How often in the last few years have investors said they’re staying out of the equity markets because of the volatility we’ve experienced recently? Many of them are waiting on the sidelines “until things are back to normal.” That raises the question: what exactly is normal for equity returns?
Usually when people think of “normal” returns, they look at historical averages. According to the Credit Suisse Global Investment Yearbook, stock markets in the developed world delivered an annualized return of 8.5% over the last 112 years. Using that average as the midpoint in a range, it seems fair to say that “normal” historical stock returns are between 6% and 11%.
You might conclude, therefore, that it will be time to get back into equities once we’ve seen a couple of years with returns in this neighbourhood. That would be signal that things have “returned to normal,” right?
To test this idea, I looked at equity index returns for Canada, the US and international developed markets (in Canadian dollars) since 1970. Sure enough, during this 42-year period, annualized returns for all three asset class returns were within our expected range: 9.1%, 10.6%, and 8.9%, respectively. But what about year-by-year returns? If you were investing throughout these four decades, what years would you have considered “normal”?
Normal annual returns are extreme
You may be shocked to learn that a portfolio with equal amounts of Canadian, US and international stocks would have posted returns between 6% and 11% exactly five times in the last 42 years. Think about that: in any given year, the chance that stock returns will be within this “normal” range was less than one in eight.
Now let’s consider the probability of more “abnormal” outcomes. If the average long-term return for stocks is 8.5%, let’s look at years where returns were a full 10 percentage points more or less than that. It turns out there were 11 years with losses of at least –1.5%, and 17 others with gains of at least 18.5%. In other words, the probability of a significant loss or a huge gain was 67%, or two years out of every three.
You can see all of these data by downloading my spreadsheet. You can also experiment by changing the upper and lower limits of what you consider “normal.” The spreadsheet will tell you how many years fall within your specified range.
In his outstanding book Debunkery, Ken Fisher looked at an even larger data set from 1926 through 2009 and found much the same result. The annualized return of US stocks over this period was 9.7%, and the simple average was 11.7%. But over one-year periods, returns almost never looked like this. Two-thirds of the calendar years produced returns of more than 20% or less than –10%. Returns were between 10% and 12% only five times in 84 years. “Normal annual returns are extreme,” Fisher writes. “It is hard to get people to accept the degree to which that’s true.”
Volatility has always been normal
There are a couple of lessons here. First, a “normal” year for stocks is one with very high or very low returns—and that has been true even since 2000. So if you are waiting until stocks can be counted on for a steady 6% to 11% return, you’re waiting in vain, because that has never happened. Sharp declines and soaring recoveries have always been normal. Any time you’re out of the market you’re safe from a sudden plunge, but you’re more likely to miss periods like the 13 months following the 2009 market bottom, when global stocks rose over 80%. Or, more recently, the period from October 2011 through this past March, when they rose almost 25% in just six months.
The second lesson is harder to hear. The “long term” is probably longer than any of us wants to admit. If you think it’s seven to 10 years, that’s fantasy. You have to look at rolling 20-year periods before there’s a very high probability of equity returns close to that 8.5% average. That means if your time horizon or temperament prevents you from thinking that far ahead, you need to dampen your portfolio with an allocation to high-quality bonds or cash. This will likely lower your expected return, but that is the inevitable trade-off between risk and reward.
People will always try to get the upside without the downside, whether it’s through market timing, stock picking or more exotic strategies. But mostly they’re fooling themselves. As Ken Fisher writes after more than three decades as a wealth manager, “To my knowledge, no one has ever achieved market-like returns without some market-like downside. If you want to achieve something close to stocks’ long-term average, you must accept downside volatility. No way around that.”
I think the second last paragraph you wrote is extremely important. When investing in common stocks you are taking a kind of risk that you are not with GICs or bonds. This might mean that if a plan depends on more than extremely conservative assumptions about equity returns that plan is at risk of bad outcomes.
What really matters is that we have the outcomes we need from our portfolios. The most conservative approach to retirement then is to simply assume the “risk free return” (what you would get from cash and GICs/Bonds) and save the amount you need to meet your needs at this rate of return.
