Every figure the Australian Portfolio Lab shows is computed in your browser from the data and rules on this page. Every default below can be changed in the tool. Nothing you enter leaves your computer.
It replays a chosen portfolio mix through every historical window of Australian financial years – FY1971 onward for mixed portfolios, earlier for Australian-shares-only mixes – applying contributions, fees, account-level tax and inflation, and shows the full range of outcomes: worst, median and best. It is general information only: it does not consider your objectives, financial situation or needs, and past performance is not a reliable indicator of future performance.
Historical asset-class returns: Andex Charts Pty Ltd, via the Vanguard Asset Class Tool (retrieved July 2026), cross-checked against Reserve Bank of Australia and Australian Bureau of Statistics published data. Bond yields and CPI: RBA/ABS published series. The financial-year figures used here reconcile with the numbers printed in the published Vanguard Index Chart.
The stress test offers three preset mixes. Its monthly figures are blended whole-portfolio returns (standard mix) and precomputed crisis-event facts (Australian-only and international-only mixes) derived from the sources above; the month-by-month returns for individual asset classes are Andex Charts' compilation and are not republished here.
The complete method and data ship with the page. The same calculation also exists in IYI's own desktop tooling, and the two are checked against a fixed file of results. How many checks there are and how many of them currently disagree is set out in full in the professionals section, on the engine page.
Term deposits are one of the three things the defensive side can hold here, alongside bonds and cash. What the tool models is a single deposit, opened for one year and rolled.
What is not modelled. Three things a real term deposit brings with it are absent here, and for anyone who might need the money before the term is up the difference is not small.
The bond returns here are government bonds. The index they stand in for, the standard Australian bond index, is about 90% government and government-related and about 9% corporate, and that corporate slice earns a little more. The tool adds it as a number you can set, under "More options", rather than building it into the data.
Australian listed property (A-REITs) can be added to the growth side under "More options". The default mix holds none. What an A-REIT distributes is not one kind of income, so the tool splits the distribution rather than taxing all of it as ordinary income.
| Fee | Default | Notes |
|---|---|---|
| Adviser and/or SMSF fee | 1.1% p.a. incl. GST (any figure 0–3%; or none), plus a fixed annual amount in dollars – $0 by default, and where an SMSF's accounting, audit and levies are entered ($4,000 p.a. is a common figure) | 15% tax credit inside super accumulation; no credit in pension; outside super only the advice cost that relates to tax advice is deductible (this tool deducts none). The fixed amount is held in today's dollars (indexed with inflation) and applies to super accounts only – outside super the field is greyed and nothing is charged. |
| Platform (wrap) fee | 0.275% p.a. incl. GST when an adviser is used | Same tax treatment as the adviser fee in super; deductible at your marginal rate outside super. |
| Industry fund | Off by default; when selected, fund fees (MER) of 0.7% p.a. flat – editable, and a figure you entered yourself is kept rather than overwritten – plus a fixed $100 p.a. | The MER receives no tax credit anywhere. The fixed dollars are held in today's dollars (indexed with inflation) and get the 15% credit in accumulation, no credit in pension. Super accounts only. It stands independently of the adviser and SMSF costs above, and any fixed amounts entered in both are added together. |
| Fund fees (MER) | Mean of the three biggest ASX ETF providers' flagship index funds, per asset class | No tax credit; subtracted from gross index returns. Editable. |
| Asset class | Funds averaged | Mean used |
|---|---|---|
| Australian shares | VAS 0.07% · A200 0.04% · IOZ 0.05% | 0.05% |
| International shares | VGS 0.18% · BGBL 0.08% · IWLD 0.09% | 0.12% |
| US shares | VTS 0.03% · IVV 0.04% | 0.04% |
| Australian listed property | VAP 0.23% | 0.23% |
| Australian bonds | VAF 0.10% · IAF 0.10% · OZBD 0.19% | 0.13% |
| Cash | AAA 0.18% · BILL 0.07% | 0.13% |
| Term deposits | Money on deposit with a bank, not a fund | 0.00% |
Where a provider has no comparable flagship fund in a class, the mean uses those available. Fund names appear here as the basis of a fee assumption, not as recommendations.
