Published online by Cambridge University Press: 04 May 2022
We link equity and treasury bond markets via an informational channel. When macroeconomic state shifts are more probable, informed traders are more likely to have valid signals about fundamentals, so that uninformed traders are less willing to trade against informed ones. This implies low volume and high volatility, that is, a high volatility–volume ratio (VVR). Central banks react to state shifts, but their actions are uncertain. Therefore, a higher state shift likelihood implies larger bond risk premia. These arguments together imply that VVR should positively predict bond excess returns. We empirically test and confirm this prediction, both in- and out-of-sample.
We are grateful to Hendrik Bessembinder (the editor) and Sunil Wahal (the referee) for insightful and constructive comments. We also thank Andrea Eisfeldt, Jean-Sebastian Fontaine, Chotibhak (Pab) Jotikasthira, Phillippe Mueller, Pierre-Olivier Weill, participants at the 2017 Conference on the Econometrics of Financial Markets, the 2017 Behavioural Finance and Capital Markets Conference, the 2017 Finance Research Network Meeting, the seminars at University of Melbourne and University of Memphis, and the brown bag seminars in UCLA and UNSW for helpful comments. We are especially grateful to Kees Bouwman for long discussions on modeling bond premia. We are very grateful to Michael Fleming for providing the data on treasury market trading volumes, to Sydney Ludvigson for updating the LN factor data to allow for up-to-date analysis, and to Allan Timmermann for the monthly bond returns. This article supersedes the article titled “Stock Market Illiquidity, Funding Liquidity, and Bond Risk Premia.”