Tag Archives: SMSF

Investing in real estate with a SMSF – traps & entanglement


There is more to investing in real estate together with a self managed superannuation fund (SMSF) than meets the eye. It can be fraught and illegal under SMSF rules. This blog looks at why.

Joint tenancy ownership compared to ownership by tenants-in-common

The title of this blog piece does not refer to investing jointly with a SMSF and this is deliberate. Co-ownership of land can be joint: viz. as joint tenants, where a surviving joint owner or owners take the interest of a joint owner who dies, or can be as tenant/s-in-common (TsIC) where each co-owner owns a discrete co-ownership interest in a fixed proportion of the whole outright ownership interest in the land which, in the case of an individual on his or her demise, will form a part of his or her estate just as an interest in land owned by a sole individual owner would.

Which type of ownership works where a (trustee of a) SMSF is a co-owner?

Joint tenancy is usually only appropriate for life partners. Investing in a joint tenancy can also work for joint trustees of a trust where, on the death of a trustee, it is appropriate that the property be legally owned by surviving trustee/s.

The point here is joint tenancy is inapt and inappropriate investing between a member of a SMSF and the SMSF obliged to deal on arm’s length basis under section 109 of the Superannuation Industry (Supervision) Act (C’th) 1993 (SIS Act) with all parties including the member when co-ownership of an asset is under contemplation. A SMSF needs to acquire assets at arm’s length and assets acquired need to have a discrete integrity which joint tenancy ownership doesn’t give.

So, if there is to be co-ownership between members or related parties and a SMSF investing in land, it needs to be as TsIC.

Related parties of a SMSF include:

  • relatives of the Members (spouse, children, siblings, etc.);
  • the (business) partners (Partners) of the Members;
  • the spouse and children of the Partners;
  • companies (Companies) controlled by the Members or any of the above (Associates);
  • the members of the SMSF (Members) themselves; or
  • trusts controlled by the Members, Associates and Companies.

(See Part 8 associates in Sub-division B of Part 8 of the SIS Act.)

Co-ownership of land between SMSF members and the SMSF as tenants-in-common

There is a further trap where SMSF members or other related parties and a SMSF contemplate co-ownership of land as TsIC where the land is residential property (RP):

Prohibition on acquisition of assets from superannuation fund members and related parties

With very limited exceptions, real estate with a residence cannot be business real property (BRP): see Self Managed Superannuation Funds Ruling SMSFR 2009/1 Self Managed Superannuation Funds: business real property for the purposes of the Superannuation Industry (Supervision) Act 1993.  A SMSF cannot acquire an asset from a related party of the SMSF (section 66(1) of the SIS Act) unless an exception applies such as the exception for BRP (permitted under para 66(2)(b) of the SIS Act).

A breach by a trustee of a SMSF of section 66 can result in criminal prosecution and imprisonment of the individual trustee/s or director/s of the trustee (TEsDRs), as the case may be, for up to one year (sub-section 66(4) of the SIS Act).

It follows that the trustee of a SMSF cannot, or likely cannot, lawfully acquire RP already owned by a member/related party of the SMSF unless the RP is BRP. This prohibition works in substance as schemes that have the result that RP of a member/related party of a SMSF is acquired by a SMSF, say indirectly via sale to the SMSF and then purchase back by the SMSF from an intermediary unrelated to the SMSF, are also caught by section 66 and are similarly prohibited: sub-section 66(3).

Implications for related co-owners who own RP as tenants-in-common with a SMSF

This has further implication when RP is acquired and co-owned where a SMSF is an established co-owner: let us say where the RP is purchased in an arm’s length sale on the open market.

The SMSF owns a part of the RP as a TsIC but section 66 prohibits the SMSF from buying more of the RP from the related TsIC who is now a co-owner too. That further purchase would be acquisition of an asset from a member/related party. The same anti-scheme rule in sub-section 66(3) again applies to prevent the SMSF acquiring a further interest owned by a related party as a TsIC indirectly through a scheme.