Tab 7 measures three common moves against one path: your settings, the mix restored to its target every year, nothing sold. Each cost is what that path finished with minus what the move finished with, one figure for every historical starting year, and the chart shows the middle 80 per cent of those figures with the median marked. The bars cannot be added: each move changes the balance the others act on. Never restoring the mix lets each asset class run at its own return, with contributions, withdrawals, fees and tax borne in proportion, so nothing pushes the weights back; the four-yearly arm is the same walk with a restore every fourth year-end. Chasing last year's winner switches the whole portfolio into a fund costing 1.5 percentage points more (adjustable) after three financial years in a row above the median return of the growth sleeve across the whole record, and back after three below; the year's flag is decided by the years before it, never by that year; the fund switched into earns the index's own returns, so the bar is the cost of the fee alone. The cash sit uses tab 3's monthly trigger: the sale in the month a 20 per cent fall (adjustable) is first crossed, cash until the same month one, two or three years later, then the mix again; outside super the sale realises capital gains tax and the repurchase resets the cost base to the price paid. These moves run on the standard share mix, as tab 3's standard mix does, because the monthly data exists for no other. The 32 starting years overlap, so adjacent stretches share 24 of their 25 years.
The optional ongoing-cost bar is measured the same way, the path without the ongoing cost minus the path with it; it shows what the cost cost, not what it bought. Tab 6's step-down runs your settings with each layer of cost and tax switched on in turn from "before any cost or tax", in this order: fund costs, platform cost, ongoing cost, fixed annual costs, investment income tax, contributions tax, and outside super capital gains tax if sold at the end; a second column gives each layer's cost alone with every other layer on. Fees show at their gross amount until tax is switched on.
The first card can take a wage instead of a contribution: the yearly before-tax contribution is then the superannuation guarantee rate, 12% of the wage from 1 July 2025, plus any extra before-tax amount entered, and the concessional cap applies as always. When the basis switches to a wage the contribution growth defaults to wage growth (the AWOTE series) where the site has it. "How it is held" is a preset of the cost controls under More options and sets nothing a visitor could not set by hand: "Adviser (% of funds) and platform" is the site's default (the adviser fee as a percentage of funds under management, including GST, and the platform cost); "Adviser (fixed fee) and platform" charges instead a set dollar amount a year, $5,000 by default, held in today's dollars and indexed with inflation, with the platform cost; "SMSF" charges a fixed yearly cost in place of ongoing advice and keeps the platform cost, because an SMSF commonly invests through one; "SMSF with an adviser" charges both; "Platform only" charges the platform cost alone; "Pooled fund, 0.7% a year" replaces the fund costs with one flat fee plus a fixed $100 a year and switches the adviser and platform costs off. Under More options an adviser's fixed fee and an SMSF's running cost are separate rows; both are charged inside super only, because the engine has one fixed-cost input and it belongs to a super wrapper. The SMSF cost defaults to $4,500 a year, the ATO's median operating-expense bundle (administration, audit, the supervisory levy and the ASIC review fee) for 2023-24, indexed with inflation as every fixed cost is. Any change made by hand under More options reads back as "Custom". A retirement age below 60 runs as entered with a note that super cannot usually be drawn before then.
Tab 8 measures events in a life against the same path as tab 7: your settings, nothing happening. Each event is a change to the yearly path the engine already runs. Being scammed removes a share of the balance at the start of the year you turn the age set (all of it by default, at 60); the contributions continue, so the bar is what the loss cost by the end of the stretch; outside super the part sold realises its share of the gain and the tax comes out of what remains. Losing a job scales the contributions from that age onward by the share of earnings kept. A break from paid work sets the contributions to nothing for the years named. A lifestage strategy lowers the growth share year by year from the age set, along one of five shapes: each shape is a percentile, across the MySuper lifecycle products APRA publishes each quarter, of the growth share at each age, scaled so that it starts at your own growth dial; the middle shape is the default. No product is named and none can be read from a shape. When two or more events are ticked a further bar shows them together in one run; it is not the sum of the others, because each event changes the balance the others act on. An event dated after the end of a stretch costs nothing in that stretch.
