The Value Spread

Abstract

We decompose the cross-sectional variance of ¢rms' book-to-market ratios using both a long U.S. panel and a shorter international panel. In contrast to typical aggregate time-series results, transitory cross-sectional variation in expected 15-year stock returns causes only a relatively small fraction (20 to 25 percent) of the total cross-sectional variance. The remaining dispersion can be explained by expected 15 -year pro¢tability and persistence of valuation levels. Furthermore, this fraction appears stable across time and across types of stocks.We also show that the expected return on value-minus-growth strategies is atypically high at times when their spread in book-to-market ratios is wide.INTUITIVELY, BOTH EXPECTED STOCK RETURNS and expected cash-£ow growth play a role in determining the market price of a ¢rm's stock and, thus, its book-to-market ratio. Firms with high book-to-market ratios have on average higher subsequent stock returns than ¢rms with low book-to-market ratios (Rosenberg, Reid, and Lanstein (1985), Fama and French (1992), and others). Simultaneously, di¡erences in ¢rms' book-to-market ratios are also related to di¡erences in future expected cash-£ow and earnings growth as well as future pro¢tability. Low-book-to-market ¢rms grow faster and are persistently more pro¢table than high-book-to-market ¢rms.The relative importance of these two contributing factors to cross-sectional variation in ¢rms' book-to-market ratios remains an open empirical question.We decompose the cross-sectional variance of ¢rms' valuation multiples using a long (1938 to 1997) U.S. panel and a shorter (1982 to 1998) international panel. Our variance decompositions show what fraction of the cross-sectional dispersion in book-to-market ratios is caused by variation in expected stock returns

Size, book-to-market ratio and macroeconomic news

“…29 The reader is referred to Campbell (1991) and Campbell and Shiller (1988a) for further details. 30 One can also use long horizon regressions instead of a VAR as in Cohen, Polk, and Vuolteenaho (2003) to decompose the returns. They argue that long horizon regression is a more appropriate approach for managed portfolios and estimating a VAR for the size-or BM-sorted portfolios might be problematic.…”

“…For example, the VAR approach might incorrectly link the dependent variables for small firms to independent variables for large firms. This is especially problematic when one uses annual data to estimate the VAR as in Cohen, Polk, and Vuolteenaho (2003). However, the violation of this assumption is less of a problem for monthly data.…”

Style Timing with Insiders

“…Because we used the FamaFrench (1993) definitions, these portfolios' returns are essentially identical to those reported on Kenneth R. French's website. 6 Previous research (see, e.g., Asness, Friedman, Krail, and Liew 2000;Cohen, Polk, and Vuolteenaho 2003;Zhang 2005) has demonstrated that the "value spread" (the relative valuations of growth and value stock portfolios) can be used to forecast the value premium. Thus, in our robustness tests, we included the value spread as an explanatory variable.…”

“…10 Several studies (e.g., Asness, Friedman, Krail, and Liew 2000;Cohen, Polk, and Vuolteenaho 2003;Zhang 2005) have reported that the relative valuations of growth and value portfolios (the "value spread") can predict the value premium. That is, when growth stock valuations are much higher than value stock valuations, the subsequent value premium is large.…”

The Stambaugh bias in panel predictive regressions

“…Traditionally, forecasting regressions have been evaluated in time-series frameworks. However, with the increased availability of data, in particular international …nancial and macroeconomic data, it becomes natural to extend the single time-series framework to a panel data setting; for instance, Cohen et al (2003) and Polk et al (2006) rely on predictive panel data regressions in some of their analyses.…”