We examine 24 global factor premiums across equity, bond, commodity, and currency markets via replication and out-of-sample evidence between 1800 and 2016. Replication yields ambiguous evidence within a unified testing framework that accounts for p-hacking. Out-of-sample tests reveal strong and robust presence of the large majority of global factor premiums, with limited out-of-sample decay of the premiums. We find global factor premiums to be generally unrelated to market, downside, or macroeconomic risks in the 217 years of data. These results reveal significant global factor premiums that present a challenge to traditional asset pricing theories.
Bibliographical noteFunding Information:
We would like to thank an anonymous referee for his great contributions in improving the paper. We also thank David Blitz, Andrea Frazzini, William Goetzmann, Winfried Hallerbach, Matthias Hanauer, Matti Keloharju, Ralph Koijen, Peter Koudijs, Martin Martens, Lasse Pedersen, Gertjan Verdickt, Jeroen van Zundert, and seminar participants at the Research in Behavioral Finance Conference (Amsterdam), Hanken School of Economics, Maastricht University, Oslo Business School, Southwestern University of Finance and Economics, Technical University Munich, and Vaasa University for valuable discussions and useful feedback. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Robeco Institutional Asset Management is a global investment management firm, which could or could not apply similar investment techniques or methods of analysis as described herein. The views expressed are ours and not necessarily those of Robeco Institutional Asset Management.