Sequence of Returns Risk: Why the First Five Years of Retirement Matter Most

Why order matters when money is coming out

While a portfolio is being built up — contributions going in, nothing coming out — the order of annual returns genuinely doesn't matter to the final balance. A string of returns like +8%, -12%, +15% compounds to the same ending value no matter what order it happens in, as long as no money is added or removed along the way. The moment withdrawals start, that stops being true. A down year that hits while you're withdrawing forces you to sell more shares at a depressed price to generate the same dollar amount of income, permanently shrinking the share count left to participate in the eventual recovery. This is sequence of returns risk: the same set of average returns, reordered, can produce a portfolio that survives comfortably or one that runs dry years early.

Two retirees, identical average return

Consider two illustrative retirees, both starting retirement with a $1,000,000 portfolio and both withdrawing $50,000 in year one, increasing that withdrawal each year to keep pace with inflation. Over a 20-year illustrative period, both portfolios experience the exact same set of annual returns — the same five bad years and the same fifteen good years — averaging out to an identical long-run annual return. The only difference is the order.

• Retiree A hits the bad years first: retirement opens with a stretch of down and flat markets in years one through five, right while the largest, least-depleted withdrawals are being taken.
• Retiree B hits the exact same bad years last: years one through fifteen are the strong years, and the downturn arrives in years sixteen through twenty, by which point in this illustration the withdrawal rate is a much smaller share of a portfolio that's had fifteen years to grow.

Retiree A's early withdrawals during down markets force the sale of a larger number of shares to produce each $50,000 payment, leaving a permanently smaller share count even once the market recovers — there are simply fewer shares left to benefit from the recovery. In this illustration, Retiree A's portfolio is at meaningful risk of running out around year seventeen or eighteen. Retiree B, drawing down the same dollar amounts in the same order of magnitude but facing the downturn only after fifteen years of growth had already built a much larger cushion, finishes the same 20-year period with a substantial balance remaining — despite both retirees experiencing the identical average annual return over the full period.

The market doesn't care about your average return. Your portfolio cares about the order the returns arrived in, and that order is the one thing you can never predict in advance.

Why this is scariest in the first five years

Sequence risk isn't evenly distributed across a retirement — it's concentrated almost entirely in the first several years after withdrawals begin. Early in retirement, the portfolio balance is at its largest, so a percentage decline translates into the largest dollar decline it will ever experience, and every withdrawal made during that decline locks in losses that a portfolio still accumulating contributions would never have to realize. A downturn that hits in year fifteen of a twenty-year retirement does far less structural damage than the identical downturn hitting in year one or two, simply because there's been less time for a smaller remaining share count to compound the damage forward.

Mitigation approaches, at a conceptual level

None of these eliminate sequence risk, but each reduces exposure to it during the most vulnerable early years:

• A cash buffer. Holding one to two years of planned spending in cash or cash-equivalents means a down year in the market doesn't force any equity sales at all that year — spending is funded from the buffer instead, giving the portfolio time to recover before those shares would need to be sold.
• A flexible withdrawal rate. Rather than withdrawing a fixed inflation-adjusted amount regardless of market conditions, some retirees reduce withdrawals modestly in down years, which directly reduces the number of shares sold at depressed prices.
• A bond tent. Deliberately increasing the bond allocation in the years immediately before and after retirement — then gradually reducing it again later — lowers portfolio volatility specifically during the window where sequence risk does the most damage, tapering back toward a more growth-oriented allocation once that early window has passed.

Each approach trades away some long-run expected growth in exchange for smoothing out the specific years where a bad sequence would otherwise do outsized, permanent damage.

Why this is easy to miss in advance

Sequence risk is hard to plan for precisely because it's invisible until it happens. A retirement projection built on an average expected return looks identical whether the bad years arrive early or late — the average is the same either way — so a straightforward projection can understate the risk facing someone retiring right before a downturn. This is part of why many advisors treat the years immediately surrounding a retirement date as the highest-stakes planning window of the entire process, even though nothing about a person's savings rate or portfolio allocation necessarily changed on the day they stopped working. The risk was always there; it simply had nothing to act on until withdrawals began.