An unsatisfactory entanglement

So the entanglement of a related party in the ownership of RP effectively prevents the SMSF from ever owning the whole of a RP it invests in as TsIC with a related party. This bears on, or should have borne on, the investment decision of the SMSF trustee to invest in the RP in the first place.

Entanglement gets worse when a SMSF has individual trustees and these individual trustees are members of the SMSF with whom the SMSF co-invests in RP. Under land law in most Australian states and territories only these individuals appear on title as registered owners of the RP. Without further steps, such as registering a caveat, the trustees of the SMSF, obliged to act at arm’s length from themselves, are poorly placed to assert co-ownership of the RP by the SMSF and to comply with mandatory covenants applicable to a SMSF including:

(b)   to exercise, in relation to all matters affecting the fund, the same degree of care, skill and diligence as an ordinary prudent person would exercise in dealing with property of another for whom the person felt morally bound to provide;

(d)   to keep the money and other assets of the fund separate from any money and assets, respectively:

  (i)   that are held by the trustee personally; or

  (ii)   that are money or assets, as the case may be, of a standard employer – sponsor, or an associate of a standard employer – sponsor, of the fund;

(e)   not to enter into any contract, or do anything else, that would prevent the trustee from, or hinder the trustee in, properly performing or exercising the trustee’s functions and powers;

from sub-section 52B(2) of the SIS Act

Entanglement disrupting sale of the TsIC interest by a SMSF

An investment in an asset which is not discretely saleable raises further section 52B covenant difficulty. The section 52B covenants continue:

(f)   to formulate, review regularly and give effect to an investment strategy that has regard to the whole of the circumstances of the fund including, but not limited to, the following:

  (i)   the risk involved in making, holding and realising, and the likely return from, the fund’s investments, having regard to its objectives and its expected cash flow requirements;

  (ii)   the composition of the fund’s investments as a whole including the extent to which the investments are diverse or involve the fund in being exposed to risks from inadequate diversification;

  (iii)   the liquidity of the fund’s investments, having regard to its expected cash flow requirements;

  (iv)   the ability of the fund to discharge its existing and prospective liabilities;

paragraph 52B(2)(f) of the SIS Act

Investing in a marooned asset

So does a trustee of a SMSF who invests in a asset that is marooned, because it can’t be readily sold without the co-operation of a co-owner or co-owners also selling, adequately deal with the risks referred to in paragraph 52B(2)(f)? Assumption that a related party TsIC will always co-operate with a co-owner trustee of a SMSF TsIC is incompatible with the section 109 of the SIS Act obligation of the trustee to act an arm’s length basis in its dealings including dealings with related parties.

Based on the section 52B covenants and section 109 the trustee/s of a SMSF should establish proper motive for making an investment as a co-owner in RP. To do that there likely needs to be either an exchange of:

  • tag along drag along rights; or
  • rights to require other TsICs to buy each other out of their interests;

so the SMSF can realise its investment in a TsIC investment interest in RP when it needs to meet its s52B(2)(f) covenants without being marooned in the investment.

The mandatory covenants in section 52B on trustees of SMSFs are between the trustee/s of the SMSF and the members of the SMSF. When they are the same people there are only occasional cases where a member would sue trustees for breach. The covenants are not civil penalty provisions.

