RBC Dominion Securities Inc. Matthew Barasch, CFA (Canadian Equity Strategist) (416) 842-7857 Matt.barasch@rbccm.com May 24, 2016 Canadian Strategy: A deeper dive into seasonality Sell in May and hold on a second Key Points Our View: The S&P/TSX demonstrates significant seasonality. However, unlike the traditional sell in May and go away, the S&P/TSX would better be defined as buy when it snows, sell on the black flies (June for those of you not familiar with these wretched insects). May, which gets a bad reputation, is actually the second best performing month of the year for the S&P/TSX (trailing on December), with all 10 sectors in the index generating a positive return over the past 35 years. All values in Canadian dollars unless otherwise noted. Priced as of prior trading day s market close, ET (unless otherwise stated). For Required Non-U.S. Analysts and Conflicts Disclosures, please see page 8. The gap in performance has been nearly 10%. If we look back over the past 3+ decades, the S&P/TSX has generated a 9.6% total return (non-annualized) from the beginning of November through the end of May (period 1) and a slight loss from June through October (period 2). Four of the five worst months of the year in terms of performance occur from June to November. The outperformance has been consistent. In any given year, period 1 outperforms period 2 83% of the time with an average gap in performance of 14% (nonannualized). The cyclical sectors show a much bigger gap in performance. Energy, Materials, Industrials, Technology and Consumer Discretionary show a much larger gap in performance from November to June vs. June to November. The non-cyclicals Financials, Telecom, Utilities, and Consumer Staples show less variation from period to period. Health Care is an outlier, falling more in the cyclical camp in terms of gap in performance. A seasonality strategy would have worked over the past 3+ decades. We explore whether or not one could employ some sort of seasonality strategy. Our findings suggest that a dynamically weighted model that looks to underweight cyclical sectors from June to November would have added value over time; however, it is dependent on how aggressively one would have been willing to overweight non-cyclicals and underweight cyclicals.
Sell in May and hold on a second Key points Exhibit 1: Monthly returns of the S&P/TSX and its 10 sectors Average monthly returns from 1981 to 2015 Month S&P/TSX Financials Utilities Telecom Over the last three-plus decades, the bulk of returns for the S&P/TSX have been generated from November to June. The November to June period has outperformed the June to November period more than 8 in 10 times since 1980. All sectors outperform from November to June vs. June to November; however, the non-cyclical sectors demonstrate a much narrower gap in performance. We found that a dynamic model that overweights non-cyclicals from June to November would have added significant value over time; however, it is dependent on how aggressively one would have been willing to overweight non-cyclicals and underweight cyclicals. Much has been written over the years about the phenomena of seasonality, which is generally defined as the propensity for equity markets to outperform from the beginning of November through the end of April versus the period from the beginning of May to the end of October. This is alternately referred to as sell in May and go away or buy when it snows, sell when it goes. We will admit to some degree of scepticism about such a strategy, as 1) we do believe investing to be harder than that ; and 2) we know of lots of mansions, but we do not know an inordinate number of season-tologists that live in them. However, we felt an exercise that explored seasonality and its fit with the S&P/TSX was relevant for two reasons: 1) intuitively we would assume that if strong and consistent seasonality did exist, it would be more pronounced in the S&P/TSX because of its heavy weighting of cyclical sectors; and 2) we are still in May, so if one were ever going to write about seasonality now would be a good time. With that in mind, we figured a good starting point would be to look at the average monthly returns for the S&P/TSX. In addition, because we wanted to see whether certain sectors demonstrated more seasonality than did others, we added the 10 GIC sectors to our study. Consumer Discretionary Consumer Staples Health Care Industrials Technology Energy Materials January 0.80% -0.16% -0.23% 0.20% 0.63% 0.69% 4.47% 1.53% 4.76% -0.21% 1.46% February 0.91% 1.21% 0.42% 0.70% 1.53% 1.50% 1.81% 1.53% -0.43% 0.82% 1.80% March 1.43% 1.90% 0.62% 1.40% 2.13% 1.23% -1.15% 1.39% -0.20% 3.42% 0.28% April 0.88% 1.49% 0.58% 0.96% 0.84% 0.95% 0.93% 1.15% 1.10% 2.63% -0.22% May 1.64% 1.47% 1.66% 1.65% 1.50% 2.34% 1.32% 1.23% 1.71% 2.35% 1.86% June -0.16% -0.23% 0.23% 0.06% 0.08% 0.97% 0.17% 0.06% -1.94% -0.08% -1.48% July 0.95% 1.81% 1.03% 0.32% 0.29% 0.97% -0.41% 1.00% 0.75% -0.35% 0.70% August 0.66% 0.29% 1.36% 0.99% 0.20% 0.51% 1.76% -0.57% 1.14% 1.72% 1.85% September -1.46% -0.31% 0.04% 0.06% -1.61% -0.05% 0.35% -1.88% -3.24% -0.80% -1.47% October 0.27% 2.10% 1.57% 3.20% 1.16% 1.29% -1.73% 1.21% 3.01% -1.33% -1.87% November 1.54% 2.40% 1.89% 1.92% 0.93% 1.67% 1.79% 0.81% 3.38% -1.24% 2.50% December 2.05% 1.92% 1.90% 3.09% 2.37% 2.23% 2.40% 2.78% 1.73% 1.44% 2.51% Note: All returns are inclusive of dividends. All returns are non-annualized. Green indicates best performer during that month, red represents worst. Source: RBC Capital Markets Quantitative Research May 24, 2016 2
