Recency bias in fund picking
Walk into any UK retail investment platform's "top funds" page and you'll see funds ranked by recent returns. Buy from this list, and you're systematically buying yesterday's winners. The academic data is brutally clear: last year's top-quartile funds typically end up below-average over the next 5-10 years. This is recency bias in action — and it quietly costs UK retail investors 1-3% per year on the actively-picked portion of their portfolios.
What recency bias actually is
Humans systematically overweight recent information when making predictions. A fund that returned 35% last year feels intuitively "good"; a fund that returned 5% last year feels "average"; a fund that returned −5% feels "bad". The natural impulse is to buy the 35% performer.
The problem: recent fund performance is mostly noise, not signal. It reflects:
- Market conditions favouring the fund's style (e.g. growth-stock outperformance helps growth funds, hurts value funds — same skill, different style)
- Random luck over short periods
- The fund's specific concentration happening to align with what worked recently
- Sometimes: genuine skill (but this is hard to identify from short-term performance alone)
Buying based on 1-year return is mostly buying what's already been bid up — classic "buy high" behaviour.
The academic data: persistence is rare
S&P SPIVA scorecards
S&P publishes annual SPIVA (S&P Indices Versus Active) data on fund persistence. Key findings for the latest UK / Europe data:
- Year 1 top-quartile funds remaining top-quartile after 5 years: ~5-8% (vs random expectation of 25%)
- Year 1 top-quartile funds remaining in TOP TWO quartiles after 5 years: ~25-30% (vs random expectation of 50%)
- Year 1 top-quartile funds dropping to bottom quartile after 5 years: ~30-40%
Translation: a fund that performed well last year is more likely to perform BELOW average over the next 5 years than to repeat its top-quartile performance. The "winners keep winning" hypothesis is empirically false in fund management.
Morningstar's "Mind the Gap" studies
Morningstar's annual "Mind the Gap" report measures the difference between:
- Time-weighted fund returns (what a buy-and-hold investor earned)
- Money-weighted investor returns (what the average investor in the fund actually earned, accounting for when they bought and sold)
The gap is the "behavioural cost". Latest UK / European data:
- Average gap across all categories: ~1-2% per year investor underperformance
- Worst categories (sector / thematic / high-volatility funds): 3-5% per year underperformance
- Best categories (boring index trackers, balanced funds): close to zero gap — investors hold through cycles
The driver: recency bias makes investors buy after good performance and sell after bad. They consistently buy high and sell low at the fund level — even when the fund itself is producing good returns.
Dalbar QAIB studies (US, applicable to UK)
Dalbar's annual Quantitative Analysis of Investor Behavior shows similar patterns. Over 20-year periods, the average investor in US equity funds underperforms the S&P 500 by 3-5% per year — almost entirely due to bad timing decisions driven by recency bias.
The UK fund tournament — how to spot the trap
Look at any UK investment platform's "Top Performers" list. They're typically displayed by:
- 1-year return (most prominent)
- Sometimes 3-year and 5-year returns shown alongside
- "Most popular" lists derived from recent net inflows
These lists naturally surface whatever happened to do well recently. They are a recency-bias trap by design.
Real example: UK fund rotation 2020-2024
| Year | UK retail top sectors | Next-year performance |
|---|---|---|
| 2020 | Tech / growth (e.g. Scottish Mortgage IT, Baillie Gifford funds) | 2021: still strong; 2022: collapsed 40-60% |
| 2021 | Tech, smaller companies, ESG funds | 2022: most fell 30-50% |
| 2022 | Energy / commodities / value funds | 2023: mostly flat or negative |
| 2023 | AI / Mag-7 / S&P 500 momentum | 2024: continued strongly |
| 2024 | More tech, US equity, momentum | 2025-26: rotation pattern continues |
UK retail investors who bought 2020's top performers (Scottish Mortgage at £15+) lost roughly 50% by 2022 (£7-£8). Those who then bought 2022's top performers (BP, Shell at £6-£8/share) saw modest gains as energy peaked. Those who bought 2023's winners (US tech via NDX trackers) did well... so far.
The fundamental problem: by the time a fund or sector appears on "top performer" lists, the news that drove its performance is fully priced in. You're buying at the local maximum.
Why recency feels so rational
The bias is hard to shake because it FEELS like good evidence:
- "This fund has a 5-star Morningstar rating" — based on recent risk-adjusted returns; about half of 5-star funds drop to 3 stars within 5 years
- "This manager has a strong track record" — usually meaning a good 3-5 year period; survivorship bias means we only hear about managers whose period happened to be good
- "The fund's strategy is working in current conditions" — tautology; we know it worked because that's why we're looking at it
- "Industry experts are recommending it" — advisers and journalists are also subject to recency bias and need to recommend something
Can you detect genuine fund manager skill?
Academic research suggests genuine fund-manager skill exists but is rare and hard to identify from performance alone:
- 10-year return data is more reliable than 1-year, but still mostly noise
- Risk-adjusted measures (Sharpe ratio, alpha) help slightly — but only over very long periods
- Process-based assessment (how the manager makes decisions, transparency, fee structure) is more informative than past returns — but harder for retail to evaluate
- Even when skill exists, it's usually small (£100-£300 alpha per year) and gets eaten by fees
The honest conclusion: very few retail investors can reliably identify skilled managers in advance. The expected value of fund-picking based on recent performance is negative.
