7 examples of how the new super tax will be calculated

A couple of people have asked why the ATO can't just use the data from superfund tax returns to calculate actual/realised earnings for the purpose of Div 296. Some have been quite insulting about the ATO in their comments, suggesting that 'you don't have to be very smart' to do this, or words like that.

So it's worth putting everyone straight on this. Super funds don't do individual tax returns for each fund investor. What the ATO has access to is fund earnings, not an attribution to each unit holder (in public funds) or member (if an SMSF). What the ATO has access to is the year-end total value of each unit holder or member's superannuation. Test this yourself - go to your MyGov account and look up in the ATO section what your super is. They'll show you what they have as of last 30 June.

Given that members make contributions and withdrawals through the year, it's not actually an easy exercise for funds to calculate what would go into a tax return for individual investors. Funds do it outside of super because they were set up to do this, but until now it's simply not been relevant for superfunds to have to do that sort of deep dive into each unit holder. Just knowing how many units they had was enough. Other than that, paying to the ATO 15% of all net income and realised gains generated by the fund is also enough.

For SMSF's, if you are the only member then, sure, what the ATO sees could be used to calculate your actual earnings for Div 296 purposes. But most SMSF's have more than one member and again all the ATO sees is the tax return for the whole fund, not each member.

It would be enormously expensive for super funds to change their systems so that they could generate an income tax return for each individual member. It could be done, but it would take a while to implement and cost all members money, not just those who would be caught by Div 296.
I can already hear someone saying, 'but they should just charge a fee to those who have over $3mn'. All well and good if they're only in the one fund, but as has been said many times there are plenty of people who have more than one fund, with no individual fund having more than $3mn. Only the ATO has the look through to all your investments to know if you're caught by Div 296. That's why it's not a superfund tax, but a personal tax based on your superannuation.

The outcry if all members of a fund had to pay for a system change because of the very small percentage who will be caught by Div 296 would be enormous - and justified.

I don't like the taxing of unrealised gains either, but the reason that Div 296 has been devised the way it has is because it balances the various interests of all participants on the system and all members. I can only ask those who comment here on FirstLinks to try to show some respect to our bureaucrats who actually have thought through all these issues, are aware of every criticism that you're throwing at the proposal, but have carefully weighed up all the competing arguments to devise a tax structure that can work with the least additional cost to superannuation fund members. This focusses the impact on those who are actually caught by and have to pay Div 296.

"show some respect to our bureaucrats":

Super funds Advice to the Treasurer: 'CRIPES, GO FIND ANOTHER TAX WORTH THE EFFORT'

Bureaucrats advice to the Treasurer should have been 'DO NOT DO IT'. [ It being taxing unrealised returns. ]

Followed by: 'AT LEAST WAIT UNTIL COST BASIS CAN BE APPORTIONED TO INDIVIDUAL MEMBERS'.

Actuaries have developed simple, rugged methods of allocating realised earnings to individual members by estimating each member's daily balance based on opening balance, contributions, withdrawals, rollins, rollouts with close balance as a check.

Super accounting systems must keep track of asset cost basis to calculate realised capital gain.

Any modestly competent actuary could devise a simple method of allocating cost basis at EOFY+0 and EOFY+1
to each member.
Reporting that to ATO requires a slight change in the Super Fund Tax Return.

My advice to Treasurer Chalmers: Ask the bureaucrats for a more efficient method of soaking the rich.

Dudley, how do you know that the bureaucrats behind closed doors haven't advised against it? But given that the government wants to go ahead, they've devised a system that meets competing ends.
Those actuarial calculations you mentioned are approximations and, if used, would trigger law suits with the ATO. I don't blame politicians or bureaucrats for wanting to avoid that.
In my view the simplest approach would be to make it compulsory for funds above $3 mn to be taken out of super. Just set a limit on how much anyone can have in. One variation could be to make it that if someone has a pension account that funds still in accumulation above $1mn get taken out.
Trying to concoct income calculations for the small proprtion of fund members who'll be caught by this is not actually helpful to anyone.

"how do you know that the bureaucrats behind closed doors haven't advised against it? But given that the government wants to go ahead, they've devised a system that meets competing ends.":

Pity Minister's attention wandered during presentation of the cons of taxing unrealised gains.
Must have else how to explain selecting the dumb option?

"Those actuarial calculations you mentioned are approximations and, if used, would trigger law suits with the ATO."

