Sequence-of-Returns Risk: Why a 2008 at 65 Wrecks a Plan

Diane retires at 65 in Phoenix, single, with exactly $1,000,000 in her 401(k) and a plan to withdraw $40,000 in year one (the classic 4% start), raising that figure with inflation each year. Her twin sister Carol retires the same day with the identical $1,000,000 and the identical $40,000 withdrawal plan. Over the next 30 years, both portfolios earn the same 7% average annual return. Diane dies at 95 with more than $1.2 million still invested. Carol runs out of money at 83.

Nothing separates them except the order in which the returns arrived. Carol’s first three years were a 2008-style crash; Diane’s bad years came in her late 80s, when her balance was small and her remaining horizon short. That gap — roughly $400,000 of divergent outcome from identical inputs — is sequence-of-returns risk. It is the single most important threat a retirement-income plan has to neutralize, and it is invisible in any “average return” projection.

Why the order of returns matters when the average doesn’t

While you are still working and contributing, sequence risk barely exists. An early crash is actually good for an accumulator — you are buying cheap shares with every paycheck, and you have decades for them to recover. The math is symmetric: a 50% loss followed by a 100% gain leaves you exactly where you started.

Retirement flips that symmetry completely. Once you stop contributing and start withdrawing, every dollar you pull during a downturn is a share sold at a depressed price — a share that can never participate in the recovery. You are forced to be a net seller at exactly the worst time. The portfolio base shrinks faster than the market falls, because the withdrawal compounds the loss.

Consider a single year in isolation. A $1,000,000 portfolio drops 30% to $700,000, then you withdraw $40,000, leaving $660,000. To climb back to $1,000,000, that $660,000 must gain 51% — not 30%. And you keep withdrawing $40,000-plus while it tries. If the recovery is slow, you may be drawing 6% or 7% of a shrunken balance without ever choosing a higher withdrawal rate. The plan breaks quietly, years before you notice.

The canonical demonstration: same average, opposite fates

Here is the comparison stripped to the essentials. Both retirees start with $1,000,000, withdraw $40,000 every year (held flat here for clarity), and earn the identical set of annual returns — just reordered. The set averages 7%. Carol’s sequence front-loads the losses; Diane’s back-loads them.

The numbers above are illustrative, but the mechanism is exact: by the end of Carol’s third year she has already withdrawn $120,000 from a base that the market had cut to roughly $467,000. Her later +22% and +18% years apply to a base a fraction the size of Diane’s, so the same percentage gain produces far fewer dollars. The average return is a tie. The outcome is not close.

The danger window: your first 5–10 years

Sequence risk is concentrated, not evenly spread. A bad return in retirement year 2 is catastrophic; the same bad return in year 25 is a footnote. There are two reasons:

1. Your balance is largest early. A 30% loss in year 1 destroys more dollars than a 30% loss in year 20, because the year-1 base is the biggest it will ever be.
2. You have the most withdrawals left to fund. A shrunken base in year 2 has to survive 28 more years of withdrawals. A shrunken base in year 28 only has to last 2 more.

Withdrawal-rate research has a well-known corollary: portfolios that make it through the first decade intact almost never fail afterward. If your money survives the danger window, the long-run average return usually carries you the rest of the way. That is why every defense below is aimed squarely at the first 5–10 years — you are buying insurance for a specific, short, high-stakes window, not for the whole 30 years.

Defense 1: a 1–3 year cash and bond bucket

The most direct fix is to never be a forced seller of stocks. Carve off 1 to 3 years of spending — on a $40,000 budget, that is $40,000 to $120,000 — into cash, T-bills, or short-term Treasuries. When markets fall, you fund withdrawals from that bucket and leave the stock sleeve untouched so it can recover. When markets are up, you refill the bucket by trimming gains.

For Carol, a two-year bucket would have meant pulling $80,000 of her first-three-years withdrawals from cash instead of selling shares into the crash. Those un-sold shares stay invested for the recovery — which is the entire ballgame. The trade-off is cash drag: holding 3+ years of spending in low-yield assets over a 30-year retirement quietly lowers your long-run return, so most planners keep the bucket at 1–2 years and refill opportunistically rather than parking 5 years in cash. The bucket mechanics across a larger portfolio are worked out in the 3-bucket decumulation guide linked below.

Defense 2: guardrails spending that flexes after down years

The 4% rule assumes you raise your withdrawal with inflation every year no matter what the market does — which is exactly what makes it fragile. Guardrails (the Guyton-Klinger approach is the best known) instead set an upper and lower band around your withdrawal rate. After a bad year that pushes your withdrawal rate above the upper guardrail, you cut spending — often by 10% — until the rate falls back in range. After strong years you can give yourself a raise.

The behavioral reality: a retiree who trims from $40,000 to $36,000 for a couple of lean years after a crash dramatically improves the odds the money lasts, because they stop selling so many shares low. A flat 10% trim on $40,000 is $4,000 — the difference between a discretionary travel year and a quiet one, not the difference between eating and not. Flexibility is the cheapest sequence-risk insurance there is, because it costs you nothing in good markets.

