Just How Sick Is the Stock Market?
China's markets set the tone for the day (and perhaps the week) with an 8.5 percent blood-letting. Global stocks followed suit, which came after last week's 5 percent tumble.
Rather than tell you that markets are oversold -- you already know that anyway, and oversold markets can become even more oversold -- I want to bring a few interesting data points to your attention.
Let's begin with China's markets. That is where much of the turmoil seems to be originating. A little context will go a long way.
One year ago, the Shanghai Stock Exchange Composite Index stood at 2209.46; it's now 3209.91. It closed last year at 3234.67, and peaked this year at 5166.35. To put those returns into percentages:
One year: +45.28 percent
Year to date: -0.77 percent
Year to date peak: +60.95 percent
Peak to trough: -37.87 percent
It is amazing that China is little changed for the year, down less than 1 percent. That gives you some idea of how absurdly inflated its markets had become in a very short time. Its easy money policies and encouragement of middle-class stock speculation (Why does that sound so familiar?) inflated a market boom that (Surprise!) is now unwinding. What those number show is quite telling.
Next, let's move to U.S. The chart below is courtesy of my colleague Josh Brown. It looks at the 5 percent, 10 percent and 20 percent declines in the Standard & Poor's 500 Index since its inception in 1957. There are several fascinating aspects to this data series.
S&P 500 1957 to present
Source: The Reformed Broker
First, let's start with the data: From all-time highs, we have seen 48 drawdowns of at least 5 percent. Of that group, a little more than a third of those became a full correction of 10 percent. (That's 17 of the 48, or 35 percent). Of all declines of 5 percent or more, less than one fifth (9 of the 48, or almost 19 percent) became drawdowns of 20 percent or more, the informal definition of a bear market.
If you had to break this 58-year period into market cycles, it is fairly easy to find the secular bull and bear markets just going by the colors. The post-World War II bull market has little of the blue or red corrections, right up until 1968-69 or so. Then we see the colorful period of 1968-1982, a brutal 16-year bear market. That is followed by the bull run from 1982 to 2000, with little in the way of blue or red. From 2000 to 2013, the wide swatch of red tells you that was a bear market. But since then we appear to be in a bull cycle. Charts such as this make it relatively easy to spot cycles visually. Indeed, this is consistent with what we know about secular bear markets, which tend to be characterized by frequent sharp rallies and severe sell offs.
Ben Carlson (using Ed Yardeni's data) looks at this data somewhat differently. He observes that the average number of double-digit drawdowns has generally been lower each decade since the 1930s. It's noteworthy that the average double-digit decline of the 2000s was even worse than the 1930s, at least in terms of intensity.
Carlson notes how the recency effect leads investors to focus on the wrong things: "Stocks are up over 200% since March of 2009. Instead of celebrating those gains, investors are constantly worried about missing the next 5-10% correction."
Finally, let's put into broader context the frequency of corrections. U.S. markets average one 10 percent correction every 20 months. On average, we should expect these declines to take 71 trading days to play out (about three months).
As discussed above, these occur more frequently during secular bear markets. In the postwar period, one quarter of these occurred during the 1970s and another fifth of them occurred during the 2000s -- both periods were part of a secular bear market.
Note that in the secular bull market that began 1982, there was but one 10 percent correction in the first five years (1982 through the 1987 crash). After the 1987 crash through the March 2000 peak, there were just two corrections of that magnitude, according to Yardeni.
As we noted last week, the economy seems to be OK, so this looks more like a correction than anything more serious, but that's just a guess on my part. We should know before the December Federal Open Market Committee meeting.
For the purposes of our discussion, I am using 10 percent for a "correction," 20 percent for "bear market," and 30 percent for "crash," and 50 percent for "generational crash." These terms are random fabrications with no real meaning other than reflecting our base-10 numeric system.
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
To contact the author on this story:
Barry L Ritholtz at firstname.lastname@example.org
To contact the editor on this story:
James Greiff at email@example.com