stocks over the long run.

■■ As noted, low volatility portfolios tend toward relatively low market betas. Baker, Bradley and Wurgler also argue that the mandate of most institutional managers compels them to maintain portfolios with betas equal to broad market benchmarks. The act of benchmarking, therefore, increases demand for stocks with betas relatively close to the market, inflating those stock prices. Stocks with betas close to their benchmark, then, tend to underperform lower-volatility, low-beta stocks over time.7

empiRical eVidence Empirical evidence suggests that the outperformance of low-volatility strategies arises from the fact that they are particularly well-insulated from implied volatility spikes because of the positive relationship between absolute volatility and volatility of volatility (“vol of vol” or VoV). This positive relationship results in low-volatility strategies typically exhibiting low absolute volatility as well as low VoV. Thus, while one would expect low-volatility portfolios to underperform in lower-volatility, rising market environments, their outperformance in higher-volatility, falling markets more than compensates. This is because the low VoV results in stable correlations across various market environments. This stability of correlations of low-volatility stocks produces tail protection and preserves some diversification benefits even during market crises.

In analyzing the performance of low-volatility strategies, we compare performance of a core index to a traditional low-volatility index in various volatility regimes, as measured by the Chicago Board Options Exchange (CBOE) Volatility Index (VIX). We define volatility regimes by levels of the CBOE VIX Index, with a VIX value less than 15 meaning Calm, 15 to 25 indicating Elevated and greater than 25 connoting Stressed. While the MSCI WMVI underperforms the MSCI World Index in lower volatility regimes, defined as Calm and Elevated (Table 2), when the market is stressed (VIX exceeds 25), the low-volatility index significantly outperforms, as expected. This asymmetric payoff profile tends to be particularly desirable for investors seeking to protect a portion of their portfolio returns during significant market downturns.

Volatility of VolatilityWhile volatility measures the historical dispersion of portfolio returns, the volatility of volatility (“vol of vol” or VoV) measures how that volatility changes over time. If the VoV of a portfolio is low, it means that the dispersion of returns changes little period over period. In contrast, a high VoV could suggest returns alternate between periods of relatively narrow and very wide dispersion.

table 2: aVeRaGe annual RetuRns

Jan 1999 – Jun 2012

ViX Regimemsci world

index

msci world minimum Volatility

index

excess performance

vs. msci world

Calm (≤15) 25.7% 22.9% – 2.8%

Elevated (15 to 25) 11.9% 9.5% – 2.4%

Stressed (>25) – 22.8% – 13.5% 9.3%

Annualized Return 2.9% 4.7% 1.8%

Source: Northern Trust Research, Bloomberg

northerntrust.com | Global Low Volatility Anomaly | 4 of 12

dRawbacks of tRaditional low-Volatility stRateGiesWhile the benefits of a traditional low-volatility strategy are evident, several important drawbacks also exist. These include the potential for increased concentration risk within individual securities and sectors; assumed stability of stock price volatility and correlations; and the natural tendency of low-volatility strategies to have greater turnover than a static index. Each of these can limit the attractiveness of a traditional low-volatility approach.

By design, a traditional low-volatility strategy is constructed to minimize portfolio return variance subject to constraints – typically by using an optimization tool or ranking scheme. However, this portfolio variance definition is typically backward-looking such that the portfolio construction exercise’s objective is to minimize historical ex ante volatility rather than forward-looking ex post volatility. Stated differently, the variance minimization process yields a portfolio that would have been the minimum variance portfolio in the past but may not necessarily be so in

the future. Only by assuming individual stock return volatilities and correlations between stock returns are stable (i.e., will behave in a manner similar to how they have behaved in the past) can one state that the low-volatility portfolio will exhibit lower volatility than a comparative core index going forward.

This assumption is not insignificant. If stock return volatilities and/or correla-tions are unstable, an investor could easily end up with a much riskier portfolio than anticipated. Worse yet, the tracking error to a comparative core portfolio could increase dramatically without an incre-mental increase in alpha expectation.