My father was a comptroller for a large multinational corporation who had to manage risk. He did not invest in the stock market and I remember asking him why. “Because it all depends on when you are putting money in and taking money out. There is no way to project what you will have. You need certainty when planning for retirement”. Now this was a multi decade period time when GICs were probably averaging 6-8%, sometimes more, so in a registered plan you did okay.
It may be prudent to be prepared in any given year to lose 40% of the equity value and to structure a portfolio so that very bad outcomes do not result if there are zero returns from equity over a decade. Just because these outcomes happened recently doesn’t mean they can’t happen again, even within the next 10 years.
The most prudent approach of all may be to invest in stock ETFs but to assume anything they give a plan is just a bonus.
If the normal PE for the S&P 500 is 15, then normal returns should be roughly 6.7% per year. That’s fairly consistent with true historical returns less inflation (since inflation affects both the cost of running a business, and the prices at which goods are sold) in the US.
Assuming some kind of normal results going forward based on historical averages and not the current earnings yield might be a little too optimistic. There is a very strong relationship between a business’s value and its earnings, and the same is true in the aggregate.
This made we wonder what would have happened to an investor that waited for a year of +6% growth before investing, and pulled all their money out (into 1% cash) any year that didn’t grow at this rate until the market started growing again.
With $1000 invested in 1970 they would have ended up missing out on ~27% against just leaving the money in. ($40.5k instead of $51.7k)
If instead they invested in any year that the previous year was positive, and stayed in 1% cash in any year that previous year was negative they would have ended with basically the same amount ($51.5), but only 1 year they would have been ahead, and on average they would have been down $3.6k.
So long story short (and as anyone here likely already knows) don’t try to time the market.
Great article! I’m newish to the world of investing. When pulling together this spreadsheet did they calculate the returns of all the stock markets over the world?
There is such a difference between Canada stocks in 2010 of 17.6% and 2011 at -8.7%… wow! Seeing the numbers I can really see how you can’t time the market!
A great read! Just tweeted!
No doubt I’m accepting some downside volatility right now.
I’ve been learning about indexing for the past three years, but never came across these facts before. An eye-opener. Great article!
@Andrew: Thanks for the insightful comment about risk management. I agree completely that anyone investing in equities should be prepared to lose 40% or more, and if that happens it will likely take years to recover. As your father understood, equities are not an appropriate investment for someone who needs a specific amount of capital on a specific date. They never have been.
@Christine: The returns in the spreadsheet for are for Canadian stocks (as measured by the S&P/TSX Composite Index), US stocks (S&P 500) and developed countries in Europe/Pacific, including Japan and Australia (MSCI EAFE Index). I assumed the investors held equal amounts of these three asset classes.
@Potato Salad: I admit the numbers shocked me, too. I expected annual returns in the neighborhood of that 8% or 9% average would be uncommon, but I had not realized they were as rare as they turned out to be.
Thanks CCP I made a note of that.
So why should one be prepared to lose 40% or more? From what I see if you invest in all markets equally, you should expect to lose almost 30% in one year, not 40%…
@Christine: Good question. The annual returns that I compiled are for calendar years, from January through December. There have been several 12-month periods spread over two calendar years when stocks have fallen by 40% or more, most recently between September 2008 and March 2009. In fact, the worst drawdowns have exceeded 50%.
So important for investors to understand that returns spend very little time at the ‘average’ … that’s why it’s critical to get ones asset allocation approximately right to moderate the extreme downmarkets.
Thanks – I’ll need to look at a breakdown of months to see this.
“That means if your time horizon or temperament prevents you from thinking that far ahead, you need to dampen your portfolio with an allocation to high-quality bonds or cash.”
What are examples of ‘high-quality bonds’? Are they bond indexes sold as mutual funds and ETFs? By cash do you mean laddered GICs? What kind of ‘average’ return could you expect from investing in ‘high-quality bonds and cash’?
William J. Bernstein, who favours index investing, wrote in “The Investors Manifesto” about the concept of “equipoise” to help determine the level of stocks one could be comfortable with. Equipoise is the exact balance between feeling happy about having gains and the regret of not watching it rise enough.