The very similar calculation of Exempt current pension income (ECPI) is an approximation and is Law.

"simplest approach would be to make it compulsory for funds above $3 mn to be taken out of super.":

When? Daily, yearly? If balance subsequently less than $3M can it be rolled back in from 'purgatory'?

"One variation could be to make it that if someone has a pension account that funds still in accumulation above $1mn get taken out.":
"Trying to concoct income calculations for the small proprtion of fund members ... not actually helpful to anyone"

Abolish:
. Super,
. Age Pension Means Tests,
. Inflation 'Stealth Tax'.
Simpler and more efficient and affordable.

Those who can not, or do not, save for retirement can live large on the Age Pension.
The rest can save in peace.

“But most SMSF's have more than one member and again all the ATO sees is the tax return for the whole fund, not each member.”

While our SMSF filed a single tax return it is not clear that the ATO does not have actual earnings for each member. SMSF Annual Return form NAT 71226-06.2024 includes in Section B, item W the gross (REALIZED) earnings for the fund and item V the assessable (REALIZED) earnings from which the tax payable is calculated.

It includes in Section F (pages 8-15) an opening and a closing account balance for EACH MEMBER of the fund. It also includes in item O allocated earnings (or losses) that contributed to each closing balance for EACH MEMBER of the fund. The latter combines both REALIZED and UNREALIZED earnings of the fund.

However since fund earnings (both REALIZED and UNREALIZED) are allocated to each member proportionally based on member balances, it should be possible to allocate separately the REALIZED and UNREALIZED earnings to each member. Alternatively this information could be requested in future tax returns.

Although Division 296 is technically neutral in that it applies to all fund types, the calculation method (unrealised gains), valuation rules, and practical payment mechanisms lead to significantly greater compliance, liquidity, and volatility issues for SMSFs.

The effect on SMSFs is that they may have to pay tax on "paper profits" without selling the asset — potentially requiring them to liquidate other assets to pay the tax. In contrast, industry and retail funds typically invest in diversified, liquid assets and report smoothed or unitised returns, which means the member is taxed only on realised earnings through unit price growth — not directly on individual asset revaluations.

Therefore it would make sense to treat SMSFs separately from industry and retail funds particularly since the ATO appears to have information about REALIZED and UNREALIZED earnings for each member of an SMSF.

If the SMSF's trustees want to incur the additional costs of generating the accounting and tax returns for individual members then, yes, maybe ...

But do you seriously think the government would countenance treating SMSF's differently to other funds?

And in any case, if one member is making contributions and the other isn't then the proportioning becomes a bit more challenging as you need to calculate the marginal earnings on those marginal contributions. And what if one is in pension mode making withdrawals, but other members aren't?
Getting exact earnings attribution is not as easy as it might look. In my book, those averaging approaches are not as accurate as unit pricing based calculations.

Be very careful what you wish for.

"If the SMSF's trustees want to incur the additional costs of generating the accounting and tax returns for individual members then, yes, maybe ...":

Already required by law. Yearly must produce audited account report for each member showing Open, Contrib, Withdraw, Rollin, Rollout, Earnings, Close. Made available to each member.
Whether for SMSF or other Super Fund.

"one member is making contributions and the other isn't then the proportioning becomes a bit more challenging as you need to calculate the marginal earnings on those marginal contributions. And what if one is in pension mode making withdrawals, but other members aren't?":

Already done considering balances for each calendar day of financial year.

"Getting exact earnings attribution is not as easy as it might look. In my book, those averaging approaches are not as accurate as unit pricing based calculations."

Both methods agree remarkable well. I have checked daily calculations against actuarial estimations based on dates of transactions.

Thank you very much for those useful examples. However, I didn't understand in the example of Pedro the politician why the increase in benefit paid was based on the growth in average earnings. An Australian Government defined benefit scheme is based on CPI.

Oh dear! Life was much simpler when we had Reasonable Benefit Limits (7xFAS).
Of course, Howard and Costello stuffed up the system completely when they
-abolished the RBL
-made super tax free in earnings and benefits
-allowed unlimited contributions (up to $1m)
On top of that, they introduced the refund of franking credits, which bastardised a system which was set up to avoid double taxation, not become the abolition of corporate tax for Australian investors.
For all this moaning about the Div 296, suck it up!
Super was never intended to be an unlimited tax shelter. It became so thanks to the stupidity of the Howard Government!
If anyone is speaking to Tim Wilson or Geoff Wilson, remind them of the facts!