The first take-away we had was that 4 of the 5 worst months of the year for the S&P/TSX occur from June through October. Further, May, which gets a bad reputation (at least from those who trumpet sell in May ), actually ranks as the second best month of the year, trailing only December. In fact, every sector of the index has generated a positive return during May going back 35 years. Buy when it snows, sell on the black flies In light of the above, let s aggregate the data for the past 35 years to get a sense of the performance differentials between the two periods. Exhibit 2: November to June has returned about 10% more per annum than has June to November S&P/TSX June to November November to June Average Return -0.1% 9.6% # Positive 22 29 # Negative 13 6 % Positive 63% 83% % Returns > 10% 11% 51% % Returns < 10% 22% 6% Average Return since 1996-0.1% 9.5% % Positive since 1996 70% 89% Note: All returns are inclusive of dividends. All returns are non-annualized. Note: June to November refers to the period from June 1 to October 31 st. November to June refers to the period from November 1 to May 31 st. Note: All data is 1981 to 2015 unless otherwise indicated. Source: RBC Capital Markets Quantitative Research Not only has the November to June period produced more than 100% of the performance of the S&P/TSX historically, but it has also infrequently generated a negative return. Further, more than half of the outcomes for this period have seen returns of greater than 10%, while only twice in the past 35 years 1982 and 2001 has there been a loss of greater than 10% during the November to June period. One thing we had assumed when we entered into this exercise was that modern markets may have arb d away any seasonality that might have existed, so we added some data on the past 20 years. The bottom line from this was that seasonality has remained relatively consistent in recent years, so our arb d assumption proved misplaced. In fact, the S&P/TSX has risen in 13 of the past 14 November to June periods. The next question we thought to ask was: how often does a November to June period outperform the subsequent June to November period (note: we assume November to June is the first portion of the year, as this follows bank year-end and year-end for most mutual funds). Let s take a look at a chart and then discuss. May 24, 2016 3
Exhibit 3: Comparison of same-year performance, 1981 to 2015 Percentage of time one period outperformed another during same year and average outperformance 90% 83% 80% 70% 60% 50% 40% 30% 20% 10% 14% 17% 11% 0% November to June June to November % Outperform Average Outperformance Note: All returns are inclusive of dividends. All returns are non-annualized. Source: RBC Capital Markets Quantitative Research More than 8 in 10 times, the November to June period outperforms the subsequent June to November period and by an average of 14% (note this is not adjusted for the difference in months; however, when we adjust for this, the differential is still robust). In fact, almost 30% of the time, the November to June period has returned at least 20% more than the subsequent June to November period. Before we move to sector seasonality, let s present a final chart that tracks performance of each period on a month-by-month basis. Exhibit 4: Month-by-month returns of seasonality strategy, 1981 to 2015 Returns indexed to 100 112 110 108 106 104 102 100 98 96 94 1 2 3 4 5 6 7 8 Months June to November November to June Source: RBC Capital Markets Canadian Equity Strategy May 24, 2016 4
