Two retirees start with identical $1 million portfolios. Both withdraw $40,000 a year for living expenses. Both earn an identical 7% average annual return over a 30-year retirement. One ends with $1.4 million remaining. The other runs out of money in year 22.
The variable is not how well they invested, what they owned, or how disciplined they were. The variable is the order their returns arrived in.
This is sequence-of-returns risk. It is the most underappreciated risk in retirement planning and one of the harder ones to defend against. This is a plain-language walkthrough of why it matters, how it works, and what disciplined retirement planning does to mitigate it.
Why averages lie about retirement income
The standard way to talk about long-term portfolio performance is the average annual return. Stocks return roughly 10% over the long run. Bonds return roughly 4%. A 60/40 portfolio returns roughly 7%.
This works fine for the accumulation phase, when you are adding money to a portfolio and not withdrawing. The average return compounds, and the order of returns matters very little to the ending value.
It breaks down in the withdrawal phase. When you are drawing money out, the order of returns matters enormously. Losing money in the first five years of retirement is much worse than losing the same amount in the last five years, because the early losses come out of a balance that compounds across the remaining decades.
Consider two return sequences, both averaging 7% per year over 30 years.
Sequence A: -15%, -10%, -5%, +5%, +8%, then twenty-five years of +10% each. Sequence B: Twenty-five years of +10% each, then +8%, +5%, -5%, -10%, -15%.
The arithmetic average is identical. The geometric (compound) average is also identical because the same numbers are multiplied in different orders.
In accumulation, both end with the same balance.
In withdrawal (drawing $40K/year inflation-adjusted from $1M starting balance), Sequence A retiree runs out of money in year 22. Sequence B retiree finishes with over $2 million. Same average, opposite outcomes.
Why this happens
The mechanics are straightforward but counter-intuitive at first.
When you withdraw from a portfolio that is also losing value, you are selling shares at depressed prices to fund withdrawals. Every share you sell at $80 is a share that is not in the portfolio to recover when the price returns to $100. The depleted base never recovers, because the recovery happens to fewer shares.
When you withdraw from a portfolio that is gaining value, you are selling shares at appreciated prices. The remaining shares continue compounding from a higher base. The portfolio funds withdrawals and still grows.
Time amplifies the difference. Losing 15% in year 1 of retirement removes 15% from a base that would have compounded for 29 more years. Losing 15% in year 29 removes 15% from a base that has already done the compounding work. The mathematical impact is dramatically different even though the dollar loss is similar.
The numbers that make this real
For a 4% withdrawal-rate retirement plan, the difference between a good sequence and a bad sequence over the first ten years can be the difference between a 30-year retirement that succeeds and one that runs out of money. The market returns over the last ten years of the retirement matter much less.
This is why retirements that started in 1966 or 2000 (right before two of the worst sustained drawdowns in modern equity history) struggled even when the long-run averages eventually recovered. Retirements that started in 1982 or 2009 (right after major drawdowns, into multi-decade bull markets) sailed through.
The retiree did not choose. The starting date was fixed by birth year. Sequence-of-returns risk is fundamentally a planning problem because the variable that drives it cannot be controlled in real time.
How sequence risk shows up in standard retirement math
Standard retirement plans use the "4% rule" or similar safe-withdrawal-rate frameworks. The 4% rule says that a 4% initial withdrawal, adjusted for inflation, has historically lasted at least 30 years across nearly any starting period.
The "nearly any" doing the work in that sentence is sequence-of-returns risk. The 4% rule barely survives a 1966 retirement start. It comfortably survives a 1982 start. The variance is enormous, and the planning decision (how much to withdraw) is made years before the variance is observable.
The implication is that a retiree planning for the average return is implicitly planning for the average sequence. Real life delivers a specific sequence, and the specific sequence is often much worse than average.
What disciplined retirement planning does
Several techniques have emerged to manage sequence risk. None eliminate it; they reduce its impact.
