So, another personal finance blog posted an article about the "false allure of compound interest" which goes about explaining that the personal finance claims about compounding interest are misleading because they often use "average rate of return" but don't take into account that most of the returns come in the later years, where you might see below average rates.

http://www.wisebread.com/the-false-allu ... d-interest

I wasn't really sure what to make of the article or whether there was a particular call to action here. Anyone else have any thoughts?

Sure, but what do you do with this information? I think the main message is "don't count your chickens before they hatch." In other words, if you're planning for retirement or some other long-term savings/investing goal, you can make assumptions about your rate of return but you have to remain acutely aware that those are assumptions, and that they could easily be off by 50% or more. So you either over-engineer and risk saving/investing too much at the expense of your current lifestyle, or you're overconfident and assume higher returns than you actually achieve, increasing the risk that you'll outlive your savings.

Compound interest is mostly BS, but it at least serves to spur peoples' interest in taking action with their financial lives.

The problems I have with "compound interest" are numerous, but the main ones are that when they say "compound interest," they rarely actually mean compound interest - they mean compound returns. Strictly speaking, compound interest would only be in relation to loan instruments, which of course are paying far too low to ever actually amount to anything meaningful in a worthwhile time span.

Compound RETURNS, on the other hand, can be much more promising, but are much more volatile, and (as noted) are returning far below the historical average often used to illustrate the compelling nature of "compound interest" (i.e., 12+%).

Finally, the timing of the returns matters immensely. There is a huge difference between earning a consistent 10% every year for 40 years, and earning a variety of returns ranging from -10% to +30% (which still average out to 10%) when you factor in contributions, and when the returns occurred. A few years of 30+% returns, followed by a decade of -10% returns, is completely different from getting those "bad" years out of the way early, before you've gotten most of your lifetime's contributions in, then getting the "jackpot" years at the tail end, right before you retire.

Kombat raises an important distinction, because some people talk about "compound interest" when referring to the stock market, but stock values don't compound the same way savings accounts do; the returns vs. interest distinction is the one to keep in mind here. It's important to remember that inflation compounds too, and that has to be factored into your assumptions.

Ultimately I think all this just reinforces the advice that it's always good to have a Plan B (and a backup Plan C), and to periodically revisit your assumptions and adjust them as best you can to reflect the current reality.

Whether the allure is potentially 'false' depends on your horizon. If you search for rolling 10-year and rolling 30-year annual returns for the S&P 500, you'll discover a couple of things:

1. Meaningful returns over a decade period are NOT a given.

2. Throughout history, there have been remarkably few 30-year windows where the average annual return was LESS than 8%. Typically, the 30-yr rolling average return is north of 10%.

Average annual return isn't really a useful measure in this case, though; the compound annual growth rate is what you want. From 1926-2010 the compound annual growth rate of the S&P 500 was 9.89%.

The problem with average annual return is best illustrated by the example that if you invest $x, and your investment doubles in value the first year, you now have $2x, or a 100% return. But if your investment loses 50% the next year you're back down to $x. Your average annual return is 25%, which sounds great but what have you actually earned? Nothing.

For more info see http://www.assetplay.net/article/basics ... erage.html.

Thanks for the important clarification, brad!