Rebalancing & Self-Balancing — Keeping Your Portfolio on Target
Why drift matters, how often to act, and the mechanical methods that beat eyeballing it.
A portfolio is built around target weights — say 70% stocks, 20% bonds, 10% gold. Markets immediately start pulling those weights apart. After a strong stock year, your "70/20/10" is 78/15/7. Rebalancing is the act of returning to target. Self-balancing is when the platform does it for you mechanically. Both matter more than people think.
Why rebalance at all
Two reasons, in order of importance:
1. Risk control. Drift means risk drift. A portfolio that started 60/40 and drifted to 80/20 is now exposed to a much deeper drawdown than you signed up for. Rebalancing pulls risk back to the level you chose.
2. Disciplined contrarianism. Selling what's gone up and buying what's gone down — automatically — captures a small "rebalancing premium" over time. The effect is modest (0.1–0.4% per year, historically) but real and free.
The order matters: rebalance for risk, not for returns. The return bonus is a side-effect.
The three rebalancing methods
1. Calendar rebalancing
Rebalance on a fixed schedule — quarterly, semi-annually, or annually. Simple, predictable, and easy to automate. Annual rebalancing captures most of the benefit with the least effort and the lowest tax/transaction friction.
2. Threshold (band) rebalancing
Rebalance when any slice drifts more than a set band — typically ±5 percentage points absolute, or ±20% relative. A 70% stock target with a 5pp band triggers when stocks hit 75% or 65%. More responsive in volatile markets; quieter in calm ones.
3. Cash-flow rebalancing (the best for accumulators)
Direct every new contribution to whichever slice is most underweight. You never sell — you only buy more of the laggard. For anyone still adding money monthly, this alone keeps a portfolio close to target without ever triggering a sell.
Self-balancing — what platforms do for you
"Self-balancing" usually means one of three behaviours, with very different implications:
(a) Auto-invest with cash-flow balancing
Trading 212 and InvestEngine pies do this on every deposit. New money skews toward underweight slices. No sells. No tax events. Effectively free.
(b) Auto-rebalance with sells
Some platforms (and most multi-asset funds like Vanguard LifeStrategy) actually sell overweight assets and buy underweight ones to hit targets exactly. Inside an ISA/SIPP this is fine. Inside a GIA, every sell is a CGT event.
(c) Target-date / lifestyle funds
The fund itself rebalances and gradually shifts allocation as you age (e.g. 90/10 in your 30s drifting to 40/60 by retirement). Set-and-forget. Slightly higher OCF, but for many people the right answer.
How often is enough
Academic studies (Vanguard, Fidelity) consistently find that:
• Annual rebalancing captures ~95% of the benefit of monthly rebalancing.
• Quarterly is fine but adds friction with little extra benefit.
• Threshold rebalancing on ±5pp bands beats time-based methods by a small margin.
• Daily/weekly rebalancing is actively counterproductive — transaction costs and tax friction outweigh any gain.
For 95% of people: annual review + cash-flow rebalancing in between is the right cadence.
Tax-aware rebalancing in a GIA
The £3,000 CGT allowance (2024/25) makes naive rebalancing in a GIA expensive. Three tactics:
1. Rebalance with new contributions only. Never sell.
2. Use the annual CGT allowance deliberately. Each year, harvest enough gain to use the £3k allowance and reset your cost base.
3. Rebalance inside the wrapper first. Do all heavy lifting in your ISA/SIPP; let the GIA drift further.
See the ISA vs GIA calculator for the long-run impact.
A worked example — annual rebalance
Target: 70% global stocks (VWRP) / 20% gilts (VGOV) / 10% gold (SGLN).
Pot at year-end: £58,400 stocks / £11,200 gilts / £4,400 gold = £74,000 total.
• Current weights: 78.9% / 15.1% / 5.9%
• Target £: £51,800 / £14,800 / £7,400
• Action: sell £6,600 VWRP → buy £3,600 VGOV + £3,000 SGLN
Inside an ISA, this is free of tax and friction. Inside a GIA, the £6,600 sale could realise gains of £1,000–£2,000 — well within the annual allowance for a single year, but costly if rebalanced sloppily across multiple accounts.
What rebalancing won't fix
• A bad target allocation. Rebalancing 100% crypto back to 100% crypto every quarter is still 100% crypto.
• Underlying fund risk. Rebalancing across two highly-correlated funds gives the illusion of diversification.
• Behavioural lapses. Selling everything in a crash is not "rebalancing".
The full system, in one paragraph
Pick a target allocation that matches your risk tolerance and horizon (see our risk guide). Build it as a pie (see our pie investing guide). Direct every deposit to underweight slices automatically. Once a year, in the same calendar month, check the drift; if any slice is more than 5pp off target, rebalance back inside your ISA/SIPP. Don't touch it the rest of the year.
Pair with the allocation calculator to set targets and the fee impact calculator to confirm rebalancing isn't quietly costing you more than it saves.