Portfolio Practice

Where the theory meets a human being who has to live with the outcome — rebalancing, withdrawals, and the gap between a fund's return and its investors'.

The part of investing that is operations, not analysis

• Asset allocation decides what a portfolio is. Everything on this page decides what actually happens to it: when it gets traded, what gets taken out, and what its owner does under stress.
• These decisions are unglamorous and, measured over decades, they routinely matter more than security selection.
• They are also where the arithmetic is most counter-intuitive — which is why each section below comes with a calculator rather than a claim.

Rebalancing: the discipline that sells winners

• Left alone, a portfolio drifts into its best performer. A 60/40 that never rebalances becomes an 80/20 in a long bull market — the risk profile changes without anyone deciding to change it.
• Rebalancing restores the intended risk, and mechanically enforces selling what rose and buying what fell. Whether it also raises return depends on whether asset returns mean-revert; the risk-control argument does not depend on that at all.
• Bands beat calendars. Trading when a weight leaves a tolerance band (say ±5 percentage points) trades less often than a fixed quarterly rule and responds when it matters. Trading costs (see costs & fees) are the reason not to be trigger-happy.
• Rebalancing is a volatility trade in disguise. It is short momentum and long mean reversion — profitable in choppy markets, a drag in a sustained trend. Knowing this stops it being a surprise.

The withdrawal phase: where the order of returns starts to matter

• While saving, the order of returns is irrelevant — the same set of annual returns in any sequence produces the same final pot.
• While withdrawing, order is decisive. A bad first decade forces selling into weakness and permanently shrinks the base that later recoveries work on. This is sequence-of-returns risk, and it is the defining risk of decumulation.
• The "safe withdrawal rate" literature — the widely cited 4%-ish starting figure, indexed to inflation — is an empirical result from specific markets and periods, not a law. Different countries, periods, fee levels and asset mixes give materially different answers.
• Flexibility is worth more than precision. Rules that reduce the draw after a bad year survive far more paths than a rigid inflation-linked withdrawal.

The behaviour gap

• Investors in a fund routinely earn less than the fund. The fund's return is time-weighted; the investors' is money-weighted, and money arrives after good years and leaves after bad ones.
• The gap is largest in the most volatile products — sector funds, leveraged products, thematic launches — precisely where the marketing is loudest. Studies of retail flows put it in the region of one to two percentage points a year, varying by market and period.
• Nothing about this requires stupidity. It is the predictable result of making decisions while afraid, using information that is mostly recent performance.
• The countermeasures are structural, not emotional: an automatic contribution, a written policy with rebalancing bands, and a decision rule agreed before the drawdown rather than during it.

Property, for completeness

• Most household portfolios are dominated by an asset the atlas otherwise skips. Its arithmetic is the same discounting logic in local clothing: a cap rate is a yield, and a mortgage is leverage.
• Positive leverage exists when the asset yields more than the debt costs — the borrowed money lifts the equity return. When rates rise past the cap rate, the same mechanism runs in reverse.
• Gross yield flatters: maintenance, vacancy, management, insurance and taxes routinely consume 20–35% of rent before any financing.

Practitioner rules

• Write the policy before you need it. Target weights, bands, contribution schedule, and what you will do after a 30% fall — decided in calm, executed in noise.
• Automate what you can. Every decision removed from the moment is a behaviour gap avoided.
• Rebalance with new money first. Directing contributions to the underweight asset restores the target without a sale, a spread or a taxable event.
• Judge the plan, not the year. A sound policy has bad years by construction; abandoning it during one converts a temporary decline into a permanent loss.
• Know which risk you are actually running. In accumulation it is volatility; in decumulation it is sequence; across both it is inflation. They call for different portfolios.