Section 100A and family trusts: what changed, what it actually means

Technically reviewed 8 October 2026 · General information only, not personal advice. Read our disclaimer.

For four decades, section 100A of the Income Tax Assessment Act 1936 was a sleeper provision. It had been on the books since 1979, and almost nobody outside specialist tax practice paid it much attention. Then in February 2022 the ATO released a draft ruling and a draft practical compliance guideline, and overnight section 100A became the most consequential trust issue in Australian private practice. The final versions, TR 2022/4 and PCG 2022/2, landed in December 2022. Three years on, the practical position is reasonably clear, with one caveat worth stating up front: the ruling is the Commissioner’s reading of the section and the guideline is a statement of where the ATO will spend its compliance effort. Neither is the law, and the ATO has flagged the guideline for review, so the framework below describes the current administrative posture rather than a settled legal test.

This is the piece I wish the ATO had written. Plain English. The risk zones explained. What the ATO actually looks for. What you should be doing about it before the next trust resolution.

A note before we start. If you have a discretionary trust and you make trust distributions every year, this matters to you. If you’ve ever distributed trust profit to an adult child at university or to a corporate beneficiary, this matters to you. If your accountant has not had this conversation with you in the last twelve months, you should ring us or ring them and have it now.

What section 100A actually does

In one sentence: where a trust distribution is part of a “reimbursement agreement,” section 100A treats the distribution as if it never happened, and the trustee is taxed at the top marginal rate plus levies on the amount instead.

That sentence sounds dry. Here’s what it means in practice. A discretionary trust distributes profit to a beneficiary on paper. The beneficiary is named in the resolution. The beneficiary’s tax return picks up the income at their marginal rate. So far, ordinary trust mechanics. Section 100A bites if there’s an arrangement, written or unwritten, where some other party effectively receives the economic benefit of that distribution. Once the elements are made out, the section operates of its own force, the named beneficiary is deemed not to have been presently entitled, and the trustee is assessed under section 99A at the top marginal rate: 47% including Medicare levy in the usual resident-trust case. It’s not a Commissioner discretion. The ATO’s role is in finding the arrangement and assessing the trustee, not in deciding whether the section applies.

The elements are set out in the section itself: a beneficiary’s present entitlement that arises out of, or in connection with, a reimbursement agreement; an agreement that provides for a benefit to someone other than that beneficiary; and a purpose of reducing someone’s tax. Each element has to be made out. A cash mismatch on its own is evidence, not a conclusion. There’s also a critical carve-out: an arrangement that is part of an “ordinary family or commercial dealing” isn’t an “agreement” at all (s 100A(13)). That carve-out is where the real fight has been for the last forty years.

One feature of section 100A worth knowing about up front. Most income tax positions sit behind a four-year amendment window, get past four years without an amendment and the position is generally locked in. Section 100A doesn’t work like that. Item 17 of the table in section 170(10) of the ITAA 1936 gives the Commissioner an unlimited period in which to amend an assessment for a section 100A position. The clock effectively never runs out. That’s why a back-catalogue of weak distributions is more dangerous than a single weak year, there’s no time-bar to hide behind.

What changed in 2022 (and what didn’t)

The provision didn’t change. What changed was the ATO’s stated interpretation of when a distribution falls outside “ordinary family or commercial dealing” and into reimbursement agreement territory. TR 2022/4 sets out the technical position (amended by Addendum TR 2022/4A1 in 2023, after the Guardian and BBlood Full Court appeals). PCG 2022/2 sets out how the ATO will allocate compliance resources, the famous risk zones.

What the ATO put on the table in TR 2022/4 was a position that “ordinary family or commercial dealing” doesn’t include arrangements that are contrived, particularly arrangements where the beneficiary doesn’t actually receive the economic benefit. That sounds obvious until you realise how much trust planning quietly relied on adult-child beneficiaries who never saw a dollar of the distribution they were named in.

The case law that’s now layered on top, BBlood Enterprises Pty Ltd v Commissioner of Taxation [2022] FCA 1112 (upheld on appeal as B&F Investments Pty Ltd as trustee for the Iluka Park Trust v Commissioner of Taxation [2023] FCAFC 89), Commissioner of Taxation v Guardian AIT Pty Ltd [2023] FCAFC 3, and the trust law context from Owies v JJE Nominees Pty Ltd [2022] VSCA 142, has reinforced rather than softened the position. The trustee’s discretion has to actually be exercised and exercised properly (the trust-law point from Owies, which is about the validity of the trustee’s decision rather than section 100A itself), and the distribution has to be one that an ordinary family or commercial dealer would make (the s 100A(13) carve-out, as elaborated in TR 2022/4).

The risk zones, in plain English

PCG 2022/2 sorts trust distribution patterns into three risk zones based on how the ATO will allocate compliance resources. (The draft guideline had a fourth, blue zone for everything in the middle; the final guideline dropped it, so an arrangement that is neither green nor red is simply assessed on its facts.) This is a compliance allocation tool, not a definitive legal answer, sitting in the green zone doesn’t legally guarantee section 100A doesn’t apply, and sitting in the red zone doesn’t legally guarantee it does. But for practical purposes, the zones are how the ATO is actually deploying its review resources, and that’s the framework most accountants now plan around.

White zone, arrangements entered into before 1 July 2014. The ATO has said it won’t generally apply compliance resources to arrangements entered into before that date unless there’s evidence of fraud or evasion. That is a statement about where the ATO will look, not a legal exemption: section 100A applied before 2014 and still applies to those arrangements. “We have always done it this way” is not a tax defence, and a pattern that started before 2014 does not carry new arrangements with it.

