The Sharpe ratio is annualised excess return divided by annualised standard deviation: Sharpe = (RETF − Rf) / σETF. For UK ETFs we use the 3-month UK gilt yield from Bank of England as the risk-free rate (Rf), and 3-year monthly returns for both the excess return and the standard deviation. Sharpe > 1 is good, > 2 is excellent, < 0 means you'd have been better in cash. Critical caveat: Sharpe treats upside volatility and downside volatility as equally bad — see Sortino for the asymmetric version.
The formula
Sharpe = (Rp − Rf) / σp where: Rp = annualised return of the portfolio/ETF Rf = annualised risk-free rate (3-month UK gilt yield) σp = annualised standard deviation of portfolio/ETF returns Annualisation: Monthly return: Rannual = (1 + Rmonthly_mean)12 − 1 Monthly std dev: σannual = σmonthly × √12
The Sharpe ratio answers: "for each 1% of annualised volatility, how much annualised return above cash did this ETF deliver?"
Inputs we use
| Input | Source | Notes |
|---|---|---|
| ETF returns (Rp) | Issuer + LSE | 36 months, GBP, total return |
| Risk-free rate (Rf) | Bank of England 3-month gilt yield, average over 36-month window | Annualised by BoE |
| Standard deviation (σp) | Same monthly returns, std dev × √12 to annualise | Sample standard deviation (n−1 denominator) |
UK risk-free rate — why 3-month gilt
The "risk-free rate" is a notional return you could get with zero risk. For UK investors:
- 3-month UK dies. Refl: the academic standard for short-term risk-free in UK studies. Reflects current Bank of England policy + market expectations.
- Bank of England base rate: close cousin, also reasonable.
- Best-easy-access savings rate: what most retail investors actually consider risk-free. Often within 50bps of 3-month gilt yield.
UK Tax Drag uses the 3-month gilt yield because:
- Academic consistency — most published Sharpe ratios use this.
- Bank of England publishes daily and free.
- Cross-comparable with international benchmarks (US Treasury 3-month, German Bund, etc).
The choice of Rf rarely changes Sharpe ratings dramatically — but the 50bps difference between gilt and best-easy-access ISA can move a borderline Sharpe by 0.05-0.10 over a 3-year window.
Worked example — VWRL Sharpe ratio
Vanguard FTSE All-World UCITS ETF (VWRL), 3 years ending April 2026.
| VWRL annualised return (3-year, GBP TR) | +11.3% |
| UK 3-month gilt yield (3-year average) | +4.2% |
| VWRL annualised standard deviation | 14.8% |
| Excess return (11.3 − 4.2) | 7.1% |
| Sharpe = 7.1 / 14.8 | 0.48 |
VWRL's 3-year Sharpe of 0.48 is roughly in line with broad-equity historical norms. Reading it: over this 3-year period, each 1% of annualised volatility delivered 0.48% of annualised excess return.
Compare to a hypothetical scenario: in a particularly good 3-year window (say 2017-2019) VWRL might show Sharpe 1.0+. In a particularly bad period (2007-2009) it might show negative Sharpe. The 3-year choice is a deliberate balance between recency and statistical stability.
What Sharpe tells you
- Sharpe < 0: you'd have been better in cash. The ETF returned LESS than the risk-free rate.
- Sharpe 0–0.5: sub-optimal — the volatility didn't pay off enough.
- Sharpe 0.5–1.0: reasonable. Most broad-market ETFs sit in this range over 3-year windows.
- Sharpe 1.0–2.0: good. Suggests the strategy delivered solid risk-adjusted returns.
- Sharpe > 2.0: excellent — typically only achievable in very calm or strongly trending markets.
- Sharpe > 3.0: rare and often a sign of a specific calm period or a strategy that hasn't been stress-tested.
What Sharpe does NOT tell you
- Sharpe treats upside and downside symmetrically. An ETF that occasionally pops 15% in a month is "punished" the same way as one that drops 15%. See Sortino for the asymmetric version.
- Sharpe assumes normal-distributed returns. Returns are usually fat-tailed (more extreme events than normal predicts). Sharpe under-estimates the risk of strategies with skewed return distributions.
- Sharpe ignores leverage. A 2× leveraged ETF can have similar Sharpe to its 1× cousin — but a much worse drawdown experience.
- Sharpe is regime-dependent. A 3-year window might cover only a bull market. Always look at rolling 3-year Sharpe to see how it varies.
- Sharpe across asset classes is misleading. A 0.8 Sharpe on equities is fine; on bonds it's exceptional; on cash management it's poor.
How to reproduce this yourself
- Download 36 months of GBP-denominated monthly closes for your ETF (issuer page or Yahoo Finance).
- Calculate monthly returns: =(Pt/Pt-1) − 1.
- Calculate annualised return: =((1 + AVERAGE(returns))12) − 1.
- Calculate annualised std dev: =STDEV.S(returns) * SQRT(12).
- Look up the 3-year average UK 3-month gilt yield (Bank of England statistics database — series IUMAJNB or similar).
- Sharpe = (annual_return − risk_free_rate) / annual_std_dev.
Cross-check: many ETF providers publish Sharpe on the factsheet. Yours should be within ~5% — small differences come from FX conversions, slightly different Rf definitions, and price source variations.
Sources and methodology
Sharpe ratio originally developed by William Sharpe (1966). Standard application follows Bodie/Kane/Marcus textbook treatment. Bank of England 3-month gilt yield is the published series IUMAJNB. The ETF Data Methodology documents all data sources. The site methodology documents the broader review process.
Related metric pages
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.