There are quite a few ETFs coming out these days marketed as “low volatility” securities. IShares has quite a host of these low-vol ETFs representing various international markets. I will disclose that I hold the iShares US minimum-vol ETF (XMU-T) in the ValueTrend Managed Equity Platform. BMO has a Canadian product that I also own a significant chunk of in my ValueTrend Managed Equity Platform. The BMO low-volatility ETF (ZLB-T) holds the 40 lowest beta stocks of the TSX large caps.
Regular readers of this blog know that I am looking for a top on the markets to occur in the next few months. Valuations, cyclical patterns and significant overhead technical resistance have influenced me to plan around this potential. It is for this reason that I have slowly been replacing higher volatility equities with lower beta securities. These ETFs and certain low-beta individual stocks are part of that plan.
For those who are unfamiliar with the “best six months” seasonal strategy, you buy the index in November, and then sell in May to hold cash for the following six months. An interesting premise was explored by BMO surrounding their low-volatility ETF mentioned above. What would happen if one were to buy the TSX index (or perhaps buy an ETF of the S&P TSX top-60 index such as Horizons S&P 60 ETF or the infamous iShares TSX S&P I-60’s) for the “best six months” seasonal cycle. Then, during the “worst six months” you buy the BMO low volatility ETF in May and sell in early November. Theoretically, you would be getting all of the beta from the TSX 60 index when you want it during the best six months, and then reducing your risk by owning low beta stocks in the more risky six months. Below is the table for that strategy as reported by BMO capital markets. The data goes back to 1996, so it is not as long a period as the seasonal studies done by Don Vialoux or Brooke Thackray (who have tested back to 1950—see Thackray’s Investor’s Guide)—but it’s a large enough sample to give us a taste of the performance over the recent flat markets, plus the tail end of the last bull market (1996-1999)
One conclusion that can certainly be drawn from the timeframe studied is that the buy/hold low beta ETF approach beat the rotational strategy between the TSX and the low beta ETF using seasonal patterns. I don’t have the precisely matching data for that period of time using the “best six months” seasonal strategy for the TSX using the index to cash, then back again–because BMO’s study started in December 1996 for some reason. But Brooke Thackray did very kindly provide me with the more typical study. His data goes from November 1st, 1996, sells every April 30th until 2012. Over his time period, the TSX buy/hold strategy averaged 6.05%, while the seasonal buy November 1, sell April 30 of each year during the time period averaged 7.11%.
My conclusion from the BMO data is that, given the time period studied (starting just before the onset of the secular sideways market), low beta has won the race for buy and hold. In both Brooke’s study using only the TSX, and in the BMO study using their low-beta ETF, the seasonal strategy beat the TSX buy/hold strategy. Perhaps that outperformance would be amplified if one were to rotate in and out of high-beta during the “best six months”, rather than just the TSX.
Either way, I believe that these studies show us seasonality does work for the TSX, and that low-beta has worked as a stand-alone strategy during the recent secular sideways market.
| Strategy 1 – Hold S&P/TSX Composite 100% | |||||||||
| Strategy 2 – Hold ZLB Low Vol 100% | |||||||||
| Strategy 3 – Hold 50% S&P/TSX, 50% ZLB Low Vol | |||||||||
| Strategy 3 – Hold ZLB Low Vol from May 1 to Oct 31, Hold S&P/TSX from Nov 1 to Apr 30 | |||||||||
S&P/TSX | ZLB | 50/50 | Seasonal Strat. | ||||||
return | 8.2% | 13.3% | 11.0% | 12.8% | |||||
st dev | 16.1% | 11.5% | 12.4% | 13.2% | |||||
sharpe | 0.32 | 0.90 | 0.64 | 0.74 | |||||
Keith Speaking in Alliston, Ontario
As we continue our book promotion tour, I’m speaking at the Notawassaga Inn on Saturday February 16th at 10AM. The topic will be on investing for income, and the technical outlook for the markets. A continental breakfast will be served at this free event. Join us if you can.
8 Comments
Hi Keith,
Thanks for suggesting this seasonal/low beta strategy. Do you think this strategy would outperform or underperform the TSX during the coming secular bull market?
Brooke’s book suggests a similar strategy: switch between consumer discretionary (high beta) & consumer staples (low beta) during the Nov & May periods. The returns seem to be around the 12-15%.
Thanks again. You, Brooke & DV have all helped me vastly improve my investment knowledge & returns.
Davey–I often trade the staples/discretionary cycle myself–and, I do think that focusing on low beta will outperform in the next few years–especially in what looks to be a coming “healthy” bear market in the next 12 months.
I just saw at least one sell signal in the media this week. Saturday’s front page of the Globe and Mail is titled “The end of fear: Why that hoard of cash is streaming into stocks “
Hi Keith:
I’d like to thank you for the many hours you spend giving us tips and guiding us through the jungle. I also follow DV, BT, and yourself in print, and BNN.
My question today is the difference between the ZLB return (13.3%) in your study and the return posted by BMO which is 2.91% with an MER of 0.350%. Does yours include a capital appreciation. What am I missing? Thanks. LB.
Hi Leo–first, thanks for the kind words. I really do put my heart into this blog-and I’m glad you get something out of it. Its my way of giving back to Technical Analysis what other teachers in the field have given to me.
As for the BMO study, I believe that they are back-testing using the formula they utilize for choosing and rebalancing their holdings. The ETF has not been around since 1996 (period of study) so they did a model comparison, had you access to their strategy or had they been offering the ETF at the time.
Hi Keith: Once again thanks for the time you put into this site. It is immensely helpful. I have run comparisons for zlb against every ETF I can think of including the indexes and this one comes out on top in capital gain, and less volatility. The average annual gain in the BMO study is about what the ETF gained in the past year. Two questions: Does the data on the above chart include distributions and secondly when something looks this promising I often find I am missing something. Is there anything else to detract from this concept of buying and holding dividend equities of lower volatility. I thought buy and hold was unfashionable.
Terry thanks for the question. I have forwarded your comment to the BMO person who sent me the study to begin with. From there, I will post the answer that he sends back to me under comments here–so keep your eye on this thread –he’s pretty good at getting back to me, so I may be posting something for you in the next few days.
I spoke to the BMO rep–the study you asked about does include distributions–it includes reinvested dividends for the TSX as well, so all-in, its looking like a fair study. As far as buy/hold, I have found that some vehicles can do well for a long time, but eventually everything falls off. Perhaps that comes from investors who pile into a strategy that is working, and over time that negates the alpha from performance. So, enjoy low-vol while you can.