Why the Order of Returns Matters
Two investors can receive identical annual returns — the exact same percentages, in different orders — and end up with dramatically different wealth outcomes. This is not a paradox. It is a consequence of the interaction between the sequence of returns and the timing of withdrawals.
For an investor who is still accumulating wealth and making no withdrawals, sequence of returns does not matter: the geometric mean return is the same regardless of the order of annual figures. An investor contributing a fixed amount each year benefits somewhat from bad early years (because they buy more units cheaply) but is not harmed by the sequence in a fundamental way.
For an investor who is drawing down their portfolio — most critically, a retiree taking regular income — the sequence of returns is critical. A severe bear market in the early years of retirement forces the retiree to sell assets at depressed prices to fund living expenses. Those sold assets cannot participate in the subsequent recovery. The portfolio is permanently impaired.
A Concrete Illustration
Imagine two retirees, each starting with £500,000 and withdrawing £25,000 per year. Both experience the same average annual return over 20 years. Retiree A faces a severe downturn in years one through three; Retiree B faces the same downturn in years seventeen through nineteen.
Retiree A is forced to sell a large number of units at rock-bottom prices in the early years, when the portfolio is largest and when those units have the most time to compound. When the recovery arrives, there are fewer units remaining to benefit from it. The portfolio may be exhausted years before death.
Retiree B faces the same downturn but has been drawing down the portfolio for sixteen years. The portfolio is smaller, withdrawals represent a larger fraction of the total, and there are fewer years remaining. The damage is real but far more manageable. In many scenarios, Retiree B survives the bad sequence comfortably.
The 4% Rule and Safe Withdrawal Rates
William Bengen's landmark 1994 research introduced the concept of a "safe withdrawal rate" — the maximum percentage of a portfolio that a retiree can withdraw annually with a high probability of the portfolio lasting 30 years. Bengen's original finding, based on US historical data, suggested 4% as the safe rate.
The 4% rule is primarily a response to sequence-of-returns risk. It is calibrated to survive scenarios in which severe bear markets arrive in the early years of retirement — exactly the worst-case sequence. A higher withdrawal rate works in scenarios with favourable early-year returns but fails in the bad-sequence scenarios that sequence-of-returns risk describes.
Implications for Asset Allocation Near Retirement
Sequence-of-returns risk has a direct implication for how portfolios should be constructed as retirement approaches and continues. Because the early years of retirement are disproportionately important, a severe equity market crash at that point is catastrophic in a way it would not be during accumulation.
This is why financial planners often recommend a "glide path" that reduces equity exposure as retirement approaches — not because equities are a worse investment over long periods, but because the downside risk of holding a large equity portfolio right at the point of greatest vulnerability is disproportionate. Holding some portion of income needs in lower-volatility assets provides a buffer: if equities fall sharply, the retiree can draw from bonds or cash rather than selling equities at the bottom.