The intuitive foundations of smart beta

Insights The intuitive foundations of smart beta Nearly half of global asset owners are now investing in smart beta 1, with interest continuing to rise yearover-year. The term smart beta encompasses an increasing variety of strategies, with many new smart beta products being launched each month around the world. Smart beta products typically aim to replicate indexes that embed the strategies underlying methodologies. Although some newer index approaches rely on detailed quantitative modelling, smart beta indexing doesn t have to be associated with complexity. In this Insights, we focus on some smart beta index construction approaches that follow relatively simple, intuitive weighting schemes. And, regardless of their methodology, all FTSE Russell smart beta indexes follow transparent, consistent rules in order to achieve the stated index objectives. Smart beta in context An increasing number of investors worldwide are using index-based approaches to construct and manage their portfolios. The objective of a manager of an index-based (or passive 2 ) investment portfolio is to replicate the index s return, before fees and costs. According to a 2017 report 3 by Moody s Investor Services, passive investments now account for US$6 trillion of assets globally and 29 percent of assets under management (AUM) in the US. Moody s predicts that the continuing adoption of index-based investment products will lead to passive funds market shares in the US exceeding 50 percent by early in the next decade. 1 Source: FTSE Russell, Smart Beta: 2017 global survey findings from asset owners. 2 Passive, index-based portfolios are passive in implementation, but still require some activity, for example when the index changes as a result of a rebalancing or due to corporate actions such as dividends, mergers, share splits and rights issues. 3 Asset Managers - Global: Passive Market Share to Overtake Active in the US No Later than 2024, Moody s, February 2017. ftserussell.com August 2017

Historically, the standard approach for passive investing has been to track a capitalization-weighted index in which the index s allocation to each constituent is determined by the constituent s market value (its number of shares outstanding multiplied by its share price). 4 In the context of indexing, smart beta is a generic term for indexes that depart from the standard market capitalization weighting method in order to achieve particular objectives, such as the generation of excess index returns, the mitigation of volatility, or diversification. But despite their change in weighting approach, smart beta indexes share important characteristics with their capitalization-weighted counterparts: They share the principle of being constructed according to a consistent set of rules, rather than relying on discretionary decisions. They are transparent, so that any market participant can look up and familiarize themselves with the details of the index s construction. A governance infrastructure also underlies any index, whether capitalization-weighted or smart beta. At FTSE Russell, the index ground rules set out the management responsibilities of the entities involved in the oversight of the index series: who acts as the benchmark administrator and the role of external advisory committees including the FTSE Russell Policy Advisory Board. Separately, the FTSE Russell Governance Board is responsible for ensuring that all the FTSE Russell indexes meet appropriate technical standards. When choosing between standard, capitalization-weighted indexes and smart beta indexes as the basis of an investment product, market participants often bear in mind questions of simplicity and transparency, cost, liquidity, capacity and governance. The main attributes of the two types of index are set out in the table below: Principal attributes of capitalization-weighted and smart beta indexes Type of index Simplicity and transparency Cost Liquidity and capacity Governance Capitalization- Weighted Simple weighting of each constituent is proportional to market capitalization Low turnoverrelated trading costs High liquidity and capacity FTSE Russell indexes have transparent ground rules and index review process Transparent index rules are disclosed Smart Beta Varies from more intuitive to more complex methodologies Transparent index rules are disclosed Potentially higher turnover-related trading costs than in capitalizationweighted indexing Varies but smart beta index providers typically prioritize liquidity and capacity in index design FTSE Russell indexes have transparent ground rules and index review process Source: FTSE Russell. 4 In practice, capitalization-weighted index constituents total shares outstanding are typically adjusted for investability or free float. This means that any shares considered unavailable for trading, for example due to being held by owners, senior executives, governments, pension schemes or other strategic shareholders, are deducted from the total. FTSE Russell The intuitive foundations of smart beta 2

