The Five Years That Decide Your Retirement

Most people, when they think about retirement, focus on the wrong number. They think about the average return their KiwiSaver or investment portfolio is expected to deliver: 5%, 7%, 9%, and assume that if the long-run average works out, they will be fine. The mathematics says otherwise. The order in which returns are experienced matters far more than the average, particularly in a roughly ten-year window straddling the day work stops. This is what financial economists call sequence-of-returns risk, and it is the single most under-appreciated risk in retirement planning.

Why Averages Lie

While saving, a market crash is genuinely an opportunity. Contributions buy more units at lower prices, and the recovery compounds in the investor's favour. The order of returns is largely irrelevant. What matters is the cumulative outcome.

Retirement reverses the mathematics. Once income is being drawn, every dollar withdrawn during a downturn is a dollar that cannot recover when the market rebounds. Two retirees can experience identical average returns over 30 years and end up in completely different places one comfortable, one running out of money, simply because of when the bad years happened.

The seminal work here is William Bengen's 1994 paper in the Journal of Financial Planning, which introduced the concept and gave us the widely-cited "4% rule".^[1] Bengen's research showed that retirees unlucky enough to retire into a poor decade (the late 1960s into the 1970s, for example) fared dramatically worse than those who retired into a strong one, even when the long-run average was the same.

Wade Pfau, professor of retirement income at The American College of Financial Services, has quantified this more precisely. His research estimates that approximately 77% of a retirement portfolio's final outcome can be explained by the cumulative return of the first ten years alone.^[2] That is an extraordinary concentration of risk into a small window of a financial life.

Approximately 77% of a retirement portfolio's final outcome is determined by the cumulative return of the first ten years. The average return over thirty years is almost beside the point.

What This Means in a New Zealand Context

The NZ Society of Actuaries' Retirement Income Interest Group expects a New Zealand male aged 65 today to live another 22 years on average; females to around age 90.^[3] For a 25-to-30-year retirement, the first decade carries disproportionate weight. Three implications follow from this for Kiwis.

Default KiwiSaver settings do not manage this risk. Default funds are now balanced, which is an improvement on the previous conservative-default era, but a "set and forget" mindset does not account for how a risk profile should evolve in the years leading up to drawdown. A balanced fund at 64 is not the same proposition as a balanced fund at 35.

Inflation is part of the sequence. Recent research has reinforced that an inflation spike in the early years of retirement permanently raises every future year's spending base.^[4] The 2022–2023 inflation experience was a textbook sequence-of-returns event for anyone newly retired.

A cash and short-duration bond buffer has real value. The "two-bucket" framework recommended by NZ actuaries, a short-term liquidity bucket alongside a longer-term invested portfolio, exists precisely so that growth assets do not need to be sold at depressed prices to fund living expenses.^[5]

The Practical Takeaway

Sequence risk does not make a strong long-term return any less attractive. It does, however, change how a portfolio should be structured in the decade either side of retirement. Reducing exposure to volatility before earned income stops, holding enough liquidity to ride out a poor market period without selling, and reviewing drawdown rates regularly are all decisions that should be deliberate, not accidental.

For anyone within ten years of retirement, this is the period where good financial planning is most measurable. It is not glamorous work: modelling, scenario-testing, and stress-testing. But it is the work that determines whether a long-run average return ever gets a chance to do its job.