One might rightly conclude that the greater the stability of historical volatilities and correlations of stocks in a portfolio, then the better a minimum-volatility strategy will perform – both on a risk/return basis and relative to a comparative core portfolio.

The use of an optimizer to construct a low-volatility strategy may also emphasize certain lower volatility securities and sectors (e.g., utilities), resulting in concentrations that are greater than a comparative core portfolio. While specific concentration limits are often placed on the low-volatility portfolio, the upper and lower bounds of these limits can be quite high. For example, in the MSCI WMVI the maximum weight of a portfolio constituent can be up to 20 times the weight in the MSCI World Index. Similarly, the sector weights can deviate as much as 5% from the core index8. Given this latitude, the tendency of the optimizer is to significantly overweight lower volatility securities and sectors leading to a portfolio that appears lopsided versus a comparative core index.

As an index strategy with periodic rebalancing, low variance portfolios will necessarily have higher turnover than a passive index. From May 31, 1988, to July 31, 2012, the MSCI WMVI had annualized turnover of 20.8% compared to annualized turnover of 2.6% for the MSCI World Index9. For some investors, this turnover may create significant tax liabilities in addition to accruing transaction costs.

Quality defined For more than 40 years, Northern Trust has applied an investment philosophy focused on identifying and developing high-quality investments across client portfolios. Our research shows that higher-quality companies — based on their management discipline and persistent profitability — better protect a portion of investors’ capital in bear markets than more-volatile, lower-quality companies. Our research shows that top-quality companies, as defined by our core quality philosophy based on our assessment of their management efficiency and profitability, typically outperformed the market. Using an empirically tested approach that identifies quality companies within each sector, our proprietary method seeks to create investor portfolios better positioned to deliver outperformance with controlled volatility.10

northerntrust.com | Global Low Volatility Anomaly | 5 of 12

noRtheRn tRust Quality low-Volatility poRtfolioTo capture the benefits of low-volatility equity strategies and to mitigate the drawbacks of tradi-tional approaches, Northern Trust developed the Northern Trust Quality Low-Volatility (QLV) solution. QLV seeks to reduce portfolio variance and to provide additional return through our proprietary quality screen. The strategy is designed to take risks in quality, low-volatility securities and to minimize other extraneous risks inherent in many passive low-volatility approaches. The combination of quality and low-volatility helps reduce return variance and mitigate challenges traditional approaches present.

Based on a simulated research portfolio, Northern Trust’s low-volatility solution QLV generally offers volatility reduction of more than 30% per year versus comparative core portfolios. It also provides exposure to Northern Trust’s proprietary quality factor, which has been shown to provide excess returns greater than those of traditional low-volatility products over the long-term.

Table 3 details the risk and return metrics for the MSCI World Index, the MSCI WMVI and the Northern Trust QLV strategy from January 1999 through June 2012. Comparing annualized

volatility, the MSCI WMVI achieved an ex post volatility that was approximately 30% less than the core MSCI World Index (11.4% versus 16.6%). At the same time, the Northern Trust QLV strategy, with its proprietary quality factor, produced a volatility of 36% less than the MSCI World Index (10.7% versus 16.6%). Likewise, the Northern Trust QLV strategy experienced lower drawdowns than the MSCI WMVI over monthly and 12-month periods. We believe this volatility and drawdown reduction is the direct result of higher quality companies having more stable returns, and thus, the VoV for QLV is lower than the MSCI WMVI.

Examined differently, we also computed the tracking error of the Northern Trust QLV portfolio to the MSCI WMVI (Table 4). While the total annual tracking error volatility is 2.5%, it has a relatively large negative correlation to the MSCI WMVI itself,

resulting in a reduction in total portfolio volatility from 11.4% to 10.7% (Table 3). (See Appendix for a proof and detailed discussion.) At the same time, the Northern Trust QLV portfolio achieved a positive active return versus the MSCI WMVI of 2.6%.