Examples of smart beta index construction approaches Although the complexity of design of smart beta indexes can vary, several indexes share the simplicity of capitalization-weighting. One category of smart beta indexes includes those constructed according to intuitive, common-sense (often called heuristic ) rules. Among the most straightforward smart beta approaches are equal-weighting and weighting by fundamental accounting measures. Example 1: Equal weight Equal weight indexes aim to increase diversification, alleviating the concentration concerns inherent in market capitalization-weighted indexes. In a capitalization-weighted index, constituents are weighted by the product of their share price and their number of outstanding shares (i.e. by their market value). By contrast, in equal weight indexes constituents in the underlying parent index receive an equal weight. Due to stock market movements, the weights of individual constituents in equal weight indexes can become unequal over time. For FTSE Russell indexes, eligible securities are reviewed annually and rebalanced to equal weight on a quarterly basis. As an approach, equal weighting helps address concerns about potential concentration risks in capitalizationweighted indexes. For example, during the internet bubble of 1999-2000, stocks from the technology sector gained particular prominence in capitalization-weighted indexes (see Figure 1, where the Russell 1000 Index of US large-cap stocks is used as an example). Such periods of concentration may have a significant effect on the index performance if the sector concerned then suffers a period of relative underperformance. Figure 1. Russell 1000 Index historical sector weights 100% 90% 80% Weight in the index (%) 70% 60% 50% 40% 30% 20% 10% 0% Jul-96 Apr-97 Jan-98 Oct-98 Jul-99 Apr-00 Jan-01 Oct-01 Jul-02 Apr-03 Jan-04 Oct-04 Jul-05 Apr-06 Jan-07 Oct-07 Jul-08 Apr-09 Jan-10 Oct-10 Jul-11 Apr-12 Jan-13 Oct-13 Jul-14 Apr-15 Jan-16 Oct-16 Consumer Discretionary Consumer Staples Energy Financial Services Health Care Materials & Processing Producer Durables Technology Utilities Source: FTSE Russell, data as of February 28, 2017. Please see the end for important legal disclosures. FTSE Russell The intuitive foundations of smart beta 3

At the constituent level, the equal weight methodology can result in a significant underweight position of a few large-cap stocks in comparison with the capitalization-weighted index. For example, as can be seen in Figure 2, where the Russell 1000 Index s Financial Services sector is used as an illustration, the Financial Services sector stocks with the highest market value, such as JP Morgan, Citigroup, US Banc Corp, American Express and BNY Mellon, receive a much smaller constituent weighting (a relative underweight ) in an equal-weighted index than in the Russell 1000 Index. These underweighted stocks are shown at the left side of the chart. At the same time, a long tail of small-cap stocks within the sector, visible on the right side of the chart, receives a small weighting boost (and an overweight position relative to the capitalization-weighted index) under the equal-weighted approach. Figure 2. Weight differences in Russell 1000 Equal Weight Index, Financial Services sector Difference in stock weight: R1EW minus Russell 1000 1% 0% -1% -2% -3% -4% -5% -6% -7% Significant underweight on a few large cap stocks Limited overweight on many smaller cap stocks JPMORGAN CHASE & CO CITIGROUP INC US BANCORP AMERICAN EXPRESS CO BANK OF NEW YORK MELLON PAYPAL HOLDINGS INC BB&T CORP PUBLIC STORAGE INC DISCOVER FINANCIAL SRVC ALLSTATE CORP FIDELITY NATL INFO SVCS M & T BANK CORP PROGRESSIVE CORP CITIZENS FINANCIAL INC REGIONS FINANCIAL CORP LINCOLN NATIONAL CORP REALTY INCOME CORP EQUIFAX INC ALLIANCE DATA SYSTEMS INVESCO LTD MID-AMER APT CMNTYS SL GREEN REALTY CORP ANNALY CAPITAL MGMT INC E TRADE FINANCIAL CORP VANTIV INC ALLEGHANY CORP RAYMOND JAMES FINANCIAL IRON MOUNTAIN INC REINSURANCE GRP OF AMER BROADRIDGE FINANCIAL INC EAST WEST BANCORP INC APARTMENT INVT & MGMT AMERICAN CAMPUS COMMUN SEI INVESTMENTS CO PEOPLES UNITED FINANCIAL LIBERTY PROPERTY TRUST ZILLOW GROUP INC CLASS C MARKETAXESS HOLDINGS INC SPIRIT RLTY CAP INC HOSPITALITY PPTYS TRUST HIGHWOODS PROPERTIES INC ALLIED WRLD ASSUR COM FIRST HORIZON NATL CORP WESTERN ALLIANCE BANCORP FIRST DATA CORPORATION WEX INC HEALTHCARE TR AMER INC BANK OF HAWAII CORP EQUITY COMMONWEALTH WHITE MTNS INS GROUP LTD ASPEN INSURANCE HOLDINGS CHIMERA INVESTMENT CORP TCF FINANCIAL CORP PROASSURANCE CORP BRANDYWINE REALTY TRUST ERIE INDEMNITY CO EMPIRE STATE REALTY CARE CAPITAL PROPERTIES QUALITY CARE PROPERTIES CNA FINANCIAL CORP FIRST HAWAIIAN INC Source: FTSE Russell, data as of December 31, 2016. Please see the end for important legal disclosures. Given that an equal weight index overweights small-cap stocks by comparison with its capitalization-weighted counterpart, index designers have to bear in mind the capacity and liquidity of the resulting index in order to ensure its suitability for practical use as an underlying benchmark for index replicating financial products such as ETFs. Index designers can address potential liquidity and capacity concerns in an equal weight index by restricting index eligibility to stocks with an adequate float-adjusted market value and to stocks whose average daily trading volume (ADTV) is above a certain threshold. FTSE Russell The intuitive foundations of smart beta 4

