Over the years, I've noted that certain subsets of market conditions - occurring together - are associated with very specific outcomes, such as oncoming recessions, abrupt market weakness, strength in precious metals, and so forth. Such indicator subsets, or Aunt Minnies, are essentially "signatures" that often have very specific implications. In medicine, an Aunt Minnie is a particular set of symptoms that is “pathognomonic” (distinctly characteristic) of a specific disease, even if each of the individual symptoms might be fairly common. Last week, we observed an Aunt Minnie featuring a collapse in market internals that has historically been associated with sharply negative market implications.
Of the 3257 issues traded on the NYSE last week, 2955 declined and just 275 advanced. The S&P 500 has now abruptly erased nearly 8 months of progress. Moreover, we observed a "leadership reversal" with new 52-week lows flipping above the number of new 52-week highs. Our broader measures of market action deteriorated to a negative position as well. Historically, we can identify 19 instances in the past 50 years where the weekly data featured broadly negative internals, coupled with at least 3-to-1 negative breadth, and a leadership reversal. On average, the S&P 500 lost another 7% within the next 12 weeks (based on weekly closing data), widening to an average loss of nearly 20% within the next 12 months - often substantially more when the Aunt Minnie occurred with rich valuations and elevated bullish sentiment.
The most recent instance was November 9, 2007, which was followed by a market loss of more than 50%, but the instances also include September 22, 2000, prior to a nearly two-year bear market decline; July 14, 1998 prior to the "Asian-crisis" mini-crash; July 27, 1990, at the beginning of the pre-Gulf War plunge; October 9, 1987, just prior to that market crash; July 2, 1981 at the beginning of the 1981-82 bear market and again in May 21, 1982, following a strong rally during that bear market, leading into a steep decline to the final lows; November 9, 1973 (just after a swift rally during the 1973-74 bear market, and leading into the main portion of that loss); and November 21, 1969, at the beginning of the 1969-70 bear market.
Given my aversion to market "forecasts," I hesitate to interpret this record as a hard prediction of what will occur in this particular instance. This is particularly true because in a handful of instances (2/9/68, 9/12/75, 10/20/78 and 4/30/04), the outcomes were fairly benign. Still, the average outcome has been awful. For that reason, the combination of unfavorable valuations and collapsing market internals is a sharp warning to examine risk exposures carefully here.
I haven't used the word "warning" for some time, and I certainly don't want to be alarmist. If your asset allocation is reasonable and you are following a well-defined discipline, do nothing. There is wisdom in ignoring the views of others and staying the course, provided that your investment strategy is in fact based on a well-defined discipline. That said, I disdain when analysts advise "staying the course" when the subtext is to disregard risk.