Sequence of returns risk

During the accumulation phase (when you're contributing but not withdrawing), the order of returns doesn't much matter. A bad year early followed by a good year later leaves you in the same place as the reverse. But the moment you start withdrawing, order becomes critical.

Why order matters in withdrawal

When you're selling assets to fund living expenses, a bad return in year 1 forces you to sell shares at a low price, permanently reducing the base that future compounding can work on. A later rebound doesn't save you — you sold too much at the bottom. Two retirees with identical long-term average returns but different order can have radically different outcomes.

How to defend against it

• Keep 1–3 years of expenses in cash or short bonds so you don't have to sell stocks in a down market.
• Use a 'bond tent' — temporarily increase bonds in the years right around retirement, then gradually reduce.
• Build flexibility into your withdrawals. Cut spending in bad years. The fixed-dollar 4% rule is brittle; a dynamic rule is robust.
• Work one more year or phase into retirement. Both directly reduce sequence risk.

The illustration, with the actual numbers

Let's slow the earlier example down and watch the mechanism. Both retirees start with $1,000,000 and withdraw $40,000 in year one, growing with 3% inflation. Retiree A hits -20% in year one: the portfolio earns -$200,000, pays out $40,000, and ends the year near $760,000. Year two's withdrawal of $41,200 is now 5.4% of the balance — the safe-looking 4% plan has silently become an aggressive one. Retiree B gets +20% in year one instead: ends near $1,160,000, and the same $41,200 is just 3.6% of the balance. Identical plans, opposite trajectories, and the divergence compounds every year after.

Note what the table implies: all three retirees can truthfully say they earned 'about 7% on average' over the period. Averages hide the order, and in the withdrawal phase the order is nearly everything. This is also why your neighbor who retired five years before you into a bull market can safely spend more than you, from an identical portfolio, forever — retirement cohorts a few years apart draw wildly different luck.

Sizing the defenses in dollars

A cash buffer sounds like a vibe until you size it. For a household spending $60,000 a year with $25,000 of Social Security, the portfolio funds $35,000 — so a two-year buffer is $70,000, roughly 7% of a $1M portfolio. That's the drag of holding cash: maybe 2-3% of forgone returns on 7% of assets, call it 0.2% a year of portfolio performance (estimate). In exchange, a 2008-sized crash in your first retirement year never forces a single stock sale at the bottom. Cheap insurance for the exact decade the plan is most fragile.

What flexibility is actually worth

• Skipping the inflation raise in down years — the mildest possible adjustment — historically rescued a large share of failing 4% scenarios.
• A 10% spending cut when the portfolio falls 20%+ (restored after recovery) raises the sustainable withdrawal rate meaningfully, often toward 5%.
• One year of part-time income early in retirement ($15-20k) can substitute for $400-500k of extra portfolio in bad-sequence cases, because it removes withdrawals at exactly the wrong moment.
• Delaying Social Security paradoxically increases early-years sequence exposure (bigger withdrawals now) but buys permanent income later — pairing the delay with a cash buffer covers both ends.

Common mistakes

• Going 100% stocks into retirement because 'stocks always recover.' They do — but your withdrawals during the recovery lock in the losses permanently.
• Overcorrecting into all bonds or cash, which swaps sequence risk for the slow-motion certainty of inflation erosion over a 30-year retirement.
• Holding the cash buffer but refusing to spend it in a crash, defeating its entire purpose. Write the rule down before you need it: when stocks are down 15%+, spending comes from cash.
• Checking the plan only at retirement. Sequence risk peaks in the five years before and after the retirement date — the 'fragile decade' deserves an annual review.

One reassuring corollary: sequence risk cuts both ways. Retire into a strong first decade and your plan is effectively over-funded for life — which is the moment to consider raising spending, gifting earlier, or de-risking the portfolio, rather than continuing to white-knuckle a 4% plan that has already won. Checking your withdrawal rate against the current balance every few years tells you which side of the luck you landed on.

Sequence risk is the reason retirement planning can't stop at 'save 25x and earn average returns.' The average will very likely show up eventually; the question is whether your withdrawals survive the order it arrives in. A modest cash buffer, a bond allocation that peaks around the retirement date, and a written willingness to cut spending 10% in bad years defuse most of it — boring tools for the most dangerous problem in retirement finance.

A last practical note for people still working: sequence risk is also why the final five years before retirement deserve a deliberate de-risking glide rather than a single dramatic reallocation on your last day. Moving from 85% stocks to 60% gradually between 60 and 65 means no single market event — and no single rebalancing decision made under stress — can define the retirement that follows.