Civil penalty provisions

In the SIS Act civil penalty provisions have these potential consequences for SMSFs:

  1. breach can lead to the Australian Taxation Office as SMSF regulator (ATO as R) issuing a notice of non-compliance (NONC) to a SMSF so it is no longer a complying superannuation fund where:
    1. non-complying superannuation funds pay 45% income tax on their assessable income; and
    2. the assessable income of a fund that becomes a non-complying superannuation fund under a NONC must include the value of the assets of the fund, less undeducted contributions, at the beginning of the income year when the fund becomes non-complying. This is a significant penalty as it effectively taxes the fund’s accumulated assets at the 45% rate: see Subdivision 295-E of the Income Tax Assessment Act 1997.
  2. intentional breach can result in criminal prosecution of TEsDRs: section 202 of the SIS Act;
  3. administrative penalties on TEsDRs (in less serious cases taken not to warrant the above): s166 of the SIS Act; and
  4. the ATO as R can give the TEsDRs directions to rectify (section 159 of the SIS Act) the breach or educational directions (section 160 of the SIS Act).

    Consequences 1 to 3 don’t apply to a breach that is solely or simply a breach of the section 52B mandatory covenants. Consequences 4 can happen though: the ATO as R can give TEsDRs a direction to rectify requiring sale of a marooned TsIC interest acquired in RP in breach of the covenants in paragraph 52B(2)(f).

    Sole purpose fails

    Even where the RP is let out under a lease entirely at arm’s length to an arm’s length tenant there could still be a sole purpose civil penalty provision problem under section 62 of the SIS Act where the purpose of an investment by the SMSF in RP was not so much to generate returns to the SMSF, or to assist a SMSF to fund the payment of SMSF benefits to members, but rather to finance SMSF member acquisition of an investment property. Not bothering to arrange the above rights for the SMSF amplifies the prospect that a SMSF auditor or the ATO as R will reach that conclusion about the illicit purpose of the trustee/s of the SMSF.

    Where the RP is acquired for a member or related party of the SMSF to live in then breach of the section 62 civil penalty provision will be yet more serious and clear cut.

    Entanglement of financing

    The need for a SMSF member and SMSF co-investors in RP as TsIC to co-operate extends further. The SMSF member borrowing with recourse or security over the property can amount to a charge over the property breaching SIS Regulations 13.14 and 13.15 and, where the recourse or security is called in, the SMSF might find itself co-investing with a financier eager to sell up the RP. In June 2011 the Commissioner and tax professionals considered these issues which were reported in National Tax Liason Group technical minutes. These can be difficult to locate on the somewhat dynamic Australian Taxation Office website so we have uploaded a copy here

    It follows that a mortgage can’t be given to the financier of the co-owning member/s of the SMSF over the RP co-owned by the SMSF. Giving security over the TsIC interest only of the member/s of the SMSF who borrow only may be possible but that security needs to be carefully target only the borrower’s TsIC interest so that it has no reach to impact or to give any recourse against the TsIC interest of the SMSF in the RP.

    Unit trust alternative?

    Investment of more than 5% of a superannuation fund in in-house assets under Part 8 of the SIS Act can give rise to breach of a civil penalty provision with the potential Consequences 1-4 described above: section 84 of the SIS Act.

    In 1999 the meaning of in house asset was widened to curtail significant investment by SMSFs in particular in related unit trusts. A popular strategy, to establish a unit trust to hold RP in which SMSFs and their related parties could hold units, could no longer be used without running into an in house asset problem.  A carve-out to in house asset treatment was extended in Division 13.3A–In-house assets of superannuation funds of the SIS Regulations for companies and unit trusts that:

    • are continuously non-geared, that is never have liabilities;
    • have assets that are not investments in other entities;
    • do not conduct a business; and
    • neither lend nor borrow

    so that SMSFs could invest in shares or units in them without these being in house assets.

    An exception in sub-paragraph 66(2A)(a)(iv) of the SIS Act means that investment in say a SIS Regulation 13.22C non-geared unit trust to hold RP is not only excluded from being an in-house asset under paragraph 71(1)(j), but its acquisition from a related party is not prohibited under sub-section 66(1).