Equal weighting is not a new index approach. In fact, one of the first commercially available index funds, launched in the 1970s by Wells Fargo, was based on an equal weight index of shares on the New York Stock Exchange (NYSE). The fund was wound up because it suffered from excessive transaction costs at the time, transactions in US shares were subject to high minimum commissions. 5 In more recent times, dramatic reductions in share dealing costs have made equal-weighting a viable proposition and the investment strategy is enjoying a resurgence of interest: roughly half of US financial advisors recently surveyed by FTSE Russell said they either use or are very likely to use an equal-weighted investment approach. 6 Example 2: Fundamental indexes A second example of an intuitive smart beta index approach is fundamental indexation. Fundamental indexes, like equal weight indexes, determine index constituents weights by some other measure than their stock price. In general, fundamental indexes measure company size using fundamental measures such as sales, operating cash flow, dividends and book value. Research Affiliates (RAFI), which popularized the strategy from 2005, notes that fundamental indexation also follows a contrarian rebalancing approach, systematically trading out of constituents whose prices have increased and increasing the allocation to securities whose prices have decreased. 7 Applying the fundamental index approach The Russell RAFI Index Series ranks and weights companies by averaging three fundamental factors: 1. Adjusted sales constituent sales are averaged over five years and then adjusted to take into account financial leverage, decreasing the index weight of companies with significant leverage 2. Retained operating cash flow the five-year average cash flow from operations, less dividends and buybacks 3. Dividends and buybacks the five-year average of dividends paid and share buybacks Empirical evidence shows that fundamental measures of companies size are relatively highly correlated with the companies market capitalization in other words, a stock that scores highly on fundamental measures tends also to have a high market value. 8 In practical terms, this means that capacity and liquidity constraints in fundamental indexes may be less of a concern than for other types of smart beta indexes. Smart beta performance in context Investors selecting smart beta indexes typically do so for a variety of reasons, such as the generation of excess index returns, the mitigation of volatility or diversification. The resulting performance of smart beta must therefore be seen in the context of the original objectives. At a headline level, the total return of both a fundamental and an equal weight version of the Russell 1000 Index exceeded that of the capitalization-weighted reference index over the period from December 1999 to May 2017 (see Figure 3). However, over the most recent 1-, 3-, 5- and 7-year periods, the capitalizationweighted Russell 1000 Index provided a slightly higher per annum total return (see Figure 4). 5 See The High Cost of Equal Weighting, Research Affiliates, May 2014. 6 Source: FTSE Russell, Smart beta: 2015 survey findings from US financial advisors. The survey included advisors with AUM greater than $20 million, at least 4% of AUM invested in ETFs and at least 20% fee-based annual revenue. 7 See https://www.researchaffiliates.com/en_us/strategies/rafi/rafi-fundamental-index.html. 8 In their 2005 paper, Fundamental Indexation, Arnott, Hsu and Moore showed that the weighted average capitalization of a fundamental index is around two-thirds as large as that of the reference capitalization-weighted index, but the concentration ratio of the fundamental index (the proportion of the overall index represented by the collective weightings of largest 100 stocks) is similar to that of the capitalization-weighted index. FTSE Russell The intuitive foundations of smart beta 5

Figure 3. Total return of Russell 1000, Russell 1000 Equal Weight and Russell RAFI US Indexes 600 500 400 300 200 100 0 Russell RAFI US Russell 1000 Equal Weight Russell 1000 Source: FTSE Russell, data from December 31, 1999 to May 31, 2017. Index levels are rebased to 100 on December 31, 1999. Past performance is no guarantee of future results. Returns shown may reflect hypothetical historical performance. Please see the end for important legal disclosures. Figure 4. Annualized total returns of Russell 1000, Russell 1000 Equal Weight and Russell RAFI US Indexes 1 year 3 years 5 years 7 years 10 years 15 years Russell RAFI US 14.82% 8.14% 15.14% 14.00% 7.81% 9.46% Russell 1000 Equal Weight 14.06% 7.28% 14.78% 13.97% 8.69% 11.26% Russell 1000 17.49% 9.83% 15.37% 14.37% 7.01% 8.02% Source: FTSE Russell, data as of May 31, 2017. Past performance is no guarantee of future results. Returns shown may reflect hypothetical historical performance. Please see the end for important legal disclosures. The divergence in performance from the capitalization-weighted Russell 1000 Index is paralleled by a divergence in the smart beta indexes risk statistics from those of the reference index (see Figure 5). Both in terms of standard deviation of return and in terms of tracking error, the Russell 1000 Equal Weight Index showed a greater divergence from the reference Russell 1000 Index than the Russell RAFI US Index. FTSE Russell The intuitive foundations of smart beta 6

