INVESTMENT STRATEGIES FOR NON-DIVIDEND PAYERS

Transcription

1 Ned Davis Research Group NDR Solutions Quarterly Investment Insight from NDR's Custom Solutions Team INVESTMENT STRATEGIES FOR NON-DIVIDEND PAYERS MARCH 2016 LUCY LIU, CFA Senior Research Analyst, Custom Research Solutions STEPHANIE O'BRIEN Research Analyst, Custom Research Solutions

2 TABLE OF CONTENTS Executive Summary... 1 Introduction... 1 Overview of U.S. and International Non-Dividend Payers... 2 Non-Dividend-Paying Stocks Are More Common in the U.S... 2 The Same Sector Can Have a Very Different Proportion of Non-Dividend-Paying Stocks Examining the Performance of Dividend Payers over Non-Payers... 5 Dividend Payers Have Outperformed Non-Payers Over the Long Run... 5 The Relative Performance Varies in Different Periods... 6 Explaining the Outperformance Under the Fama-French Framework Investment Strategies for U.S. Non-Payers... 9 Strategy A: Mimicking Dividend Payers... 9 Strategy B: Investing in Companies That Have Justifiable Reasons to Not Pay Dividends Comparing the U.S. Non-Payers' Strategies Both Strategies Have Outperformed the U.S. Market Both Strategies Exhibit Robust Performance Across Most Regimes Both Strategies Have Slightly Better Risk Profiles Than the U.S. Market Holdings Comparison Conclusion... 15

3 This edition of NDR Solutions Quarterly discusses investment strategies for non-dividend-paying stocks, which can be helpful for investors with taxable investment accounts. As originally stated, the goal of this publication is to provide our client base with unique investment ideas, tools, and analytical frameworks each quarter. Through this publication, the NDR Solutions team can showcase the varied skill sets and diverse experience we regularly provide to clients with direct engagement, customized output, and tailored solutions. As always, we welcome any feedback, as well as questions about our services, and hope that you find this publication insightful, unique, and useful. EXECUTIVE SUMMARY In the past two decades, dividend payers have outperformed non-dividend payers over the long run. However, after taking away some of the key qualities of dividend-paying stocks, such as lower market beta, higher market cap, and cheaper valuation, the outperformance of dividend-paying stocks over non-dividend paying stocks may be subdued, especially for the U.S. market. In the U.S., the percentage of stocks that are non-dividend-paying is much higher than it is in the rest of the developed world. Because of this, the U.S. market provides greater opportunities to form non-dividend-paying stocks investment strategies. Two investment strategies for non-dividend-paying stocks are discussed in this publication. The first strategy aims to select non-dividend-paying stocks that resemble dividend-paying stocks, and the second strategy screens for companies with justifiable reasons not to pay a dividend. The two strategies are quite different in nature, but both are able to generate comparable or better returns than an equal -weighted portfolio of dividend-paying stocks. INTRODUCTION Income-oriented investments have been quite popular given the U.S. demographic and the low interest rate environment. However, those investments may not be tax efficient for investors using taxable accounts. In reality, investors can have a good amount of investments in taxable accounts for various reasons. For individual investors, tax-deferred investment accounts, such as 401(k) and IRA are great investment vehicles. However, those accounts typically come with restrictions, such as early withdrawal penalties and contribution limits. For investors who prefer higher liquidity and flexibility of their wealth, they may not be willing to or be able to invest all their savings in those accounts. For institutional investors, some institutions such as foundations and endowments may enjoy tax-exempt status, but others may need to pay tax on investments too. Given the tax situation, it makes sense for investors to allocate as much high-yielding investment into tax-exempt or tax -deferred accounts, and keep low-yielding or non-yielding investments in the taxable account. In terms of stock investing, creating investment strategies for non-dividend-paying stocks can be just as important as creating investment strategies for dividend-paying stocks. In this paper, we present the following: Differences between non-dividend-paying stocks and their major contributing sectors in the U.S. versus the rest of the developed world. The performance spread between dividend-paying stocks and non-dividend-paying stocks under the Fama-French framework. Two investment strategies to pick non-dividend-paying stocks. 1

