The order of returns decides retirements that started the same

Most explanations of sequence of returns risk stop at the point where it starts to matter. This one carries on.

Why the order stops being neutral

While you are contributing and not withdrawing, the same set of annual returns produces the same final balance regardless of their order. Once you are withdrawing, that ceases to be true, because each withdrawal permanently removes capital that cannot participate in any later recovery. A fall early in retirement therefore compounds with the withdrawals taken during it.

The same fall late in retirement, applied to a portfolio that has already delivered years of income, does far less damage.

The mechanism, stated concretely

Withdrawing a fixed real amount from a portfolio that has fallen means selling a larger proportion of the remaining units. Those units are gone, so when prices recover, the recovery applies to a smaller holding. This is the same arithmetic that makes selling in a fall costly for any investor, applied systematically every year.

The arithmetic is straightforward: it is why the first five to ten years after stopping work are described as the fragile period.

What reduces the exposure

Holding one to three years of planned spending in cash allows withdrawals to be taken from cash after a fall rather than from the portfolio. The cost is a lower expected return on that portion, which is a real and calculable price for the protection.

Reducing withdrawals in poor years, even temporarily, has a large effect because it preserves the units that would otherwise be sold. A more cautious allocation around the transition into retirement addresses the same risk from the other direction.

Guaranteed income changes the arithmetic

Spending covered by a state pension, defined benefit scheme or annuity does not require selling anything, so it is immune to sequence risk. A retiree whose essential costs are all covered by guaranteed income can suspend portfolio withdrawals entirely in a bad year. That flexibility is worth a great deal and is the practical argument for a guaranteed floor.

How much floor is appropriate depends on circumstances and warrants regulated advice.

It applies before retirement too

A large fall in the final years before stopping work reduces the capital available to convert into income, with little time to recover. This is the reasoning behind lifestyling and target-date approaches that reduce risk as a target date approaches. Those approaches assume a particular retirement date and a particular use of the money, which may not match your plan.

Where a default fund is doing this automatically, it is worth knowing what date and what assumptions it is using.

The right answer depends on your tax situation, which this cannot see.

Monitoring rather than predicting

Nobody can forecast the sequence, so the defence is structural rather than predictive. An annual review comparing the portfolio to the plan catches a bad sequence early, when adjusting withdrawals is still effective. Deciding in advance what adjustment you would make removes the need to decide it during a fall.

For most households, this is general information about a well-documented risk, not advice about how to draw your own retirement income.

The takeaway

The fragile years are the ones either side of stopping work. Structure the plan so a bad start does not force selling.

The decision is rarely about picking the best option — it is about avoiding the expensive one.