Sector seasonality Now let s move to the 10 sectors of the S&P/TSX to determine whether or not any of them might demonstrate any seasonality. Given that the index appears to have a good degree of seasonality, some of the sectors would probably have to as well, but we thought it would be interesting to identify those that worked best. Exhibit 5: The cyclical sectors are the most sensitive to seasonality Performance from 1981 to 2015 for 10 sectors of the S&P/TSX ranked by difference in seasonal performance 16% 14% 12% 14% 12% 11% 11% 11% 11% 10% 8% 6% 8% 7% 6% 4% 3% 2% 0% -2% -4% -6% Materials Health Care Industrials Technology Energy Consumer Disc. Consumer Stap. Financials Telecom Utilities Difference in Performance November to June June to November Note: All returns are inclusive of dividends. All returns are non-annualized. Source: RBC Capital Markets Canadian Equity Strategy All of the sectors demonstrated some seasonality; however, with the exception of Health Care, which would normally slot into the non-cyclical segment of the market, the cyclicals Materials, Industrials, Technology, Energy and Consumer Discretionary were more sensitive to seasonality than were the non-cyclicals Staples, Financials, Telecom and Utilities. In fact, none of the cyclical sectors have generated a positive return from June to November over the past 35 years, with results ranging from 0% for Technology to -3.2% for Materials. Could one capitalize on this? The question from our lens would be is there some way to capitalize on this? That is, could one create a trading strategy that offered meaningful value over and above one s normal strategy based on seasonality? To undertake this exercise, we tested the weights of the cyclical and non-cyclical sectors to see whether or not consistent overweights and underweights in each of the two groups could have added value over time. Our findings suggest that dynamic weightings from November to June added little value; however, going underweight cyclicals from June to November added significant value over time and this value was relatively consistent. May 24, 2016 5
Exhibit 6: Dynamic model market weight November to June; 20% underweight cyclicals June to November, 1981 to 2015 Annual Return November to June June to November Dynamic Model 10.3% 9.6% 0.6% Passive Model 9.5% 9.6% -0.1% Note: All returns are inclusive of dividends. All returns are non-annualized. Source: RBC Capital Markets Canadian Equity Strategy In the above example, we have underweighted cyclicals from June to November by 20% (adding the exposure to non-cyclicals, including Health Care) and then market weighted all sectors from November to June. By taking this aggressive underweight position in cyclicals from June to November, one would have added about 80 basis points of annual performance over the past 35 years, while the strategy would have been successful about 70% of the time. Before we cash-in our fundamental hats and go straight to the seasonality mansion, we would note two things: 1) the model is dependent on how aggressively one is willing to go overweight non-cyclicals and underweight cyclicals from June to November; and 2) the model is skewed to some degree by the late 1990s and the Nortel-effect. If we remove the period from 1996 to 2002 when the impact was most acute in our model, the ~80 basis points of annual outperformance indicated above narrows to about 40, which is still meaningful albeit not as robust. Exhibit 7: Annual performance of dynamic model relative to S&P/TSX, 1981 to 2015 6.0% 5.0% 4.0% 3.0% 2.0% 1.0% 0.0% 1981 1985 1989 1993 1997 2001 2005 2009 2013-1.0% -2.0% Note: All returns are inclusive of dividends. All returns are non-annualized. Source: RBC Capital Markets Canadian Equity Strategy Why might seasonality exist Before we close the book on this exercise, we felt it would be worthwhile exploring some of the reasons why seasonality might exist. Summer liquidity challenges: This is probably the most commonly cited rationale for poor summer results. As many go away on holidays for parts of the summer months, liquidity declines and stocks and markets become prone to wilder swings. Add to this a May 24, 2016 6
general slowdown in deal activity mergers and acquisitions, new issues and the market may also be lacking catalysts. Poor third quarter earnings: Because of the summer liquidity challenges noted above and because some industries shut down for the summer, third quarter earnings may tend to be less inspiring than other quarters of the year. Bad stuff just seems to happen a lot in the summer and early fall: While there is not necessarily a good reason that big market crashes have tended to occur in September and October, they have, and this has had some impact on the data. We suppose the stock market could have crashed in November 1987 rather than October, or Lehman Brothers could have gone under in January as opposed to September, but bad stuff just seems to happen more from May to November than it does from November to May. The New Year brings optimism: Corporate budgets are flush with new spending, people are optimistic heading into the New Year, and whatever may or may not have gone wrong in the prior year is forgotten. We would add that in Canada, RRSP season kicks into high gear right after New Year, which adds another level of buying to the market. Shoulder seasons and jewellery demand: Because of the S&P/TSX s heavy exposure to commodities, it might be expected to be more susceptible to the ebbs and flows of demand seasonality. Both so-called shoulder seasons for