4 OVERVIEW OF U.S. AND INTERNATIONAL NON-DIVIDEND PAYERS Non-Dividend-Paying Stocks Are More Common in the U.S. To distinguish established non-dividend-paying stocks from those that eliminate dividends possibly due to temporary financial conditions, we divide equity markets into the following four 1 mutually exclusive dividend categories based on companies previous and current calendar years dividend payments: dividend payers, dividend non-payers, dividend initiators and dividend eliminators. To qualify as dividend payers (or nonpayers), the dividend payments for both previous and current calendar year need to be non-zeros (or zeros). For the U.S. stock market, which is represented by the MSCI U.S. investable market universe 2, about 48% of the stocks are non-dividend-paying on average, as is shown in the chart below. For the international market, which is represented by the MSCI World ex. U.S. investable market universe, about only 18% of the stocks are non-dividend-paying on average. Namely, non-dividend-paying stocks are more common in the U.S. than in the international market. The top three contributing sectors for U.S. non-dividendpaying stocks historically are Information Technology, Health Care, and Consumer Discretionary, while the top three contributing sectors for international non-dividend-paying stocks are Industrials, Materials, and Financials. 1. This approach is different from NDR s SBOX_202_SP5 product, where dividend eliminators are considered non-payers and dividend initiators are considered dividend payers. 2. The MSCI investable market universe includes large-, mid-, and small-cap stocks U.S. vs International Non Dividend Payers and Top Contributing Sectors MSCI U.S. IMI and MSCI World ex U.S. IMI universes are used for U.S. and international stocks. The numbers in the labels correspond to the average counts of non-payers through out the history ( absolute counts and % of all dividend categories ). U.S. Non-Payers (1382, 48.2%) -- Information Technology Non-Payers (404, 13.9%) -- Health Care Non-Payers (261, 9.2%) -- Consumer Discretionary Non-Payers (256, 8.8%) World ex U.S. Non-Payers (833, 18.3%) -- Industrials Non-Payers (139, 2.9%) -- Financials Non-Payers (130, 2.8%) -- Materials Non-Payers (123, 2.8%) NSQ2016Q1_01 Source: MSCI 2

5 The Same Sector Can Have a Very Different Proportion of Non-Dividend-Paying Stocks The chart below expands from the top three contributing sectors mentioned above, and shows all ten sectors contribution to non-dividend-paying stocks in the U.S. and the international markets. NSQ2016Q1_02 It is true that the top contributing sectors for non-dividend-paying stocks in the U.S. and in the international market has something to do with sectors benchmark weights in the two markets. For example, as shown in the chart below, the Information Technology sector has a higher weight in the U.S. market, while the Industrials and Materials sectors have higher weights in the international market. This could lead to the Information Technology sector being the top non-dividendpayers' contributing sector for the U.S. and the Industrials and Materials sectors being the top dividend-payers' contributing sectors for the international market. However, the sector benchmark weights alone cannot explain why some sectors, such as Consumer Discretionary, which have similar weights in the U.S. and in the international market, do not have similar shares of non-dividend-paying stocks in the two markets. NSQ2016Q1_02 3

6 The chart below looks at non-dividend-paying stocks more closely, by comparing the percentage of stocks in each sector that are non-dividend-paying in the U.S. versus in the international market. Note that in the U.S., about 83% of the technology stocks are non-dividend-paying, while in the international market, only 25% of the technology stocks are. In the U.S., about 80% of the health care stocks are non-dividend-paying, while the number is only 28% in the international market. For the consumer discretionary stocks, about 55% of the stocks are non-dividendpaying in the U.S., while only 16% are non-dividend-paying in the international market. The same sectors can have a very different makeup of dividend-paying vs. non-dividend-paying stocks in different markets. This can also lead to different levels of sector contributions for non-dividend-paying stocks in different markets Monthly Data to % of Stocks in Each Sector That Are Non-Payers ( Historical Average ) MSCI U.S. IMI MSCI World ex U.S. IMI Note: MSCI IMI Universe includes small caps. Source: MSCI Energy Materials Industrials Consumer Discretionary Consumer Staples Health Care Financials Information Technology Telecommunic ation Services Utilities NSQ2016Q1_03 4

