Retirement & Drawdown
Retiring into a bad year: sequence risk, measured
- Near retirement, the order of returns matters more than the average.
- Two illustrative portfolios with identical average returns end the first decade of retirement £101,200 apart — the only difference is which year the −20% arrived.
- Three questions reveal your own exposure before the date does.
- Watching for this is a daily job — and between annual statements, nobody is doing it.
The colleague who went back to work
Most people planning retirement don't know the term for this risk. They know the story.
A colleague — careful, senior, well pensioned — retired on schedule into a bad market year. Eighteen months later she was back at her desk three days a week, "by choice". Everyone understood what had happened without quite being able to say what had happened.
If your own date is circled — 63, say, five years out — that's the story that nags. You've saved consistently, the statements look healthy, and on average markets grow. The question that remains: is the plan on track if the average doesn't arrive in the right order?
That question has a number.
One concept: sequence-of-returns risk
Sequence-of-returns risk is the risk that the order in which returns arrive — not their average — decides whether a portfolio funds a retirement.
While you're saving, order barely matters. A bad year early even helps: your contributions buy in cheaply, and decades remain for the recovery to compound. That's why the risk stays invisible for thirty working years.
The moment withdrawals begin, it reverses. A bad first year shrinks the pot and forces you to sell more units at depressed prices to fund the same spending. What later recovers is permanently smaller, because part of it was spent at the bottom. A bad tenth year lands on a pot that has already had nine years of growth and withdrawals at fair prices.
Same returns, same average, different order — different life. This is the most exposed moment in the whole journey: the pot is at its largest, the buying years are ending, the spending years are about to begin. And it can't be seen on an annual statement — a statement reports what happened; sequence risk lives in what happens next, and in what order.
I've watched stressed paths land on a risk screen in the morning: the same book of positions, run through the same set of bad scenarios, every single day, before anyone has had coffee. Not because anyone expects the bad path — because the one morning it shows up, you want to have already measured it. That discipline is standard across every major institution. What has always struck me is that private investors, at the most exposed moment of their financial lives, get none of it.
The number: two paths, one decade, £101,200 apart
Take an illustrative saver — call her Margaret — retiring with £800,000 and drawing £32,000 at the start of each year (a 4.0% initial rate). Both portfolios below earn the same ten annual returns. Only the order differs.
View the figures as a table
| Point | Bad year first | Bad year last |
|---|---|---|
| At retirement | £800,000 | £800,000 |
| Year 1 | £614,400 | £844,800 |
| Year 2 | £669,760 | £910,336 |
| Year 3 | £688,781 | £957,386 |
| Year 4 | £696,188 | £897,625 |
| Year 5 | £737,248 | £926,218 |
| Year 6 | £754,616 | £992,582 |
| Year 7 | £700,937 | £1,018,217 |
| Year 8 | £729,142 | £1,065,115 |
| Year 9 | £780,799 | £1,188,082 |
| Year 10 | £823,678 | £924,866 |
The lesson is in the two endpoints. With no withdrawals, order is irrelevant — both paths finish at £1,303,500. Start drawing £32,000 a year and the same decade leaves the "bad year first" path £101,200 behind. That gap is more than three years of Margaret's spending, created by nothing but sequence.
And the size of the first-year fall sets the scale of the damage — the house normal market / stressed market / real-world shock view, applied to year one alone:
- Normal-market bad year, −10% → £980,100
- Stressed market, −20% (global equities fell roughly −18% in 2022¹) → £823,700
- Real-world shock, −35% (2008-scale for an equity-heavy book) → £589,000
None of this says a bad first year is coming. It says the same portfolio, the same average, can fund or fail to fund the same retirement — and the difference is measurable today, at 59, rather than discoverable at 64.
Ask this of your own portfolio
- How much of your first five retirement years' spending is exposed to equities today?
- When was your plan last tested against a bad first year — not an average year, a bad one?
- Who would notice the drift in that exposure before you did?
The honest limits
History constrains this analysis; it doesn't predict. The two paths illustrate arithmetic, not a market forecast. Real retirements adapt — spending flexes, part-time income appears, plans change — and the illustration deliberately fixes withdrawals and ignores inflation and tax to isolate one effect. It gives ranges, not fate. What it removes is not uncertainty, but blindness to it.
Who is watching for this?
Sequence risk lives in the gap between annual statements. Your exposure to a bad first year isn't a fixed property of your portfolio — it drifts as markets move, as funds rebalance (or don't), and as your date gets closer. A review is a snapshot; the watching is the point. For most people five years from a circled date, the honest answer to "who is checking this?" is: no one.
Under the hood
The chart is a deterministic illustration: one identical set of ten annual returns applied in two orders, £32,000 withdrawn at the start of each year, growth applied to the remaining balance. No inflation, tax, fees or spending flexibility is modelled — the point is to isolate sequence, not to simulate a full plan.
On the platform, the same question is answered properly: the consolidated portfolio is run through stressed paths daily, downside is measured as Expected Shortfall at 97.5% (the average of the worst 2.5% of outcomes) alongside stressed ES, and results are checked against each client's pre-agreed limits — alerting the moment a limit is breached, in time to act. Data window, scenario set and assumptions: see the methodology page.
Common questions
My provider sends an annual statement — isn't that enough? A statement tells you what happened over the last year. It doesn't tell you whether you're on track for your date, what a bad first year would do to the plan, or who is watching in between statements. Those are three different questions, and a statement answers none of them.
Does sequence risk matter if I'm still ten years out? Less — and that's precisely the trap. While contributions continue, early bad years are absorbed and even help, so the risk stays invisible. It switches on as the pot peaks and withdrawals approach. Five years out is when the exposure becomes measurable and worth measuring; at the date itself, it's simply live.
Can sequence risk be eliminated? It can be measured, and it can be watched. Different portfolio shapes carry measurably different exposure to a bad first five years, and each shape has its own trade-offs — in growth given up, in inflation risk taken on. Those trade-offs can be quantified for your numbers. Which trade-off to accept is, and stays, your decision.
Sources
- MSCI, MSCI World Index — 2022 calendar-year performance (index factsheet / end-of-day index data), msci.com.
- Private Hedge, Methodology — measures, scenarios and assumptions, privatehedge.co.
Is your wealth on track?
A free 20-minute risk assessment with a senior risk manager gives you the one number most investors never see: how far your wealth could fall — property included — against the loss you’d actually accept. No report, no pitch; just where you stand.
Read-only. One flat fee. Nothing to sell you.
Book your free risk assessment →This article is information and general commentary, not financial advice or a personal recommendation. The views are the author’s own, in a personal capacity — not those of any current or former employer. Private Hedge Limited is not authorised or regulated by the Financial Conduct Authority; we never recommend specific investments and never handle client money. Figures are illustrative unless a source is given; past performance is not a reliable indicator of future results.