The standard advice about retirement investing is to shift steadily out of growth assets and into conservative ones as you age, and it is repeated so consistently that it rarely gets tested. This page tests it, by running four allocation strategies through identical conditions and reporting what each one produces. The results do not support a simple answer in either direction, which is itself the finding worth having. Defensive allocation genuinely helps in the one scenario that actually destroys retirements, a serious downturn in the opening years while withdrawals are already running, and the strategy that helps most is the one almost nobody recommends: starting conservative at retirement and becoming more growth-oriented over time. That works because the risk being defended against is concentrated at the beginning, so protection belongs where the danger is rather than spread evenly across thirty years. The difficulty is what the same protection costs the rest of the time. Conservative assets return less every single year whether or not a downturn ever materialises, and compounded across a long retirement the shortfall becomes very large indeed. There is also a more uncomfortable result underneath all of it. At a withdrawal rate that looks entirely reasonable, none of the four strategies survives a bad start at all. They differ only in how long the failure takes, which means the allocation debate is being conducted about the wrong variable.
| Strategy | No downturn | Downturn years 1 to 3 | Downturn years 15 to 17 | Downturn at the end |
|---|---|---|---|---|
| All growth, never change | $3,000,332.78 | Out year 17 | $427,641.54 | $1,187,840.34 |
| Static 60/40 | $1,346,627.03 | Out year 20 | $151,177.10 | $684,682.34 |
| Declining glide, 80% to 20% | $1,242,251.55 | Out year 18 | $233,155.98 | $810,217.09 |
| Rising glide, 30% to 70% | $915,462.28 | Out year 22 | Out year 30 | $430,015.91 |
Staying in growth wins three of the four scenarios outright and loses the one that matters most. No strategy wins everywhere, which is why this is a judgement rather than an optimisation.
| Strategy | Years survived after a bad start | Against all growth | Final balance with no downturn | Given up |
|---|---|---|---|---|
| All growth, never change | 17 | the baseline | $3,000,332.78 | the baseline |
| Static 60/40 | 20 | 3 more | $1,346,627.03 | $1,653,705.75 |
| Declining glide, 80% to 20% | 18 | 1 more | $1,242,251.55 | $1,758,081.23 |
| Rising glide, 30% to 70% | 22 | 5 more | $915,462.28 | $2,084,870.51 |
Five extra years of survival costs over two million dollars in the case where the downturn never comes. That is the trade, stated plainly.
| Year | All growth | Static 60/40 | Declining glide | Rising glide |
|---|---|---|---|---|
| 1 | 100.00% | 60.00% | 80.00% | 30.00% |
| 5 | 100.00% | 60.00% | 71.72% | 35.52% |
| 10 | 100.00% | 60.00% | 61.38% | 42.41% |
| 15 | 100.00% | 60.00% | 51.03% | 49.31% |
| 20 | 100.00% | 60.00% | 40.69% | 56.21% |
| 25 | 100.00% | 60.00% | 30.34% | 63.10% |
| 30 | 100.00% | 60.00% | 20.00% | 70.00% |
Percentage held in growth assets. The two glide paths cross at about year 15, which is roughly where sequence risk stops dominating.
| Year | Declining: growth weight | Declining: closing | Rising: growth weight | Rising: closing |
|---|---|---|---|---|
| 1 | 80.00% | $635,780.00 | 30.00% | $693,230.00 |
| 2 | 77.93% | $500,778.48 | 31.38% | $594,625.23 |
| 3 | 75.86% | $389,174.77 | 32.76% | $503,808.58 |
| 4 | 73.79% | $377,639.80 | 34.14% | $492,823.51 |
| 5 | 71.72% | $364,249.09 | 35.52% | $480,740.29 |
By the end of year three the rising path holds $114,633.81 more, which is the entire source of its later advantage.
The declining path, which is what most retirement products implement, performs worst of the three defensive options against the risk it is supposed to address.
Against a downturn in years 1 to 3 it lasts until year 18, just one year longer than staying entirely in growth. A static 60/40 manages year 20 and the rising path reaches year 22.
The reason is straightforward once stated. A declining path is at its most aggressive precisely when the portfolio is largest and the sequence risk is highest, and becomes conservative only after the danger has passed.
All growth: runs out in year 17 after a bad start, and finishes with $3,000,332.78 if no downturn arrives.
Static 60/40: year 20, and $1,346,627.03.
Declining glide: year 18, and $1,242,251.55.
Rising glide: year 22, and $915,462.28.
The ordering reverses completely between the two columns. Whichever strategy you prefer, you are choosing which of those columns to optimise.
Protection against a downturn is paid for in every year, while the benefit only arrives if the downturn actually happens.
The rising path buys 5 additional years of survival against a bad start. If no downturn comes it finishes $2,084,870.51 behind staying in growth.
That is not an argument against it. Insurance is usually a bad deal in expectation and still worth buying when the uninsured outcome is intolerable, and running out of money at 82 is intolerable in a way that dying with less is not. But it should be recognised as insurance rather than optimisation.