7 EXAMINING THE PERFORMANCE OF DIVIDEND PAYERS OVER NON- PAYERS Dividend Payers Have Outperformed Non-Payers Over the Long Run As shown in the charts at right and below, dividend payers have outperformed non-dividend payers in both the U.S. and the international ex. U.S. markets since 6/30/1994 3, in terms of both absolute and risk-adjusted returns. This is consistent with our U.S. strategy team s findings for the S&P 500 stocks (SBOX_202_SP5), where dividend payers have outperformed non-payers, dividend growers have outperformed dividend non-changers, and dividend non-changers have outperformed dividend cutters. 3. This is the earliest date of MSCI IMI universe in our database. 5,012 3,981 3,162 2,512 1,995 1,585 1, Returns of MSCI U.S. IMI Stocks by Dividend Policy Payers ( Gain/Annum: 12.1%, Std Dev Ann: 15.5%, Sharpe Ratio: 0.62, Turnover :27.9%, Avg # Stocks: 1347 ) ---- Growers ( Gain/Annum: 12.4%, Std Dev Ann: 13.7%, Sharpe Ratio: 0.72, Turnover :152.0%, Avg # Stocks: 501 ) ---- Non-Changers ( Gain/Annum: 11.5%, Std Dev Ann: 16.3%, Sharpe Ratio: 0.55, Turnover :137.4%, Avg # Stocks: 794 ) ---- Cutters ( Gain/Annum: 4.6%, Std Dev Ann: 21.8%, Sharpe Ratio: 0.09, Turnover :175.7%, Avg # Stocks: 52 ) Non-Payers ( Gain/Annum: 9.2%, Std Dev Ann: 25.1%, Sharpe Ratio: 0.27, Turnover :49.7%, Avg # Stocks: 1385 ) Initiators ( Gain/Annum: 7.8%, Std Dev Ann: 17.8%, Sharpe Ratio: 0.30, Turnover :221.7%, Avg # Stocks: 38 ) Eliminators ( Gain/Annum: -0.9%, Std Dev Ann: 36.2%, Sharpe Ratio: -0.09, Turnover :235.3%, Avg # Stocks: 18 ) Source: MSCI, S&P Capital IQ Compustat Monthly Data to (Log Scale) 5,012 Portfolios are Equal Weighted, Rebalanced Quarterly. Total Returns are in USD. The Major Dividend Categories Listed in the Chart are Mutually Exclusive. Dividend Payers Include Dividend Growers, Cutters and and Non-Changers. Dividend Growers are Stocks with Positive Change of IDC ( for International Stocks ) or Compustat ( for U.S. stocks ) Indicated Annual DPS from the Previous Year End ( in Local Currency ). 3,981 3,162 2,512 1,995 1,585 1, ,413 1,259 1, Returns of MSCI World ex U.S. IMI Stocks by Dividend Policy NSQ2016Q1_04 Payers ( Gain/Annum: 6.5%, Std Dev Ann: 16.2%, Sharpe Ratio: 0.24, Turnover :36.5%, Avg # Stocks: 3324 ) ---- Growers ( Gain/Annum: 8.7%, Std Dev Ann: 15.5%, Sharpe Ratio: 0.39, Turnover :147.3%, Avg # Stocks: 1209 ) ---- Non-Changers ( Gain/Annum: 4.3%, Std Dev Ann: 17.1%, Sharpe Ratio: 0.10, Turnover :155.7%, Avg # Stocks: 1745 ) ---- Cutters ( Gain/Annum: 3.9%, Std Dev Ann: 17.9%, Sharpe Ratio: 0.07, Turnover :172.3%, Avg # Stocks: 370 ) Non-Payers ( Gain/Annum: 1.9%, Std Dev Ann: 24.3%, Sharpe Ratio: -0.03, Turnover :75.8%, Avg # Stocks: 851 ) Initiators ( Gain/Annum: 4.5%, Std Dev Ann: 19.9%, Sharpe Ratio: 0.10, Turnover :202.2%, Avg # Stocks: 168 ) Eliminators ( Gain/Annum: 2.9%, Std Dev Ann: 25.2%, Sharpe Ratio: 0.02, Turnover :218.1%, Avg # Stocks: 74 ) Source: MSCI NSQ2016Q1_04 Monthly Data to (Log Scale) Portfolios are Equal Weighted, Rebalanced Quarterly. Total Returns are in USD. The Major Dividend Categories Listed in the Chart are Mutually Exclusive. Dividend Payers Include Dividend Growers, Cutters and and Non-Changers. Dividend Growers are Stocks with Positive Change of IDC ( for International Stocks ) or Compustat ( for U.S. stocks ) Indicated Annual DPS from the Previous Year End ( in Local Currency ) ,413 1,259 1,