On tab 8 a self-managed fund can buy a commercial property. At the age set a share of the balance (all of it by default) leaves the invested account as the deposit; the price follows the loan-to-value ratio (70% by default, so the member finds 30% of the price plus the purchase costs) and the rest is borrowed. The loan can be interest only, the default, or principal and interest over a term you set (25 years by default), the repayment re-set each year on the balance, that year's rate and the years still to run; only the interest is deductible and the principal comes out of the fund's liquid money. The rate is the Reserve Bank's standard variable owner-occupier housing rate for each financial year plus a margin (1.5 points by default) standing for the premium a limited recourse borrowing arrangement carries. Each year the property earns a gross yield (6% of its value by default) less the empty months, pays running costs (25% of the full rent by default, charged whether or not it is let) and interest; the net rental income is taxed at the fund's 15%, and the result flows into the invested account mid-year, like a contribution. The value grows at the year's inflation plus a margin (3 points by default). The empty period begins a set time after purchase (one year by default) and runs for the years and months set (two years by default). A year in which the invested account ends below zero is a shortfall: the member puts the shortfall in, and the bar counts that money as part of the cost. With the forced sale on, the property is sold at the start of the next year at a discount to its value (20% by default, the figure a Reserve Bank research discussion paper, RDP 2022-03, assumed for commercial property collateral in a stress test; a working paper, not a settled result), the loan repaid and the fund's capital gains tax paid at 10% on a gain held more than twelve months, 15% otherwise; without it, the member keeps topping up and the property is kept. The loan is limited recourse, so neither a sale nor the horizon can leave the fund below nothing on the property. A rental loss is worth 15% against the fund's other income, which assumes the fund has other assessable income at least as large as the loss. Contributions count as that income when they are tax deductible, so a fund still receiving employer or salary-sacrifice contributions will generally have it; a fund whose only asset is the property and which receives nothing would carry the loss forward instead, which is not modelled. Purchase costs (transfer duty and the rest, 5% of the price by default, a judgement) come out of the deposit and join the cost base. The purchase happens at the start of the year, so that year's rent, interest and value growth all run. The money a member puts in during a shortfall restores the account to nothing and earns no return of its own; it is counted in the cost bar and in the money-weighted return. A property still held at the end of the stretch enters the final balance at its value less the loan, with no sale and no tax. The scenario runs in super accumulation only and cannot be combined with a drifting mix or a cash sit. Platform and fund costs are charged on the invested account, not on the property. Defaults and their sources are in the configuration file's smsf_property block.
Tab 11 runs two households with the same money over every historical stretch of the years shown, from the buying age set (30 by default). The home's value follows, by default, what prices actually did (the published index below); a switch runs instead the what-if framing of a value growing at each year's inflation plus a margin (2 points by default), with the rent then growing with inflation. The buyer pays a deposit (20% by default) and purchase costs (5% of the price, a judgement standing for transfer duty, which on a $1 million home runs from about $31,000 in Queensland to $55,000 in Victoria, plus conveyancing), borrows the rest on a variable principal-and-interest loan at the Reserve Bank's standard variable owner-occupier rate less the discount of each era (or the visitor's own), with the repayment re-set each year on the balance, the year's rate and the remaining term (30 years by default), pays the running costs of owning (1% of the home's value a year by default: rates, insurance, maintenance, a judgement), and owns a home whose value follows the OECD's analytical house price index for Australia (from the ABS's series, financial-year means from FY1971). The renter pays a rent that starts at a share of the price (3.5% by default, a judgement; the research's anchor is 3.15%, a mean private rent over a mean dwelling price for 2019-20, two ABS populations that differ) and follows the ABS CPI Rents index (from FY1973), so the rent tracks rents, not the home's value, and invests the deposit and purchase costs on day one plus, each year, what owning would have cost more than renting, outside super at the marginal rate; in a year renting costs more, the buyer invests the difference instead. At the end the buyer holds the home's value less selling costs (2.5% by default, a judgement) less the loan still owing, plus any investment after CGT; the home itself, a main residence, bears no capital gains tax, while the renter's investment account is valued after the CGT a sale would bear. The home and the rent are set at the start of the stretch, on the index levels of the year before it, so the home's value takes the same number of yearly steps as everything else. The invested money runs on the standard share mix at tab 1's growth dial with the site's cost settings. A home is also a place to live, which neither figure prices.
