“The past is never dead. It’s not even past.” -William Faulkner

Academics have been trying to explain stock returns for decades with ever more elaborate models. The classic story—Markowitz portfolios, CAPM betas, Fama-French factors—assumes investors rationally update beliefs when new information arrives. While a massive improvement on what came before (nothing), that tidy picture was never quite right. Enter behavioral finance, which points out that humans are messy. We chase hot stocks, panic at the bottom, and overweight vivid stories.

The paper, “Memory and the Cross-Section of Stock Returns” by Aynur Tosun takes the behavioral idea one step further: it argues that investors don’t just misinterpret new information—they literally misremember the past. Our brains retrieve prior experiences based on cues (recent volatility, a big volume spike, or a return that “feels familiar”), and those retrieved memories distort expectations. The result is systematic mispricing: stocks investors “recall fondly” tend to get overvalued and then underperform, while those buried in mental filing cabinets get undervalued and outperform.

Memory as a Trading Model

The core idea is one borrowed from cognitive psychology: associative memory. You don’t remember all past experiences equally; instead, you recall the ones most similar to what’s happening now. If a stock just had a sharp up month, you’re more likely to recall other times it surged—and then anchor on the returns that followed.

The author operationalizes this with a new metric called Memory-Based Return Expectation (MEM). Four versions are built, depending on the “cue” investors might latch onto:

• MEM1: Last month’s return.

• MEM2: Return volatility

• MEM3: Abnormal trading volume

• MEM4: A combination of all three

To construct MEM, the past 15 years of stock-level data are searched for months that resemble today’s cues. Similar months are weighted more heavily, and the one-month-ahead returns following those months are averaged. That average becomes today’s memory-based expected return.

If investors overweight MEM, they’ll chase high-MEM stocks (which then underperform) and avoid low-MEM stocks (which then outperform).

The Evidence: Memory Hurts Your Portfolio

The paper runs the usual battery of asset pricing tests: sorts, regressions, robustness checks. The punchline: high-MEM stocks earn lower future returns. Further, low-MEM stocks (which earn higher returns) are not riskier by conventional metrics. If anything, they have lower downside betas and tail risk.

In other words, this doesn’t seem to be a risk premium. It’s mispricing driven by how brains dredge up past experiences.

Why It Matters for Traders

For quants, MEM is a shiny new anomaly. You can sort stocks on memory-based expectations and build a contrarian strategy. For behavioral finance folks, it’s more evidence that markets are shaped not just by what information investors see, but by what their brains choose to recall.

But there’s a deeper lesson: the market doesn’t just misprice risk, it misprices memories.

Think about it. If investors overweight the wrong “mental samples,” they’ll systematically overvalue stocks that look like past winners. MEM is essentially a formalized version of the old trader’s adage: “the market has a long memory for rallies, short memory for crashes.”

Practical Takeaways

1. Don’t trust your recall. If you’re mentally anchoring on “what happened last time,” odds are your memory is biased toward similar, vivid episodes.
2. Look where memory is weakest. The alpha comes from low-MEM stocks—names investors’ brains aren’t readily retrieving.
3. Expect noise in small caps. The effect is magnified where arbitrage is costly. Translation: if you’re trading microcaps, other people’s faulty recall is half your edge.
4. MEM isn’t momentum. In fact, it predicts the opposite—high memory-based expectations lead to lower returns. It’s contrarian, not trend-following.

But obviously, don’t expect this to last forever. If MEM makes its way into the quant factor zoo, returns will likely get arbitraged down. But the underlying truth—that humans overweight similar past experiences—will keep shaping markets in various ways.

Conclusion: The Market’s Faulty Hard Drive

Finance has always been about balancing models of rationality with the reality of human behavior. This paper reminds us that investors don’t have infinite hard drives storing objective histories. We have faulty, biased memory banks.

Stocks that trigger “this feels familiar” moments in investors’ minds get mispriced. MEM formalizes that intuition and shows the effect is real, significant, and persistent.

So, the next time you hear an investor justify a trade by saying, “Last time this happened, the stock ripped higher,” remember — they’re just consulting their brain’s messy archive. The smart move might be to bet against them.

Disclaimer

This document does not constitute advice or a recommendation or offer to sell or a solicitation to deal in any security or financial product. It is provided for information purposes only and on the understanding that the recipient has sufficient knowledge and experience to be able to understand and make their own evaluation of the proposals and services described herein, any risks associated therewith and any related legal, tax, accounting, or other material considerations. To the extent that the reader has any questions regarding the applicability of any specific issue discussed above to their specific portfolio or situation, prospective investors are encouraged to contact HTAA or consult with the professional advisor of their choosing.

Except where otherwise indicated, the information contained in this article is based on matters as they exist as of the date of preparation of such material and not as of the date of distribution of any future date. Recipients should not rely on this material in making any future investment decision.