8 The Relative Performance Varies in Different Periods The chart below provides a more detailed view of the relative performance of dividend payers versus non-payers in the U.S. and the international market. The green relative performance line represents the U.S. market, and it is a ratio of U.S. equal-weighted dividend payers total return index to the U.S. equal-weighted non-payers total return index, starting at 1:1. Over the long run, U.S. dividend payers have outperformed non-payers, as evidenced by the relative performance line rising from 1:1 to around 1.8:1. Most of the outperformance of U.S. dividend payers is concentrated between 1994 and After 2004, U.S. dividend payers exhibit relatively flat performance over the non-payers. The orange relative performance line represents the international market, and for the international market, most of the dividend payers outperformance is concentrated in recent years. For the periods 1994 to 2000 and 2002 to 2008, the international non-payers outperformed the international dividend payers. 2.7 Monthly Data to Dividend Payers vs Non-Payers Relative Performance MSCI U.S. IMI Dividend Payers / Non-Payers MSCI World ex U.S. IMI Dividend Payers / Non-Payers Relative performance is measured by the dividend payers total return index divided by non-payers total return index (in USD). Start at 1: Source: MSCI NSQ2016Q1_05 6

9 Explaining the Outperformance Under the Fama-French Framework Dividend payers tend to have lower beta, larger market cap, and cheaper valuations than non-payers. When taking these dividend payers' features away, the excess returns of U.S. dividend -paying stocks over non-payers may no longer look significant. The tables below use the Fama-French three-factor model to examine the statistical significance of the monthly excess returns of dividend payers over non-payers. Monthly returns of the MSCI IMI growth index 4 over the MSCI IMI value index are used to represent the growth minus value factor, and monthly returns of the MSCI small cap index over the MSCI large cap index are used to represent the small cap minus large cap factor. The negative coefficient estimates for the "MSCI IMI", "Small Cap Minus Large Cap" and "Growth Minus Value" factors indicate that dividend payers in general have lower beta, larger market cap, and cheaper valuation than non-payers. The zero p-values of those estimates indicate that those dividend payers' features are not negligible. The coefficient estimate of the intercept represents monthly excess returns of dividend payers over non-payers. For the period to 2016, the estimate of the excess returns of dividend payers over non-payers in the U.S. market is positive; however, it has a p-value of 0.219, which is greater than 0.1, indicating the excess returns of the dividend payers versus non-payers are not very different from zero. In the international market, the estimate of the excess returns of dividend payers has a p-value of 0, meaning the excess returns are considered different from zero and statistically significant in the international market, even after taking the beta, market cap, and valuation factors into account. 4. All indices are total return indices in USD. 5. MSCI small cap indices start in 2001 in our database. REGRESSION ANALYSIS SUMMARY Coefficient Estimate Std. Error T-Value P-Value intercept MSCI U.S. IMI Small Cap Minus Large Cap Growth Minus Value Ned Davis Research Group T_NSQ201603XX1.1 REGRESSION ANALYSIS SUMMARY Coefficient Estimate Std. Error T-Value P-Value intercept MSCI World ex U.S. IMI Small Cap Minus Large Cap Growth Minus Value Ned Davis Research Group T_NSQ201603XX1.2 7