natural gas and heating oil demand fall in the May to November period. Meanwhile, the bulk of Indian wedding season falls in the November to May period, as does the holiday season, which should tend to stoke demand for gold. Conclusion The S&P/TSX has demonstrated significant seasonality (June to November vs. November to June) over the last three-plus decades, with the November to June period generating more than 900 basis points of average annual outperformance vs. the June to November period. Furthermore, all sectors within the S&P/TSX demonstrate varying levels of seasonality, with the cyclicals (Materials, Industrials, Technology, Consumer Discretionary, and Energy) more pronounced than non-cyclicals (Financials, Utilities, Telcos, and Consumer Staples). A dynamic model that focuses on underweighting cyclicals from June to November would have added value over the past 35 years; however, the magnitude to which one was willing to underweight cyclicals plays an important role in the success of the model. May 24, 2016 7
Required disclosures Non-U.S. analyst disclosure Matt Barasch (i) is not registered/qualified as a research analyst with the NYSE and/or FINRA and (ii) may not be an associated person of the RBC Capital Markets, LLC and therefore may not be subject to FINRA Rule 2241 restrictions on communications with a subject company, public appearances and trading securities held by a research analyst account. Conflicts disclosures The analyst(s) responsible for preparing this research report received compensation that is based upon various factors, including total revenues of the member companies of RBC Capital Markets and its affiliates, a portion of which are or have been generated by investment banking activities of the member companies of RBC Capital Markets and its affiliates. Distribution of ratings For the purpose of ratings distributions, regulatory rules require member firms to assign ratings to one of three rating categories - Buy, Hold/Neutral, or Sell - regardless of a firm's own rating categories. Although RBC Capital Markets' ratings of Top Pick/Outperform, Sector Perform and Underperform most closely correspond to Buy, Hold/Neutral and Sell, respectively, the meanings are not the same because our ratings are determined on a relative basis (as described above). Distribution of Ratings RBC Capital Markets, Equity Research As of 31-Mar-2016 Investment Banking Serv./Past 12 Mos. Rating Count Percent Count Percent BUY [Top Pick & Outperform] 887 51.78 258 29.09 HOLD [Sector Perform] 722 42.15 115 15.93 SELL [Underperform] 104 6.07 8 7.69 Conflicts policy RBC Capital Markets Policy for Managing Conflicts of Interest in Relation to Investment Research is available from us on request. To access our current policy, clients should refer to https://www.rbccm.com/global/file-414164.pdf or send a request to RBC Capital Markets Research Publishing, P.O. Box 50, 200 Bay Street, Royal Bank Plaza, 29th Floor, South Tower, Toronto, Ontario M5J 2W7. We reserve the right to amend or supplement this policy at any time. Dissemination of research and short-term trade ideas RBC Capital Markets endeavors to make all reasonable efforts to provide research simultaneously to all eligible clients, having regard to local time zones in overseas jurisdictions. RBC Capital Markets' equity research is posted to our proprietary website to ensure eligible clients receive coverage initiations and changes in ratings, targets and opinions in a timely manner. Additional distribution may be done by the sales personnel via email, fax, or other electronic means, or regular mail. Clients may also receive our research via third party vendors. RBC Capital Markets also provides eligible clients with access to SPARC on the Firm s proprietary INSIGHT website, via email and via third-party vendors. SPARC contains market color and commentary regarding subject companies on which the Firm currently provides equity research coverage. Research Analysts may, from time to time, include short-term trade ideas in research reports and / or in SPARC. A short-term trade idea offers a short-term view on how a security may trade, based on market and trading events, and the resulting trading opportunity that may be available. A short-term trade idea may differ from the price targets and recommendations in our published research reports reflecting the research analyst's views of the longer-term (one year) prospects of the subject company, as a result of the differing time horizons, methodologies and/or other factors. Thus, it is possible that a subject company's common equity that is considered a long-term 'Sector Perform' or even an 'Underperform' might present a short-term buying opportunity as a result of temporary selling pressure in the market; conversely, a subject company's common equity rated a long-term 'Outperform' could be considered susceptible to a short-term downward price correction. Short-term trade ideas are not ratings, nor are they part of any ratings system, and the firm generally does not intend, nor undertakes any obligation, to maintain or update short-term May 24, 2016 8
