Sequence of returns risk: why the first years of retirement matter most

Sequence of returns risk is the danger that poor investment returns arrive early in retirement, just as you start taking money out. While you are saving, the order of returns does not change where you end up. Once you are withdrawing, it does. In our worked example, two people retire with $450,000, earn exactly the same 2.5% a year on average over 30 years, and withdraw the same $309 a week. The one with three bad years at the start runs out at 89. The one with the same bad years at the end still has $76,220 at 95. Across our 1,000 simulated markets, the first ten years explained most of the difference between plans that lasted and plans that did not. Rules checked 12 November 2026.

What sequence risk is

If you invest a lump sum and never touch it, the order of returns does not matter. A fall of 12% followed by a rise of 14% leaves you in the same place as a rise of 14% followed by a fall of 12%, because multiplication works in any order.

Withdrawals change that. When markets fall early, you have to sell more units to fund the same pay, at low prices. Those units are gone, so they are not there to recover when markets bounce back. Late in retirement the balance is smaller and fewer years are left, so the same fall does less harm.

That is why the years just before and just after you stop work are sometimes called the "risk zone". Your savings are at their largest, and you are about to start drawing on them. The Retirement Commission's 2025 Review of Retirement Income Policies (page 67) names "how markets will perform" alongside "how long they will live" as the uncertainties retirees have to plan around.

Worked example: same average, different order

Ana and Ben both retire at 65 with $450,000. Both take $16,127 a year ($309 a week) in today's dollars. That is the level amount that lasts exactly to 95 if the fund earns a steady 2.5% a year, the official assumption after 65 in Schedule 7A of the Financial Markets Conduct Regulations 2014.

Both earn the same 30 returns, averaging exactly 2.5% a year:

• Ana: -12%, -8% and -4% in her first three years, then 3.7% a year.
• Ben: the same returns in reverse: 3.7% a year, then -4%, -8% and -12% in his last three years.

Had neither withdrawn anything, both would have $521,103 in today's dollars at 95. Same returns, same end point. It is the withdrawals that make the order matter.

Ana's pay could safely have been $260 a week; Ben's $362. That is a 39% difference from the order of returns alone. After 89, Ana lives on NZ Super only.

What 1,000 simulated markets show

Real markets do not come in neat patterns, so we checked the same idea across the 1,000 markets our calculators use. Each market draws a return each year around a 2.5% average (after fees and tax) with a yearly swing of 6%, which is our assumption, and inflation of 2%. For each market we found the highest level weekly pay from $450,000 that lasts to 95, then grouped the markets by how their first ten years went.

To isolate the order effect, we then took only the 137 markets whose 30-year average return came out close to the 2.5% assumption (a real return within 0.2 points of 0.5%). Over the whole 30 years they were all about the same. Even so:

• The quarter with the worst first decade (about -1.5% a year after inflation) could typically pay $285 a week, and $309 a week lasted in only 3% of them.
• The quarter with the best first decade (about 2.2% a year) could typically pay $343 a week, and $309 lasted in 94%.

Same long-run return, very different retirement. The first decade decides most of it.

Why the swing matters as much as the average

Notice that the typical 30-year real return across all 1,000 markets is 0.3% a year, a little below the 0.5% you get from 2.5% less 2% inflation. That is not an error. When returns swing up and down, the average you actually compound at is lower than the simple average of the yearly returns. A fall of 10% needs a rise of about 11% to get back to where you were. The bigger the swing, the bigger this drag, which is one reason higher-risk funds do not always deliver more spending power in retirement.

How our calculators allow for it

A single fixed-return projection, such as the "about $X a week until 90" on a KiwiSaver statement, has no sequence risk in it: every year earns the same. That is fine for a quick estimate, but it gives the middle answer, which half of all real outcomes fall short of.

Our calculators run your plan through 1,000 markets, each with its own order of good and bad years, and report the pay that holds in 9 of 10. That builds sequence risk in. In the example above, the fixed-return level is $309 a week, but the pay that holds in 9 of 10 markets is $236. The $73 gap is mostly the price of not knowing which order your returns will come in.

See your own figure on the retirement pay calculator, which also shows the typical market and a very poor one.

What you can do about it

You cannot choose your sequence, but you can soften it.

1. Be flexible in the early years. A rule that trims spending after poor years protects the balance when it matters most. Our guardrails guide shows one.
2. Keep a cash buffer. A year or two of withdrawals in cash or a defensive fund means you are not forced to sell growth assets after a fall.
3. Think about risk around retirement. A lower-risk fund narrows the range of early outcomes. Our fund type guide shows the trade-off: lower risk lowers the typical result but helps the worst ones.
4. Keep some work income. Part-time work in the first years means smaller withdrawals while the balance is largest. See working after 65.
5. Retire a little later. Each year you delay is one more year of saving and one less of drawing. When can I retire? tests this to the month.
6. Remember NZ Super. NZ Super does not depend on markets. The bigger it is relative to your spending, the less sequence risk matters.

What does not help

Moving to cash after a fall tends to lock in the loss. And no forecast can tell you in advance which sequence you will get. The protection comes from planning for a poor one.

This is general information, not financial advice. A licensed financial adviser can help you set up a buffer or a drawdown plan.

Questions

What is sequence of returns risk?

The risk that poor returns arrive early in retirement, while you are withdrawing. Because you sell more units at low prices, early falls do lasting damage, even if the long-run average return turns out fine.

Does sequence risk matter while I am still saving?

Much less. With no withdrawals, the order of returns does not change the final balance. It starts to matter as your balance grows and once you begin drawing down.

How long does the danger period last?

In our simulations the first ten years of retirement explained most of the gap between plans that lasted and those that did not. The years just before you retire matter too, because that is when your balance is largest.

Is sequence risk worse if I retire before 65?

It can be. Before 65 there is no NZ Super underneath, so withdrawals are larger relative to your savings, and KiwiSaver stays locked until 65. Our retiring before 65 guide covers the bridge years.

Does a KiwiSaver statement projection include sequence risk?

No. Statement projections use a fixed return every year, so they show a middle outcome. Our calculators test 1,000 markets and show the pay that holds in 9 of 10.