10 The chart below illustrates the statistical significance of excess returns of dividend payers over non-payers in detail, using three-year rolling regressions instead of a single regression for the period from 2001 to The green dots represent the excess returns of dividend payers over non-payers from the rolling three-year regressions ending on 12/31 each year for the U.S. market, and the orange dots represent the excess returns for the international market. Statistically insignificant excess returns are grayed out in the chart. For the U.S. markets, almost all excess returns throughout different rolling three-year timeframes in history are not statistically significant, after removing the three key features of dividend-paying stocks. However, for the international markets, the dividend payers excess returns in recent years are statistically significant. Monthly Excess Returns of Dividend Payers Over Non-Payers Monthly Data Monthly Excess Returns (%) Note: Fama-French three factor model is used to calculate monthly excess returns on a rolling 3-year basis ending on 12/31 each year. Excess returns with p values greater than 0.10 are considered statistically insignificant, and are grayed out. MSCI World ex U.S. IMI MSCI U.S. IMI Year Source: MSCI NSQ2016Q1_07 8

11 INVESTMENT STRATEGIES FOR U.S. NON-PAYERS Given non-dividend paying stocks are more common in the U.S. than in the international market, and there is strong evidence showing that non-payers are not necessarily inferior to dividend payers in the U.S. market purely because they are not paying any dividends, we will be focusing on developing investment strategies for U.S. non-payers in this publication. Strategy A: Mimicking Dividend Payers If one is happy with the performance of dividend-paying stocks, but the concern is paying tax on dividends, then why not create a non-dividend-paying portfolio mimicking similar features of the dividend-paying stocks? MSCI U.S. IMI Non Payers Strategy A As mentioned in the analysis above, dividend payers tend to have lower beta, larger market cap, and cheaper valuations than non-payers. By creating a scoring system that ranks nonpayers based on these three features, one may be able to pick a basket of non-payers that resembles dividend payers. The chart below shows a strategy that uses such a scoring system. The best tile (tile 5 in the chart) represents the strategy, namely the basket of non-dividend-paying stocks that match the three features of dividend payers most closely. The performance of this strategy has been quite impressive, with 14.3% gain per annum and 0.67 Sharpe ratio from 1994 to Monthly Data to (Log Scale) 5,623 3,162 1, Quintile Portfolio Gain/Annum Std Dev Ann Sharpe Ratio Turnover Avg # Stocks 1 0.9% 38.6% % % 28.9% % % 22.7% % % 19.5% % % 17.5% % 231 Shaded areas represent NDR defined bear markets. Portfolios are Equal Weighted, Rebalanced Quarterly. Total Returns are in USD. The strategy is based on the overall scores of the following factors: 3-Year Rolling Market Beta ( Small Values = Low Scores ), Market Capitalization ( Large Values = High Scores ), EBITDA to EV ( Large Values = High Scores ). 5,623 3,162 1, ,778 Strategy A Best / Worst Quintile ( the Higher the Quintile Number, the Higher the Values, the Better ) 1,778 Source: MSCI NSQ2016Q1_08 9

12 Strategy B: Investing in Companies That Have Justifiable Reasons to Not Pay Dividends Another approach is to think about when it makes sense for a company not to pay a dividend, and then invest in these types of companies. When does it make sense for a company not to pay a dividend? One answer: when a company is very profitable and has great growth opportunities, it makes sense for it to plowback more to the business than distributing earnings to the shareholders. In this strategy, we use return on equity (ROE) to represent companies profitability, and the past three years earnings per share (EPS) growth trend 6 to gauge if companies have good growth opportunities. In addition, we would like to avoid companies with high accrual ratios and avoid glamourous companies with very high growth anticipation 7 from The Street. As shown in the chart below, this strategy (tile 5) has also shown impressive performance for the period 1994 to 2016, and the spread between the best and worst tile is even more consistent throughout the history than the first strategy. 6. Measured by rolling three-year z-score of trailing twelve months earnings per share. 7. Measured by estimated year-over-year sales per share growth. 10,000 5,623 3,162 1, MSCI U.S. IMI Non Payers Strategy B Quintile Portfolio Gain/Annum Std Dev Ann Sharpe Ratio Turnover Avg # Stocks 1-6.4% 33.5% % % 26.7% % % 23.7% % % 22.1% % % 21.1% % 203 Monthly Data to (Log Scale) 10,000 5,623 3,162 1, Shaded areas represent NDR defined bear markets. Portfolios are Equal Weighted, Rebalanced Quarterly. Total Returns are in USD. The strategy is based on the overall scores of the following factors: Return on Equity ( Large Values = High Scores ), 3-Year EPS Growth ( Large Values = High Scores ), Estimated Year over Year SPS Growth ( Small Values = Low Scores ), Accruals Ratio ( Small Values = Low Scores ) ,000 5,623 3,162 1, NSQ2016Q1_8 Strategy B Best / Worst Quintile ( the Higher the Quintile Number, the Higher the Values, the Better ) Source: MSCI ,000 5,623 3,162 1,