Based on a simulated research portfolio, the Northern Trust QLV portfolio outperformed the traditional low-volatility indexes in all volatility regimes, with outperformance increasing as implied volatility increased. Thus, the addition of the proprietary quality screen helps QLV deliver additional return and, perhaps more importantly, more correlation stability, providing added protection in stressed markets.

table 3: Risk-RetuRn metRics

Jan 1999 – Jun 2012

msci world index

msci world minimum Volatility

index

northern trust Quality low

Volatility

Annual Geometric Return 2.9% 4.7% 7.3%

Annualized Volatility 16.6% 11.4% 10.7%

Return/Risk 0.2 0.4 0.7

Maximum 12-Month Drawdown

– 46.7% –37.2% – 32.2%

Source: Northern Trust Research, Bloomberg

table 4: tRackinG eRRoR

northern trust QlV to msci wmVi

Active Return 2.6%

Tracking Error 2.5%

Correlation – 0.4

Source: Northern Trust Research, Bloomberg

northerntrust.com | Global Low Volatility Anomaly | 6 of 12

These results resonate for two reasons. First, history shows that stressed markets occur not infrequently. Second, a primary objection to low-volatility portfolios is an over-reliance on portfolio optimizers and assumptions about the stability of the covariance matrix. With the addition of a quality factor, the Northern Trust low-volatility solution over-weights stocks with more stable return volatilities, resulting in a higher-returning, more stable portfolio with lower ex post volatility.

As discussed, low-volatility portfolios tend to have more stable VoV, resulting in an asymmetric payoff with incremental outperfor-mance during periods of market stress. The addition of a quality factor stabilizes correlations even further, as high quality companies have a lower VoV while simultaneously providing incremental alpha. Chart 3 shows the VoV for the Northern Trust QLV portfolio as well as the global low-volatility and cap-weighted MSCI indexes. The VoV of Northern Trust’s QLV is significantly lower than that of the MSCI indexes based on the application of a low-volatility approach, as well as the incorporation of our proprietary quality factor. The lower VoV for Northern Trust’s QLV is the direct result of investing in higher quality companies.

table 5: aVeRaGe annual RetuRns

Jan 1999 – Jun 2012

ViX Regimemsci world

index

msci world minimum

Volatility index

excess performance

vs. msci worldnorthern trust

QlVperformance

vs. msci world

Calm (≤15) 25.7% 22.9% – 2.8% 24.6% – 1.1%

Elevated (15 to 25) 11.9% 9.5% – 2.4% 11.7% – 0.2%

Stressed (>25) – 22.8% – 13.5% 9.3% – 10.3% 12.5%

Annualized Return 2.9% 4.7% 1.8% 7.3% 4.4%

Source: Northern Trust Research, Bloomberg

Key attributes of Northern Trust’s quality, low-volatility approach:■■ Outperformance due to high quality factor and effective capture of low-volatility anomaly■■ Significant reduction in absolute volatility due to portfolio construction which seeks to

minimize portfolio variance■■ Low VoV due to quality focus, which provides more stable correlations and additional

protection against extreme market events

3.0%

3.5%

4.0%

4.5%

5.0%

5.5%

6.0%

6.5%

Source: Northern Trust Research, Bloomberg

Northern Trust Quality Low Volatility

MSCI World Index

MSCI World MinimumVolatility Index

Ann

ualiz

ed “

Vol o

f Vol

”

Northern Trust QLV “Vol of Vol”Increase Due to Lack of Quality FactorIncrease Due to Lack of Low Vol Approach

chaRt 3: Volatility of Volatility

northerntrust.com | Global Low Volatility Anomaly | 7 of 12

noRtheRn tRust QlV in a multi-stRateGy poRtfolioMany investors considering low-volatility equity strategies also are exploring altering asset allocation frameworks to reduce equity exposure, increase fixed income exposure and lower total portfolio volatility. Based on our analysis, the combination of the proprietary quality screen and unique portfolio construction process of the Northern Trust QLV increases annualized return.