Figure 5. Index risk statistics Standard Deviation 3 years 5 years 10 years Russell RAFI US 10.3% 9.9% 16.4% Russell 1000 Equal Weight 11.6% 10.6% 18.1% Russell 1000 10.3% 9.6% 15.4% Tracking Error 3 years 5 years 10 years Russell RAFI US 2.6% 2.4% 3.0% Russell 1000 Equal Weight 3.9% 3.5% 5.2% Russell 1000 0.0% 0.0% 0.0% Source: FTSE Russell, data as of May 31, 2017. Past performance is no guarantee of future results. Returns shown may reflect hypothetical historical performance. Please see the end for important legal disclosures. The deviation of the smart beta indexes return and risk measures from those of the reference index should be considered in terms of the stated objectives of the two different smart beta approaches. For the Russell RAFI US Index, whose objective is to select securities based on fundamental measures, rather than stock price, the concentration level in the top ten securities is similar to that of the capitalizationweighted reference index in this case, the Russell 3000 Index (see Figure 6). At the same date, the Russell RAFI US Index also exhibited an improvement in fundamental scores vis-à-vis its capitalization-weighted counterpart, the Russell 1000 Index 9 (see Figure 7). Although it had a higher priceearnings ratio, the Russell RAFI US Index had lower price-to-sales and price-to-book ratios than the reference index, as well as a higher dividend yield. For the Russell 1000 Equal Weight Index, the effective number of securities (Effective N) in Figure 6 is markedly higher than that of the reference index, the Russell 1000, while the concentration in the top ten holdings is markedly lower than that of the reference index. 10 These statistics, together with the relatively high active share, show that the index is meeting its primary objectives of reducing concentration and increasing diversification. Figure 6. Active share and concentration measures for the Russell 1000 Equal Weight and Russell RAFI US Indexes Index-level statistics Russell RAFI US Russell 1000 Equal Weight Russell 1000 Russell 3000 Active share 0.24 0.53 n/a n/a Effective N 159 766 162.5 189.3 Number of securities 1485 982 1000 3000 Weight in top 10 17.8% 2.4% 15.9% 17.2% Source: FTSE Russell, data as of May 31, 2017. Active share for the Russell RAFI US Index is measured relative to the Russell 3000 Index and Active Share for the Russell 1000 Equal Weight Index is measured relative to the Russell 1000 Index. Please see the end for important legal disclosures. 9 The Russell 1000 Index of large-cap stocks is used here for purposes of comparison (the Russell RAFI US Index draws from a starting universe of the Russell 3000 Index). 10 The Russell 1000 Equal Weight Index uses a modified version of the equal weight methodology, in which sector weights are first equalized, then constituent weights are equalized within each Sector. A pure, constituent-level equal weight index would have an Effective N equal to the number of constituents in the starting universe (in this case, 1000). FTSE Russell The intuitive foundations of smart beta 7

Figure 7. Valuation measures for the Russell RAFI US Index Index Price to sales ratio Price to book ratio Price earnings ratio Dividend yield (%) Russell RAFI US 1.23 2.44 27.45 2.37 Russell 1000 2.04 3.10 23.95 1.97 Source: FTSE Russell, data as of May 31, 2017. Please see the end for important legal disclosures. Conclusion Smart beta doesn t have to be complex. While some non-traditional index approaches rely on more involved quantitative modelling, some popular smart beta index construction approaches follow simple, intuitive methodologies. Equal-weighted and fundamental indexes are two such approaches: they use a straightforward construction methodology in order to address potential concentration or valuation concerns inherent in capitalization-weighted indexes. When choosing between standard, capitalization-weighted indexes and smart beta indexes, market participants often bear in mind questions of simplicity and transparency, cost, liquidity, capacity and governance. Index users should also evaluate smart beta in the context of the original objectives, such as the generation of excess index returns, the mitigation of volatility, or diversification. This information is as readily available as it is for capitalization-weighted indexes and should not be an obstacle to considering smart beta alongside more traditional index options. FTSE Russell The intuitive foundations of smart beta 8

