DP17592 Book-to-Market, Mispricing, and the Cross-Section of Corporate Bond Returns
A corporate bond’s book value divided by its market price strongly predicts its return from actual transactions occurring at least eight days after observing the signal. Bonds with the 20% highest “bond book-to-market ratios” outperform their lowest quintile counterparts by 3%-4% per year, other things equal. The finding controls for numerous attributes tied to liquidity, default, microstructure, and priced asset risk, including yield, credit spread, structural model equity hedges, bond rating, and maturity. If an efficient markets story explained the 3%-4% spread, we would not observe (as we do) rapid decay in the ratio’s predictive efficacy with implementation delays beyond one month, efficacy across the bond-type spectrum, and an inability of microstructure, factor risk, and bond attributes to account for the anomaly.