Short answer: sequence-of-returns risk is the danger that bad returns arrive early in retirement, while you are withdrawing. Before you withdraw, the order of returns doesn’t change where you end up. Once you are taking money out, it does: losses early force you to sell more of a shrinking portfolio, and those shares aren’t there for the recovery. Two retirees with exactly the same returns in a different order can end up a million dollars apart.
Same returns, different order
Hypothetical return sequences, not forecasts. Two retirees each start with $1,000,000 and withdraw $40,000 at the start of every year (4%, held constant in today’s dollars). Both get the same 30 yearly returns after inflation: three losing years of −15%, −10% and −5%, and 27 years of +6%. Retiree A gets the losses first; Retiree B gets them last.
| Portfolio value | A: losses in years 1 to 3 | B: losses in years 28 to 30 |
|---|---|---|
| After 3 years | $625,480 | $1,056,031 |
| After 10 years | $584,592 | $1,231,982 |
| After 20 years | $488,049 | $1,647,426 |
| After 30 years | $315,156 | $1,447,928 |
Without withdrawals, both portfolios would end at exactly the same $3,504,640, because multiplication doesn’t care about order. The withdrawals are what make the order matter. Retiree A sold $120,000 of investments at depressed prices in the first three years, and the money left had to recover from a much lower base. Neither retiree ran out here, but A ended with less than a quarter of B’s balance and far less room for error.
What history shows
The same effect shows up in real data. Morningstar’s 2025 report found that the starting safe withdrawal rate for a 50/50 stock-and-bond portfolio over rolling 30-year periods since 1927 ranged from 3.9%, for people who retired in late 1968, to 10.5% for those who retired in mid-1982 (The State of Retirement Income: 2025). Same portfolio, same rules; the difference was mostly what the market did in the first years after each start date.
Why it matters more for early retirees
- A longer road. A portfolio damaged early has to last 40 years or more, not 30.
- No pension yet. Social Security usually starts decades later, so the portfolio carries the whole load through the riskiest years.
- Stock-heavy portfolios. Many FIRE savers hold mostly stocks, which swing more than a balanced mix.
Ways to reduce the risk
None of these removes the risk; each trades something for a bit more safety.
| Approach | How it helps | What it costs |
|---|---|---|
| A lower starting withdrawal rate | Sells less in the bad years. See the 4% rule vs 3.5%. | A bigger target and more years of work. |
| Flexible spending | Cutting withdrawals after losses leaves more invested for the recovery. Morningstar’s 2025 research finds flexible approaches support higher starting rates. | Income that rises and falls. |
| Some bonds or cash for the early years | Lets you draw on assets that haven’t fallen as much while stocks recover. | Lower expected long-run growth. |
| Part-time income at first | Every dollar earned is a dollar not sold in a down market. | Not fully retiring. |
Limits of this example
The sequences above are invented to isolate one effect. Real returns don’t come in neat blocks, inflation varies, and taxes and fees reduce what you keep. The FIRE calculator assumes the same return every year, so it can’t show sequence risk; its results are a starting point, which is why its estimates are labelled as estimates.
Related reading: the Roth conversion ladder for funding the early years, and the three-fund portfolio for setting a stock-and-bond mix.
