Two portfolios earn an identical average annual return over thirty years. One receives its worst years at the start, the other at the end. During accumulation, this makes almost no difference. Once withdrawals begin, the difference can be enormous.
The reason is arithmetic rather than psychology. Selling assets during a decline removes shares permanently, so fewer remain to participate in the recovery. A portfolio drawn down during an early bad stretch never gets those units back, regardless of what the market does afterward.
This is sequence-of-returns risk, and it’s the single largest structural difference between saving and spending a portfolio.
Why the Timeline Makes It Sharper?
The arithmetic applies to any retirement and bites harder on a longer one. Anyone pursuing financial independence ahead of a conventional timeline is asking a portfolio to survive forty or fifty years of withdrawals rather than twenty-five or thirty.
That extends the window during which a poor early sequence can do permanent damage, and it removes several of the buffers a later retirement has available: a shorter horizon, contractual pension income arriving sooner, and more flexibility to return to work if needed.
None of that makes an early exit unworkable. It does mean the early years carry more weight, and planning that treats all years equally is understating the risk.
Where the Risk Concentrates?
The exposure isn’t spread evenly across retirement. It clusters at the point where the portfolio is largest relative to remaining contributions.
Analysis of what’s sometimes called the portfolio size effect describes accumulating extra bonds in the years leading up to retirement, then spending down that volatility-dampening reserve during the first half of retirement, producing a bond allocation that finishes higher than in early accumulation but lower than at the portfolio’s peak.
The shape that produces is often described as a bond tent: fixed income builds toward the retirement date, holds through the early years, then declines. It’s the inverse of the conventional advice to reduce equities steadily with age.
The logic is that a 20% decline matters far more when a portfolio is at its largest and withdrawals have begun than when it’s small and contributions are still arriving.
What the Glidepath Research Found?
The counterintuitive part is what happens to equity exposure afterwards.
Research published in the practitioner literature used Monte Carlo simulation across multiple capital market assumptions, including a baseline set, a low-interest-rate scenario, and one calibrated to more optimistic historical averages.
The work found that a rising equity glide path through retirement reduced retirement risk, with earlier findings noting that spending bonds first does not increase shortfall risk.
So the sequence runs opposite to intuition. Lower equity exposure at the point of maximum vulnerability, rising equity exposure as the dangerous window passes.
The mechanism is straightforward once stated. If the early years go badly, the conservative allocation limits the damage. If they go well, the portfolio has grown enough that later volatility matters less.
Practical Mitigations
Several approaches address the same risk from different angles:
- A cash and bond reserve covering two to five years of spending, so withdrawals need not come from equities during a decline
- Flexible spending rules that reduce withdrawals in poor years rather than holding a fixed inflation-adjusted amount
- Contractual income covering essential costs, which reduces how much the portfolio has to produce
- Part-time income in the early years, which cuts withdrawals during the highest-risk window
- A glidepath decided in advance, so the allocation shifts on a schedule rather than in response to conditions
The second is often the most powerful and the least discussed. A retiree who can reduce spending by 10% in a bad year has changed the arithmetic substantially, and that flexibility does more than most allocation adjustments.
The Trade-offs Nobody Mentions
Every mitigation costs something. A large bond allocation at retirement means lower expected returns in the median case, which is the case that usually happens. Insurance against a bad sequence is paid for in every sequence that turns out fine.
The original research on rising glidepaths noted a further concern: an approach that leaves an investor holding higher equity exposure in later life may sit poorly with risk tolerance at that age, even where the arithmetic supports it.
That tension is real and there’s no clean resolution. A plan someone abandons at seventy because it feels wrong is worse than a slightly suboptimal one they follow.
What Allocation Can’t Fix?
Sequence risk can be reduced. It cannot be removed, because nobody knows which sequence they’ll get and the tools available all involve giving something up.
What’s achievable is knowing the exposure exists, where in the timeline it concentrates, and which lever is cheapest to pull.
For most people the cheapest lever isn’t allocation at all. It’s retaining some flexibility over spending and over when the withdrawals actually start.
