Systematic Relative Strength

Not too long ago, we showed a market fair value estimate done by the analysts at Morningstar. Another take on it comes from Kelley Wright at Investment Quality Trends. Their valuation method is based on where dividend yields are for individual companies, based on the typical range for that particular company. What’s nice about this particular measure of valuation is that it has been statistically tested. A couple of excerpts from the story by Mark Hulbert at Marketwatch:

Wright classifies each of the stocks that make the grade into four categories according to how its current dividend yield compares to the historical range of that stock’s yield.

The “Undervalued” category contains those stocks with yields at or close to the high end of their respective ranges, while the “Overvalued” category contains stocks with relatively low yields.

The statistic that is most relevant to market timers, I found upon analyzing the data, is the percentage of stocks that make it into this “Overvalued” category. At the 95% confidence level that statisticians often use to determine if a correlation is genuine, this Overvalued percentage is inversely related to the stock market’s return over the next several years.

That is, a high percentage of Overvalued stocks is a bad omen, while a low percentage is a good one.

Currently, Wright’s Overvalued category contains just 35 of the 254 stocks in his dataset, or 13.8% of the total.

That is well below the five-decade average of 21%. In fact, it is lower than 69% of comparable readings back to the mid 1960s.

…

My finding suggests that the stock market is poised to produce above-average returns over the next couple of years.

It’s interesting to me that sentiment is so negative toward equities that stock funds have been seeing outflows, despite a number of measures suggesting that stocks may have reasonable returns going forward.