Tab 10 answers two questions with what the superannuation guarantee alone, 12% of the income, built over every historical stretch, from nothing. Starting out: university from an age (18 by default) with some years of study (three) on a small income ($15,000 a year), then a set income ($100,000); or an apprenticeship from a younger age (16) on a starting income ($60,000) that grows faster than wages (5 points a year) until it has multiplied by a set factor (doubled), then with wages. Retraining mid-career: from a deciding age (40) on an income now ($100,000), stay, or study for some years (six) on a share of it (50%) and earn a set income afterwards ($200,000), with an optional further stage. All incomes are today's dollars and, by default, hold their value in today's money through each stretch (grown with prices, the CPI), so a $100,000 income is $100,000 of today's money in every year; a switch grows them with wages (the AWOTE series) instead, and because wages have outpaced prices an income that grows with wages buys more at the end of a stretch than the same figure does today. Where the guarantee on an income exceeds the concessional cap, the excess is left out of super and the page says by how much. Contributions are taxed and capped as everywhere else; the invested money runs at tab 1's growth dial on the standard share mix with platform and fund costs, and the adviser fee and fixed costs under More options only when ticked, since a person starting from nothing rarely pays them and at 1.1% the fee would take a quarter of a guarantee-only balance. A tick adds three typical paths from ABS statistics: a construction from two ABS statistics, the level by highest qualification from Employee Earnings (August 2025: a bachelor degree or above about 1.70 times, a certificate III or IV about 1.40 times a person with no post-school qualification) and the shape by age from Personal Income in Australia (2022-23), multiplied and read at band midpoints; no ABS table publishes the cross, so the paths are typical rather than measured. On the typical degree path the person studies from 18 to 20 on a part-time wage and works at the graduate profile from 21; on the typical trade path the person earns 60, 70, 80 and 90 per cent of the certificate wage in four apprentice years (a judgement; the award pay guide's percentages are not reproduced because the Fair Work Ombudsman's material is licensed for non-commercial use only). A degree's study debt is repaid from take-home pay, not from super, so the bars do not show it.
Tab 9 takes one yearly amount a household could spare after tax and runs it three ways over every historical stretch of the years shown. Every household holds the same loan on the same schedule: a balance owing over a term (25 years by default), the required repayment an annuity on the balance, the years still to run and that year's rate, re-set each year as the rate moves. Paying extra shortens the loan; it does not reduce the required repayment, which is how a lender treats it. So each household's outlay is the same, the required repayment plus the spare amount, and the arms differ only in where the spare amount goes. Each arm is reported net of the loan it still owes at the end of the stretch: the extra-repayment arm usually owes nothing, the two invested arms owe whatever the schedule has not yet repaid. Both households end with the same house, so the house itself cancels and is left out. A control decides how each bar counts the loan. Taking the loan still owing off each arm, the default, means a bar can be negative when the loan outlives the stretch; counting the loan repaid instead adds the same starting loan to every arm, so no bar is negative and the differences between them are identical to the cent. The rate of each year is the Reserve Bank's standard variable owner-occupier rate less the discount borrowers of that era actually obtained (nothing before 1996, a cheaper basic product in the late 1990s, a package discount from 2004 that has drifted up to about 2.5 points), read from a schedule derived from the Reserve Bank's tables, its Bulletin, the ACCC's home loan inquiry and the Bank's Statement on Monetary Policy; a visitor who knows their own discount can enter it instead. Once the extra repayments have cleared the loan, the spare amount and the repayment no longer made are both invested, outside super by default or into super if the visitor chooses, for the remaining years. Before April 1986 the advertised rate was a regulated ceiling and therefore the rate paid; actual paid rates have been published only since mid 2015; the discounts between are read from the sources above and are a derivation, not a published series. As salary sacrifice, the same take-home is grossed up at the marginal rate into a before-tax contribution and run through the ordinary super accumulation path from a zero balance: 15% on the way in (30% with Division 293 on), the fees and fund tax used everywhere else; the part of the contribution above the concessional cap is invested outside super after tax, and the cap is assumed otherwise unused. Invested outside super, the same after-tax amount goes into a personal account at the marginal rate and is valued after the capital gains tax a sale at the end would bear under current rules. All three grow the yearly amount by the contribution growth setting; the invested arms carry the site's cost settings and run on the standard share mix. The marginal rate of this tab follows the rate set for the whole site until a rate is chosen here. Super is locked until preservation age; the bars show what history did with the same money in three places.
