History Rhymes Premium?

Does economic history rhyme in that similar economic/financial conditions precede similar equity factor performance? In their March 2025 paper entitled “Regimes”, Amara Mulliner, Campbell Harvey, Chao Xia, Ed Fang and Otto Van Hemert present a way to characterize the current economic/financial regime and relate this characterization to future factor returns. They consider seven input variables: (1) S&P 500 Index level; (2) 10-year U.S. Treasury note yield minus 3-month U.S. Treasury bill (T-bill) yield; (3) WTI crude oil price; (4) copper price; (5) T-bill yield; (6) VIX level; and, (7) U.S. stocks-bonds 3-year daily correlation. Specifically, they each month:

1. Compute for each input variable its annual change and its z-score relative to a rolling 10-year history, constraining the z-score to a range of -3 to +3.
2. For each variable, measure the similarity of its current z-score to its past z-scores for all dates based on squared difference.
3. For each date, add the squared differences across the seven input variables to compute a global similarity score for that date (lowest values are most similar to now).
4. For each past date, measure next-month gross performance of each of six long-short equity factors (market, size, value, profitability, investment and momentum).
5. Sort past dates into fifths (quintiles) based on global similarity scores relative to the current date.
6. For each quintile of past dates and each factor, take for the current date a long (short) position in the factor if its average next-month past performance is positive (negative) and reform an equal-weighted portfolio of the resulting six factor positions for the next month.

They focus on the quintiles of past dates with the most and least global similarities to the current date. Using daily and monthly data for the seven economic/financial input variables and for the six equity factors during 1985 through 2024, they find that:

(more…)