Hold a multi-year cash buffer. Two to three years of living expenses in cash or short-term bonds at the start of retirement means the first market correction does not force you to sell equities at depressed prices. The equity portfolio gets time to recover. This is sometimes called a "bucket strategy" and it dramatically reduces sequence risk at the cost of a few percentage points of long-run return on the cash buffer.
Use a glidepath. Start retirement with relatively conservative equity exposure (maybe 50% equity) and increase it over time as the sequence risk window closes. After 10 to 15 years of successful withdrawals, the remaining horizon is shorter and the portfolio can re-add equity exposure. Counter-intuitive: the optimal asset allocation gets MORE equity-heavy as retirement progresses, not less.
Adjust withdrawals dynamically. Rather than fixed real-dollar withdrawals (the strict 4% rule), use a dynamic framework that withdraws less during drawdown years and slightly more during strong-return years. This is sometimes called the "guardrail" approach. It costs lifestyle flexibility but dramatically extends portfolio survival.
Delay retirement by a year or two in bad starting environments. A retiree with the option to work one more year during a market drawdown can dramatically improve outcomes. The drawdown often recovers within that year, the portfolio rebuilds, and one year of additional contributions plus delayed withdrawals compounds across the remaining decades. The flexibility itself is a meaningful planning lever.
Diversify income sources. Social Security, pension, partial annuity, dividend income, rental property. Each provides some non-portfolio cash flow that reduces the pressure to sell equities for living expenses. The less the portfolio is being drawn down, the smaller the sequence-risk exposure.
A worked example
A 65-year-old retiree with $1 million wants to plan for 30 years. Standard advice: withdraw 4% per year ($40K) inflation-adjusted, 60/40 portfolio.
Standard plan, average sequence: portfolio survives 30 years with $1M+ remaining.
Standard plan, bad sequence (start 1966 or 2000 analog): portfolio runs out by year 22. Retiree spends the last 8 years on Social Security alone.
Sequence-defended plan (three-year cash buffer + glidepath + dynamic withdrawals): same bad sequence start, portfolio lasts 30+ years even with sub-average returns through year 12.
The difference between the two is not investment skill or stock picking. It is plan design.
What individual investors can do
If you are within 10 years of retirement, sequence-of-returns risk is the highest-leverage thing you can plan around.
Audit your equity exposure. If your retirement plan assumes 7% average returns and your portfolio is 90% equity, you are accepting a wider distribution of sequences than your plan can survive. Reducing to 60% or 70% equity in the years immediately preceding retirement reduces sequence risk at modest expected-return cost.
Build the cash buffer before you need it. The two-to-three-year cash buffer is most effective when it is established before retirement starts, not built up during retirement. If you need to retire next year, start moving cash into the buffer now.
Plan flexibility into your spending. If 100% of your retirement spending is fixed costs, your sequence risk is enormous because you cannot reduce withdrawals during drawdowns. If 30% of your spending is discretionary (travel, dining, gifts), you have a natural buffer.
Consider a partial annuity for the fixed-spending floor. Some retirees use a low-cost income annuity to cover essential expenses, then run an equity portfolio for everything else. The fixed-income floor eliminates sequence risk for the survival portion of spending.
The bottom line
Sequence-of-returns risk is real and substantially under-discussed relative to its impact. A retirement plan that uses average returns and ignores sequence is a plan that depends on getting lucky with the year you happened to retire.
Disciplined retirement planning treats sequence risk as a primary design variable rather than an afterthought. Cash buffers, glidepaths, dynamic withdrawals, and diversified income sources are the structural defenses. None of them eliminate sequence risk, but together they shrink the bad-sequence outcomes substantially.
This is educational content, not personalized retirement-planning advice. The right sequence-defense strategy for your situation depends on your assets, income sources, time horizon, and spending flexibility. Consult a fiduciary financial advisor for retirement planning tailored to your circumstances.