In my last blog post, we found that overweighting Canadian stocks by about 30% (relative to a global market capitalization equity weight of about 3%) had historically resulted in a less risky portfolio. Obviously, this analysis gave no guarantee of the minimum risk portfolio going forward, but it did provide a useful starting point for discussion.
This week, we received our first email from an investor who wanted to know why our model ETF portfolios didn’t have a higher allocation to Canadian stocks. As Canadian stocks have outperformed global stocks by a wide margin during 2016, we figured that it was only a matter of time before investors started to favour Canadian stocks once again.
This brings me to my next argument for overweighting Canadian stocks:
Reason #2: Behavioural biases may be reduced with a 30% (or more) overweighting to Canadian stocks.
This is one of the most important, but often most underappreciated aspects of the global equity mix decision. Over the past five calendar years, global stocks have outperformed Canadian stocks by an impressive 11.54% per year. Recency bias (the tendency to base our investing decisions on more recent events) has caused investors to question why on earth they would have any portion of their portfolio invested in Canadian stocks (or even more recently, why they don’t have more invested in Canadian stocks).
Global stocks have not always been the investor darlings that they were at the end of 2015. If we look at the fifteen calendar years prior to 2011, we find that Canadian stocks outperformed global stocks by an average of 6.52% per year. During that time, it was not uncommon for Canadian investors to dismiss global stocks entirely and instead opt for a 100% allocation to Canadian stocks.
Annual Return Differences in Canadian Stocks vs. Global Stocks: 1988 to 2015
Between 1996 and 2010 (the same fifteen year period), a $100,000 investment in Canadian stocks would have grown to $453,510. During the same period, a $100,000 investment in global stocks would have grown to only $182,207 (for a difference of $271,303). Most Canadian investors that chose a market capitalization-weighted portfolio (with only about a 3% weighting in Canadian stocks) would have been kicking themselves for their decision. This would have been especially true if they knew how their friends and family had been invested.
Using 2014 year-end data from the International Monetary Fund’s Coordinated Portfolio Investment Survey and the World Federation of Exchanges, Morningstar estimated the average Canadian equity content of a Canadian investor to be around 62% (Vanguard Canada estimated this figure to be around 59% at the end of 2012). If the average Canadian investor had a similar home-bias to these figures between 1996 and 2010, it would have made the cocktail discussions even more intolerable for a market capitalization-weighted investor.
Sure, maybe the returns of Canadian stocks will soon fall short of their global counterparts once again. But what if they continue to outperform over the next five, ten, or even fifteen years?
Fifteen years is a long time to be consistently wrong. Investors need to choose an asset mix that they can stick to through the good times and bad. Personally, I’m comfortable with overweighting Canadian stocks by 30%, but I would never argue with an investor who allocated 50% of their stock allocation to Canadian stocks and 50% to global stocks, for behaviour reasons.
During periods when global stocks have been outperforming Canadian stocks, I receive a growing number of questions asking why my model ETF portfolios are skewed so heavily towards Canadian stocks (as one might imagine, I receive the opposite feedback when Canadian stocks are outperforming global stocks). The argument usually goes something like this: Canadian stocks make up just over 3% of the global equity markets – shouldn’t Canadian investors simply hold a 3% equity allocation to Canadian stocks?
I’ll let you all in on a dirty little secret: no one knows the optimal amount of Canadian stocks that an investor should hold. Not me, not the talking heads on BNN, not anyone. All we can hope to do as mere mortal investors is to make a reasonable guess that we are comfortable with, and stick with it over the long term.
As mentioned, Canadian stocks make up just slightly more than 3% of the global equity markets. But in my model ETF portfolios, I overweight Canadian stocks by about 30% (resulting in a 33% allocation, or 1/3 equity weighting towards Canadian stocks). There are a number of good reasons for doing so, which I’ll cover in a series of blog posts.
Reason #1: The minimum risk portfolio has historically overweighted Canadian stocks by about 30% (relative to a market capitalization-weighted index portfolio).
If we assume that the future expected returns for Canadian stocks are similar to global stocks, rational investors would prefer a blend that results in the least amount of portfolio risk. In the graph below, I’ve calculated the historical volatility for various balanced portfolios between 1988 and 2015 (with a 60% allocation to stocks and a 40% allocation to bonds). Each portfolio along the curve overweights Canadian stocks by varying amounts.
Minimum Risk Portfolio: January 1988 to December 2015
Notes: Canadian equities are represented by the MSCI Canada Index; global equities are represented by the MSCI ACWI Index. Canadian bonds are represented by the FTSE TMX Canada Universe Bond Index.
Sources: MSCI Indices, Dimensional Returns 2.0
To the far left of the graph, you’ll find a balanced portfolio that invests 100% of its equity component in Canadian stocks (you may also notice that it has the highest risk, or volatility, of all the portfolios). Most investors realize that this is an extreme position, and that significant diversification benefits can be achieved by investing some of their equities into global stocks.
As we move further along the curve to the right, we find that the portfolio risk bottoms out at around a 30% overweight to Canadian stocks (a similar mix to my PWL model ETF portfolios). As we continue to the far right of the curve, we reach the market capitalization-weighted index portfolio. Indexing purists will often argue that this is the most optimal portfolio, as it is the most diversified and requires no active decision-making. Contrary to what they believe, this portfolio has historically been more volatile than a portfolio that overweight Canadian stocks by 30%, or even 50%.
In their 2014 research paper that followed a similar methodology as the analysis above, Vanguard Canada was quick to point out that their analysis was backward-looking and particularly dependent on the time period examined. The specific asset allocation (the mix between stocks, bonds and other asset classes) also affected the historical results. Without knowing what the future holds, a reasonable starting point would arguably be between a 20% and 40% overweight to Canadian stocks.
In my next blog post, I’ll look at a number of other reasons that it may make sense for a Canadian investor to overweight Canadian stocks within the equity component of their portfolio.