A pension must pay a minimum out every year. On tab 8, in the pension setting, two bars compare what happens to the part of those payments the household does not spend (all of it by default; the share spent is a control). In one arm it is saved outside super at the rate set on tab 5. In the other it goes back into super as an after-tax contribution to a second, accumulation account, which pays 15% on its earnings and the same fees as the pension, and is merged into the pension at the end of the year its balance reaches the figure set ($35,000 by default, in today's money, between $5,000 and $500,000). Nothing goes back in once the member passes 75 and nothing above the yearly after-tax contribution cap; whatever is turned away is saved outside super instead. Nothing goes in at all once the total in super reaches the transfer balance cap ($2,000,000 in today's money): the law reads the balance at the previous 30 June and the tool reads the balance at the start of the year, which is close but not the same. The bring-forward rule, which lets some people put up to three years of the cap in at once, is not modelled, so a household that would use it is understated. The transfer balance cap also limits how much may be moved into a pension when the two accounts are merged; that limit is not modelled, so a household near the cap is flattered.
How much a household spends is a control, and there are two rules. By default both households spend the same number of today's dollars every year and save the rest, so their spending is identical and the whole difference between the bars is the strategy; a year whose payment is smaller than that amount is spent in full, and only in such a year do the two differ. The other rule spends a share of each payment. Under that rule the two households no longer spend the same amount, and the reason matters: putting the payments back into super makes the pension larger, a larger pension must pay a larger statutory minimum, and the household spends the same share of that larger payment. Both bars therefore count what has already been spent as well as what is left, in the dollars of the year it was spent, the same way this site totals a pension everywhere else. A dollar spent stops earning, so under the share rule at a high share the two households finish close together.
The second bar is what a non-dependant receives if the member dies at the end of the stretch. Money in super is part tax free and part taxable: an after-tax contribution is tax free, earnings are taxable, and a pension's tax-free proportion is fixed on the day it starts, which is why a merge matters. It re-starts the pension on the two balances put together and works the proportion out afresh. On a death benefit paid to a non-dependant the taxable part bears 15% plus the 2% Medicare levy; the tax-free part and money held outside super bear none. The pension is taken to start wholly taxable by default, which is the common case for a balance built from employer contributions and the least favourable starting point; the share that is tax free at the start is a control. Nothing here is advice, and the bars do not weigh up whether a re-contribution suits anyone: they are the arithmetic of two accounts over the same history.
Tab 14 takes an amount now, an amount each year, or both, and runs them two ways over every historical stretch of the years shown. Set against a loan, the money compounds at the rate you enter with nothing taken off, because interest saved is not taxed. Invested, it goes into a personal account outside super at tab 1's growth dial and cost settings and is shown after the capital gains tax a sale at the end would bear. The page states how many of the stretches the invested side finished ahead in. A loan rate is certain and a return is not, so the two bars are not the same kind of number, and the page says so.
