The four most dangerous words in long-term financial planning are, "Let's assume I'm average."
Averages can play an extremely useful role in grounding our expectations and showing us what's realistic and what's not. They're not, however, an infallible guide to how you should conduct your affairs, because they don't show you the gap between worst and best cases.
A Canadian who dresses for the year-round average temperature will freeze in February and swelter in August. Someone who assumes they will encounter only average results in planning for retirement may encounter far more distressing problems.
A good example is the length of your retirement. It's easy to assume you will have an average lifespan and pass away at 79 (for men) or 83 (for women). Based on those numbers, most of us might plan for 15 to 20 years of life after work.
Unfortunately, it's not that simple. The expected life span of someone who reaches 65 is considerably higher than the life expectancy for the country as a whole, because the all-inclusive national figure is dragged down by those unfortunate souls who pass away young.
Statistics Canada tells us that a man who reaches 65 can expect to live another 18.5 years, while a woman at that age can look forward to another 21.6 years. But those, too, are just averages.
A substantial number of Canadians will live far longer; that's what an average implies. To account for the very real possibility that you will still be chugging away well into your 90s, a well thought-out financial plan should prepare you for at least 30 years of retirement – much longer than the averages would initially suggest.
Averages can also be a treacherous guide when assessing how long your portfolio will last in retirement.
Judging from my e-mail, many readers figure that since a combination of half stocks and half bonds has historically produced an average annual return of roughly 8 to 10 per cent, they should be able to count on withdrawing their money in retirement at roughly that same pace with little or no risk of running out of money.
That, too, is not so simple. First, it ignores the impact of inflation and fees. Second, it sidesteps the issue of how variable those returns will be.
The Vanguard Group Inc., the U.S.-based mutual fund operator, did a fascinating study in 2011 of how a balanced portfolio of half stocks and half bonds would have fared during recessions since 1926. It found that, on average, the portfolio would have produced similar returns during both downturns and expansions.
At first glance, that looks like strong evidence for the dependability of returns from a balanced portfolio. But look further and you find there is an enormous variation in how a balanced portfolio fared during various recessions.
In the downturns of 1926, 1953, 1960 and 1981, the balanced portfolio produced strong returns. In the recessions of 1937, 1973 and 2008, it performed miserably.
In retirement, this variability becomes a problem because you're drawing down your portfolio at the same time as you're experiencing market volatility.
During this stage of your life, the sequence of returns you encounter can have a big effect on how long your portfolio will survive.
An investor who enjoys an early string of successes builds up a substantial buffer that can easily withstand subsequent downturns as well as the steady erosion of withdrawals.
In contrast, someone who gets hit by early losses, but has to keep withdrawing funds, may run out of money, although his or her average annual returns over the same period are exactly the same, before accounting for withdrawals, as the more fortunate investor.
What should an, um, average investor take away from all this?
One important point is to build in a margin of safety. Rather than planning for a 20-year retirement, prepare for 30 years of life after work. Rather than assuming you can draw down your retirement portfolio at 8 per cent a year, count on something more like 4 per cent.
On a deeper level, take volatility into account when you plan for the long haul. Look at the spread of possible outcomes, not just the average.