    Non-geared unit trust compared to co-investing in residential property as tenants-in-common

    This is a significant advantage over investing in an interest as a TsIC in RP. So a SIS Regulation 13.22C non-geared vehicle should be seriously considered as an alternative to investing with a related party in RP as a TsIC. Still a SIS Regulation 13.22C non-geared unit trust is nevertheless a challenging structure for indirect SMSF investing in RP as:

    1. the compliance requirements, especially those that cause abrupt loss of the in house asset exclusion in SIS Regulation 13.22D are daunting (albeit the problems with investing as a TsIC in RP are covertly so and are all across the SIS Act , as this post illustrates); and
    2. units in a non-geared unit trust that don’t amount to all of the units in the trust still have the same propensity to be marooned assets of the SMSF unless the investing SMSF can compel all other unit holders to buy or drag along when the SMSF needs to realise its investment.

Individual trustees of a SMSF – useful or a hindrance to SMSF decision making?

In 2008 I wrote an article published in SMSF magazineSeparatedDot

Having individuals as trustees of SMSFs – companies versus individuals as trustees of SMSFs 

which mainly looked at when having individuals as trustees of a SMSF can prove a false economy.

A recent Full Federal Court case authoratively sets out reality checks for SMSFs with individual trustees: Frigger v Trenfield (No 3) [2023] FCAFC 49 concerned an accountant, Mrs. Frigger and her husband, Mr. Frigger, who had been made bankrupt.

The bankrupts took action against their official receiver to require the official receiver to treat assets, which had been sequestrated by the official receiver, as the property of their superannuation fund, the Frigger Superannuation Fund (FSF). The aim of the bankrupts was to bring these assets within sub-paragraph 116(2)(d)(iii)(A) of the Bankruptcy Act (C’th) 1966 as assets not divisible amongst the creditors of undischarged bankrupts.

SMSF assets need to be owned by the trustees

Regulation 4.09A(2) of the Superannuation Industry (Supervision) [“SIS”] Regulations contains the following prescribed standard:

A trustee of a regulated superannuation fund that is a self managed superannuation fund must keep the money and other assets of the fund separate from any money and assets, respectively:

(a) that are held by the trustee personally; or

(b) that are money or assets, as the case may be, of a standard employer-sponsor, or an associated of a standard employer-sponsor, of the fund

Despite the above Regulation 4.09A(2), the comparable covenant in paragraph 52(2)(g) of the SIS Act 1993 and trust law principles that forbid trustees from mixing their own property with property held on trust, the bankrupts ran bank accounts and held assets, including rental earning real estate, in their own names which the bankrupts claimed were assets of the FSF in Frigger v Trenfield (No 3). These assets had not been put into the names of all trustees being Mr and Mrs Frigger, their children Mr Michael Frigger and Ms Jessica Frigger jointly either expeditiously or at all.

In my article I looked at the work needed and likely cost to keep assets in the name of the trustees on changes of trustee of a SMSF. The FSF, where there were numerous changes of trustee and numerous assets including real estate, was a chronic case of the imperative to keep fund assets in the name of the trustees and the significant effort and costs required my article considered. The bankrupts and the other trustees of the FSF didn’t act on the imperative made plain below in this post.

Further accounts, tax and SMSF returns and other records relied on by the bankrupts to show that the assets were owned by the FSF, and so attracted sub-paragraph 116(2)(d)(iii)(A) protection from sequestration, were found by the Full Federal Court to be deficient and insufficient to convince the Full Federal Court that the assets were assets of the FSF.

How trust property must be held. protected and made good by individual trustees of a SMSF

In the course of the lengthy joint judgement in Frigger v Trenfield (No 3) the Full Federal Court (Allsop CJ, Anderson And Feutrill JJ) elaborated on how individuals, who are co-trustees of a trust such as a SMSF must reach decisions and hold, protect and, if need be, make good the property of the trust. This elaboration is an aide-memoir for individual trustees of a SMSF:

  1. Decisions of co-trustees must be unanimous: Luke v South Kensington Hotel Co (1879) LR 11 Ch D 121 at 125. In the case of In the Estate of William Just (deceased) (No 1) (1973) 7 SASR 508 (Estate of Just), where money had been paid into a bank account held in the name of one of two co-trustees, Jacobs J said (at 513–514):
… In the case of co-trustees of a private trust, the office is a joint one. Where the administration of the trust is vested in co-trustees, they all form, as it were, but one collective trustee and therefore must execute the duties of the office in their joint capacity. Sometimes, one of several trustees is spoken of as the active trustee, but the court knows of no such distinction: all who accept the office are in the eyes of the law active trustees. If anyone refuses or is incapable to join, it is not competent for the other to proceed without him, and if for any reason they are unable to appoint a new trustee in his place, the administration of the trust must devolve upon the court. Though a trustee joining in a receipt may be safe in permitting his co-trustee to receive in the first instance, yet he will not be justified in allowing the money to remain in his hands longer than reasonably necessary. The proper course is to pay trust money into a joint bank account in the names of both or all the trustees. …
  1. In general, a trustee must discharge the duties and exercise the powers of trustee personally. Where there are co-trustees, in the absence of unanimous agreement, actions taken independently of the other co-trustees lack authority and do not bind all trustees: see, e.g., Lee v Sankey (1872) LR 15 Eq 204 at 211; Astbury v Astbury [1898] 2 Ch 111 at 115–116; Pelham v Pelham & Braybrook [1955] SASR 53 at 57. A co-trustee is not and cannot be bound by a decision of the majority and each co-trustee must turn his or her mind to the exercise of the applicable power and decide on the action to be taken: see, e.g., Cock v Smith (1909) 9 CLR 773 at 800; Re Billington [1949] St R 102 at 111, 115.
  2. As the authors of Ford and Lee: The Law of Trusts (looseleaf at 13 February 2020, Thomson Reuters) (Ford and Lee) observe (at [9.11090]): “A consequence of the unanimity rule is that trust business cannot be transacted except at a meeting at which all the trustees, or their delegates, are present; and that where the trustees cannot agree about a course of action the status quo prevails.” In the case of a regulated superannuation fund, if the superannuation entity has a group of individual trustees, the trustees must keep, and retain for at least 10 years, minutes of all meetings of the trustees at which matters affecting the entity were considered: s 103(1) SIS Act.
  3. It may also be possible for co-trustees to ratify an action taken by another trustee without prior agreement: Meeseena v Carr (1870) LR 9 Eq 260 at 262–263; Hansard v Hansard [2015] 2 NZLR 158 (Hansard v Hansard) at [47] citing Thomas Lewin and others, Lewin on Trusts (18th Ed, Sweet & Maxwell 2018) at [29-209]. However, for ratification to be effective, the ratifying co-trustee(s) must know of the essential detail of the act or decision in question. It must be more than passive acquiescence to a decision made by another trustee. The act of ratification must show that the co-trustee(s) considered the exercise of power as trustee and consented to the action taken. Thus, “[s]ubsequent approval of financial statements [by all trustees] may therefore not be sufficient to amount to ratification of actions taken without the unanimous approval of trustees”: Hansard v Hansard at [51]. Something more than mere approval of financial statements would be necessary to demonstrate the required ‘act of ratification’.