Daniel Kahneman won a Nobel prize for his work on prospect theory and one thing he found is that a loss is about twice as bad in terms of emotional impact as a gain gives one a feeling of good. He likened losing money to feeling physical pain. Thats why I like to see the number, in dollars, of a 40% decline in the equity side before I consider how much I might be giving up by not having more equity.
@Darby: By “high-quality” bonds I mean those issued by the Canadian government or blue-chip corporations, as opposed to high-yield (junk) bonds issued by emerging countries or companies with weak balance sheets. Most bond index funds contain only high-quality (or “investment-grade”) bonds unless the word “high-yield” is explicitly in the fund’s name.
GICs are similarly safe investments, though I would not consider them “cash”: they are far closer to short-term government bonds. Strictly speaking, “cash” would be a high-interest savings account or money market fund.
Using the same Credit Suisse data I used for stocks, global bonds have returned 4.8% annually over the very long term. However, right now a broad-based bond index fund has a yield to maturity of about 2.25%. High interest savings accounts are paying about 1.25%.
@Andrew: I love the term, “equipoise.” :)
Kahneman’s work has some interesting implications for investors who watch their portfolios every day. If we assume that daily market returns are about equally likely to be up or down, then what happens if you watch the markets every day? When they go up you get one unit of pleasure. When they down, you get two units of pain. Up one, down two, up one, down two, over and over. Eventually you feel horrible even if the markets have gone up overall.
Great post! Thank you so much.
Looking at the Credit Suisse PDF, can you point me to the page/paragraph/section where you obtained the 8.5%? Thanks!
@Bob: The data are on page 57.
great article and one of my favourite subjects. Longterm equity returns are meaningless. What is more normal is equity returns of zero against inflation-save for your dividends-for periods lasting 20-30 years. Investors should realize that they won’t get rich from equity returns and being aggressive with all stock portfolios is not worth the risk. History says there r not 10% waiting for you.
And because most investors can’t handle even modest volatility, i hold the opinion that the masses would do better in a portfolio more weighted to bonds. 60-40, even 70-30. Instead of seeing your portfolio with 50 and 60% retracements you might be looking at 10-20% losses. Holding some high income in your equities will help as well. Recently my portfolio only went back 2% during a 20% canadian equity pullback.
The only financial plan that works, is the one that an investor can stick to.
Great great article. Puts things in perspective.
I think markets are like what you described simply because of the two human emotions: Greed and Fear. Greed will drive markets to inexorable heights sometimes, and then fear would drag markets down to inexplicable levels as well. Rarely do you see the middle ground. After all, people like to follow crowds and participate in the overall economic sentiment. Hence, human nature always dictates.
But regardless of the reason, indexing long term will make sure we capture whatever returns will be out there.
Great article. Just wondering if the returns include dividends?
@Amir: Yes, all of the index returns include reinvested dividends.
In looking at my own year returns on my Cdn Equity portfolio, I use EXCEL’s XIRR (Internal Rate of Return) for Stock Purchased, Stock Sold and Dividends Rec’d. All items are tagged with their specific date stamp. I have a 16 yr record, 1997 to 2012. My first year was a lucky buy (IPL) which netted me 197%. Then, tings settled down and I was not so lucky as i continued to add to my portfolio. Disregarding that first year, my worst return was in 2008 at -24.75%. My best year was in 2003 at +25 .03%. 2012 is logging a sluggish 2.26% to date. By combining all the time stamped data for 1997 to 2012, my XIRR is 10.67% on my spreadsheet, I can look at $ returns, or returns as percentages.
Excellent article I’ve been an index investor for 8+ years and from experience agree and have seen this type of activity/return in my portfolio. The article and spreadsheet clearly portrays the details nicely.
Which indexs did you evalute to get the numbers in your spreadsheets? The typical ishares indexs appear to be off by 2 – 3% thus making the annual returns worse in all cases.
@Jeff: The index data I used was the S&P/TSX Composite, the S&P 500, and the MSCI EAFE, all measured in Canadian dollars. If you are comparing these to ETF returns, the problem might be that you are looking at US-listed ETFs that express returns in US dollars.
I’d love to see an update of the spreadsheet for the years 2012, 2013 and eventually 2014!