What is not modelled, and it matters: borrowing to invest, splitting a loan into deductible and non-deductible parts, and the deductibility of interest. The wider question people usually mean by this comparison turns on exactly those, and this exhibit answers the return comparison alone. It says so on the tab as well as here.
Tab 15 takes a lump sum and runs it two ways over every historical stretch of the years shown: put in on day one, and put in as equal instalments a year apart, the first on day one and the last one, two or three years later (up to five), so "spread over three years" is four instalments and "spread over one year" is two. Money still waiting sits in cash at each year's cash return, taxed as income at the account's rate (the marginal rate outside super, the fund's rate inside), pays no fee, and what it earned goes in with the last instalment, so a spread arm puts in the lump plus what its own waiting money earned, and the lump at once puts in the lump alone. The instalments are stated in the dollars of the day they arrive, which is what a lump in hand is. The starting balance and yearly saving from tab 1 are set aside on this tab so the bars show the lump alone; everything else about the run (growth dial, costs, tax, and outside super the capital gains tax a sale at the end would bear) is the engine's own. The page counts the stretches in which putting the lump in at once finished ahead of the longest spread.
What is not modelled: inside super a lump is an after-tax contribution, and neither the yearly cap on such contributions nor the rule that lets three years' caps be used at once is applied. The tab says so.
Tab 16 runs a starting amount and a yearly saving of its own (tab 1's account type, growth dial, costs and contribution growth apply; its balance and saving do not) over every historical stretch of each length from the shortest to the longest you set, three to seven years at first and up to twenty, and counts in how many of them the money finished at or above a goal you enter: the balance at the end of the stretch is compared with the goal, and what it did in between is not counted. Nothing new is worked out: each stretch is an ordinary run. The goal is a dollar figure and is compared in the same dollars as the bars, today's dollars unless the before-inflation view is chosen, and outside super after the capital gains tax a sale at the end would bear when that box is ticked. Shorter lengths have more stretches than longer ones, because each begins in a different year and the record ends in the same year.
This page says what the money did. It does not say what a deposit would have bought or whether a sum is enough for anything: no house-price series feeds it (the dwelling price series on the rent-or-buy tab is not used here, on purpose), and the First Home Super Saver scheme is not modelled.
Tab 17 runs the same stretches as the other tabs for money outside super and shows what a sale at the end left after capital gains tax under both rule sets side by side, whatever the rules selector on the left says. The two rule sets are the ones every after-tax figure on the site has used since the capital gains toggle shipped. Current rules: half the gain at the marginal rate at sale, the final year's additions undiscounted. 2027 rules: every year's cost indexed by the CPI from its own year, no discount, and the indexed gain taxed at the higher of the marginal rate at sale and 30%. A loss pays nothing under either, and indexation never turns a gain into a loss.
The rules were read from the amending Act itself, the Treasury Laws Amendment (Tax Reform No. 1) Act 2026, No. 49 of 2026, on the Federal Register of Legislation. Christoph signed that reading on 5 September 2026, and the tab's figures appeared on his word. For the law, read the Act there rather than this page. The reading found no contradiction with the two rule sets, and two gaps of scope, both stated on the tab. The whole stretch is treated as beginning after 1 July 2027: under the Act, money already held on 30 June 2027 keeps the discount on its growth to that date and only the later growth falls under the new rules, so for such money the 2027-rules figure here overstates the tax. And the Act's exemption for people who received a listed income-support payment in the year of sale is not modelled. Two approximations: the 30% minimum is applied to the whole indexed gain at the higher of the marginal rate and 30%, where the Act tops up the tax actually attributable to the gain, the same figure when one marginal rate applies to the whole gain; and the cost base is indexed with the site's yearly CPI series where the Act uses quarterly index numbers. Outside this site altogether: the new-dwelling and affordable-housing discounts, the quarantining of rental losses, small business concessions, trusts and companies. Complying superannuation funds' own discount is untouched by the Act.