For more information about our indexes, please visit ftserussell.com. 2017 London Stock Exchange Group plc and its applicable group undertakings (the LSE Group ). The LSE Group includes (1) FTSE International Limited ( FTSE ), (2) Frank Russell Company ( Russell ), (3) FTSE TMX Global Debt Capital Markets Inc. and FTSE TMX Global Debt Capital Markets Limited (together, FTSE TMX ) and (4) MTSNext Limited ( MTSNext ). All rights reserved. FTSE Russell is a trading name of FTSE, Russell, FTSE TMX and MTS Next Limited. FTSE, Russell, FTSE Russell MTS, FTSE TMX, FTSE4Good and ICB and all other trademarks and service marks used herein (whether registered or unregistered) are trademarks and/or service marks owned or licensed by the applicable member of the LSE Group or their respective licensors and are owned, or used under license, by FTSE, Russell, MTSNext, or FTSE TMX. All information is provided for information purposes only. Every effort is made to ensure that all information given in this publication is accurate, but no responsibility or liability can be accepted by any member of the LSE Group nor their respective directors, officers, employees, partners or licensors for any errors or for any loss from use of this publication or any of the information or data contained herein. No member of the LSE Group nor their respective directors, officers, employees, partners or licensors make any claim, prediction, warranty or representation whatsoever, expressly or impliedly, either as to the results to be obtained from the use of the FTSE Russell Indexes or the fitness or suitability of the Indexes for any particular purpose to which they might be put. No member of the LSE Group nor their respective directors, officers, employees, partners or licensors provide investment advice and nothing in this document should be taken as constituting financial or investment advice. No member of the LSE Group nor their respective directors, officers, employees, partners or licensors make any representation regarding the advisability of investing in any asset. A decision to invest in any such asset should not be made in reliance on any information herein. Indexes cannot be invested in directly. Inclusion of an asset in an index is not a recommendation to buy, sell or hold that asset. The general information contained in this publication should not be acted upon without obtaining specific legal, tax, and investment advice from a licensed professional. No part of this information may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission of the applicable member of the LSE Group. Use and distribution of the LSE Group index data and the use of their data to create financial products require a license from FTSE, Russell, FTSE TMX, MTSNext and/or their respective licensors. Past performance is no guarantee of future results. Charts and graphs are provided for illustrative purposes only. Index returns shown may not represent the results of the actual trading of investable assets. Certain returns shown may reflect back-tested performance. All performance presented prior to the index inception date is back-tested performance. Back-tested performance is not actual performance, but is hypothetical. The back-test calculations are based on the same methodology that was in effect when the index was officially launched. However, back- tested data may reflect the application of the index methodology with the benefit of hindsight, and the historic calculations of an index may change from month to month based on revisions to the underlying economic data used in the calculation of the index. This publication may contain forward-looking statements. These are based upon a number of assumptions concerning future conditions that ultimately may prove to be inaccurate. Such forward-looking statements are subject to risks and uncertainties and may be affected by various factors that may cause actual results to differ materially from those in the forward-looking statements. Any forward-looking statements speak only as of the date they are made and no member of the LSE Group nor their licensors assume any duty to and do not undertake to update forward-looking statements. FTSE Russell 9

About FTSE Russell FTSE Russell is a leading global index provider creating and managing a wide range of indexes, data and analytic solutions to meet client needs across asset classes, style and strategies. Covering 98% of the investable market, FTSE Russell indexes offer a true picture of global markets, combined with the specialist knowledge gained from developing local benchmarks around the world. FTSE Russell index expertise and products are used extensively by institutional and retail investors globally. $12.5 trillion is currently benchmarked to FTSE Russell indexes. For over 30 years, leading asset owners, asset managers, ETF providers and investment banks have chosen FTSE Russell indexes to benchmark their investment performance and create investment funds, ETFs, structured products and index-based derivatives. FTSE Russell indexes also provide clients with tools for asset allocation, investment strategy analysis and risk management. A core set of universal principles guides FTSE Russell index design and management: a transparent rules-based methodology is informed by independent committees of leading market participants. FTSE Russell is focused on index innovation and customer partnership applying the highest industry standards and embracing the IOSCO Principles. FTSE Russell is wholly owned by London Stock Exchange Group. For more information, visit ftserussell.com. To learn more, visit ftserussell.com; email info@ftserussell.com; or call your regional Client Service Team office: EMEA +44 (0) 20 7866 1810 North America +1 877 503 6437 Asia-Pacific Hong Kong +852 2164 3333 Tokyo +81 3 3581 2764 Sydney +61 (0) 2 8823 3521 FTSE Russell