Rebalancing a portfolio in practice

What rebalancing is actually for

A portfolio drifts. Whatever has risen becomes a larger share of the total, and the mix you deliberately chose slowly turns into a mix chosen by recent performance. Rebalancing is the act of trimming what has grown and topping up what has not, so the portfolio's risk stays close to what you intended.

The primary purpose is risk control, not extra return. Sometimes rebalancing improves outcomes, sometimes it reduces them — it depends on whether markets mean-revert or trend over the period in question, which is unknowable in advance. Anyone presenting rebalancing as a reliable return enhancer is overstating it.

Step one: write down the target, not just the intention

You cannot rebalance without a target. Before anything else, record the allocation you intend to hold, at a level of detail you can actually check: broad buckets rather than individual tickers, expressed as percentages, with a note on why each bucket is there. If you cannot state a target, the drift you are worried about has nothing to be measured against.

The main methods

• Calendar rebalancing. Review on a fixed schedule — annually, semi-annually or quarterly — and correct any drift found. Simple, predictable, easy to keep to. It can trade when nothing needs correcting, and ignore drift between dates.
• Threshold (band) rebalancing. Act only when a holding moves more than a set distance from its target, for example a few percentage points absolute or a set proportion of the target weight. More responsive to what is actually happening; requires monitoring and can cluster trades in volatile periods.
• Combined calendar-and-threshold. Check on a schedule, but trade only if a band has been breached. This is the version most people find sustainable, because it bounds both the effort and the trading.
• Cash-flow rebalancing. Direct new contributions into the underweight holdings, and take withdrawals from the overweight ones. Often the cheapest method available to an individual because it can correct drift without selling anything.

No method is universally best. Wider bands and less frequent reviews mean lower costs and more drift; tighter bands mean tighter control and more trading. The honest framing is a trade-off, not an optimum.

The frictions that decide whether it is worth doing

• Tax. In a taxable account, selling can realise a gain and create a tax liability that dwarfs the benefit of the correction. In a tax-sheltered account the same trade may cost nothing. This single difference changes the sensible answer more than any other factor, and it is country- and account-specific.
• Trading costs. Commissions, spreads, platform charges, and currency conversion where holdings are priced in another currency — see currency risk and hedging for how conversion quietly adds up.
• Minimums and fractions. Some products cannot be traded in small amounts, which makes precise weights impossible and small corrections uneconomic.
• Settlement and timing. Sale proceeds may not be available immediately, so a rebalance can leave you briefly out of the market or holding cash.
• Behaviour. Rebalancing asks you to sell what has done well and buy what has not. That is uncomfortable, which is precisely why a written rule beats a judgement call in the moment.

A workable procedure

Rebalancing when part of the portfolio is automated

If some capital sits in an automated or "AI" strategy, treat that allocation as one bucket in the overall portfolio and rebalance the bucket, not the strategy's internal positions. Interfering with a running system mid-cycle usually breaks the very behaviour you allocated to it. Decide in advance what share of total capital that bucket may occupy, and how you will respond if it grows or shrinks past that share. The failure modes of automated allocations are covered in the risks of AI trading.

When not to rebalance

• When the drift is inside your bands. Doing nothing is the correct action far more often than people expect.
• When the trade's tax and dealing costs plainly exceed the risk being corrected.
• When your circumstances have changed — then the job is to revisit the target itself, ideally with advice.
• When you are reacting to a headline. A rule written calmly is worth more than a decision made anxiously.

Where platform mechanics matter

Costs, fractional dealing, settlement times and conversion charges differ between providers, and they directly affect how practical a rebalancing policy is. Verify the entity and the published cost schedule before assuming anything, using the pre-deposit checklist. Our methodology sets out what our reviews examine, and the long-term investors path collects the related reading in order.

The limits of this guide

This is general education. It contains no recommended allocation, no suggested band width and no claim about which method performs best, because those depend on your goals, timeline, tax position and account types, and any specific figure here would be invented precision.

Speak to a regulated financial adviser about whether your target allocation is suitable, and to a qualified tax adviser in your country before selling anything in a taxable account — the tax consequences frequently determine whether a rebalance is worth doing at all. Where trusts, pensions or cross-border estates are involved, take legal advice too.

The takeaway

Rebalancing is maintenance, not strategy. Write down a target and a tolerance, review on a schedule, use contributions and withdrawals before sales, account for tax and costs honestly, and accept that most reviews should end with no action. The discipline of the rule matters more than the precision of the weights.