In the latest episode of the Canadian Couch Potato podcast, I speak with Robert J. Shiller, professor of economics and finance at Yale University, and winner of the 2013 Nobel Prize in Economics.
Prof. Shiller was in Toronto recently for the launch of the BMO Shiller Select US Index ETF (ZEUS), a fund that selects US stocks using a methodology based on Shiller’s CAPE ratio. The acronym stands for cyclically adjusted price-to-earnings: unlike traditional P/E ratios, which usually use only the previous 12 months of earnings, CAPE uses the average of the previous 10 years. The idea is to smooth out any short-term aberrations and provide a more useful measure of a company’s value.
The CAPE ratio is widely followed, not just for individual companies, but for the US market as a whole. During the interview I mention that a 2012 study by Vanguard found the CAPE ratio was the one of the most useful tools for estimating long-term stock returns—although even then, it explained less than half of subsequent performance.
Prof. Shiller has also lent his name to the Case-Shiller Home Price Indexes, which track real estate prices in major US cities. During our interview we chat about whether home ownership has historically been a good investment, both in US and here in Canada.
More promises of reward without risk
This episode’s “Bad Investment Advice” segment focuses on a new book called Slash Your Retirement Risk, by Chris Cook, founder of an advisory firm in Dayton, Ohio. The book begins by making a common claim: “Traditional buy-and-hold investment strategies that emphasize return on your investments don’t work anymore.” It goes on to suggest an alternative designed to “maximize the equity portion of your portfolio to capitalize on market upsides, while protecting against dramatic losses on the downside.”
Here’s the basic strategy: you start with a portfolio of 11 ETFs covering all of the economic sectors—healthcare, utilities, real estate, financials, and so on. You stay invested until the S&P 500 declines by 10%: as soon as it hits that threshold, you sell all your equities and move to bonds. Then you wait for the S&P 500 to recover by 15%, at which point you buy back the equities.
The idea is to limit your losses during a major drawdown—like the ones that followed the bursting of the tech bubble in 2000 and the financial crisis of 2008–09—without missing the whole recovery.
We’d all love to capture the returns of the market without having to endure all of that pesky risk, and there’s no shortage of investment gurus who make that promise. They’re inevitably well-armed with backtests showing you how well you would have done if you’d just been able to anticipate the huge drawdowns of the early 2000s. Funny how the strategies are only revealed after the fact.
On the path to becoming an enlightened investor, everyone needs to accept that there is no reliable way to capture equity returns without also accepting the risk of gut-wrenching short-term losses. The only way to reduce those losses is to diversity with high-quality bonds, GICs or cash, all of which will lower your expected return. That’s the trade-off.
Loonie tunes
In the “Ask the Spud” segment, I answer a question from Abe, who wants to understand the pros and cons of currency-hedged ETFs. These are designed to reduce or eliminate the impact of currency exchange rates for investors holding foreign equities. When the loonie soars—as it did between May and September—returns on US an international equities suffer, and currency hedging starts to look attractive.
I’ve written a lot on this topic (most recently here), so regular readers will know I advise against it. First, it’s not very precise. For example, over the three years ending September 30, the iShares S&P 500 Index ETF (IVV) returned 10.76% annually in U.S. dollars. One should expect a currency-hedged Canadian ETF tracking the same index would deliver a very similar return. But both the iShares and Vanguard versions delivered 9.95% annually over that period, for a difference of 81 basis points a year. Roughly half that difference is the result of higher fees and foreign withholding taxes, but even a tracking error of 30 or 40 basis points will erode returns over time.
The second reason to avoid currency hedging is that it can actually increase volatility. That’s why it’s misleading to say hedging “eliminates currency risk,” because this implies that non-hedged ETFs are more risky. In fact, there’s evidence that being exposed to multiple currencies—and especially the U.S. dollar—can actually lower the volatility of an equity portfolio for Canadian investors.
I haven’t listened to this episode yet – looking forward to in on my drive home now :). As a longtime reader, I can guess what your stance will be on currency hedging and the CPP investment board agrees with you. If any fund has a long term investment approach with Canadian dollars it would be CPP.
Funny coincidence that Reuters just posted about CPP and currency hedging today.
https://ca.reuters.com/article/businessNews/idCAKBN1DA1R9-OCABS
“The fund does not hedge against currency movements, saying that while they may impact its results in the short-term, it does not expect them to have a significant impact on its long-term performance.”
Looking forward to another great podcast when I am on my way home, you always have insightful guests. Thank you Dan!
Hi Dan,
While I’m willing to accept that being unhedged is best in the long term, I wonder if it’s a good strategy when you’re in the drawdown phase.
