I’ve resisted discussing my asset allocation for some time, but now is finally the time to describe it and the reasons behind it. I’m not holding it out as a model portfolio; I believe it’s suitable for me and not necessarily anyone else. With that warning, enjoy the discussion and feel free to ask questions or take some pot-shots.
The main reason why I haven’t discussed my asset allocation in detail before now is that I don’t want readers to treat it like a model portfolio and follow it blindly. Everyone’s situation is different. The need for variation in asset allocation from one investor to the next usually isn’t huge, but there is no one-size-fits-all portfolio.
A secondary reason for keeping my asset allocation to myself is that I can’t be sure I’ll never change it. For example, I recently made the minor change of switching from an ETF of small-cap stocks to an ETF of small-cap value stocks. I don’t have any plans to change my current allocation, but I may think differently in the future.
To start with, my portfolio is 100% allocated to stocks. This means that my returns are more volatile than portfolios containing some bonds and real estate. I don’t recommend 100% stocks for everyone, but I believe it works for my situation.
I have a fully paid-for house and have no debts of any kind. I get job offers fairly regularly, so I’m not overly concerned about the continuity of my pay cheque. I also have a cash buffer that bounces around between 3 and 6 months of my family’s spending. This stable base makes it reasonable to handle the volatility of stocks.
On the emotional side, I handled the tech crash in the early 2000s and the financial crisis of 2008-2009 without any panic selling or sleepless nights. So, I’m confident I don’t need the security of a lower-volatility portfolio.
I’m also prepared to adjust my retirement date based on stock market performance. Before retiring, I’ll certainly sell a sizeable block of stocks and hold the proceeds in safe investments like cash or GICs. If it seems like a bad time to be selling stocks, I’ll just keep working until I can sell some stocks. It’s true that many people have a retirement date forced upon them for health or other reasons, but I expect to have enough savings to cover my expenses indefinitely before I reach age 65.
To decide on an allocation among Canadian, U.S., and international stocks, a starting point is the percentages of the world stock market capitalization each represents. According to a Harvard blog, Canada and the U.S. have 4% and 40% of the world’s stock market capitalization, respectively. So, we have
4% Canada, 40% U.S., 56% international
However, I live in Canada, and while it doesn’t make sense to have too much of a home country bias, my spending is correlated with the fate of Canadian companies. So, I chose to bump up my Canadian allocation to 30%. There’s nothing magical about this figure; it just seems about right. Leaving the other two categories in roughly the same proportion gives us
30% Canada, 30% U.S., 40% international
I spend a good chunk of my time in the U.S., so for the same reason I increased my allocation to Canada, I increased my allocation to the U.S. to 45%. Again, there isn’t much to justify this exact figure; it just seems about right. This gives
30% Canada, 45% U.S., 25% international
There are a number of ETFs that cover my 3 allocation categories. I use low-cost index ETFs for each category. For Canadian stocks I chose Vanguard Canada’s VCN. There are several other possibilities, but I like that VCN includes some smaller-capitalization stocks and more stocks than many other ETFs. I also like Vanguard.
For the U.S. and international allocations, I decided to go with ETFs from the U.S. rather than Canada. This has the advantage that the MERs are lower than they are with Canadian ETFs, but a disadvantage is that I have to do currency exchanges between Canadian and U.S. dollars. Fortunately, Norbert’s Gambit is a clever way to reduce currency exchange costs.
For international stocks, I went with Vanguard’s VXUS. This ETF covers the entire world other than the U.S. It does include a little bit from Canada. So, my actual allocation to Canada is more like 31% than 30%.
I chose to use my U.S. allocation to get some added exposure to small-cap stocks and value stocks. So, my U.S. allocation is split between VTI and VBR, both from Vanguard. VTI holds an index of all U.S. stocks. VBR holds all U.S. small-cap value stocks.
Here is the final allocation to index ETFs:
There is nothing magical about these exact percentages. The key thing is to stick to them. It’s far too easy to see one ETF shoot up in price and suddenly decide to increase your allocation there. It’s better to mechanically rebalance back to your chosen percentages occasionally.
Having only 4 ETFs may seem overly simple and that there is room for finer slicing and dicing of categories. However, there are always added complications when you try to use an asset allocation in real life. It’s better to start with something simple.
One complication I have is that my wife and I have 9 accounts. Why not consolidate some of them you may ask? I can’t. My wife and I each have one RRSP, one TFSA, one LIRA, and one non-registered account. The ninth account is her spousal RRSP. I treat the 9 accounts as a single portfolio, which simplifies things somewhat, but handling this many accounts adds some complication.
Another complication is that I keep as much of my VTI, VBR, and VXUS in RRSP accounts as possible. This is because the U.S. imposes a dividend withholding tax on stocks held in our TFSAs and non-registered accounts. Following this rule sometimes forces me to make extra trades when I’m rebalancing the portfolio.
I have a group RRSP at work that I joined to get the company matching contributions. The MERs of the funds offered are extremely high, and my lowest cost option among stock funds is a Canadian index fund. So, I treat it as part of the 30% allocation to Canada.
My wife and I still hold some Berkshire Hathaway stock that has significant unrealized capital gains, so I’m not inclined to sell until it’s convenient from a tax perspective. For now, I treat it as part of my U.S. allocation which reduces the amount of VTI and VBR we hold.
There are other complications, but you get the idea; it’s best to start with a simple allocation.
A particularly good feature of my allocation is its low cost. Here are the MERs:
Blended Portfolio MER: 0.08%
This is a very low yearly cost, and you may wonder what would be the big deal if it were 0.25%, or 0.5%, or higher. Because your investments grow for a very long time, it’s best to think of costs over a much longer time than just one year. These costs aren’t just charged on new money; the same money gets hit year after year. So, the costs build up.
I use what I call the MERQ (Management Expense Ratio per Quarter century). Thinking about costs over 25 years gives a better perspective. My portfolio’s blended MERQ works out to 2.0%. This means that every dollar I save that stays in my portfolio for 25 years grows to 2% less than it would have without MER costs. But for an MER of 0.5%, the MERQ is a much more painful 12%.
MERs aren’t the only portfolio costs. There are trading commissions, bid-ask spreads, trading costs within the fund, currency exchange spreads, and dividend withholding taxes to consider. Be sure to account for all costs when choosing among portfolio options. In my case, my portfolio is large enough that the MERs and international withholding taxes within VXUS dominate the other costs.
So there you have it. I have a fairly simple, low-cost portfolio of index ETFs. It suits me, but your mileage may vary.