"made super tax free in earnings and benefits":
"allowed unlimited contributions (up to $1m)":
Creating an alternative to stuffing money into tax free home improvement and tax free Age Pension.

"refund of franking credits, which bastardised a system which was set up to avoid double taxation, not become the abolition of corporate tax for Australian investors":
Can not trust all foreigners to pay tax after running off with capital and dividends.

"Super was never intended to be an unlimited tax shelter.":
Was set up to be. Then government saw the rivers of gold and changed narrative of what it was set up to be.

“Then government saw the rivers of gold and changed narrative of what it was set up to be”

Totally agree that the so called overarching “purpose of superannuation” as enshrined in recent legislation was never envisioned by the founding fathers but formed an indispensable justification for introducing taxable thresholds (TBC, Div 296) into superannuation that was touted to be "tax free in retirement" . Those rivers of gold eventually became too enticing to successive cash strapped governments to be ignored so the subterfuge began.

The big question I want answered is how the credits for losses will be applied. I also want to know happpens if a loss takes the super balance below $3M. Will that loss not be counted against future gains that takes the super ballance over $3M?

The explanatory memorandum for the Bill (which will need to be reintroduced) has the following.

An individual will have transferrable negative superannuation earnings for an income year if the amount of superannuation earnings worked out under section 296-40 for the year are less than nil and either or both of the following are true: the TSB at the end of an income year is greater than $3 million or an individual’s TSB immediately before the start of the year is greater than $3 million.

Why not add another column showing how much tax (and medicare) levy the person would pay on the relevant amount of super income if it were held outside the Superannuation System.
Also in regard to defined benefits funds it is worth mentioning that a portion of their superannuation pension is taxed whereas accumulation fund pensioners receive drawings on their pension fund tax-free. The defined benefits pensioners do receive a tax offset which helps reduce the tax however any extra income they receive outside of the pension is taxed differently as accumulation pensioners still have access to the tax-free threshold which is denied the defined benefits pensioner.

It is incorrect that beneficiaries of Government Defined Benefits Schemes have not contributed to those schemes. This includes federal MPs that are members of the Parliamentary Contributory Superannuation Scheme: https://www.finance.gov.au/government/superannuation/parliamentary-contributory-superannuation-scheme. Large defined benefit pensions are a function of large salaries. The alternative - an accumulation superannuation scheme - would have required large annual contributions by the employer (Government) which was not required for the Defined Benefits Schemes.

Your defined benefit examples suggest there will be a significant change in fund valuation for defined benefit recipients when the family law calculations are applied. Currently Pedro’s and Patricia’s valuations would be $4 million ($250,000x16). The valuations jump to $6,070,560 and $6,076,075, respectively. These figures are an approximate multiple of 24. This suggests it would only take a $125,000 a year pension to reach the $3m threshold, capturing many more defined benefit pensioners than under current valuations.

I think you will find the valuations are very case specific. I’m a retried public servant - age 65 and drawing a css pension. My pension is valued less using the family law method by about 6%.

I wouldn’t be generalising too much and certainly not based on the Pedro and Patricia examples which are probably at the extreme end, even for defined benefit schemes.

Howdy, just wondering how did you get a valuation for your pension?

OldbutSane

Hi, the proposed methodology for Div 296 valuation of certain public sector superannuation schemes (including CSS) is detailed in "Family Law (Superannuation) (Methods and Factors for Valuing Particular Superannuation Interests) Approval 2025". The regulation is available at https://www.legislation.gov.au/F2025L00281/asmade/text/2

The document details the methodolgy to be used for valuing various CSS scenarios (pension, contribution phase etc). If you spend a bit of time reading through the relevant part, the actual calculation is pretty straight forward- basically a simple formula just requiring reference to various factors in look up tables.

Caveat - this is all based on my understanding of the previously proposed legislation which the government has yet to reintroduce to this session of parliament. So things might change..

Joan’s adjusted TSB at the end of the year is calculated to be $4.05 million by adding her total withdrawals of $250,000 and deducting her total contributions of $300,000 from her 2025–26 TSB of $4.1 million.

This does not make sense. Why would you add the withdrawals and not subtract? Should the adjusted TSB not be $4,150, 000.00 ($4,100,000 plus $300,00 minus $250,000)?

