2007 Volatility: Below Historical Average

I keep hearing how there was an unusual amount of volatility in 2007.

That assessment is wrong. We have become used to the aberrationally low volatility of recent years, but historically,  we are just below average, according to Bespoke Group:

"The average absolute daily price change in the S&P 500 by year. In 2007, the average worked out to 72 basis points, which means that, on average, the S&P 500 had a daily move (up or down) of 0.72% versus an average of 0.75% since 1928. The last time the market was this ‘placid’ was in 1996."

SPX Chart:



Think B.I.G., December 31, 2007 at 11:41 AM


Print Friendly, PDF & Email

What's been said:

Discussions found on the web:
  1. Joe commented on Jan 3

    It looks like the average is inflated from outliers in the 29-32 era. What if we started this thing from 1942 instead…

  2. Philippe commented on Jan 3

    Volatility for the indices may look subdued save for the stocks as individually they have shown a variance out of historical proportions (throughout the year 2007 +/- 10 Pct was genuine)
    No wonder as the markets had to improve on few stocks and short sellers to be proved on the same.

  3. Hank commented on Jan 3

    1928 is a curious starting point, unless one is deliberately trying to include the volitilty associated with the ’29 crash to inflate the average. Also, without having the data in front of me, I would suspect that a chart broken out by quarter would show Q3 and Q4 to be well above the historical average.

  4. Ross commented on Jan 3

    Beginning in 1928 does distort a bit.

    Speaking of 1928, Turner Classic Movies is running a 1928 film featuring Marion K. Davies. She was an underated actress. They are running her films all day.

    I’ll give anyone an extra biscuit or glass of ethanol ( your choice ) if you can tell me who she was married to in the 1930’s.

    Ta ta

  5. michael schumacher commented on Jan 3

    this is an honest guess, no google:

    Hearst?? or Orson Wells??


  6. John Forman commented on Jan 3

    Among technical analysts there is a study known as the Average True Range (ATR) which is basically the same as what’s been described here. When looking at the monthly chart of the S&P 500, the 14-period monthly ATR (as a % of price) bottomed in January of 2007. That reading was right about where it started 1996. It had actually gotten lower at the end of 1993, the low reading for quite some time (my data only goes back to 1980). We start 2007 with a reading right about where things bottomed out in 1981/82 and 1985. By way of comparisson, after the ’87 Crash, the ATR reading peaked out 80%-90% higher than it ended 2007, and the highest level reached in early 2003 was better than twice what it is now.

  7. Ross commented on Jan 3

    MS, Actually she never married Hearst but it would now be deemed a common law marriage.
    Legend has it that during the depression, Davies mortgaged her real estate and bailed out Hearst. TODAY my guess is that you hock your securities to bail out your real estate mortgages.
    You win.

  8. Suge Knight commented on Jan 3

    John Forman, next time after writing all that non-sense, write your conclusion, so we understand what your point is. Thanks, re-read it, think, and ask yourself if you’re making a point.

  9. Michael C. commented on Jan 3

    I think it just goes to show, as pertaining to the market, most people have a memory of only a few years.

  10. kio commented on Jan 3

    I am afraid that there is some misunderstanding of the term “volatility”. One should differ random fluctuations associated with volatility and trends.
    According to your (borrowed) definition volatility of 1% (p.p.) would be measured if an 1% positive return would be observed for every day of a year, i.e. total linear growth would be around 220%. Obviously, according to the accepted definition, one should consider than deterministic trend as a year of high volatility. I would disagree.
    Detrending must be the first step in volatility estimates, RMS or ATR.
    Then one can find that the years of high volatility since 1990 were associated with strong trends – positive before 2000 and negative one after 2001.
    Actually, 2007 has very high volatility because of very weak trend – around 7% annual return.

  11. Abhay Avachat commented on Jan 3

    Why just average ? A median along with average would have been much better information. Looking at the chart, it seems like the median would be lower than the 2007 number.

  12. jw commented on Jan 3

    Purely anecdotal, but I’ve noticed over the past year or two that CNBC uses whatever graphics/charts that make it seem like the markets are volatile.

    Like, “you can’t miss a minute of CNBC’s coverage during these volatile times we live in!”

    For example, they started using one minute charts to inflate a stock’s reaction to news, which does indeed make it look like the stock is going wild… except the chart’s peaks and valley’s cover a range of 75 cents.

    And when one minute charts are boring they switch to charts showing a period covering 30 seconds of trading.

  13. Eric Davis commented on Jan 3

    Always feel free to get rid of any data that doesn’t feed your thesis.
    I actually don’t give a crap to get into it, but if you(various commenters) go for a 10 year time frame the median is “Higher”. and since 1970 is maybe a couple basis points lower. Maybe we need to generate a chart with every stock market anywhere since the renaissance.

    But I digress, as I have heard, the term “volatility” seems to be commonly used synonymously with “going down”, but saying that would be predictive. “volatility” and bear market, also seem interchangeable, since saying “Bear” is heresy on “The street”

    On a similar note, I’d love for anyone who says a sub 100k job number, isn’t “Bad”… could please define what a “Bad job number” is…. narcissus says “when you lose your job”

    Damn, how do I get a biscuit…..

  14. Old Ari commented on Jan 3

    Gore Vidal quoted Orson Welles, as to the meaning of “Rosebud”, in the movie about Hearst, it involved Marion.

  15. luker commented on Jan 3

    you mean it wasn’t a sled????? LOL

  16. Ross commented on Jan 3

    OK Eric, you get a biscuit also. I enjoy your posts. No one wants the ethanol? Aged 3 weeks!

  17. Greg Feirman commented on Jan 3

    But that analysis misses the fact that 2007 was essentially two different years: the first half and the second half. I bet the second half was volatile by historical standards. It’s the first half that made the year look normal overall.

  18. PTodd commented on Jan 3

    I would prefer to look at the median and not the mean.

    And nice to compare 2007 to 1928. If I recall, 1929 was kind of a downer. Might be a hidden warning?

Posted Under