13 One explanation of the consistent spread between the best and the worst tile is that the factors used in this strategy have relative low correlation to each other, and some factors undesirable performance in certain years can be compensated by other factors desirable performance in those years. The chart below shows each factor s performance year by year, measured by the total returns of the factor s best tile minus the worst tile. In general, green means a factor performed well in a year, while red means vice versa. It is hard to find a year during which all the factors had undesirable performance in any year from to 2015, and each factor seems to have its own good versus bad performance pattern. Using traditional correlation analysis for the full history, as shown in the table below, the correlations among the four factors are quite low, ranging from to is the earliest year when all the factors have performance stats. The accruals factor starts in 1998, the estimated year-over-year sales per share growth factor starts in 1996, the return on equity factor starts in 1994, and the threeyear earnings per share factor starts in NSQ2016Q1_9 MSCI U.S. IMI NON PAYERS QUANT FACTORS CORRELATION (06/30/ /29/2016) Return on Equity 3-Year EPS Growth Estimated Year over Year SPS Growth Accruals Ratio Return on Equity Year EPS Growth Estimated Year over Year SPS Growth Accruals Ratio Copyright 2016 Ned Davis Research Group. Further distribution prohibited without prior permission. All rights reserved. See NDR disclaimer at For data vendor disclaimers refer to Ned Davis Research Group T_NSQ201603XX1.1 11

14 COMPARING THE U.S. NON-PAYERS' STRATEGIES The above section shows the performance of the two nonpayers' strategies on their own. In this section, we will compare the two strategies performance with the broad U.S. market and with the dividend payers portfolio, and analyze the strength and weakness of those strategies. Both Strategies Have Outperformed the U.S. Market The charts below show the two non-dividend-paying strategies and the dividend payers portfolio s performance relative to the broad U.S. market. For comparison purpose, the performance is calculated from equal-weighted, quarterly rebalanced, total return indices. From 1994 to 2016, both of the non-payer strategies have outperformed the customized U.S. market benchmark, in terms of both absolute and risk-adjusted returns. The two strategies have also outperformed the dividend payers portfolio in terms of the absolute returns, and achieved comparable risk-adjusted returns. Note that the relative performance line for Strategy A, which tries to resemble the dividend payers portfolio, and the relative performance line for the dividend payers portfolio look correlated, while the relative performance line for Strategy B looks quite different. This confirms that Strategy B is quite different from Strategy A, and Strategy A succeeds in mimicking the dividend payers portfolio to a certain degree. Relative Performance Against MSCI U.S. IMI Equal-Weighted Total Return Index Monthly Data to (Log Scale) MSCI U.S. IMI Non-Payers Strategy A Equal-Weighted Total Return MSCI U.S. IMI Non-Payers Strategy B Equal-Weighted Total Return MSCI U.S. IMI Payers Equal-Weighted Total Return Source: MSCI Name Gain/Annum Ann Std Dev Downside Dev Sharpe Ratio Max Drawdown (%) MSCI U.S. IMI Non-Payers Strategy A Equal-Weighted Total Return 14.3% 17.5% 11.2% % MSCI U.S. IMI Non-Payers Strategy B Equal-Weighted Total Return 13.7% 21.1% 12.9% % MSCI U.S. IMI Payers Equal-Weighted Total Return 12.1% 15.5% 10.2% % MSCI U.S. IMI Equal-Weighted Total Return 10.8% 19.8% 12.9% % NSQ2016Q1_11 12

