THE best test of a value model like BV is how it fits real market history. How does it fit? Is the fit close, or loose? Does the model differentiate between normal, euphoric and dejected markets?
Of the value models based on traditional ratios like price/earnings, price/equity and price/dividends, price/earnings fits the closest. But the PE ratio can mislead investors at market lows, when earnings collapse, leaving PEs as high in the trough as they were at the peak. This amounts to an invitation into Pitfall #4, refusing to buy because the market isn’t cheap enough, and missing the first upmove.
To address this shortcoming, some investors calculate a cyclically-adjusted PE ratio. They calculate a long-term moving average of earnings, usually ten years, to smooth out peaks and valleys. Thus adjusted, the PE ratio no longer reacts to economic swings.
On the chart above, I show the cyclically–adjusted PE ratio (violet) since 1990, along with the regular PE ratio (green) and BV (white) — each ratio solved for price, or capitalized, to indicate where each ratio finds fair value.
Notice how PE deceives. It shows fair value falling in 2001 and 2002, even though the market was heading to a cyclical low and becoming cheaper. Later on, in 2007, PE said the market was not far above fair value, despite four and a half years of appreciation. In short, the PE ratio said the market was expensive at the low and cheap at the high.
Compared to this less than helpful performance, the CAPE ratio was as regular as a railroad. Investors might assume this is an improvement over PE’s bad driving. But note how poorly the CAPE ratio fits the actual market; it suffers from underestimation. In fact, the CAPE ratio stopped modelling reality back in the 1980s.
For further comparison between Brock Value and other valuation ratios, view BV vs Classic Ratios 1919-2014 and 1980-2014 (below).