  4. Property of the trust must be held jointly and it is a breach of the trustees’ duties not to ‘get in’ the trust property and hold the legal (or where applicable equitable) title to that property jointly or otherwise hold the property under joint control: Lewis v Nobbs (1878) 8 Ch D 591 (Lewis v Nobbs) at 594; Guazzi v Pateson (1918) 18 SR (NSW) 275 at 282. As Hall VC explained in Lewis v Nobbs, the rationale for the duty is to ensure that trust property is not dealt with improperly by one of the co-trustees or without the agreement of all co-trustees.
  5. Clause 140 of the FSF Trust Deed required the trustees of the FSF to ensure that money received by the fund was, amongst other things, deposited to the credit of the fund in an account kept with a bank chosen by the trustees. That is, chosen by the co-trustees by unanimous agreement.
  6. While there may be circumstances in which property is conveyed to, or acquired, by one of a number of co-trustees as trust property with the consent of the other trustees, it remains the duty of all trustees to ensure that title to the property is ultimately convey (sic.) to and held by all co-trustees jointly within a reasonable time thereafter: Estate of Just at 513–514. Any inconvenience that might result from the changing composition of the co-trustees from time to time does not absolve the trustees of that duty: see, e.g., Trustees of the Kean Memorial Trust Fund v Attorney-General (SA) [2003] SASC 227; 86 SASR 449 at [94].
  7. Further, in the absence of an express provision permitting mixing, it is also a duty of a trustee to keep personal and trust property separated: Associated Alloys at 605 [34]. A trustee has a positive duty “to distinguish the piece of property he … acquires from other similar things which he may obtain for himself or in which he may be interested”: Van Rassel v Kroon (1953) 87 CLR 298 at 302–303 (Dixon CJ); Heydon and Leeming, Jacob’s Law of Trusts in Australia (8th ed, LexisNexis, 2016) at [17-02]. It is also trite that a trustee cannot unilaterally repudiate the trust and appropriate the trust property: Ford and Lee (looseleaf as at 14 May 2021) at [17.4530].
  8. A co-trustee is obliged, as part of the duty to get in trust property, to bring proprietary claims against another trustee who, in breach of trust, has misappropriated or mixed trust property with his or her own property. Likewise, a trustee who has participated in such a breach of trust has an obligation, notwithstanding his or her own wrongdoing, to make good the trust property and, if necessary, to institute proceedings against other trustees who participated in the wrongdoing to make good the loss: Young v Murphy [1996] 1 VR 279 at 282–284, (per Brooking J), 300 (per Phillips J), 319 (per Batt J).
  9. The above principles have significance in this appeal because the evidence before the primary judge was to the effect that Mrs Frigger alone held the legal title to the funds in the BW1 and BOQ2 accounts, Mr Frigger alone held the legal title to the funds in the BOQ1 account and Mr and Mrs Frigger jointly held the legal title to the securities in the Main Portfolio. Also, Mrs Frigger held the legal title to the Como and Bayswater properties. Therefore, the legal title to the disputed assets was not held by the co-trustees of the FSF jointly immediately before the sequestration orders were made or by the sole trustee of the FSF (H & A Frigger) immediately after the sequestration orders were made. There was no evidence before the primary judge that any of the individual co-trustees had taken any steps at any time to ‘get in’ the trust property and have funds held in a bank account in the joint names of the four individual trustees (or HAF), to transfer the securities in the Main Account into a CHESS holding account held in the joint names of the four individual trustees (or HAF), or to transfer the Como and Bayswater properties into the joint names of the four individual trustees (or HAF).