Tab 12 takes the same take-home given up in each of the years before a release age and runs it three ways over every historical stretch of the years shown: used as catch-up contributions and taken out at the release age to pay down the home loan, paid straight onto the loan as it is earned, or invested outside super. Money put in before tax is taxed at 15% on the way in rather than at the marginal rate, which is why more of it goes in for the same take-home; it is then taxed on its earnings inside the fund like any other super money, and comes out at the release age. Every household is on the same loan schedule as tab 9, with the same choice of how a bar counts the loan. Unused concessional cap space can be used only by someone whose total super was under a limit ($500,000, a flat figure the law does not index, so it does not move along the stretch either) at the previous 30 June, and only within the years the law allows; this tool reads the balance when the catch-up begins, which is close but not the same, and leaves a stretch out when the test fails, saying how many. Room carried forward accrues over a limited number of years, so the space that can be claimed is capped at five years of the concessional cap; anything entered above that is reported as left out rather than quietly used. Because this is the one case where the law allows more than a year's cap in a year, the usual cap is lifted by exactly that room for these contributions and by nothing else. The release age cannot be set below 60, since below it a lump sum is not tax free and this exhibit takes the whole account out at once. The cap space itself is entered in today's money and carried along the stretch's inflation path, like every other dollar figure here. What counts as a condition of release, and the rules for carrying room forward, are set out by the Australian Taxation Office; the page links to them rather than restating them, and the figures used here (five years of room, a $500,000 balance limit, and the site's own concessional cap) were read on 4 September 2026.
Tab 13 moves a share of each year's concessional contributions to a partner's account and compares what the couple held at the end with keeping everything in one account. The concessional cap is applied before anything is split, so splitting cannot put more into super than one person could. A split is made after the financial year ends; this tool treats the money as landing in the partner's account in the year it was earned, and since it is invested either way the difference is whose account earned that year's return. With nobody starting a pension the two arms are the same to the dollar, which is the finding: the money is in the same investments, taxed the same way, whichever account holds it. A difference appears only when something else changes. A fixed cost, if one is set, is charged to every account that exists, so splitting carries two of them. If a pension is started at the age set, that account's earnings stop being taxed but the minimum must be paid out, and what is paid out is invested outside super at the marginal rate, where its earnings are taxed at that rate rather than 15%; starting a pension earlier is therefore not free, and the bars show it. A split cannot go to an account that has left accumulation, so splitting stops in the year the partner reaches that age and the whole contribution stays with the member from then; and nothing is contributed to an account already in the retirement phase, so contributions stop in the year the member reaches it, in both arms alike. The couple therefore contributes the same money in the same years whichever arm is shown. What may be split, and by when, is set out by the Australian Taxation Office; the figure used here, at most 85% of a year's concessional contributions, was read on 4 September 2026. The partner's own account starts empty by default, so the bars show what the splitting itself does.
Tab 3 also states how many of the stretches that triggered a sale the selling finished ahead in, which is a count rather than a verdict, and carries a tick for an inherited package: money that arrives and is not added to, so the contribution is set to nothing for that tab alone.
"The cost of selling out" simulates an investor who sells everything in
the month their portfolio is first down the chosen percentage from its peak
(checked monthly from 1970 onward), then stays in cash for the rest of the
window – fees and cash-income tax continue, and any contributions keep
arriving into cash. The optional third bar sells at the lowest point of the
same fall instead. It offers three preset mixes – the standard 45/55
growth mix, Australian shares only, and international shares only, each
with bonds defensive; the custom splits under "More options" apply to the
other views. The GFC drawdown figures quoted beside the chart (Australian
shares −48%; international shares in AUD −38%; the standard 45/55 blend at
100% growth −39%; the default 70% growth portfolio −27%; peak to trough,
Jul 2007 – Jun 2009) come from the same monthly data, shipped with the
site in data/facts.json; an automated repository check
fails if the figures on this page drift from that file.
Built by In Your Interest Financial Planning. In Your Interest Financial Planning Pty Ltd, ABN 28 094 300 464 is Authorised Rep. No 308161 of Fiduciary Duty Advisers Pty Ltd AFSL No 527434. Questions and corrections are welcome – contact us.