Is Value Back
Value stocks posted their seventh best day in 20 years on September 10, while momentum stocks had their worst two-day stretch in a decade. Our friends at Alpha Architect crunched the numbers and concluded this was a five-sigma event for value and momentum stocks.
Financial commentators were quick to call it a paradigm shift. “Is the great value-stock rotation finally upon us?” asked Barron’s. “Value stocks are acting like a tightly wound spring that has started to uncoil.”
Though we are delighted by this development, we have no crystal ball on whether this really is the start of a big shift or merely a temporary respite from the grinding 12-year period of growth outperforming value.
And though we can’t say if the spring has really started to uncoil, we can examine the historical data to understand just how tightly that spring is wound.
The last 12 years have been brutal for value relative to growth. This recent stretch of market history has been extreme by historical standards. Below are the trailing 12-year compounded relative returns for every month dating back to the Great Depression for the cheapest 30% of the market minus the most expensive 30% of the market (in blue) for both value-weighted and equal-weighted portfolios from the Ken French library. We also include Value minus the overall market (in orange).
Figure 1: Trailing 12 Year Value Relative Performance (Value Weighted)
Source: Ken French Library
Value-weighted portfolios of the cheapest stocks have underperformed growth portfolios by 151.8% in the last 12 years, which was only exceeded briefly at the peak of the dot-com bubble. What’s more, the same portfolio has underperformed growth in 9 of the last 12 calendar years, which has never happened historically.
Value’s underperformance is primarily a result of the changing multiples for growth and value stocks. Growth stocks are now more expensive than value stocks by margins not exceeded since the peak of the dot-com bubble and the Great Depression. In both of those other two extreme periods of market history, valuation spreads between the most expensive and cheapest stocks had widened significantly. And in both cases, while the ride up the spreads was painful for value relative to growth, when spreads eventually mean reverted, value dramatically outperformed.
Figure 2: Market to Book Value Spreads and 3-Yr Fwd Relative Returns for Growth and Value
Source: Ken French library. Spreads on BE/ME for top 30% growth and top 30% value stocks. Relative returns are on the same 30% portfolios for value – growth over the next three years.
Implicit in high relative spreads are high relative expectations. In the last 12 years of steadily rising relative valuations, surprisingly little has changed for value stocks. They have compounded about +74% over those 12 years and remain at about the same trading multiples the’ve been at througout the growth cycles. During these 12 years, value has traded at about 1.1x book and around 7x EBITDA while growth has gone from 2.5x book and 11x EBITDA to 4x book and 18x EBITDA.
Figure 3: Absolute Valuation Multiples of Growth and Value Stocks in the US over Time
Source: Capital IQ. All listed North American stocks by 25th percentile, median, and 75th percentile valuation breakpoints.
Growth stock financial statements have dramatically outperformed during this growth rally, so the multiples have also risen dramatically, baking in an expectation about the future drawn from these recent past trends.
But the higher expectations get, the less likely they are to be met in an uncertain future. And the higher relative expectations become, the less likely relative performance expectations are to be met in an uncertain future.
Expectations have grown ever higher for growth, causing value investors to worry that the last 12 years of value’s relative underperformance, which falls well into the far extremes of historical data in its duration, scale and consistency, might be the signal that value investing no longer works.
But value investing is not plausible simply because it fits a very specific, knowable “true” trend that we can calculate with perfect percision. To the contrary, it has probably worked and worked best when other market participants do lose faith in it from time to time.
And value has actually done fine over the 12 years on it’s own, compounding annually in the high single digits. What’s more exceptional is the distinctive rise in expectations for growth stocks. On a historical basis, we can see that this rise in expectations has coincided with most periods of value relative underperformance. This phenomenon has been one of the most consistent historical signals of decreased returns to growth on a forward-looking basis.
And for those who remain fully invested for the long haul, unpopular assets after periods of poor relative returns and measurably lower relative expectations have most regularly made for the best long-term investments. Value is, admittedly and encouragingly, at a very rare historical extreme combination of those attributes right now: completely unpopular, at a crisis of faith, deprived of short-term results, and at extreme low relative valuations compared to much more popular segments of the market. These are the ingredients of mispricing.