Articles - Journaltalk

One might suppose that a theory of the stock market that has clearly not worked at all for over two decades might, in financial circles, at last be widely understood to be invalid. We should not, then, have investment companies and financial advisors maintaining portfolios based on the theory. One such theory is that of stock market momentum, which states that stocks that have performed well in the past tend to continue to perform well in the future. I chart portfolio cumulative returns versus time, graphically revealing momentum's demise circa year-end 2002, as well as that of two other celebrated factors, market capitalization and the book-to-market ratio, less than a decade later. This article is at odds with articles that were published in 2014, 2021, and 2024 by fund manager Clifford Asness and by Eugene Fama and Kenneth French. Those articles acknowledge some of the underperformance by the factors but hold out for their continued viability. But admitting that there has been a decline in factor profitability is one thing; rigorously testing for ``regime change'' is another. I reuse the hypothesis testing approach of Fama and French but apply it to datasets of greater relevance and demonstrate, by that and other means, that said declines in factor performance are actually gross and reflect abrupt structural changes in the market of the sort that, together, constitute what is called regime change. I also relate the present article to several previous articles that, in their own way, also bark up the regime-change tree.