Green zone, low risk. The classic green zone arrangement is a trust distribution to an adult beneficiary who actually receives the cash, retains it, and uses it for their own purposes. A distribution to a non-working spouse who genuinely receives and controls the money. A distribution to an adult child who is over 18, a tax resident, and gets the cash. The ATO has said it will not generally apply compliance resources to green zone arrangements.

Between green and red. The final guideline has no named zone for the middle ground, so an arrangement that is neither clearly green nor clearly red is assessed on its facts. Most loan-back arrangements, most arrangements where the cash benefit is shared between family members in a way that’s normal but not perfectly clean, sit here. Not safe, not at high risk. Plan to be able to demonstrate the dealing was ordinary if asked.

Red zone, high risk. The ATO has said it will apply compliance resources to red zone arrangements. The classics: a distribution to an adult child who never sees the cash because the trustee retains it; a distribution that gets reimbursed back to a parent for past expenses (school fees being the textbook example); a distribution to a beneficiary on a low marginal rate where the cash is then routed back to a beneficiary on the top rate; round-robin distributions; arrangements where the cash flow doesn’t follow the paper.

The red zone scenarios that catch most people

The two patterns I see most often in clean-up work are these.

The “adult child at uni” distribution where the cash never moves. Family trust distributes $30,000 to an 18-year-old uni student. On paper, the student declares it and pays tax at the lower marginal rate. In practice, the cash stays in the trust’s offset account and is used by the parents for ordinary household spending. Section 100A red zone. The ATO position is that this is exactly the kind of contrivance the section was written for.

The corporate beneficiary loop where the cash benefit accrues elsewhere. Family trust distributes to a bucket company at the corporate rate. Bucket company holds an unpaid present entitlement against the trust. The trust spends the money on lifestyle expenses for the parents. The bucket company holds an entitlement that nobody intends to pay, while the parents have had the economic benefit. Red zone, and on top of that the broken structural integrity of the bucket company arrangement is its own problem.

A red zone distribution doesn’t automatically mean a section 100A assessment. It means the ATO will look at it. If the distribution was genuinely “ordinary family or commercial dealing”, for example, the parent spent the money on the child’s tuition, accommodation and living costs, the dealing might still be defensible. But the burden is on the trustee to be able to demonstrate that. Without contemporaneous evidence, the position becomes very hard to defend.

What the ATO is actually looking for in practice

In the engagements we’ve seen go through review, the ATO’s questions tend to focus on three things.

First, where the cash actually went. Bank statements, transfers, the pattern of distributions over multiple years. If the cash didn’t follow the paper, that’s the strongest single signal of a red zone arrangement.

Second, why the named beneficiary was chosen. Was it because the trustee, exercising discretion in good faith, decided this beneficiary should benefit? Or was it because the beneficiary’s marginal rate was lower? Both motivations can be present, but the dealing has to be capable of being characterised as ordinary, not as a pure tax-rate-arbitrage.

Third, what records exist. Trust resolutions need to be in place before 30 June each year. The resolution has to actually exercise discretion (not be a pro-forma “to all default beneficiaries”). The cash flows need to be reconcilable to the resolutions. Where there are loan accounts, the loan accounts need to be properly recorded and managed.

If the answers to all three questions are clean, you are in a far better position to explain the arrangement. Records can prove what happened; they cannot rescue an arrangement that doesn’t work. If any of the three are messy, the exposure is real.

What you should be doing now

If your trust makes distributions every year (and most do), here’s the operational checklist.

1. Run a 100A risk audit on the last three years of distributions. Map the named beneficiary against the actual cash flow for each distribution. Identify any distributions that don’t sit clearly in green or white. Document the dealing for any that don’t.

2. Tighten the resolution practice. The trustee must actually exercise its discretion within the time the deed and the tax law require, which for an ordinary 30 June trust means by the end of the income year. Whether the deed demands a signed document by that date is a question of the deed. Our practice is a signed, dated resolution before 30 June every year, because it ends the argument before it starts. Never back-date one. We see resolution practice fall apart most often in years when the family has had a busy June or a personal disruption, and a missed or back-filled resolution creates exactly the uncertainty you do not want in a trust year.

3. Match the cash to the paper. This is our firm’s risk rule rather than a safe harbour in the legislation: where a distribution is made to a beneficiary, we want the cash to follow within a reasonable timeframe, because a benefit that lands with the named beneficiary is the clearest answer to the question section 100A asks. Where it can’t, the entitlement should be properly recorded and managed, either as a payable, or, where appropriate, on a complying loan footing.

4. Don’t make the obvious red zone distributions any more. Adult-child-at-uni-where-cash-stays-in-trust is the pattern that most needs to stop. If the child genuinely needs and receives the money, that’s fine. If they don’t, the distribution shouldn’t have gone there.

5. Document the dealing. If the distribution was for a specific purpose (paying for the child’s education, contributing to a deposit, supporting a family business), write it down at the time. Contemporaneous evidence is the strongest defence available.

Good records explain the arrangement. They do not make a bad one work.

The conversation we have with clients now is different to the one we had four years ago. The ATO has been clearer than it has been in three decades about what it expects, and what it doesn’t accept. The firms that have already adjusted are operating in a clean zone. The firms that haven’t are sitting on a back catalogue of distributions that are hard to defend. If your accountant hasn’t walked you through your trust’s exposure under PCG 2022/2 and TR 2022/4, that’s the conversation to have before the next 30 June.

If you’re not sure where you sit, ring us. The 100A audit takes us a couple of hours. The cost of doing it is rounding error against the cost of a top-marginal-rate assessment on a multi-year back-catalogue, and on section 100A there’s no four-year amendment window to hide behind.

Raise higher.