Asset Allocation with a Twist
As regular readers of this blog know, I’m not a big fan of owning bonds for savings that won’t be needed for at least three years. It may not be for everyone, but I invest 100% of my long-term savings in stocks (with no leverage). For money I’ll need in less than three years, like University tuition for one of my sons, I buy bonds that expire close to the date I’ll need the money.
This means that I don’t get the advantage of rebalancing a portfolio between stocks and bonds. By trading to maintain constant percentages in stocks and bonds, investors are forced to buy low and sell high (as long as they actually do the rebalancing). Too often investors delay rebalancing when an asset class is priced low and the news is full of doomsday stories.
My latest idea is to change my portfolio to include fixed allocations to different stock asset classes. I’m sure that other people have thought of this before, but it’s the first time I have seriously considered committing my money to it.
My first thought is to include low-cost index ETFs for Canadian and U.S. large and small cap stocks as well as possibly a few company stocks. I would define a percentage allocation to each stock class and rebalance whenever the percentages went outside of preset bounds.
All simulations I’ve ever done indicate that rebalancing gives a portfolio a small boost, but not enough to overcome the penalty of including an asset class with low expected returns, like bonds. Thus, I want to stick to stocks. I’m interested in reader feedback on this idea along with suggestions for which ETFs (and possibly individual stocks) make sense to use.
This means that I don’t get the advantage of rebalancing a portfolio between stocks and bonds. By trading to maintain constant percentages in stocks and bonds, investors are forced to buy low and sell high (as long as they actually do the rebalancing). Too often investors delay rebalancing when an asset class is priced low and the news is full of doomsday stories.
My latest idea is to change my portfolio to include fixed allocations to different stock asset classes. I’m sure that other people have thought of this before, but it’s the first time I have seriously considered committing my money to it.
My first thought is to include low-cost index ETFs for Canadian and U.S. large and small cap stocks as well as possibly a few company stocks. I would define a percentage allocation to each stock class and rebalance whenever the percentages went outside of preset bounds.
All simulations I’ve ever done indicate that rebalancing gives a portfolio a small boost, but not enough to overcome the penalty of including an asset class with low expected returns, like bonds. Thus, I want to stick to stocks. I’m interested in reader feedback on this idea along with suggestions for which ETFs (and possibly individual stocks) make sense to use.
For my retirement, I'm entirely invested in equities as well. For my son's education, we use an 80%/20% equities/bond ratio. The rationale being that I need at least a little cushion for him (he's 2) though I know it costs me a bit in the longer term. I can choose to delay my retirement if things don't go well, he can't delay his education really. It also helps me sleep at night which is important to me.
ReplyDeleteNovice: You still have lots of time before you need the money for your 2-year old son's education, but eventually you'll have to decide when to shift entirely into guaranteed investments. This can be surprisingly difficult. I put everything in a bond about 3 years before needing the money. You may choose a particular time now, but it will be hard to pull the trigger when the time comes if stocks happen to be down.
ReplyDeleteHi Michael,
ReplyDeleteI've been using this approach - 100% stock across asset classes - with no regrets over the past few years.
There's lots of examples of sample portfolios - check out http://www.canadiancapitalist.com/introducing-the-sleepy-portfolio/ for one, just ignoring the bond content... :-) Also, the Couch Potato Portfolio is another example.
The 'granddaddy' of them all is the CoffeHouse Portfolio - http://www.coffeehouseinvestor.com/. Highly recommended.
Here in Canada I used TD eFunds when the amount invested was little and then moved on to ETFs (Barclays iShares and some Claymore ETFs) when the lower MER justified the commission charged.
My current allocation is 25/25/25/25 between Canada/US/EAFE/EEM. I apply the same allocation to my son's RESP but since I follow the same approach as you to short-term money I'll convert it to bonds shortly after he enters high-school.
Hope this helps.
Fernando: Thanks. I'm also interested in the next level of analysis. Do you have a fixed rule for when you rebalance and how you rebalance? Have you done any analysis of how your rebalancing has improved returns compared to buy-and-hold?
ReplyDeleteCC: "Slicing and dicing" looks like a good idea to take advantage of volatility by rebalancing, but it usually costs more in MERs and tracking error. So, I'm not inclined to slice up the asset classes too much. I first looked at iShares and was surprised that there is no ETF covering a representative sample of all Canadian stocks. So, there is no choice (within iShares) other than to separate large and small cap stocks.
ReplyDeleteThe comment above is a reply to Canadian Capitalist's comment:
DeleteI don't slice and dice personally. I simply use broad-market indices to capture stock asset classes. Many investors do split their allocation and overweight value and small-cap stocks. William Bernstein has sample portfolios in his book. I believe DFA constructs slice-and-dice portfolios -- their website might have more information.
RE: MJ - thanks for the reminder of something else for me to concerned about when he's in his teens ;)
ReplyDeleteI think what I'll do is start the active rebalancing on his 13th birthday and bump bonds up by 25% until he's all in by 16. If I tie it to his birthdays and not to the market, I'm hoping that I can avoid getting greedy at the table / scared off.
11 years is a long ways away, so we'll see what happens. I use Td efunds as well, they might not even be around in a decade.
Thicken: I'm not sure I understand the question. I'm 100% in stocks and want to keep it that way. If I switched to a multi-ETF system, I might choose to give them all an equal percentage.
ReplyDeleteI never understood the logic of rebalancing based on time. It makes more sense to me to rebalance when the allocation is off by some percentage.
The comment above is a reply to Thicken My Wallet's comment:
DeleteWhat % are you thinking about?
I rebalance once a year but I am in mixed equity and fixed income in my RSP. Outside my RSP, its all equity by way of a dividend fund which causes a sector specific rebalancing issue since all my stocks are enrolled in DRIPS which, over time, drags the portfolio into higher yield industries (hmmm... just came up with a blog topic...)
The data (here: http://www.assetcorrelation.com./ ) suggest to me that holding some Utilities and Healthcare stocks/funds might be useful. In addition to international diversification.
ReplyDeleteAlso, perhaps splitting into currency hedged and non-hedged would add enough diversification to justify the costs? For instance in TD e-series tdb904 tdb902 you'd pay 0.15 for hedging.
You could do some mix and match of Canada, US and EAFE as your core with possibly some EEM and small, depending on your risk tolerance. William Bernstein seemed to have written a bunch on all this at www.efficientfrontier.com or his "Four pillars of Investing" book. I recall that his analysis indicated that you "might" be able to get an additional 1 percent by rebalancing every year or two on some reasonable stock allocation versus just letting the stock allocations drift by themselves.
ReplyDeletePerhaps people shouldn't go out of their way to too do much of this stock rebalancing, given its minimal value and given the trading costs. Rather, perhaps people should just rebalance when they have to make a trade anyway, like making an RRSP contribution.
Thicken: I haven't made any final decisions, but my first thought was 50% Canadian, 50% US, and each one divided into three equal-sized allocations. So, not too much slicing and dicing.
ReplyDeleteThe comment above is a reply to Thicken My Wallet's comment:
DeleteSorry, I meant what are you looking at in terms of dividing equity allocations (to pick up on CC's comment on slicing and dicing)?