I’ll be retiring next July. I have a modest Couch Potato nest egg. 70% bonds and GICs and 30% ETF (the ones you recommand).
I’ll have to withdraw about 5% of my nest egg every year for the first five years to cover my expenses (then a small defined benfit pension plan will kick in). The plan is to sell bonds/GICs and ETF each year, keeping the allocation 70/30.
What if the canadian dollar goes up 20% in those 5 years ? Then the value of the equity part of my PF will sink accordingly. That would leave me with a smaller nest egg to draw from. Should I switch to hedged ETFs for these 5 years ? Or even for the whole retirement period ?
Thanks for your thoughs on that topic.
Thanks for the podcast. The stat on the annual return of .5% for real estate in the US in the last 100 years was thought provoking especially given that I just moved to Abbotsford B.C. where the market thinks paying $700,000 for a two bedroom townhouse makes sense. No thanks.
Keep up the good work!
Another great podcast Dan!
I really liked listening to your piece on Hedging! Excellent!
keep up the great work.
Another informative podcast.
It is interesting that I have read several blogs recently about currency hedging, and I have never really considered it as much of an issue. I have always agreed with you in that it just doesn’t seem to make a huge difference in the long term, and overly complicates an already complex topic (investing).
As well, I have always followed the diversification= decreased volatility way of thinking. As you put it above, “In fact, there’s evidence that being exposed to multiple currencies—and especially the U.S. dollar—can actually lower the volatility of an equity portfolio for Canadian investors.” Exactly.
Great work Dan! Did you discuss the impact foreign investment may be having on real estate prices? Is it something to consider or just a part of the story being propagated to explain the ludicrous process we’re seeing? Thanks!
Hi Dan, can you clarify this statement please: “but even a tracking error of 30 or 40 basis points will erode returns over time.”
By definition couldn’t the tracking error result in an increase in returns over time? Over the long term wouldn’t this even out in theory? As you stated if fees or taxes are higher that will reduce the returns, but how over the long term will a tracking error always be bad?
Thanks.
Bob
@252 sleeps: This is the idea I was getting at in the podcast when I said it’s wrong to think of hedging as eliminating a risk. It simply changes your risk exposure. While it is true that a rising Canadian dollar would hurt your foreign equity returns in retirement, it is also true that a falling Canadian dollar would give them a boost. Unless you feel that one of these outcomes is more likely (and there is simply no reason to have any confidence in such a forecast) you are not reducing your risk by using currency-hedged funds.
In any case, with only 30% of your portfolio in equities, the impact of foreign exchange rates is not likely to have a significant effect. Good luck in your retirement!
@Jon: Thanks for the comment, and glad you liked the podcast. Prof. Shiller and I did not discuss the specific reasons for the lofty real estate prices in some Canadian cities. I think it’s fair to say that foreign investment is only part of the story, and probably not a particularly large part. My understanding is that foreign buyers represent only a fairly small percentage of home sales in Canada’s major cities. I think low interest rates and a widespread comfort level with debt are bigger factors.
@CJBob: I was assuming here that the tracking error would always be negative—or at least the long-term tendency would be negative—because that is what their performance shows. I am not aware of any meaningful period that showed a significant positive tracking error with a currency hedged ETF (i.e. the hedged ETF outperformed its benchmark index).
I hope it is clear that I am not suggesting that unhedged ETFs will outperform hedged ETFs over most periods: this is not true at all. It’s just that unhedged ETFs tend to track their index more closely, so all other things being equal, they are a better choice.
Hi Dan, excellent podcast, as usual.
I’m curious then, would one investment strategy benefit over another with going hedge vs. unhedged. For example, would placing hedged ETFs in non-registered accounts basically win out against unhedged in the long run just due to tax implications? I’m just trying to grasp their effective function with offering both options and what their ideal purpose is.
Thanks.
@IndexFan: I don’t recommend using hedged ETFs at all, but if you must, they are likely to be better in a tax-sheltered account, as the currency hedging strategy can result in capital gains that get distributed to investors at the end of the year.
Dan, what do think about the absence of any low-cost, indexed mutual funds in Canada? For example, in the US you can buy the balanced 60/40 Vanguard Admiral fund for all of 7bps. As we know, the decision to buy mutual funds or ETFs is the packaging, not the product.
Mutual funds offer investors many advantages, such as automatic reinvestment, automatic re-balancing for the balanced fund, and you don’t need a discount brokerage account and are therefore, commission-free. Are any providers considering such funds in Canada? It seems like a huge hole ready to be filled in my opinion.