trade ideas. Short-term trade ideas may not be suitable for all investors and have not been tailored to individual investor circumstances and objectives, and investors should make their own independent decisions regarding any securities or strategies discussed herein. Please contact your investment advisor or institutional salesperson for more information regarding RBC Capital Markets' research. Analyst certification All of the views expressed in this report accurately reflect the personal views of the responsible analyst(s) about any and all of the subject securities or issuers. No part of the compensation of the responsible analyst(s) named herein is, or will be, directly or indirectly, related to the specific recommendations or views expressed by the responsible analyst(s) in this report. The Global Industry Classification Standard ( GICS ) was developed by and is the exclusive property and a service mark of MSCI Inc. ( MSCI ) and Standard & Poor s Financial Services LLC ( S&P ) and is licensed for use by RBC. Neither MSCI, S&P, nor any other party involved in making or compiling the GICS or any GICS classifications makes any express or implied warranties or representations with respect to such standard or classification (or the results to be obtained by the use thereof), and all such parties hereby expressly disclaim all warranties of originality, accuracy, completeness, merchantability and fitness for a particular purpose with respect to any of such standard or classification. Without limiting any of the foregoing, in no event shall MSCI, S&P, any of their affiliates or any third party involved in making or compiling the GICS or any GICS classifications have any liability for any direct, indirect, special, punitive, consequential or any other damages (including lost profits) even if notified of the possibility of such damages. May 24, 2016 9
Disclaimer RBC Capital Markets is the business name used by certain branches and subsidiaries of the Royal Bank of Canada, including RBC Dominion Securities Inc., RBC Capital Markets, LLC, RBC Europe Limited, RBC Capital Markets (Hong Kong) Limited, Royal Bank of Canada, Hong Kong Branch and Royal Bank of Canada, Sydney Branch. The information contained in this report has been compiled by RBC Capital Markets from sources believed to be reliable, but no representation or warranty, express or implied, is made by Royal Bank of Canada, RBC Capital Markets, its affiliates or any other person as to its accuracy, completeness or correctness. All opinions and estimates contained in this report constitute RBC Capital Markets' judgement as of the date of this report, are subject to change without notice and are provided in good faith but without legal responsibility. Nothing in this report constitutes legal, accounting or tax advice or individually tailored investment advice. This material is prepared for general circulation to clients and has been prepared without regard to the individual financial circumstances and objectives of persons who receive it. The investments or services contained in this report may not be suitable for you and it is recommended that you consult an independent investment advisor if you are in doubt about the suitability of such investments or services. This report is not an offer to sell or a solicitation of an offer to buy any securities. Past performance is not a guide to future performance, future returns are not guaranteed, and a loss of original capital may occur. RBC Capital Markets research analyst compensation is based in part on the overall profitability of RBC Capital Markets, which includes profits attributable to investment banking revenues. Every province in Canada, state in the U.S., and most countries throughout the world have their own laws regulating the types of securities and other investment products which may be offered to their residents, as well as the process for doing so. As a result, the securities discussed in this report may not be eligible for sale in some jurisdictions. RBC Capital Markets may be restricted from publishing research reports, from time to time, due to regulatory restrictions and/ or internal compliance policies. If this is the case, the latest published research reports available to clients may not reflect recent material changes in the applicable industry and/or applicable subject companies. RBC Capital Markets research reports are current only as of the date set forth on the research reports. This report is not, and under no circumstances should be construed as, a solicitation to act as securities broker or dealer in any jurisdiction by any person or company that is not legally permitted to