15 Both Strategies Exhibit Robust Performance Across Most Regimes The historical period from 6/30/1994 to 2/29/2016 can be broken down into four different pairs of regimes: equity bull market and equity bear market, commodity bull market and commodity bear market, high volatility market and low volatility market 9, and rising real interest rates 10 market and falling real interest rates market. MSCI U.S. IMI Non-Payers Strategy A Equal-Weighted Total Return MSCI U.S. IMI Non-Payers Strategy B Equal-Weighted Total Return MSCI U.S. IMI Payers Equal-Weighted Total Return Source: MSCI The chart on the top right shows the two non-dividend paying strategies' excess gain per annum over the U.S. market in different regimes, and the chart on the bottom right shows the excess gain per annum of the dividend payers portfolio over the U.S. market. Both of the non-dividend-paying strategies have outperformed the U.S. market in most of the regimes, except that Strategy B has underperformed the U.S. market by about -1.2% per annum in the commodity bear market. The shapes of the two strategies look quite different: Strategy B has performed better than Strategy A in equity bull market, commodity bull market, and high volatility market; while Strategy A has performed better than Strategy B in equity bear market, commodity bear market, and low volatility market. NSQ2016Q1_12 The dividend payers portfolio has a more similar shape to Strategy A, except that it has performed better in the falling real interest rates market than in the rising real interest rates market, indicating one aspect of the dividend payers portfolio that Strategy A may not be able to fully replicate. Unlike the two non-dividend-paying strategies, the dividend payers portfolio has underperformed the U.S. market in equity bull markets by about -3.6% per annum and in rising real interest rates environment by about -0.9% per annum. 9. VIX index is used to define high versus low volatility market. The historical average of the VIX index during the 6/30/1994 to 2/29/2016 period is When the VIX index is higher than 21.2, the market is considered a high volatility market. Vice versa, for the low volatility market. 10. Note that the real interest rates are used here instead of the nominal interest rates. The nominal yields of 10- year U.S. treasuries minus the U.S. inflation rate is used to represent the real interest rates. NSQ2016Q1_12 13

16 Both Strategies Have Slightly Better Risk Profiles Than the U.S. Market For the period from 6/30/1994 to 2/29/2016, the chart on the top right shows the maximum drawdowns of the two non-dividend-paying strategies minus the maximum drawdown of the U.S. market in the following regimes: equity bear market, commodity bear market, and high volatility market. A positive number indicates the maximum drawdown is smaller than that of the U.S. market, which is considered desirable. As shown in the chart, the two non-dividend-paying strategies display less maximum drawdown than the U.S. market in equity bear market and the high volatility market. However, their maximum drawdowns are slightly higher, though not significantly higher, than the U.S. market in commodity bear market: Strategy A experienced 0.3% more drawdowns, and Strategy B experienced 1.5% more drawdowns in commodity bear market. Source: NSQ2016Q1_12 MSCI U.S. IMI Non-Payers Strategy A Equal-Weighted Total Return MSCI U.S. IMI Non-Payers Strategy B Equal-Weighted Total Return MSCI U.S. IMI Payers Equal-Weighted Total Return In general, Strategy A displays a slightly better risk profile than Strategy B, and has slightly fewer drawdowns than Strategy B in all of the three regimes. As for the dividend-paying portfolio shown in the chart on the bottom right, it exhibits a less impressive risk profile in high volatility market, but shows a better risk profile in the commodity bear market than the two non-dividend-paying strategies. NSQ2016Q1_12 14

17 Holdings Comparison Last but not least, one may be interested in seeing those strategies sector exposure. Sector exposure analysis on portfolios holdings is a good and a common tool to use; however, the true nature of the portfolios may elude this type of analysis. The chart below shows the sector weights of the current holdings of the two equal-weighted non-dividend-paying strategies' holdings, alongside those of the equal-weighted dividend payers portfolio. Though the two non-dividend-paying strategies are quite different in nature, the sector breakdown of the strategies are more similar to each other: Technology, Health Care, and Consumer Discretionary sectors have higher weights in the two strategies holdings. For the dividend-paying portfolio, even though it shared a lot of common features with Strategy A, its sector exposure is quite different. Financials, Materials, and Industrials sectors have higher weights in the dividend payers portfolio s holdings. NSQ2016Q1_13 CONCLUSION In summary, unlike dividend-paying stocks, non-dividend-paying stocks are more tax efficient for investors with taxable accounts. Investment strategies can be developed to enhance non-dividend-paying stocks performance and reach a comparable or even better returns than the dividend-paying stocks. It is also possible to develop a strategy that resembles key features of the dividend-paying stocks. For more information or requests for further research, please contact our custom research team at NDRSolutionsPub@ndr.com 15

