In our last blog/video, we introduced the all-equity ETFs from iShares and Vanguard. These ETFs make it easy to gain and maintain exposure to global stock markets with the click of a mouse, eliminating the hassle of juggling several ETFs in your all-equity portfolio.
Vanguard and iShares don’t offer their services for free though.
The MERs for their all-equity ETFs are slightly higher than the weighted-average MERs of their underlying holdings. Consider this modest surcharge as the price of admission for their professional asset allocation and rebalancing services. In my opinion, that’s a bargain for most investors.
Then again, there are those who might prefer to squeeze every last penny out of their portfolio costs. If that’s you, you may want to try skipping the value-add of an all-equity ETF, and simply purchase the underlying ETFs directly, in similar weights. If you take on the task of rebalancing back to your targets each month when you add new money to your portfolio, you should be able to mimic an all-equity ETF for a lower overall MER.
That’s the goal anyway. But it’s still going to take time, money, or both to keep your asset allocations on track each month. Let’s look at three potential strikes against trying to reinvent an all-equity ETF on your own, as well as one potential play that may serve as a suitable compromise.
Strike One: The potential cost savings are minimal.
For example, let’s say you’ve got $10,000 to invest. Instead of investing it in the Vanguard All-Equity ETF Portfolio, or VEQT, you could divide it up among VEQT’s component funds. The estimated cost savings might let you rent an extra movie each year, but are the savings really worth it? The extra time you’ll need to spend on rebalancing may not leave you much time to even enjoy your movie.
For larger amounts, the fee savings start adding up, but only if you can buy and sell ETF shares at zero commission as you rebalance. If not, you can forget about it.
Strike Two: Managing a portfolio of four ETFs (instead of just one) will be more difficult.
Sticking with our VEQT example, a DIY investor would either need to visit Vanguard’s website monthly to collect the individual ETF weights within VEQT, or use the market cap data from the FTSE and CRSP index fact sheets to determine how to allocate each of the underlying ETFs. They would then need to calculate how many ETF units to buy or sell across various accounts to get their portfolio back on target, and place multiple trades to get the job done.
If you were to perfectly execute this strategy each month, you may be able to eke out slightly higher returns than VEQT. If we compare VEQT’s past returns to the aggregate performance of its four underlying ETFs since February 2019, we see our component strategy did outperform VEQT during this time – which is good news.
But again, this assumed you made no mistakes along the way, rebalancing your portfolio like clockwork each month. With a busy schedule and constant distractions, could you honestly say you’d never miss a beat?
Strike Three: You’ll be working with stale data.
Let’s discuss whether we mere mortals really could flawlessly replicate VEQT’s returns by buying and rebalancing its underlying funds. While you can probably come close, perfection seems out of reach.
Consider, for example, Vanguard’s resources versus your own. Vanguard has constant daily inflows and outflows from thousands of investors, and they can use these flows to continuously fine-tune their target asset mix. They also have up-to-the-minute market cap index information to improve their aim.
You’ve got you, rebalancing monthly, based on more limited index information. It usually takes a couple weeks before Vanguard updates their website with VEQT’s asset class weightings from the previous month-end. The FTSE index data is also only publicly available several days after month-end, and the CRSP index data is only available several days after the end of each quarter. This means DIY investors will never be working with the most current ETF target weightings when rebalancing their portfolios each month.
Again, you can still probably come close. But there will be random differences, which will lead to positive or negative tracking error. In other words, if you’re unlucky, through no fault of your own, negative tracking error could potentially cost you more than the reduced MER may save you.
Bottom line, even if you perfectly execute your own buy-and-rebalance-monthly strategy, there’s still a risk you might underperform an all-equity ETF anyway, even after fee savings.
So does this mean we’ve struck out on building a cheaper all-equity ETF portfolio? Actually, there is one more play to consider.
You could ditch your three foreign equity ETFs for a single fund. Just like VEQT, the Vanguard FTSE Global All Cap ex Canada Index ETF (VXC) also weights its underlying U.S., international, and emerging markets equity regions by market capitalization. This makes it a decent substitute for the VUN, VIU, VEE combination.
The main difference is that VXC’s U.S. equity allocation excludes micro-cap stocks, while VUN includes them. However, this small difference is not expected to have a material impact on VXC’s returns.