*Proxy for bond is total return of EU 10-year bondSource: Northern Trust Research, Bloomberg

75% MSCI World, 25% Bond*MSCI World Index

60% MSCI World, 40% Bond*

Northern Trust Quality Low Volatility MSCI World Minimum Volatility IndexX

9.0% 10.0% 11.0% 12.0% 13.0% 14.0% 15.0% 16.0% 17.0%2.0%

3.0%

4.0%

5.0%

6.0%

7.0%

8.0%

X

Annualized Volatility

Ann

ualiz

ed R

etur

n

chaRt 4: efficient fRontieR

Jan 1999 – Jun 2012

table 6: deViations /RetuRns/dRawdowns

annualized Jan 1999 - Jun 2012

annualized standard deviation

annualized Return

max 1-month drawdown

max 12-month drawdown

MSCI World Index 16.6% 2.9% – 18.9% – 46.7%

75% MSCI World, 25% Bond* 12.5% 3.4% – 14.1% – 36.4%

MSCI World Minimum Volatility Index

11.4% 4.7% – 15.9% – 37.2%

Northern Trust Quality Low Volatility

10.7% 7.3% – 14.0% – 32.2%

60% MSCI World, 40% Bond* 10.0% 3.6% – 11.2% – 29.6%

*Proxy for bond is total return of EU 10-year bond

Source: Northern Trust Research, Bloomberg

northerntrust.com | Global Low Volatility Anomaly | 8 of 12

It also further reduces portfolio volatility versus alternative strategies, including tradi-tional low-volatility strategies or a mixed portfolio of equity and bonds.

Compared to executing a mixed portfolio targeting set allocations of equity and bonds, the Northern Trust QLV portfolio provides additional excess returns driven by the proprietary quality screen. Based on our back test, from January 1999 through June 2012, QLV offers a volatility level between that of the 75% equity/25% bonds and

60% equity/40% bonds portfolios. It also offers more than a 100% increase in annualized returns versus either portfolio. Additionally, the QLV portfolio has managed turnover of approximately 60%, which helps control costs. A portfolio targeting a set allocation to cash and equity may experience greater turnover and increased transaction costs in order to maintain the targeted allocation across asset classes.

Further research shows that low-volatility equity strategies may be beneficial in a multi-strategy portfolio context. They allow investors such as those following a liability-driven investment structure to achieve overall return targets while freeing a portion of the risk budget to apply to less-liquid strategies, such as hedge funds.

Based on Northern Trust research, the annualized returns of Northern Trust QLV portfolio are in line with those of hedged equity strategies, as measured by the HFRI Hedged Equity Index. The drawdown of low-volatility equity portfolios tends to lie between that of cap-weighted equity indexes and hedge funds. Thus, the QLV product may be an efficient substitute for equity hedge fund exposure, providing similar returns, significantly increased transparency and liquidity, and reduced counterparty risk at the expense of slightly higher drawdowns.

table 7: RetuRn/Risk, dRawdowns

Jan 1999 – Jun 2012

msci world index

northern trust Quality low

VolatilityhfRi hedged

equity

Annual Geometric Return 2.9% 7.3% 7.1%

Annualized Volatility 16.6% 10.7% 9.8%

Return/Risk 0.2 0.7 0.7

Maximum Monthly Drawdown

– 18.9% – 14.0% – 9.5%

Maximum 12-Month Drawdown

– 46.7% – 32.2% – 26.7%

Source: Northern Trust Research, Bloomberg, Hedge Fund Research, Inc.*

* HFRI returns are shown net of fees. Based on research by Burton Malkiel and Atanu Saha (“Hedge Funds: Risk and Return,” 2004), hedge fund returns are positively skewed based on reporting, backfill and survivorship biases. Based on these biases, we believe the HFRI net returns are approximately comparable to the MSCI World and Northern Trust QLV gross returns.

