The S&P 500 is fighting (unsuccessfully at this point) for its eighth positive session in a row today (the longest winning streak for the index in exactly two years). But looking under the hood at the past two sessions, there has been some underlying weakness.  Although Monday and Tuesday saw the S&P 500 rise 17.5 bps and 28.4 bps, respectively, net daily breadth (advancers less decliners) was negative on both days.  Whenever we hear talk of weak breadth on market up days, comparisons are usually made to the 1999/2000 period right before the Dot Com bubble’s peak.  While it has been very uncommon for the S&P 500 to be up on back-to-back days when breadth was negative, not all (or even most) of the prior occurrences were isolated to just the period leading up to the Dot Com peak.

In the table below, we show 19 prior times that the S&P 500 rose in back-to-back sessions with negative daily breadth readings on both days and no other occurrences in the prior three months.  The most recent occurrence was back in June 2021, and one thing that stands out is just how bad breadth has been in this period.  On Monday and Tuesday, cumulative breadth was at -267, and the only prior instance with worse breadth over the course of the two up days was in July 2015.  While returns were outright negative for the following six months after that July 2015 occurrence, as a whole, these past occurrences with gains on negative breadth have not been an especially bearish or bullish signal. On both an average and median basis, performance was generally in line with the S&P 500’s performance for all periods since 1990.

In looking at the table above, on most of the days when the market was up and breadth was negative, the magnitude of the gains was very small, and in many cases, the S&P 500 didn’t even move a tenth of one percent (10 bps) on either day.  With that in mind, we filtered the table above to show only days when the S&P 500 was up at least 10 bps on each of the days when breadth was negative.  Adding in that criteria, the 19 prior occurrences get whittled down to just six, and in this case, four of the six occurrences were in the months leading up to and after the Dot Com peak.  While forward returns over the next week were positive all six times, average and median returns over the next one and three months were actually negative.  Longer-term, six and twelve-month returns were split with a wide variance between average and median performance.  These past examples suggest that while weak breadth on back-to-back positive days for the S&P 500 is not an outright negative, it’s hardly a positive indicator either.