To use this approach, you would invest 30% in VCN and 70% in VXC, rebalancing the portfolio back to these target weights each month. The weighted-average MER of a VCN/VXC combo is also similar to the fees of our four-fund approach.
A back-test of this strategy also yielded similar results, but with tighter tracking and much less effort.
All things considered, I still prefer the simplicity of an all-equity ETF. But if you just can’t stand the thought of potentially leaving some cash on the table, a two-fund approach might be an appropriate compromise for you and your portfolio.
MER difference is only one component to performance. Timing is another. Beside unavoidable tracking error there is also frequency. Your calculations are based on unspoken assumption that individual and Vanguard would rebalance with the same frequency and, therefore, the same accuracy. This is a pretty strong assumption.
Is there a publicly available research comparing performance of investments rebalanced daily vs monthly vs quarterly?
@Anatoli – My calculations for the back-testing assumed monthly rebalancing. I’m not sure if you read this article / watched the video, but one of my main arguments for sticking with an all-equity ETF was that an individual investor does not have the index data available to rebalance on a daily basis (so this is a strike against trying to outsmart an all-equity ETF).
H Justin, thanks a lot for this. In the owning multiple ETF’s instead of a single asset allocation ETF option, what do you think about owning the US market ETF component directly (VTI or ITOT). This would lower the combined weighted MER further, do you think this is worth the trouble, even if Norberts Gambit is used for currency conversion?
@Simon – You’re very welcome. In my opinion, this would only be worthwhile for larger RRSP portfolios (where the investor would also benefit from the reduced foreign withholding tax drag on dividends received by VTI or ITOT in this account type).
What are the optimal component weights for an all equity portfolio? There is a discrepancy between the component weights of VEQT and XEQT, and lately my XEQT has been outperforming my VEQT (although that could reverse at some point). The Canadian economy, being at most 3% of the world’s economy, suggests a much lower weighting of VCN or XIC would be appropriate.
By the way, I really appreciate your work. When I don’t have time to immediately read an email, I make sure I flag it for reading later. None of your posts are ignored. Thanks.
@Herman – An investor would first need to define what “optimal” means to them (i.e., lowest risk?, highest after-tax returns?, highest risk-adjusted returns?, highest risk-adjusted after-tax returns?, etc.). Even after doing so, many of these outcomes can’t be determined in advance.
Historically, the lowest risk asset mix has differed, depending on the specific time period. Since 1970, it’s been around 24% Canadian stocks, 76% foreign stocks. Over the past 20 years or so, it’s been around 1/3 Canadian stocks. Both VEQT and XEQT have similar allocations to these “optimal” weights, so they seem reasonable to me (I wouldn’t argue with an investor if their allocation to Canadian stocks was anywhere between around 10% to 40%).
A 3% Canadian equity weighting is likely not optimal for most Canadian investors though. It’s never been the lowest risk portfolio over rolling 20-year time periods since 1970, and it is less tax efficient than portfolios with a higher allocation to Canadian stocks (for most investors).
My goal would be to maximize risk-adjusted after-tax returns. But as you indicate, it’s not possible to determine what portfolio compositions would achieve such a goal, in advance. I suppose one could always augment XEQT or VEQT with more of one of their component ETFs to tilt the portfolio towards the US, ex-NA developed, or emerging markets. Thank you for your feedback.
The difference between VCN+VXC vs. VCN+VUN+VIU+VEE is a lot smaller than the difference between the iShares equivalents. I wonder if Vanguard Canada will ever drop the MER on VUN to compete better with XUU, and perhaps optimize VEE’s foreign withholding tax drag to directly compete with XEC to boot.
@Bill – I expect Vanguard will eventually lower VUN’s MER and update the structure of VEE to compete with BlackRock.
Thanks for the video and blog.
I am happy with my XAW+VCN portfolio with a reduced MER.
I just need to add money to the ETF that is underperforming (would be VCN at the moment), doesn’t take much time.
And I usually only invest when I have about $5k in cash to reduce trading costs.
I m glad you mentioned this option as well!
@Jacky – It sounds like you have a very disciplined process in place – keep up the great work! :)