At this withdrawal rate, none of the four strategies survives a bad start. They fail in year 17, 18, 20 and 22.
No rearrangement of assets converts failure into success here. The allocation only changes how long the failure takes, and a five year delay on a thirty year retirement is not a solution.
Lowering the withdrawal rate does change the outcome, and our sequence of returns risk calculator shows where that threshold sits. Spending less is unwelcome advice and it is the only lever here that decides whether the money lasts.
Moving to a conservative fund at 65 assumes the money is being spent at 65, and usually it is not.
Someone retiring at 65 may be drawing on that balance into their nineties, so a large part of it has an investment horizon measured in decades. Treating the whole balance as short-term money because of the owner's age confuses when you stop working with when each dollar gets spent.
The more useful framing is to match assets to the time until they are needed. Money for the next three years belongs in cash, and money for year twenty-five does not. Our retirement bucket strategy calculator builds that structure explicitly, and our investment time horizon risk calculator shows how the risk of a loss falls with horizon.
Test your own numbers before adopting any shape, because the answer moves considerably with the withdrawal rate and the gap between the two asset returns.
If every strategy fails under a bad start, that is the finding, and the allocation question can wait until the withdrawal rate is fixed. If several survive, the comparison becomes a real choice about how much expected wealth you will trade for a better worst case.
Our client portfolio projection calculator shows the range of outcomes around any single path, which is worth seeing before committing to one.
It depends entirely on what you are trying to achieve, and this page shows the trade rather than a recommendation. On the worked example, a rising glide path lasts 5 years longer than staying fully in growth if a downturn hits immediately, and finishes $2,084,870.51 behind it if no downturn arrives. Neither result makes one of them correct.
Partly, and less than people expect. On the worked example a downturn in years 1 to 3 exhausts every one of the four strategies tested. The rising glide path lasts until year 22 against year 17 for all growth, so defensive allocation delays failure rather than preventing it.
An allocation that starts conservative at retirement and becomes more growth-oriented over time, which is the reverse of conventional advice. The logic is that sequence risk is concentrated in the first decade, so that is where the protection belongs, and later years have less remaining time for a downturn to do damage.
None of them, across all scenarios. Staying in growth wins comfortably in three of the four cases tested and fails earliest in the fourth. The rising path is best in the scenario that actually ends retirements and worst everywhere else. Choosing between them is a judgement about which risk matters more to you, not an optimisation.
The withdrawal rate, and it is not close. On the worked example every allocation fails under a bad start at this rate, differing only in whether that happens in year 17 or year 22. Lowering the withdrawal changes whether the portfolio survives at all, which no allocation choice does.
Switching at 65 assumes the money is being spent at 65, and often it is not. Someone retiring at 65 may be drawing on that balance into their nineties, which is a long investment horizon rather than a short one. The relevant question is when each dollar will be spent, not what age you are when you stop working.
Because the protection is paid for continuously while the benefit only arrives if a downturn actually happens. Conservative assets return less every single year, and over 30 years that compounds into a very large number. On the worked example the gap between all growth and the rising path is $2,084,870.51 when no downturn arrives.
They are illustrative assumptions you should replace with your own. Real portfolios do not deliver a fixed number outside downturn years, downturns vary in depth and duration, and conservative assets can fall too. The purpose is to compare strategies under identical conditions rather than to forecast any of them.
If you've found a bug, or would like to contact us, or learn more about James Graham and Calculate.co.nz.
Calculate.co.nz is partnered with Interest.co.nz for New Zealand's highest quality calculators and financial analysis.
Calculate.co.nz is the sister site of CalculatorHub.com, the world's largest calculator website by tool count.
All calculators and tools are provided for educational and indicative purposes only and do not constitute financial advice.
Calculate.co.nz is proudly part of the Realtor.co.nz group, New Zealand's leading property transaction literacy platform, helping Kiwis understand the home buying and selling process from start to finish. Whether you're a first home buyer navigating your first property purchase, an investor evaluating your next acquisition, or a homeowner planning to sell, Realtor.co.nz provides clear, independent, and trustworthy guidance on every step of the New Zealand property transaction journey.
Calculate.co.nz is also partnered with Health Based Building and Premium Homes to promote informed choices that lead to better long-term outcomes for Kiwi households.
Calculate.co.nz is hosted in Auckland by SiteHost New Zealand.
All content on this website, including calculators, tools, source code, and design, is protected under the Copyright Act 1994 (New Zealand). No part of this site may be reproduced, copied, distributed, stored, or used in any form without prior written permission from the owner.
About & trust: Why Calculate is NZ's most comprehensive · By the Numbers · How we compare · Editorial standards · How we keep data current · NZ finance glossary · Research & data · Financial literacy NZ · About · Privacy policy · Terms of use
Reviewed and maintained. Last reviewed 2026-08-15 and checked on a twice-monthly cycle against IRD, RBNZ and Stats NZ. How we keep data current.
© 2026 Calculate.co.nz. All rights reserved. Building free NZ calculators since 2011.