A new tight and general bound on return predictability

We propose a novel upper bound on the predictability of asset returns. This bound is tighter than the bound proposed by Ross (2005) because it takes into account not only the volatility of the pricing kernel but also the correlation between the pricing kernel and trading strategies that exploit predictability. It is also at least as tight as the bound proposed by Huang and Zhou (2017). We apply our bound to study the predictability of returns on currencies of emerging and developed economies from 1994 to 2016. We find evidence of return predictability in excess of the bound, especially for emerging markets currencies. This implies either market inefficiency or, alternatively, that investors either can become very risk-averse or price currencies using a model radically different from the CAPM. In contrast, the evidence of excess-predictability is much weaker under the wider bound proposed by Ross (2005).