Why passive index funds are the default answer
For UK retail, the standard advice is to default to low-cost index funds (VWRL, CSPX, SWDA, etc.) precisely BECAUSE they:
- Don't try to pick winners
- Don't suffer recency-bias-driven manager changes
- Charge low fees (0.07-0.22% OCF)
- Are typically held through full cycles — minimal investor behavioural gap
The cost of NOT recency-biased fund-picking: roughly the OCF differential between active funds (often 0.7-1.5%) and index funds (0.1-0.2%). On a £100,000 portfolio over 30 years, that's roughly £30,000 of cumulative cost savings, plus avoidance of the 1-3% behavioural underperformance that comes with active fund-picking.
If you must pick active funds — better heuristics
For investors who genuinely want active management (some sectors, some strategies), here are better-than-recency heuristics:
1. Low cost first
Filter out everything above 0.5% OCF unless there's a specific reason. Cost is the most reliable predictor of long-run net performance (in any category).
2. 10+ year track record, same manager
If the manager changes, the previous track record means nothing. Look for managers with 10+ years on the same fund. Even then, 10 years is a small sample.
3. Look at process, not results
Can you understand the fund's process? Is it consistent and disciplined? Or does it look like recent results explained backwards?
4. If buying actively, time it contrarian
Want to buy a value fund? Buy after value has had a bad period (e.g. value funds in 2020 after years of underperformance). Want a small-cap fund? Buy after small caps have lagged (e.g. UK small caps in 2024 after a brutal 2022-2023). Buying after underperformance is the recency-bias antidote.
5. Diversify across managers if going active
If you're committed to active management, holding 5-8 funds rather than 1-2 reduces the impact of any single fund's recency-driven mistakes. But this defeats much of the appeal of active management.
Anti-recency checklist
Before clicking "buy" on any fund:
- ✓ Am I buying this because of its 1-year or 3-year return? (If yes, pause.)
- ✓ Has this fund / sector / strategy been featured prominently in financial media recently? (If yes, the news is priced in.)
- ✓ Would I still buy if its 1-year return was −15% instead of +25%? (If no, I'm buying recency, not the fund.)
- ✓ Can I explain WHY this fund will outperform over the next 10 years, not just the last 1-3? (If no, I'm guessing.)
- ✓ Is the OCF below 0.5%? (Higher OCF compounds against you regardless of skill.)
- ✓ Would a global index fund (VWRP) serve the same purpose at lower cost? (Usually yes.)
Worked example: the real cost of recency-driven picking
Two UK retail investors, each contributing £500/month to an ISA from age 30 to 60 (30 years, £180,000 total contributions).
Investor A: Buys whatever's top-of-the-pops
- Changes funds every 2-3 years based on recent winners
- Pays average 1.0% OCF on actively-managed funds
- Suffers ~2% per year behavioural gap (buying high, selling low)
- Gross market return: 7%/year
- Net return after costs + behavioural gap: 4%/year
- End-of-30-years portfolio: ~£348,000
Investor B: VWRP and forget
- Buys VWRP at outset, never changes
- Pays 0.22% OCF
- Zero behavioural gap (rebalances mechanically, doesn't react to news)
- Gross market return: 7%/year
- Net return: 6.78%/year
- End-of-30-years portfolio: ~£596,000
The gap
Investor B is £248,000 wealthier — despite making no skilled decisions, no clever fund picks, no market-timing calls. Just: low cost + no recency-biased behaviour. The "do nothing" approach beats the "stay on top of it" approach by a factor of 1.7x.
This isn't a hypothetical — it's roughly the gap that Mind the Gap, Dalbar, and SPIVA studies consistently find between disciplined passive investors and active fund-pickers.
Frequently asked questions
What about funds with consistent multi-decade outperformance?
They exist but are rare. The most famous (Magellan, Berkshire, Lindsell Train UK Equity) all eventually went through extended weak periods. The challenge for retail: identifying genuine 30-year skill ahead of time is essentially impossible. Once a manager is famous for their skill, their fund is usually expensive and crowded.
Should I sell my recency-bought funds and switch to index?
If you bought based on recent performance, the underlying recency-bias mistake is past. The question is whether the fund holds you back going forward. Generally: yes, switching to a low-cost global index fund is the standard recommendation. Watch for CGT implications outside an ISA/SIPP and any fund exit fees.
Is some recency information useful?
Yes — recent extreme losses are sometimes a signal of fund problems (manager change, strategy drift, scandal). But "recent strong performance" is rarely useful as a buy signal. Asymmetric: pay attention to recent BAD news, ignore recent GOOD news.
What about thematic ETFs (AI, clean energy, biotech)?
Thematic ETFs are recency bias products. They launch after a theme has had good performance, attract flows at peak enthusiasm, and typically underperform broad markets going forward. The data is consistent across decades and themes. Avoid unless you have a specific conviction backed by analysis beyond "AI is the future".
What if I want exposure to specific themes anyway?
If conviction is real, do it through individual companies (not thematic ETFs), with limited position sizing (1-3% of portfolio per theme), and with explicit acceptance that this is speculative/non-core allocation. Don't dress up a momentum bet as "diversified investing".
Related guides
How UK Tax Drag holds itself to account
Every page is reviewed against the editorial standards, written from primary sources, sourced openly, and corrected publicly. No affiliate revenue. No sponsored content. No paid placements.