Market breadth is a widely discussed indicator in equity market analysis. It explains how broad-based a rising market is supported. Usually, it is considered a warning sign when the number of companies moving the market higher decreases. Recently, this indicator has again attracted much attention as the intra-market dispersion, especially in the U.S., has increased rapidly. While the S&P 500 is trading close to an all-time-high, large parts of it have entered bear market territory meaning that a diminishing number of mega-caps holds up the aggregate market.
The following graphics take a closer look at the phenomenon and compare the historical development of market breadth with significant drawdowns to examine how well it has served as a leading indicator.
In line with the spirit of this blog, the attached PDF also provides more information on the "Bear-Market Shading" function used as well as a link to the underlying dataset (click here to directly get the GitHub link).
There are different ways of defining market breadth. One approach is to look at the equal-weighted index's performance relative to the market-capitalization-weighted index's performance. For example, the following chart illustrates the relative performance of the S&P 500 against the S&P 500 equal weight.
The grey stripes in the background (Bear-Market Shading) indicate drawdowns of more than 7% compared to the previous 252 Day high. It shades only the time from the beginning of the drawdown to the through.
The indicator seems to follow some longer-term trends, but while the outperformance of the market-cap weighted index often coincided with bear market conditions, it is hard to see how it could be used as a predictor.
Another more promising approach is to look at the percentage of index members reaching new highs. This example defines the latter as trading above the 200-day simple moving average. This indicator is naturally correlated with overall market conditions as the number of companies trading above their previous highs will, by definition, fall during a bear market.
The chart below uses the same Bear-Market Shading as the previous visualization. While it is not particularly easy to read, the correlation between the grey drawdowns and the blue percentage of companies trading above the medium-term moving average is pretty straightforward.
It also shows why today's market conditions are unusual and why market participants are probably rightfully worried. This bull market was exceptionally broad-based, with almost 100% of companies trading above their 200 days moving average at some point. However, market breadth has deteriorated at an unusually high speed since then. Nevertheless, it is also noteworthy that this indicator is notoriously unreliable.
A diminished number of index members can carry markets for a long time, and as can be seen above, similarly low levels of breadth often didn't translate into bear markets.
So what if we examine this a bit further by studying the times when conditions were most closely aligned with those observable today? The following chart shades all periods when the following conditions were met:
Over the past 21 years, and not counting today, this constellation occurred only four times, and it was always followed by a significant correction three times.
The average 90 days return following such conditions was -3.9% compared to +2.5% for all other periods. The number of samples increases if the last state is relaxed, allowing for similar periods when markets traded no more than 5% below all-time-high (second scatter plot). Still, mean and median return over the following three months stays negative at -1.8% and -1.6%. What's more, the standard deviation of returns realized over the 90 days following similar market conditions was lower than the standard deviation of returns observed over the same time horizon during all other periods.
How concerned should we be?
Probably more than usual!
As with any market timing indicator, market breadth is notoriously unreliable guidance, and investors should be suspicious of visualizations or publications raising the impression that a significant drawdown is inevitable. For a long-time, the post-Covid bull market was exceptionally broad-based. This could also justify the currently observable speed of the deterioration of breadth.
As the charts above show, the market breadth indicator tends to be rather contemporary than leading and can reverse quickly. Nevertheless, today's market conditions are exceptional on many levels (see also this post on the topic of increasing valuation dispersion) which makes any suspicious intra-market move a bit more frightening than it would probably be otherwise.
Over the past 21 years, similar a similar co-occurrence of buoyant markets and relatively low and declining market breadth has seldom occurred and was almost always followed by negative cumulative returns over the subsequent three months.
.jpg)
