Bengen's 4% rule (William Bengen, 1994) found that a US retiree could withdraw 4% of starting portfolio value in year 1, adjusted for inflation each year, with very high probability of not running out over 30 years. Applied to UK conditions, the safe withdrawal rate is typically 3.5-3.8% — slightly lower than the US 4% — due to higher historical UK inflation and lower long-run real equity returns. For a £500,000 UK pension portfolio at 60/40 allocation, this translates to £17,500-£19,000/year of inflation-adjusted spending, sustainable for 30+ years with ~90-95% probability based on historical data.
The original Bengen research
William Bengen's 1994 paper analysed every 30-year retirement period from 1926-1993 in US data. He tested various withdrawal rates and asset allocations. Key findings:
- A 60/40 stock/bond portfolio could sustain a 4% inflation-adjusted withdrawal for 30 years in 100% of historical 30-year windows.
- 4.5% worked in 96% of windows.
- 5% worked in 76% of windows.
- Optimal allocation: 60-75% equities for maximum survival probability.
Subsequent research (Trinity Study, Otar) extended and refined these findings — generally confirming 4% as a reasonable starting safe withdrawal rate for US retirees.
Why UK is different
| US (Bengen's data) | UK | |
|---|---|---|
| Long-run real equity return (1926+) | ~7% | ~5-6% |
| Long-run inflation | ~3% | ~3.5-4% (higher due to oil shocks, 1970s) |
| Bond return (long-run real) | ~2.5% | ~2.0-2.5% |
| Tax efficiency of withdrawals | 401(k) tax-deferred | Pension drawdown taxed at marginal rate above 25% lump sum |
| Healthcare cost burden | Significant retiree expense | NHS reduces healthcare cost risk |
| Currency / FX exposure | Mostly USD-denominated | GBP base, often global investments |
Net effect: UK retirees historically face slightly higher inflation pressure and slightly lower real returns. The safe withdrawal rate adjusts down accordingly — from 4% to roughly 3.5-3.8%.
The mechanic — how it actually works
- Year 1: withdraw 3.7% of starting portfolio value. Example: £18,500 from a £500k portfolio.
- Year 2: withdraw the same £ amount as year 1, adjusted for inflation. If inflation was 3%, year 2 withdrawal: £19,055.
- Year 3: same inflation adjustment from year 2. If inflation was 2.5%: £19,531.
- ...continuing for 30 years.
The strategy: portfolio invested for growth (60-75% equity), withdrawing income at a sustainable rate, surviving sequence-of-returns risk by maintaining the strategy through bad market years.
What 3.7% looks like in practice
£500,000 portfolio, 65-year-old, 30-year horizon
| Starting portfolio value | £500,000 |
| Year 1 withdrawal (3.7%) | £18,500 |
| Year 5 withdrawal (assumed 3% inflation each year) | £20,851 |
| Year 10 withdrawal | £24,217 |
| Year 20 withdrawal | £32,569 |
| Year 30 withdrawal | £43,801 |
| Cumulative real (today's £) spending over 30 years | ~£555,000 |
Portfolio expected to grow at ~5% real per year (60/40 allocation in UK historical data). Survives in roughly 90-95% of historical 30-year periods.
Sequence-of-returns risk — the killer
Bengen's findings assume the retiree stays the course through bad markets. The danger: bad market years EARLY in retirement deplete the portfolio so fast it can't recover.
Two retirees, same average returns over 30 years:
- Retiree A: bad returns years 1-3, good returns years 28-30. Portfolio drops 35% in early years; withdrawals compound the loss. Portfolio depleted by year 22.
- Retiree B: good returns years 1-3, bad returns years 28-30. Portfolio grows 30% early; later bad years are insignificant. Portfolio still intact at year 30.
Same long-run returns. Wildly different outcomes. This is sequence-of-returns risk.
Mitigating sequence risk
Three strategies to reduce sequence-of-returns risk:
- Cash/gilt ladder buffer: hold 2-5 years of spending in short-dated gilts or cash. During market downturns, draw from the ladder, not equity. See the gilts in retirement guide and the sequence risk guide.
- Glide-path equity allocation: start retirement with 50-60% equity, glide up to 70-80% over 10 years. Counter-intuitive but reduces early-retirement loss exposure.
- Variable withdrawal rules: reduce withdrawals in bad years (e.g. "Floor and Ceiling" or "Guyton-Klinger" rules). Reduces income smoothness but improves survival probability.
What the 4% rule does NOT account for
- Healthcare costs in late retirement. NHS helps but social care can be expensive.
- State Pension as inflation hedge. If you have State Pension + DB + 4% rule, you have multiple income streams — less stress on the drawdown.
- Inheritance objectives. 4% rule depletes the portfolio over 30 years. If you want significant inheritance, you need a lower withdrawal rate.
- Tax implications. Withdrawals beyond the 25% tax-free lump sum are taxed at marginal rate. Net spending is less than gross withdrawal.
- Behavioural drift. Many retirees can't stick to the strategy through bad years.
Is 4% still safe in 2026/27?
Several factors affect the calculus today:
- Higher inally favo: bond income is better than 2020-2022 era. Marginally favourable for SWR.
- Equity valuations: US equity Shiller P/E is high; UK and EM are moderate. Mixed.
- Demographic / longer life expectancy: retirements often last 30-40 years now. May warrant lower SWR.
- Inflation uncertainty: 2022-2024 inflation shock reminded everyone that low-inflation regimes aren't guaranteed. Conservative SWR makes sense.
Bottom line: 3.5-3.8% remains a reasonable starting SWR for UK retirees in 2026/27. Adjust down to 3.3% for very conservative profiles; up to 4.0% for retirees with strong other income sources.
Sources and methodology
Bengen W. (1994), "Determining Withdrawal Rates Using Historical Data." Trinity Study (Cooley, Hubbard, Walz, 1998). UK-specific research from FCA, IFS, and academic papers. For complex retirement income strategies, see the tax adviser editorial recommendation. Regulated retirement income advice requires an FCA-authorised IFA. The methodology page documents sources.
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.