It's the same for all example calculations.

Why? Because that's the formula in Div 296.

Why is that the formula? Because Div 296 is taxing a proxy for "earnings".
Opening balance (B0) + contributions (C) - withdrawals (W) + earnings (E) = closing balance (B1).

Rearrange this and you get:
E = B1 + W - C - B0

So, you add back the withdrawals and subtract the contributions, just as the formula says. The people at Treasury are quite bright after all!

"The people at Treasury are quite bright after all!"

Then why are they aiming to tax "earnings" that may evaporate to nothing the day after the tax is paid?

"why are they aiming to tax "earnings" that may evaporate to nothing the day after the tax is paid?":

Can not fathom the difference between realised opening and closing balances and unrealised balances.
Which is unbright.

“Can not fathom the difference between realised ... and unrealised balances. Which is unbright.”

How bright do you need to be to fathom the difference from the annual tax returns that SMSFs are obliged to file.

"How bright do you need to be to fathom the difference from the annual tax returns that SMSFs are obliged to file.":

Very.
SMSF earnings, withdrawals, contributions, rollouts, rollins are all actual / realised, usually cash.
SMSF open and close balances are unrealised valuations.
SMSF tax return does not report open and close asset cost basis valuations.

No reporting of realised balance = cost basis.

“Very”
I reviewed at our SMSFs last tax return and it includes:
Section D: Income tax calculation statement
This section sets out the taxable income for the current financial year and includes the REALIZED earnings of the fund.
Section F: Member information
This section includes an opening and a closing account balance for each member of the fund. It also includes allocated earnings (or losses) that contributed to each closing balance.
The latter includes dividends and interest received as well as the increase in (unrealized) market value of investments (shares) held by the SMSF. Hence the closing balance includes both REALIZED and UNREALIZED earnings of the fund.
Not sure how illiquid assets such as real estate (eg.farms) are treated in a tax return but expect that periodic valuations may be required so that this may be reported to the ATO.
Thus it appears that the difference between realized and unrealized balances can be fathomed from the tax return that our SMSF filed.

"Hence the closing balance includes both REALIZED and UNREALIZED earnings of the fund.":

Both open and close balances are the total of realised and un-realised VALUATIONS of the fund at EOFY.

Cash at bank is always realised (real) valuation. Cost basis of $1 is always $1.
Shares have a market value (last market transaction) and are un-realised (imaginary) valuation.
Properties have an estimated value (valuer's guess) and are un-realised (imaginary) valuation.

Shares and property assets have a ('historic') cost basis (price when purchased) - realised valuation.
Current format SMSF tax return has no place to record cost basis (realised valuation)

Only instance where cost basis (realised valuation) and be determined from current format SMSF tax return is when all assets are cash only.
All else is un-realised valuation.

Thanks Firstlinks the explanation is sorely needed. Rang the ATO to explain and they cannot as bill not passed, advised to contact Treasury as its their bill and they must explain your query. Did this and no reply. Have a question on this. If you have only have shares in your SMSF and are retired. The value sits at $1m but at the tax time it has grown to $4m and its earnings are $300.000 in Div only. Therefore the total value= $4.3m. the % calc = (4.3m-3m)/4.3= 30.2%. So 30.2%x15%=4.53% or 0.0453. So in this case is the 4.53% mult by the total capital gain for the tax year which is $3.3m or is it mult by the earnings only $300,000?
They are saying it would be simpler for this Govt to just tax you 30% on any earnings over the $3m asset value.

According to the following calculator: https://www.smsfalliance.com.au/div-296-calculator/
Adjusted Total Super Balance: $4.3M
Starting Total Super Balance: $1M
Growth: $1.3M
Percentage of Growth which is taxable: 30.23%
Taxable Growth: $392,990
Tax at 15%: $58,948.50
So according to this calculator the 4.53% is mult by the capital gain above $3m for the tax year which is $4.3M-$3M=$1.3M (the calculations assume that there were no contributions or withdrawals during the year)

“according to this calculator the 4.53% is mult by the capital gain above $3m for the tax year”

