Recency Bias
Recency Bias is a cognitive bias in which investors and decision-makers place disproportionate weight on recent events, trends, or data points relative to longer historical evidence, leading them to extrapolate near-term patterns indefinitely into the future, overreact to recent performance (either positive or negative), and underestimate the likelihood of mean reversion to long-run historical norms. In financial markets, recency bias drives momentum in investor behavior, contributes to bubble formation during bull markets, and exacerbates panic selling during downturns.
Key takeaways
- Recency bias causes investors to over-allocate to recently outperforming asset classes and strategies, often near market peaks, and to under-allocate after drawdowns, near troughs.
- The phenomenon is documented in fund flow data: investors consistently pour money into equity funds after strong market performance and redeem during downturns, buying high and selling low.
- Performance chasing — selecting funds based on recent returns rather than long-term risk-adjusted metrics — is a manifestation of recency bias that typically produces below-average results for investors.
- Irrational exuberance, as described by Robert Shiller, is driven partly by recency bias: investors extrapolate recent high returns indefinitely and assume that stocks 'always go up.'
- Professional investors are not immune: portfolio managers' return expectations measured by surveys move with recent market performance, consistent with systematic recency-driven updating.
Explanation
Recency bias emerges from a fundamental constraint of human cognition: the availability heuristic. Psychologists Tversky and Kahneman demonstrated that people estimate the probability of events based partly on how easily examples come to mind — recent events are more vivid and accessible in memory than distant ones, inflating their perceived probability. In a financial context, an investor who has just experienced three consecutive years of double-digit equity returns finds it natural to assume that high returns will continue, while someone who has just experienced a severe bear market struggles to imagine a recovery, even if the long-run base rate strongly suggests one.
The consequences of recency bias at the portfolio level are well-documented and economically significant. Dalbar's annual Quantitative Analysis of Investor Behavior study consistently finds that the average U.S. equity fund investor earns roughly 2–4 percentage points less per year than the funds they invest in, primarily because they buy after strong recent performance and sell after losses. This 'behavior gap' represents a massive destruction of long-term wealth relative to the returns available to disciplined investors who maintain their allocations through market cycles.
In professional investment management, recency bias manifests in manager selection, risk budgeting, and scenario planning. Investment committees that review recent performance and adjust allocations accordingly can inadvertently inject recency bias into institutional processes — increasing risk exposure near peaks and cutting it near troughs. Strategic asset allocation frameworks are designed partly to counteract recency bias by establishing long-horizon return assumptions anchored to fundamental valuation metrics (CAPE ratios, credit spreads, yield levels) rather than recent realized returns.
The interaction between recency bias and mean reversion bias creates interesting psychological dynamics. In trending markets, recency bias dominates: investors extrapolate trends. After large drawdowns, recency bias temporarily reinforces pessimism, but as the market begins to recover, mean reversion thinking may reassert — investors expect a 'snapback' to prior levels. The balance between these two biases partially determines the speed and character of market recoveries. Quantitative strategies that exploit these behavioral patterns — such as intermediate-term momentum (12-1 month) and long-term mean reversion — are essentially harvesting the systematic nature of recency bias in aggregate investor behavior.
Example
During the 2017–2021 period, U.S. growth and technology stocks dramatically outperformed value stocks, generating five-year cumulative returns of over 200% in the Nasdaq 100 versus roughly 80% in the Russell 1000 Value Index. By early 2021, U.S. institutional and retail investors had sharply increased their growth allocations, with the typical 60/40 portfolio heavily overweight U.S. large-cap growth. This concentration reflected recency bias: investors extrapolated the recent dominance of growth stocks indefinitely. In 2022, rising real interest rates triggered a sharp rotation: the ARK Innovation ETF (a proxy for speculative growth stocks) fell over 70%, while value stocks broadly outperformed. Investors who had chased recent growth performance entered 2022 with the highest growth allocations in a generation, precisely when the regime was reversing.
Related terms
Asset Allocation Availability Heuristic Cap Confirmation Bias Equity Fear And Greed Index Irrational Exuberance Mean Reversion Mean Reversion Bias Mental Accounting Quantitative Analysis Speed