Explaining Realized Returns: Roll’s R2 40 Years Later

Can we explain asset price movements even with ex post observations of their drivers? Reinterpreting Roll’s (1988) AFA Presidential Address, “R2,” as a modern empirical exercise, we show that conventional macroeconomic, cash-flow, expectations, and flow variables deliver out-of-sample R2 values that are often negative and otherwise below 10% for monthly U.S. equity returns and are unavailable at the daily frequency. We use generative AI to extract newspaper-stated reasons from Wall Street Journal market summaries and filter out price and return language. These reasons explain around 30–40% of daily and monthly returns out of sample. Their content maps to interpretable themes, including corporate news, macroeconomic conditions, monetary policy, sentiment, and geopolitics.