I actually meant to say that “according to this calculator the 4.53% is mult by the EARNINGS above $3m for the tax year”. “Earnings” are different to capital gains because they track market movements not actual gains which may fall in subsequent tax years.
One reason why this tax is so deeply flawed is that there is no mechanism for refunding the tax collected if the shares fall in value to below $3m (or even below $1m) in a subsequent tax year or are disposed of at a loss, meaning that the tax should never have been collected.
The situation gets worse if the cycle repeats in subsequent tax years because div 296 tax will be collected in every year that there is positive movement above $3m with no refund being given if the value falls below $3m in a subsequent year, meaning that you can end up paying the same tax multiple times even if the shares never grow above $4m.
Hence this tax may be bad news if you have shares whose values track side-wise over many years (eg. Bendigo Bank shares peaked above $17 in 2007 and never recovered) because you can end up paying the tax multiple times for the same upward movements ($3m to $4m) with no refunds for downward movements ($4m to $3m).

Re the tax treatment of a fall in asset values:

Consider the extremely artificial case where you have a pool of assets worth $3m (or less) at 30 June 2025. With no net contributions or withdrawals, the value jumps to $4m at 30 June 2026. Then it drops back to $2.5m in 2027, then $3.5m, $3m, $3.5m, $2.5m, and (finally) $4.5m in 2032.

Div 296 tax in 2026 is 15% x (4 - 3) / 3 x $1m = $49,995 after rounding the taxable percentage.

In 2027, Div 296 records a loss of $1m (remembering that it's only concerned with amounts over $3m). No, there's no refund.

In 2028, the gain of $0.5m above $3m reduces the recorded loss to $0.5m. No tax is payable.

In 2029, the loss increases to $1m, and it reduces to $0.5m again in 2030. Back up to $1m loss in 2031. Finally, in 2032, the gain of $1.5m above $3m wipes out the $1m loss and results in tax of 15% x (4.5 - 3) / 3 x $0.5m = $37,500.

So, apart from not having a refund mechanism, Div 296 is quite reasonable in its treatment of falls in asset values.

I appreciate the thrust of the article gives results under various examples.
In all examples, other than Patricia, only a round to 4 decimal places (not round down or round up) will yield the correct % for calculating the Taxable Super Earnings subject to Div 296 tax.
In addition, Patricia's Super Earning s should be $462,662 and NOT $462,562 (the difference between $6,538,737 and $6,076,075).
It is the detail which will get you in the end.
DetailDaddy

Many thanks to ASFA for providing this information which Trustees of many Defined Benefit Funds have failed to do.
In the case of Patricia, she would currently pay $78,536 tax on her $250,000 pension. Patricia would also pay 45% tax on any other income from employment or savings. Her taxable income on any capital gains she made would be reduced by 50%.
By comparison, a person whose total TSB was $6 million ($1.9 million in pension mode and $4.1m in an accumulation account) would, prior to Div 296, be paying about $42,750 in tax less any franking credits. Assuming 15% tax on $285,000 (earnings rate of 7%). Personal income outside of super would be taxed at marginal rates, not 45%.
Patrica’s pension will increase based on the CPI whereas the retiree’s super earnings will increase much faster.
The retiree also has the flexibility of withdrawing funds from super to help their grandkids buy a home or to cover the RAD if, for health reasons, they or their partner requires nursing home care but they don’t need to leave their home.

I’m not convinced that Family Court actuarial tables are appropriate for determining the balance for defined benefit pensions. My reason is that they make general assumptions about the longevity of the pensioner and their spouse.
A pensioner’s balance (and Div 296 tax liability) will fall if their spouse dies before them. However, taxes paid previously on the expectation of their spouse outliving them are unlikely to be refunded by the ATO.

I forgot to state that Patricia’s after tax pension income would be $171,464. If she had another $100k in investment income her total after tax income would be around $235,000
The retiree with a super balanced of $6 million and $100k in investment income would earn about $450,000 after tax.

I totally agree.

Given that these pensions are already taxed at the recipients marginal tax rate (less a 10% up to the TBC, if applicable) it seems like the family law method will way overvalue them and these retirees would be miles over taxed, especially when they don't have the flexibility to remove money from the system like accumulation retirees do.

Don't forget that you pay "double tax". Div 296 taxes increases in value (unrealised gains). You pay tax again when you realise an asset - at 10% of the gain. Not just Fred when he sells the farm, but all examples above are hit, except the MPs. There is no credit for tax paid on unrealised gains.

Can someone please confirm my understanding that the same "gain" (related to assets above $3m) is taxed twice. Once as an unrealised gain and second once realised.

