Do Historical Stock Market Returns Matter?

A Bold Claim on Stock Market Returns

A couple of weeks back, Financial Times published an article titled “Past Performance Is a Public Enemy.” To support that assertion, the author showed that, over the past 20 years, stocks that led the US equity charts in a given calendar year rarely repeated the feat during the next year. In fact, they tended to trail the averages.

That’s a fine warning for new investors who are tempted by Nvidia’s NVDA success, but not, I think, of much use to this column’s readers, most of whom learned long ago that trees do not grow to the sky. Also, the evidence is sorely incomplete. There are many ways to attempt to use stock prices to forecast future returns besides comparing calendar-year performances.

Happily, academic researchers have studied this topic for decades. This column will summarize their findings, when addressing the following question: Should one ignore past stock market returns when investing, as the Financial Times author stated, or can such information sometimes be useful?

Stage 1: Short-Term Prices

Almost 60 years ago, future Nobel laureate Eugene Fama debunked the promises of stock market chartists. In a landmark paper called “The Behavior of Stock-Market Prices,” Professor Fama demonstrated that “the series of price changes [among individual stocks] has no memory, that is, the past cannot be used to predict the future in any meaningful way.”

The academic community quickly embraced the paper, as it had long considered technical analysis to be a modern form of reading tea leaves, following astrology, or interpreting entrails. Its finding was, however, less universal than was initially believed. Fama had only considered daily movements (and for a limited number of stocks over a limited period, at that). Changes in equity prices could well be random over 24 hours but predictable over longer periods.

Stage 2: Long-Term Prices

Which, two decades later, another future Nobel laureate documented. In “Does the Stock Market Overreact?”, Richard Thaler and his co-author, Werner De Bondt showed that, when extending the analysis from one day to three years, the outcomes dramatically changed. This time, past performance truly was meaningful, as portfolios formed by buying the previous 36 months’ losers cumulatively outgained the overall stock market by almost 20 percentage points, while portfolios formed by the winners trailed the averages.

As the authors themselves admitted, their exercise tipped its hat to legendary value investor Ben Graham. Not directly, as Graham picked stocks by studying financial statements, while Thaler and De Bondt skipped the hard work by simply selecting the worst recent performers. But the two methods largely reached the same place…