Defense 3: a lower starting withdrawal rate (3.3–3.5% vs 4%)

The 4% rule came from historical data that included some brutal starting points. But if you retire into rich valuations — high price-to-earnings markets where future returns are statistically lower — a growing body of research suggests starting at 3.3% to 3.5% instead.

Every dollar you don’t withdraw in years 1–5 is a dollar that isn’t sold low. Dropping from $40,000 to $33,000 means 17.5% fewer shares liquidated each year through the danger window. The cost is real — you live on $7,000 less from day one — but for a retiree with little flexibility and no income floor, a conservative start is the simplest way to defang an early bear market. Pair it with guardrails and you can raise spending later if the market cooperates.

Defense 4: a TIPS ladder or annuity income floor

The most robust defense removes essential spending from the market entirely. Build a floor — Social Security plus a TIPS ladder, a single-premium immediate annuity, or both — that covers your non-negotiable bills: housing, food, insurance, Medicare premiums. The 2026 Medicare Part B base premium alone is about $185/month per person (CMS, Medicare Premiums & Cost Sharing 2026), and that is a bill that arrives in every market.

With a floor in place, your portfolio is only on the hook for discretionary spending. That changes everything about sequence risk: in a crash year you can simply pause portfolio withdrawals and live on the guaranteed floor, never selling a single share low. The income floor converts a forced-seller problem into an optional-seller choice. The annuity-versus-TIPS-ladder trade-off — guaranteed lifetime income versus inflation-protected control of principal — is the subject of the dedicated comparison linked below.

Which defense fits which retiree

• Plenty of discretionary spending, comfortable adjusting: guardrails plus a 1–2 year cash bucket. You have room to trim in down years, so flexibility does the heavy lifting cheaply.
• Retiring into a high-valuation market, average flexibility: start at 3.3–3.5% and add a cash bucket. The lower rate is your buffer; the bucket keeps you from selling low.
• Tight budget, low risk tolerance, fixed essential costs: build an income floor first (Social Security + TIPS ladder or annuity), then run the remaining portfolio for discretionary spending only.
• Most people: a combination — an income floor for essentials, a small cash bucket for the next 1–2 years, and guardrails on the discretionary portfolio.

What most people miss

The most common and most dangerous misunderstanding is treating sequence risk as a forecasting problem. It is not. You cannot know whether your first three years will look like Diane’s or Carol’s — nobody can — and trying to time your retirement date around market valuations usually backfires. Sequence risk is a structural problem with a structural fix: you build a portfolio that doesn’t force you to sell stocks low, regardless of which sequence you draw.

Three other widely missed points:

• The risk is asymmetric in time, so spend your defense budget early. Don’t spread a thin cash buffer across 30 years. Concentrate protection in years 1–10, the only window where a crash can actually break the plan.
• RMDs can force selling in a crash. Once required minimum distributions begin — age 73 for those born 1951–1959, age 75 for those born 1960 or later under SECURE 2.0 §107 — you must take the distribution even in a down market. The cash bucket should hold your RMD-year spending so you satisfy the RMD from cash, not from depressed shares. The penalty for a missed RMD is 25% of the shortfall under SECURE 2.0 §302 (cut to 10% if corrected within the two-year window on Form 5329), so you cannot simply skip it.
• Reverse dollar-cost averaging is the real enemy. Accumulators benefit from buying low; retirees are harmed by selling low. Every defense here is ultimately one idea — stop being a forced seller in a downturn.

A quick worked example of the cash-bucket defense

Return to Carol. Suppose at 65 she had split her $1,000,000 into a $80,000 cash bucket (two years of spending) and $920,000 in stocks. Her first three years still crash. But instead of selling $120,000 of shares into the decline, she funds years 1 and 2 entirely from cash, and only has to sell a partial year in year 3. Roughly $90,000 of shares that Carol-without-a-bucket sold at the bottom stay invested for the recovery in her solid years 4 onward.

That single change — same market, same average return, same total spending — is frequently the difference between a portfolio that lasts to 95 and one that fails at 83. The market did not change. Carol’s behavior in the market changed, because her structure removed the forced sale.

The decision lever

Sequence-of-returns risk is decided in the first 5–10 years, and the lever you control is whether a bad market can force you to sell stocks low. Set the structure before you retire, not after the crash: choose how many years of spending sit in cash (1–3), whether your withdrawal rule flexes (guardrails) or runs flat (4% rule), where your starting rate lands (3.3–3.5% into rich valuations vs 4%), and how much of your essential spending sits on a guaranteed floor (Social Security + TIPS ladder or annuity). Pick at least two of the four defenses, weighted toward whichever the danger window most threatens for your budget. Build it once, and the order your returns happen to arrive in stops being able to wreck the plan.