You are quite right. The tax on realised gain, when you sell, is paid by the fund because the fund owns the asset. The tax on unrealised gains - Div 296 - is paid by the individual and is applied to the annual increase in value across all your super accounts and compiled by the ATO. They are two quite separate and unrelated taxes paid by two separate taxpayers. Although it is a tax liability against your personal TFN, you can request your super fund to pay it.

Fred the farmer may face no capital gains tax (or more accurately his SMSF) when the farm is eventually sold. If the asset supports a retirement income stream then all investment returns are tax free. When there are substantial increases in asset values an account balance for a retirement income stream can be well above $3 million, even though a much lower Transfer Balance Cap applied when the retirement income stream was established. The Transfer Balance Cap limits money in, higher balances after can occur due to increase in property or share prices.

The problem the farmer may encounter if he holds the asset in his SMSF in pension phase to take advantage of its tax-free status, is that these funds require the member to withdraw a mandated pension each year in cash and that mandated amount increases with age.
There will come a time where the fund has insufficient liquidity to provide that cash and pension funds cannot accept new contributions. Failure to meet this strict requirement, means the fund is taxed as an accumulation account and CGT will then apply.

You will note that the retired Politicians in these examples, Pedro and Patricia, have the biggest Balances in these examples despite contributing $Zero to their Taxpayer funded Government Defined Benefits Schemes and despite being only 55 years old.

According to the AFR, UNFUNDED or UNDERFUNDED Federal Government Defined Benefits Schemes (which include retired Politicians) alone now cost approx. $20 Billion in tax payer funds each year. And growing. And that's in addition to the Future Fund where over $200 billion of taxpayer funds has been set aside to fund these UNFUNDED or UNDERFUNDED Federal Government Defined Benefits Schemes which only Public Servants, Politicians and Judges benefit from.

ALL these Schemes need to be the subject of a Royal Commission so that their costs are reined in to something manageable and sustainable just like they were in the Private Sector. There are proven solutions to be utilised from the Private Sector's experiences, some rorts to be removed, and a cap on these tax payer funded benefits introduced. The exception is Military such Schemes which should be treated sympathetically as putting your life on the line for your Nation is fundamentally different to going to civilian work each day.

IF Superannuants are to be attacked in the name of Revenue raising then EQUITY demands that these Government Schemes, which are COSTS to all Taxpayers, be brought under control and similarly treated (ie. penalised Retrospectively and without Grandfathering).

Perhaps if the Politicians and Public Servants who benefit from these UNFUNDED or UNDERFUNDED Federal Government Defined Benefits Schemes were treated like everyone else, and had had to save for their retirements and take the same Risks with their retirement funds and ride the same ups and downs, they would be less likely to constantly fiddle with everyone else's retirement.

Angus. Is the $20 billion dollars in government outlays the gross or net amount? Any defined benefit pensioner who had income of $45k from part time employment or other investments in addition to their defined benefit pension would be paying 32.5% of their pension in tax.
Maybe the net cost to government revenue is only 2/3 of the figure quoted.
And has been pointed out by others the unfunded/underfunding of these schemes is a result of the government not contributing its share in the past.
As I mentioned before, because defined benefit pensions are taxed at marginal rates Patricia’s after tax pension income would be $171,464. If she had another $100k in investment income her total after tax income would be about $235,000
The retiree with a super balanced of $6 million and $100k in investment income would have an after tax income of around $430,000 assuming an earnings rate of 7% and no franking credits.

Just negotiate a max amount you can have in super….and the excess out each year after 30 June tax reconciliation. Simple. The money is then invested elsewhere and/or wasted in more Australian housing. The real question is ‘what is a fair amount for people to have as a maximum amount in super with its tax concessions?’

I reckon 2 times the transfer balance (which is indexed sort of) which would allow the transfer balance amount even following a 50% crash in stocks.

Simple’s and fair to avoid taxpayers funding extreme wealth tax free

Where can you easily work out the Family Law value of a CSS pension?

Not easy to find, but there is a legislative instrument made under the authority of the Family Law Regulations which sets out the scheme specific factors that have been approved for the various Commonwealth schemes
https://www.legislation.gov.au/F2025L00281/asmade/text/2

I have been asking a similar question, what is the calculation for a CSS pension?
CSS and the tax department don't know.
So if anyone knows please post it here, even the family law formula would help more than the 16 times figure.
Thanks OldbutSane for the post.