northerntrust.com | Global Low Volatility Anomaly | 9 of 12

low-Volatility anomaly and actiVely desiGned stRateGiesInvesting in strategies seeking to benefit from the low-volatility anomaly may provide an alternative method to manage risk while still maintaining portfolio return targets. The return profile of a low-volatility strategy may prove particularly useful for investors seeking hedged equity-like exposures with increased liquidity and transparency, or those following a liability-driven investing framework seeking to efficiently employ their risk budget.

We believe an actively designed strategy seeking to mitigate drawbacks of traditional low-volatility approaches can best exploit the low-volatility anomaly. The Northern Trust Low Volatility solution seeks to minimize portfolio variance while providing incremental return and drawdown protection by incorporating our proprietary quality screen and portfolio construction.

foR moRe infoRmationTo learn more about the Northern Trust QLV solution, please contact your relationship manager or visit northerntrust.com.

u.s. anomaly also eXistsOur analysis found similar data supporting the low-volatility anomaly’s existence in U.S. markets. It also revealed the benefits of investing in low-volatility equity strategies through an actively designed approach. Our companion piece, “The Low-Volatility Anomaly in U.S. Markets,” explores this anomaly.

northerntrust.com | Global Low Volatility Anomaly | 10 of 12

appendiXWe can show mathematically that Northern Trust’s quality factor lowers the volatility of the Northern Trust QLV portfolio versus the MSCI World Minimum Volatility Index.

Recall that the volatility of the MSCI WMVI was 11.41% and the volatility of the tracking error between the QLV product and this index was 2.46%. The correlation between the tracking error and the MSCI index was –0.40. We will assign these values to variables: sMSCI = 11.41%sTE = 2.46%rMSCI,TE = –0.40

Where MSCI stands for the MSCI WMVI and TE stands for the tracking error between the QLV portfolio and this index. By definition:

s2A + B = s2

A + s2B + 2(sA x sB x rA,B)

So then,

s2QLV = s2

MSCI + s2TE + 2(sMSCI x sTE x rMSCI,TE)

s2QLV = 0.11412 + 0.02462 + 2(0.1141 x 0.0246 x –0.40)

s2QLV = 0.011389

sQLV = 0.011389 = 10.67%

This is the same value reported in Table 4. So, while uncorrelated variances are additive, the negative correlation between the tracking error of the QLV portfolio to the MSCI WMVI causes the overall volatility of Northern Trust’s QLV portfolio to be lower than the MSCI WMVI. This is the direct result of the addition of the Northern Trust proprietary quality factor.

√

northerntrust.com | Global Low Volatility Anomaly | 11 of 12

end notes

1 Sharpe, W., “Capital Asset Prices: A Theory of Market Equilibrium under Conditions of Risk,” The Journal of Finance, Vol. 19, No. 3, 425 – 442, 1964

2 Ang, A., Hodrick, R., Xing. Y, and Zhang, X, “High Idiosyncratic Volatility and Low Returns: International and Further U.S. Evidence,” Journal of Financial Economics, Vol. 91, 1 – 12, 2008

3 Frazzini, Andrea and Pedersen, Lasse H., “Betting Against Beta,” 2011

4 Baker, M., Bradley, B., and Wurgler, J., “Benchmarks as Limits to Arbitrage: Understanding the Low-Volatility Anomaly,” Financial Analysts Journal, Vol. 69, No. 1, 2011

5 Mitton, T. and Vorkink, K., “Equilibrium Under-Diversification and the Preference for Skewness,” The Review of Financial Studies, Vol. 20, No. 4, 1255 – 1288, 2007