18 WENJIA ( LUCY ) LIU, CFA Senior Research Analyst Wenjia ( Lucy ) Liu, CFA, Senior Research Analyst, is a member of the Custom Research Solutions group. She specializes in global markets, asset allocation, and investment strategy research, and was the lead developer of the ETF Asset Allocation Model. Lucy joined NDRG in Prior to joining NDRG, Lucy worked at Boston Common Asset Management, where she conducted fundamental research on international stocks. Before coming to the U.S., she worked as an analyst for American International Assurance Company, where she conducted analysis for the bancassurance division and provided training for salespeople. Lucy earned her Master of Business Administration degree with a focus in finance from Carnegie Mellon University, Tepper School of Business, in She received her dual bachelor degrees in Law and Economics from Peking University. Lucy is a CFA charterholder and is a member of the CFA Institute. STEPHANIE O BRIEN Research Analyst Stephanie O Brien, Research Analyst, Custom Research Solutions, performs a variety of research projects for NDRG clients. Stephanie joined the firm in 2008 and previously worked for Ned Davis, the firm s President and Senior Investment Strategist, developing indicators, models, and studies, as well as building and maintaining charts and reports. Stephanie graduated cum laude from the University of South Florida with a Bachelor of Science in Economics. 16

19 NED DAVIS RESEARCH GROUP (800) VENICE 600 Bird Bay Drive West Venice, FL (941) BOSTON 50 Federal Street 6 th Floor Boston, MA (617) ATLANTA 400 Northridge Road Suite 515 Atlanta, GA (678) SAN FRANCISCO 50 California Street Suite 1500 San Francisco, CA (415) LONDON 8 Bouverie Street London EC4Y 8AX +44 (0) NDRG EDITORIAL BOARD Ned Davis Senior Investment Strategist Tim Hayes, CMT Chief Global Investment Strategist Joseph Kalish Chief Global Macro Strategist DISCLAIMER Ed Clissold, CFA Chief U.S. Strategist Brian Sanborn, CFA Director of Research Applications The data and analysis contained herein are provided as is and without warranty of any kind, either expressed or implied. Ned Davis Research, Inc. (NDR), d.b.a. Ned Davis Research Group (NDRG), any NDRG affiliates or employees, or any third-party data provider, shall not have any liability for any loss sustained by anyone who has relied on the information contained in any NDRG publication. NDRG disclaims any and all express or implied warranties, including, but not limited to, any warranties of merchantability, suitability or fitness for a particular purpose or use. NDRG s past recommendations and model results are not a guarantee of future results. This communication reflects our analysts opinions as of the date of this communication and will not necessarily be updated as views or information change. All opinions expressed herein are subject to change without notice. NDRG or its affiliated companies or their respective shareholders, directors, officers and/or employees, may have long or short positions in the securities discussed herein and may purchase or sell such securities without notice. Using any graph, chart, formula or other device to assist in deciding which securities to trade or when to trade them presents many difficulties and their effectiveness has significant limitations, including that prior patterns may not repeat themselves continuously or on any particular occasion. In addition, market participants using such devices can impact the market in a way that changes the effectiveness of such device. Further distribution prohibited without prior permission. For data vendor disclaimers, refer to Copyright 2016 (c) Ned Davis Research, Inc. All rights reserved. Generate alpha. Identify risk. Choose Ned Davis Research. Founded in 1980, Ned Davis Research Group is a leading independent research firm with clients around the globe. With a range of products and services utilizing a 360 methodology, we deliver awardwinning solutions to the world s leading investment management companies. Our clients include professionals from global investment firms, banks, insurance companies, mutual funds, hedge funds, pension and endowment funds, and registered investment advisors. Macro Fundamental Idea Technical Sentiment 360 APPROACH