While putting together my **S&P 500 Fun Facts post**, I created an Excel spreadsheet to analyze the returns of the S&P 500 Index. I got the original S&P 500 returns as well as the inflation numbers in Ibbotson’s **Stocks, Bonds, Bills, & Inflation**. I then used those numbers to create 5, 10, and 20-year rolling period real returns. After I posted my findings in a couple of posts (**here** and **here**), I decided to create some bar charts to help illustrtate the impact of time on returns. These charts present the information in a way you just can’t get from looking at a column of numbers.

First we’ll start by looking at 1-year annual real returns:

**S&P 500 Returns Annual Real Returns**

**1926 – 2006**

Out of the 81 one-year periods from 1926 – 2006, 26 years experienced negative REAL returns. By “real” I mean returns that are adjusted for inflation using that particular year’s CPI. There were 38 years in which the real return was less than 10%. Real returns exceeded 10% in 43 of the 81 years (53% of the time).

Now let’s move to 5-year holding periods and see what happens:

**S&P 500 5-Year Rolling Period Real Returns**

**1926 – 2006**

By 5-year rolling periods, I mean beginning in 1926 and going out 5 years through 1930, for a 5-year period. The next 5-year period starts at 1927 and goes through 1931. The last rolling 5-year period was 2002 – 2006. As you can see from the bar chart, there were a lot less 5-year rolling periods with negative returns. In fact, of the 77 periods 5-year rolling periods, 19 had negative average annual real returns throughout the period. Over sixty-one percent of the periods (47 out of 77) had average annual real returns of less than 10%. The average annual real return for ALL 77 5-year rolling periods was 7.03%.

Here’s what happens when we go to 10-year holding periods:

**S&P 500 10-Year Rolling Period Real Returns**

**1926 – 2006**

Here’s the breakdown of the area in the circle, which shows the impact that inflation has on returns:

What the graphic above shows is that from 1965 – 1974, the average annual rate of return for the period was 1.24%. However, due to inflation over the same period, the average annual REAL return was actually -4.62% over the ten-year period. Bottom line: inflation has a very real impact on returns.

Finally, let’s take a look at 20-year holding periods:

**S&P 500 20-Year Rolling Period Real Returns**

**1926 – 2006**

Notice that of the 62 20-year rolling periods, NONE of them had negative annual average real returns. The worst 20-year period was from 1962 – 1981, which had an average annual real return of just .52%. The rest of the periods were much better than that.

It’s amazing to me how time smooths out the individual bumps in the road.

DISCLAIMER: As with everything, past returns are no guarantee of future returns. Invest at your own risk.

I’ve seen some similar data before, but of course it was buried in 10 pages of yada, yada, yada…

I like how you present this in a very straight-forward and easy to understand manner!

I’m sure I will refer this post to others in the future. Thanks for doing the legwork!

I’ve recently seen similar info in a book by Paul Merriman. He aggregates similarly to this, but over a longer period. He also then further indicates what happens when you diversify. That is, he creates model portfolios with incremental increases in stock/bond splits, and in domestic stock/international stock/bond splits, etc. He eventually includes growth and value stocks for domestic and international large-cap, mid-cap, and small-cap indices. It is extremely instructive to see the effects of proper diversification and long-term investing. Good work!

This is an excellent presentation of the data. The key to me here is the recognition that investing is a long-term proposition. Investors need to be prepared to weather a sustained down market like the 1965-1974 period you highlight. All investors should ask, if we went through another period like 1965-1974, would I hold steady, or would I sell?

This is a misleading way to look at returns. Sure the annual returns are smoothed out, but the total returns aren’t. Your charts make it seem that the variation in returns decreases with time when in fact it increases.

Andy, I’m not sure I understand your comment. The return numbers are for the entire (rolling) period specified – one year (with significant variation), five years (with somewhat less variation), ten years and 20 years, not annual returns for the years included. If you look at the standard deviations of the periods as well (though he doesn’t present them here, but you can find them easily enough), you find that the standard deviation decreases, which indicates that the real risk of the investment decreases for the period as the period increases.

I don’t think the point here is that the volatility of any given year decreases, but that the likelihood of negative returns decreases to essentially zero with a long enough time horizon.

Andy,

Misleading? I don’t think so.

This is pretty good, similar to Bogle’s analysis.

What’s not explicitly mentioned though, is that your actual returns depend on how much you have invested at a given time. This market isn’t an annuity table. So if you’re adding to your investments you want a bear market early on. But a retirement can be doomed by an early bear market – you’ll have to sell investments in a down market and it can be quite difficult to recover.

Risk is as important as returns, but it isn’t as sexy to talk about.

I use to see the same kind of blogs.But,the way you explained is really clear and genuine.