6 Barberis, N. and Huang, M., “Stocks as Lotteries: The Implications of Probability Weighting for Security Prices,” NBER Working Paper No. 12936, 2007

7 Baker, M., Bradley, B., and Wurgler, J., “Benchmarks as Limits to Arbitrage: Understanding the Low-Volatility Anomaly,” Financial Analysts Journal, Vol. 69, No. 1, 2011

8 MSCI Global Low-Volatility Indexes Methodology, January 2012

9 MSCI

10 “Proprietary Quality Philosophy: Focusing on Quality Companies to Seek Outperformance, Controlled Volatility,” Northern Trust Research, 2012

northerntrust.com | Global Low Volatility Anomaly | 12 of 12

Q 52087 (4/14)

© 2014 Northern Trust Corporation. Head Office: 50 South La Salle Street, Chicago, Illinois 60603 U.S.A. Incorporated with limited liability in the U.S. Products and services provided by subsidiaries of Northern Trust Corporation may vary in different markets and are offered in accordance with local regulation. Northern Trust Asset Management comprises Northern Trust Investments, Inc., Northern Trust Global Investments Limited, Northern Trust Global Investments Japan, K.K., NT Global Advisors, Inc. and investment personnel of The Northern Trust Company of Hong Kong Limited and The Northern Trust Company. This material is directed to professional clients only and is not intended for retail clients. For Asia-Pacific markets, it is directed to institutional investors, expert investors and professional investors only and should not be relied upon by retail investors. For legal and regulatory information about our offices and legal entities, visit northerntrust.com/disclosures. The following information is provided to comply with local disclosure requirements: The Northern Trust Company, London Branch; Northern Trust Global Services Limited; Northern Trust Global Investments Limited. The following information is provided to comply with Article 9(a) of The Central Bank of the UAE’s Board of Directors Resolution No 57/3/1996 Regarding the Regulation for Representative Offices: Northern Trust Global Services Limited, Abu Dhabi Representative Office. Northern Trust Global Services Limited Luxembourg Branch, 2 rue Albert Borschette, L-1246, Luxembourg, Succursale d’une société de droit étranger RCS B129936. Northern Trust Luxembourg Management Company S.A., 2 rue Albert Borschette, L-1246, Luxembourg, Société anonyme RCS B99167. Northern Trust (Guernsey) Limited (2651)/Northern Trust Fiduciary Services (Guernsey) Limited (29806)/Northern Trust International Fund Administration Services (Guernsey) Limited (15532) Registered Office: Trafalgar Court Les Banques, St Peter Port, Guernsey GY1 3DA. Issued by Northern Trust Global Investments Limited.

Past performance is no guarantee of future results. All material has been obtained from sources believed to be reliable, but its accuracy, completeness and interpretation cannot be guaranteed. This information does not constitute investment advice or a recommendation to buy or sell any security and is subject to change without notice.

Important Information Regarding Hypothetical Returns – Where hypothetical portfolio data is presented, the portfolio analysis assumes the hypothetical portfolio maintained a consistent asset allocation (rebalanced monthly) for the entire time period shown. Proprietary quality dividend scores are evaluated on a quarterly basis. Hypothetical portfolio data is based on publicly available index information. All information is assumed to be accurate and complete but is not guaranteed. Hypothetical portfolio data contained herein does not represent the results of an actual investment portfolio but reflects the historical index performance of the strategy described which were selected with the benefit of hindsight. Components of the hypothetical portfolio were selected primarily utilizing actual historic market risk and return data. If the hypothetical portfolio would have been actively managed, it would have been subject to market conditions that could have materially impacted performance and possibly resulted in a significant decline in portfolio value.

IRS CIRCULAR 230 NOTICE: To the extent that this message or any attachment concerns tax matters, it is not intended to be used and cannot be used by a taxpayer for the purpose of avoiding penalties that may be imposed by law. For more information about this notice, see http://www.northerntrust.com/circular230.