@Greg: I couldn’t agree more, and this is something I have lamented for many years. As much as I think the Tangerine and TD e-Series funds are good products, they are very expensive compared to their US counterparts. They are also restricted to clients of those two firms.
The reason we can’t access low-cost index mutual funds in Canada is simply because no fund provider wants to go down that road. To make any money on low-cost index funds you need to have enormous scale, and that is incredibly hard to build. Even Vanguard Canada doesn’t want to do it, as they would rather focus on their ETF business. I think it’s a shame, too, but from a business perspective it’s hard to blame them.
@CJBob even if the tracking error had an average of zero in the long run, it would still reduce returns. For example, earning 5% for five years followed by 7% for five years results in less total return than earning 6% for ten years (albeit the difference isn’t much!).
Hi CCP
Can you shed some light on the disconnect us investors hear on BNN for example and the returns we get?
So much talk for the past while on BNN etc of how overvalued the markets are especially the US market compared to historical.
So I went to the Ishares canada website to see how well our investments have done.
Annualized Returns up to Oct 31, 2017
XIC 10 YR 3.77 SINCE INCEPTION FEB 16 2001 6.58
XSP 10 YR 6.05 SINCE INCEPTION MAY 24 2001 3.47
XIN 10 YR 1.63 SINCE INCEPTION SEP 6 2001 3.63
I and most likely other retail investors have read and heard stock markets have historically returned around 10% a year however the returns above are no where near that number, especially considering we are told we get the market return minus a small fee when we invest in ETF’s and with the hearsay of the markets being overvalued it doesn’t make sense…
We also hear how poor the Fixed Income investors have done the past few years due to low intesest rates however the 5 YR annual return for XBB to oct 31 is 2.72% which is better than what XIN has produced per year in the past 10 years.
My question is if the markets are overvalued compared to past why has our returns been so far below the past for the past 10 years and since inception of the funds above back in 2001?
SP November 18, 2017 at 12:03 pm #
@CJBob even if the tracking error had an average of zero in the long run, it would still reduce returns. For example, earning 5% for five years followed by 7% for five years results in less total return than earning 6% for ten years (albeit the difference isn’t much!).
________________________
Well, sure, using your example. But the opposite is also possible where the return is higher for the first 5 years and then lower for the last 5 years resulting in a higher return!
I think what you mean is it COULD reduce returns, not that it WOULD reduce returns.This was my original point – there seems to be a mental bias that the tracking error is negative and I don’t understand why this would be. As far as I can tell it could be in either directions.
@CJBob: The track record of currency hedged funds includes consistently negative tracking errors. It’s not a mental bias.
In your “Bad Investment Advice” section the ROI method was a similar idea to the Faber paper “A Quantitative Approach to Tactical Asset Allocation” back in 2007. I believe that timing model was back tested much farther and of course now, forward tested after its release. Last I looked at it the results were positive but I don’t employ this method myself (yet).
BTW, good podcast!
@Simple: These may be of interest:
https://canadiancouchpotato.com/2013/12/30/the-failed-promise-of-market-timing/
http://www.cambriafunds.com/gaa (the ETF originally based on Faber’s strategy: note the dismal performance compared to just about any equity benchmark)
Hi Dan,
Thanks for the great podcasts.
Clearly a lot of respect needs to be paid to Dr. Shiller but I’m still interested in your thoughts/criticism on the BMO Shiller Select US index fund (ZEUS).
I’ve been reading some behavioural psych and economics and I’m clearly no expert but this is where my mind is going (I know, you’re probably rolling your eyes): After reading Misbehaving by Richard Thaler he talks about regression to the mean with stock and market price. Thaler describes a study where they took all stocks on the NYSE and and made portfolios of “winners” and “losers”. Say the top 35 and bottom 35 (I believe using P/E ratio but he is not explicit). The “loser” portfolio outperformed the “winners” and was less risky – regression to the mean! While I’m currently 100% invested in the TD index mutual fund strategy you so kindly advocate for, I did wonder about a mutual fund or ETF that did exactly the above as it would seem not so risky and potentially fruitful for future investing if after I max out my RRSP and TFSA.
I thought maybe ZEUS would do this but after looking at it in detail they start with “100 stocks with the highest 10-year CAPE” and then doing some “momentum filter – Sector Neutral” fandangling.
Doesn’t this seem backwards? It wouldn’t you want to make a “loser”/Low CAPE fund? Does anything like this exist beyond just doing it yourself? Because I’m way too lazy for that…
Unfortunately this ETF is set for termination. Curious if there’s any other CAPE based funds out there? – I thought it was a great idea.