Recency Bias
Whatever happened last feels like what will always happen next.
Simple Definition
Recency bias is the tendency to give recent events disproportionate weight when forming expectations — treating what just happened as the most reliable guide to what will happen next, even when long-run history tells a very different story.
Why Humans Behave This Way
The brain is a pattern-recognition machine, and it prioritises fresh data. Evolutionarily, this made sense: if you saw a predator in a certain location yesterday, the probability of finding a predator there today is higher than average. The recent event was genuinely informative. In stable environments with short feedback cycles, recency bias is an efficient shortcut. In financial markets — where conditions change over decades, cycles are long, and recent data is often the least predictive of long-run outcomes — this shortcut becomes a liability.
Psychologists explain recency bias partly through the "availability heuristic" — the tendency to judge the probability of events by how easily they come to mind. Recent events are more mentally available than distant ones. A crash that happened three months ago feels more "real" and probable than one that happened twenty years ago, even though the long-run data on crash frequency may assign them equal probability. Emotional vividness compounds the effect: the memory of recent financial pain or pleasure carries far more psychological weight than statistically identical events from the distant past.
Recency bias interacts destructively with the structure of financial returns. Equity markets experience long periods of strong returns punctuated by sharp, relatively brief downturns. Recency bias causes investors to extrapolate the bull phase into permanent prosperity (driving valuations dangerously high at peaks) and to extrapolate the bear phase into permanent decline (driving valuations dangerously low at troughs). Both errors are mirror images of the same cognitive flaw: treating the most recent experience as the most probable future state.
Everyday Example
Your city has had an unusually cold July for three years in a row. You conclude that summers are now permanently colder and cancel your summer holiday plans.
Three years of data has overridden decades of meteorological history. The same cognitive error occurs in financial markets constantly: three years of rising prices feels like a permanent new reality; three consecutive monthly losses feels like a trend that will never reverse. In both cases, recent experience has been awarded more predictive power than it actually possesses.
How It Affects Investors
Recency bias manifests most visibly in flows into and out of investment funds. After sustained bull markets, fund inflows are highest — investors extrapolate recent returns forward and expect them to continue. After sustained bear markets, fund inflows are lowest or negative — investors extrapolate recent losses forward and expect them to continue. Data from major equity markets shows that fund inflows peak near market tops and trough near market bottoms, consistently.
In bear markets, recency bias is lethal to long-term plans. An investor who has watched their portfolio decline for twelve months has twelve months of recent evidence that "markets go down." Their expectation for the next twelve months is shaped primarily by this recent experience. They sell — exactly when long-run data suggests they should be buying more. The historical base rate of equity market recovery from bear markets is dramatically more positive than recency bias allows investors to believe in the moment.
In bull markets, recency bias creates dangerously optimistic return expectations. Investors who have experienced five years of strong equity gains begin to assume that 15–20% annual returns are normal, fail to maintain adequate cash reserves, take on leverage, and reduce their asset allocation to bonds. When the inevitable mean-reversion arrives, these investors are maximally exposed at the worst moment.
Professional investors and fund managers are equally susceptible. Research shows that analyst earnings estimates exhibit systematic recency bias — following strong earnings quarters, analysts over-estimate future earnings growth; following weak quarters, they over-estimate future earnings declines. This creates predictable patterns of estimate revision that sophisticated investors can exploit, but which most investors are instead harmed by.
How It Damages Wealth
- 1
Selling after crashes: recency bias makes recent losses feel like a permanent new state, driving investors to exit equities at precisely the moment long-run history suggests they should be holding or adding.
- 2
Buying after bull runs: extrapolating recent strong returns produces dangerously optimistic expectations, leading to overexposure to equities at peaks when valuations are stretched and prospective returns are lowest.
- 3
Abandoning long-term plans: one or two bad years of returns on a fifteen-year investment plan feel like evidence the plan has failed, when they are statistically expected variations in any long-run equity journey.
- 4
Chasing recent winners: performance-chasing — moving money into the asset class or fund that performed best last year — is recency bias in pure form. Studies consistently show that last year's top-performing assets are among the weakest performers over the subsequent three years.
- 5
Over-weighting recent economic narratives: an investor who heard "inflation is permanent" repeatedly in 2022 may structurally under-weight assets that outperform in declining-inflation environments for years afterward, missing a full recovery cycle.
How To Avoid This Bias
Study long-run market history before making any asset allocation decision. Read about how markets have behaved over full cycles — booms, crashes, recoveries — not just the last three to five years. Data from the full 20th and 21st centuries provides a far more accurate base rate than any recent period.
Anchor your expectations to long-run historical averages. Global equity markets have historically delivered roughly 7–10% annualised real returns over very long periods. Use these long-run averages as your baseline expectation, not recent performance.
Review your investment plan before selling after a crash. Ask: does the recent decline change my fifteen-year investment thesis? If the answer is no — and for diversified long-term investors it almost never should be — recency bias is the likely driver of the sell impulse.
Track performance-chasing explicitly. Keep a record of any fund or asset class moves you make based primarily on recent performance. Review the subsequent one and three-year outcomes. Most investors find their performance-chasing decisions systematically underperform their buy-and-hold decisions.
Use a written investment policy statement that specifies your asset allocation and rebalancing rules in advance. Mechanical rebalancing — selling what has recently done well and buying what has recently done poorly — forces you to act against recency bias systematically.
Discuss major decisions with someone who has lived through at least one full market cycle. Investors who experienced the 2008 crash, the 2000 dot-com bust, or the 1987 crash have a richer base of historical experience that counteracts the power of recent data.
Read historical accounts of past market panics during a current market panic. The similarity between the emotions investors reported during the 1929 crash, the 1987 crash, and the 2020 crash is striking — and serves as a powerful reminder that the current experience is not unprecedented.
Frequently Asked Questions
Key Takeaways
- 1
Recency bias treats the most recent market experience as the most reliable guide to the future — but long-run data is almost always a more accurate predictor of long-run returns than recent performance.
- 2
Performance-chasing — moving capital into last year's best-performing assets — is recency bias in its most wealth-destructive form, systematically capturing bad performance after missing good performance.
- 3
The antidote is anchoring expectations to long-run historical averages, not recent results, and having a written investment policy that defines rebalancing rules in advance.
- 4
In bear markets, recency bias is most dangerous: recent losses feel permanent when history shows they are almost always temporary for diversified long-term investors.
- 5
Reviewing long-run market history during a market panic is one of the most reliable emotional antidotes to recency bias — the similarity between past panics and current ones is a powerful reality check.
Related Concepts
Put knowledge into action: