Sequence-of-returns risk: the thing that actually breaks early retirement

Here is the fact that keeps retirement researchers up at night: two people can retire with the same pot, follow the same plan, and earn the exact same average return over thirty years — and one of them dies with millions while the other runs out of money in their seventies. Nothing separates them but luck about timing. This is sequence-of-returns risk, it is the single most important idea in spending down a portfolio, and it has been lurking behind every post in this FIRE series. This one drags it into the open, because if you understand only one risk before you retire early, make it this one.

Why the order of returns matters when you’re spending

While you’re saving, the order of your returns barely matters. A crash early, a crash late — as long as the average is the same, you end up in roughly the same place, and an early crash is arguably a gift because you’re buying cheap. Averages are all that count when money is only flowing in.

Flip to withdrawing and that comfortable truth collapses. Now you’re selling assets every year to live on, and a crash early in retirement is a catastrophe a late crash never is. The reason is brutally simple: when the market falls and you sell to fund your spending, you liquidate a larger number of shares to raise the same amount of cash — you’re selling cheap. Those shares are gone. When the market recovers, they aren’t there to recover with. You’ve permanently shrunk the base that the rebound acts on. Do that in your first few retired years and you can hollow out a portfolio so badly that no later bull market can refill it. The order suddenly matters enormously, because withdrawals turn a temporary paper loss into a permanent, realised one.

A tale of two retirees

Picture two retirees, Anna and Ben. Both start with £1,000,000, both withdraw £40,000 a year (rising with inflation), both experience the identical set of annual returns over their retirement — the very same numbers. The only difference is the order.

Anna gets unlucky: her first few years are a nasty bear market, the good years come later. Ben gets the mirror image: strong early years, the bad stretch arrives once he’s well established. Same returns, same average, same withdrawals — reshuffled.

Ben sails through. His early gains lift the portfolio well clear of his withdrawals, so by the time the bad years arrive he’s drawing from a much bigger, well-cushioned pot and barely notices. Anna is in trouble from the start: she’s selling into falling markets in years one, two and three, cashing out shares at their cheapest to fund her spending, and by the time markets recover she has far fewer shares left to ride the rebound. On identical average returns, Ben can die a millionaire while Anna faces running dry — purely because of the sequence. That is the whole risk in one story, and it is why an average return is a dangerously incomplete way to plan a retirement.

Why early retirees are the most exposed

This risk hunts everyone who draws down a portfolio, but early retirees stand directly in its path, for reasons that stack.

Their horizon is enormous — a forty-year-old might need the pot to last fifty years, giving a bad sequence far more time and opportunity to strike. They usually can’t lean on a state pension or state healthcare backstop for decades — the very gap that makes barista FIRE attractive — so the portfolio is carrying the whole load alone in exactly the fragile early years. And the lean FIRE crowd are most exposed of all, because a bare-bones budget offers no spending to cut when the market turns — they must keep selling at the worst possible time. The 4% rule we examined earlier is, at heart, an answer to sequence risk: that rate is low precisely because it had to survive history’s worst starting sequences, not its average ones.

What you can actually do about it

The good news is that sequence risk is well understood, and there is a toolkit for blunting it. None of these is magic; together they meaningfully change the odds.

Hold a cash buffer. Keep a year or two of spending in cash or short bonds, and in a crash you spend that instead of selling stocks into the fall. You give the equity portion time to recover rather than crystallising the loss. It’s the simplest defence and one of the most effective.

Build a bond tent. This is the elegant one, and it has serious research behind it. In 2014, retirement researchers Wade Pfau and Michael Kitces published work in the Journal of Financial Planning showing that a rising equity glide path — starting retirement relatively conservative and getting more aggressive over time — could reduce the risk of ruin. It’s counterintuitive: you hold more bonds right around your retirement date, when sequence risk peaks, forming a protective “tent” of bonds over the danger years, then spend those bonds down and let equities grow back as the risky window passes. You are most defended exactly when you’re most vulnerable.

Stay flexible on spending. The single most powerful lever. A retiree who can cut spending in a downturn — skip the big trip, defer the new car — sells far fewer shares at the bottom. Formal versions like the Guyton-Klinger “guardrails” turn this into rules: trim withdrawals after bad years, allow raises after good ones. Flexibility is what lets a plan bend in a storm instead of snapping.

Stress-testing beats hoping

Notice what every one of those defences has in common: you have to plan them before the bad sequence arrives, because once you’re three years into a crash your options have already narrowed. And you can’t plan for a risk you can’t see — which is why “my portfolio averages 7%” is such a dangerous sentence. Averages hide sequences, and sequences are what actually kill plans.

This is precisely where a single deterministic projection lets you down and a proper stress test earns its keep. A Monte Carlo simulation runs your plan through thousands of different orderings of returns — good sequences, terrible ones, and everything between — and reports not a comforting single number but a probability: in what fraction of possible futures does your money outlast you? Strata’s planning tools do exactly this, including a dedicated withdrawal-phase survival test that puts your drawdown through the wringer of many market paths rather than one flattering average. Seeing that your plan survives 95% of sequences — or, sobering but useful, only 70% — is worth more than any point estimate, because it prices in the one risk that a single average pretends doesn’t exist.

That brings us nearly to the end of the tour. In the final post we pull everything together into the practical act of working out your own FIRE number and tracking the climb toward it. Or, if watching your plan survive a few thousand bad-luck sequences sounds more reassuring than hoping for an average, you can build your portfolio and stress-test it yourself.