carry on the business of a securities broker or dealer in that jurisdiction. To the full extent permitted by law neither RBC Capital Markets nor any of its affiliates, nor any other person, accepts any liability whatsoever for any direct or consequential loss arising from any use of this report or the information contained herein. No matter contained in this document may be reproduced or copied by any means without the prior consent of RBC Capital Markets. Additional information is available on request. To U.S. Residents: This publication has been approved by RBC Capital Markets, LLC (member FINRA, NYSE, SIPC), which is a U.S. registered broker-dealer and which accepts responsibility for this report and its dissemination in the United States. Any U.S. recipient of this report that is not a registered broker-dealer or a bank acting in a broker or dealer capacity and that wishes further information regarding, or to effect any transaction in, any of the securities discussed in this report, should contact and place orders with RBC Capital Markets, LLC. To Canadian Residents: This publication has been approved by RBC Dominion Securities Inc. (member IIROC). Any Canadian recipient of this report that is not a Designated Institution in Ontario, an Accredited Investor in British Columbia or Alberta or a Sophisticated Purchaser in Quebec (or similar permitted purchaser in any other province) and that wishes further information regarding, or to effect any transaction in, any of the securities discussed in this report should contact and place orders with RBC Dominion Securities Inc., which, without in any way limiting the foregoing, accepts responsibility for this report and its dissemination in Canada. To U.K. Residents: This publication has been approved by RBC Europe Limited ('RBCEL') which is authorized by the Prudential Regulation Authority and regulated by the Financial Conduct Authority ('FCA') and the Prudential Regulation Authority, in connection with its distribution in the United Kingdom. This material is not for general distribution in the United Kingdom to retail clients, as defined under the rules of the FCA. However, targeted distribution may be made to selected retail clients of RBC and its affiliates. RBCEL accepts responsibility for this report and its dissemination in the United Kingdom. To Persons Receiving This Advice in Australia: This material has been distributed in Australia by Royal Bank of Canada - Sydney Branch (ABN 86 076 940 880, AFSL No. 246521). This material has been prepared for general circulation and does not take into account the objectives, financial situation or needs of any recipient. Accordingly, any recipient should, before acting on this material, consider the appropriateness of this material having regard to their objectives, financial situation and needs. If this material relates to the acquisition or possible acquisition of a particular financial product, a recipient in Australia should obtain any relevant disclosure document prepared in respect of that product and consider that document before making any decision about whether to acquire the product. This research report is not for retail investors as defined in section 761G of the Corporations Act. To Hong Kong Residents: This publication is distributed in Hong Kong by RBC Capital Markets (Hong Kong) Limited and Royal Bank of Canada, Hong Kong Branch (both entities which are regulated by the Hong Kong Monetary Authority ( HKMA ) and the Securities and Futures Commission ('SFC')). Financial Services provided to Australia: Financial services may be provided in Australia in accordance with applicable law. Financial services provided by the Royal Bank of Canada, Hong Kong Branch are provided pursuant to the Royal Bank of Canada's Australian Financial Services Licence ('AFSL') (No. 246521). RBC Capital Markets (Hong Kong) Limited is exempt from the requirement to hold an AFSL under the Corporations Act 2001 in respect of the provision of such financial services. RBC Capital Markets (Hong Kong) Limited is regulated by the HKMA and the SFC under the laws of Hong Kong, which differ from Australian laws. To Japanese Residents: Unless otherwise exempted by Japanese law, this publication is distributed in Japan by or through RBC Capital Markets (Japan) Ltd. which is a Financial Instruments Firm registered with the Kanto Local Financial Bureau (Registered number 203) and a member of the Japan Securities Dealers Association ( JSDA ). Registered trademark of Royal Bank of Canada. RBC Capital Markets is a trademark of Royal Bank of Canada. Used under license. Copyright RBC Capital Markets, LLC 2016 - Member SIPC Copyright RBC Dominion Securities Inc. 2016 - Member Canadian Investor Protection Fund Copyright RBC Europe Limited 2016 Copyright Royal Bank of Canada 2016 Copyright RBC Capital Markets (Hong Kong) Limited 2016 All rights reserved May 24, 2016 10