It is on the CSC site if you know where to look. Not every call centre person would know to go to the particular page and look at Step 2.
https://www.csc.gov.au/members/manage-super/Your-home-and-family/Separation-and-Divorce

The reply below by Old Super Hand (to which there was no "reply" button) does not give an actual method for calculation. It simply tells you that for a fee ($170 for members) CSS will provide the information you need to give an actuary, etc. CSS does not do the valuations.

If anyone who knows of an online calculator, a link would be appreciated - or maybe it is too complicated/their are too many variables to do this.

Thanks for the examples. Hard to understand the reaction to this being so strong when it is such a small part of gains. There is always the option to take it out of super and see if you can effect a lower tax rate. There are structural changes required to balance the fed budget and such a small change seems reasonable. I also thought the article by Harry Chernay was very well balanced. https://www.firstlinks.com.au/case-for-the-3-million-dollar-super-tax

The reaction(s) are because:

1. They have changed the rules. Superannuation is a long term goal. You can't force people along the Superannuation pathway, and then change the rules when the goalposts are in sight. If you want to bring in something like this because you are too incompetent to manage your budget, then it should be for people entering the workforce now;

2. It may be a "small part of gains", but these aren't necessarily liquid gains. People may not have the cash to pay the tax, even though "it is only $15 000". Similar to pensioners living in multi million dollar water front properties who can't affors the thousands of dollars in rate payments

"They have changed the rules. Superannuation is a long term goal."
Totally agree that changing the rules on someone who has followed the rules for 40 years or more is indefensible and can only be made fair by grandfathering the changes in the same way that CGT was grandfathered in September 1985.
"It may be a "small part of gains", but these aren't necessarily liquid gains"
Not only may the gains be illiquid, they may also include LOSSES if the asset(s) was/were acquired at a higher cost base compared to its value on 30 June. If you are going to determine tax liability on what is essentially MARKET VOLATILITY then the system must be sufficiently sophisticated to refund with interest incorrectly paid tax when unrealized gains fail to materialize or turn negative.

You have calculated earnings @ around 6% pa. This seems to be on the low side, just above fixed term deposit. If SMSF's were getting this sort of return it's hardly worth running your own SMSF. Most self directed SMSF investors I know are targeting 9 % to 13% capital return and 4% to 5% dividends. So somewhere around 13% to 18%. I would like to see the figures redone with varying rates of return, Division 296 tax is significant, especially if you calculate unrealised gains over 10 year period with CAGR of 15%. Do the calculation, it's frightening how much division 296 tax you will pay. It also significantly reduces the compounding of your SMSF. I am totally against the tax in its present form.

I have seen many calculations on how much to pay, but only limited to the first year of implementation. I wonder how it will work for subsequent years. Assuming for year 2025/2026, the TSB was $3M on 30 Jun 2025, and $3.2M on 30 Jun 2026, the 15% tax was paid. Again, assuming for year 2026/2027, the TSB on 30Jun 2027 was $3.2M, despite fluctuations during the year. Do I have to pay the 15% tax? Note that there was no change in the year for the TSB, but above the $3M threshold. Thanks in advance for clarifications from experts.

Div 296 tax is based on "earnings" - which is essentially based on the difference in TSB between the start and end of the year. Assuming there were no contributions or withdrawals made during the year, and you ended up with the same TSB of $3.2m at the beginning and end of the year, your earnings would be $3.2m - $3.2m = Nil. As a result, there would be no Div 296 tax to pay. However, if you only ended up back at $3.2m because you received $200,000 in pension payments during the year, those payments get added back into the end of year (adjusted) TSB figure. So your earnings would then be $3.4m-$3.2m = $200,000. As per above, you then calculate the taxable portion of the earnings as ($3.2m - $3m)/3.2m = 6.25% x $200,000 = $12,500. The tax is then $12,500 x .15 = $1,875.

"If you ended up back at $3.2m..." the $200K pension payment is being deducted from the EOFY total of $3.4M. That $3.4M is reflective of a $200K increase in TSB, which of itself is taxable. The pension payment is not being taxed.

Can we please not using the word "earnings", because it disguises what's really going on here

On idiocy, complexity and discrimination alone this tax is a very ugly pig wearing lipstick! Thanks for the worked examples though!

What this tax is clearly about is taking some fattened pigs to market.