So the bankrupts fell short of the exacting requirements on individual trustees.

Majority individual trustee decisions and the unanimity rule?

The above passage from Frigger v Trenfield (No 3) does not countenance a SMSF trust deed that allows individual trustees to reach decisions by a majority of the individual trustees.

It is questionable whether a SMSF trust deed allowing for majority trustee decisions, not in conformity with the unanimity rule, satisfies paragraph 17A(1)(b) of the SIS Act. There is the prospect that a superannuation fund governed by such an (attempted) SMSF trust deed may not be (qualify as) a SMSF. That may be so unless a deeming clause in the trust deed overrides the decisions by majority clauses in the trust deed in which case individual trustees are obliged to comply with the unanimity rule nonetheless so that the superannuation fund can be regulated as a SMSF.  

No scope for equivocal individual ownership of SMSF assets

Clearly there can be no presumption that the Australian Taxation Office, a tribunal or a court will accept that an asset not in the name of a SMSF is an asset of the SMSF. The bankrupts may have hoped that, by having individual trustees, the FSF was usefully positioned to assert ownership by the SMSF of multiple properties in the names of one or more trustees but not all of them.  But the above principles set out in Frigger v Trenfield (No 3) mean that individual trustees of a SMSF, in particular, need to produce valid minutes of meetings of trustees and documents that establish and explain why an asset, and particularly an asset not in the name of all of the trustees, is an asset of a SMSF before an asset will be inferred to be an asset of a SMSF with individual trustees.

Advantages of a corporate trustee

A corporate trustee, especially a corporate trustee that acts in no other capacity, is better placed to assert ownership of any asset in the name of the corporate trustee and won’t be at the same risk of mixing trust property as a practical matter. Although corporate trustees need to keep records of their decisions, corporate trustee can have streamlined decision making procedures which need not require meetings of directors and their minuting where individual trustees can’t streamline their decision making.


Frigger v Trenfield (No 3) is a stark reminder that ownership of assets of a superannuation fund by the trustees needs to be unequivocal should a member of the fund go bankrupt. There are numerous breaches of standards beyond Regulation 4.09A(2) and bankruptcy troubles that can arise where assets of a SMSF are not kept separate from assets that are not successfully.

The member’s say over where their SMSF death benefits are to go

On 15 June 2022 the High Court in Hill v Zuda Pty Ltd [2022] HCA 21 ruled against a challenge to earlier court decisions that a binding death benefits nomination (BDBN) made under the governing rules (GRs) [trust deed] of a self managed superannuation fund (SMSF) need not lapse after three years of the nomination as a comparable nomination made under a retail or industry superannuation fund must under Regulation 6.17A(7) of the Superannuation Industry (Supervision) Regulations.

Non-lapsing SMSF BDBNs confirmed

The High Court’s decision confirmed that this regulation, which operates under the force of sub-section 59(1A) of the Superannuation Industry (Supervision) Act 1993 (SIS Act), did not apply to or limit the effectiveness of a BDBN made as non-lapsing more than three years earlier by a member of a SMSF under GRs of the SMSF that allowed for non-lapsing BDBNs.

External discretions and directions

It follows that other limitations in section 59 and Regulation 6.17A also have no application to a SMSF. Among them is the original rule in sub-section 59(1) which broadly, with exceptions, invalidates GRs that allow a person other than the trustee to exercise a discretion under the GRs of a fund. A related provision, which is similarly confined to retail and industry funds, and especially clearly so, following Hill v. Zuda; is section 58 of the SIS Act which broadly, with exceptions, invalidates a GR under which a person may direct the trustee (to act or not act in a certain way in the conduct of the fund).

Implications for death benefits directions

Without the reservation for superannuation funds other than retail and industry funds it would be thought that death benefits agreements or “SMSF wills” included in, or which took effect under, the GRs of a SMSF which prevailed over, or overrode, a power or discretion of a trustee to decide on which dependant of a deceased member was to take death benefits of the member would fail. It seems indisputable now that trustees of SMSFs and Small APRA Funds, can be effectively directed and bound by members on where their death benefits are to go under valid GRs whether that is by nomination, direction or in some other way .

It follows that SMSF GRs are free to allow for members, either not as or not acting in their capacity of trustees or as directors of a trustee, to control decisions over to whom their death benefits are to be paid. This is starkly different to a retail or industry fund that cannot outsource the duties and obligations of its trustee to determine who is take the benefits of a deceased member unless the BDBN requirements in Regulation 6.17A are strictly complied with.

Too much latitude to a SMSF then?

Many SMSF GRs are and will be unconstrained or vague in setting out process of where death benefits of a deceased member are to go without the stricture of Regulation 6.17A. Accordingly SMSF GRs that give a member a binding and final say over to whom their death benefits are to be paid need to be carefully considered when a death benefits decision is to controlled by a member in any way.  A reckoning should be taken as to whether the SMSF GRs, that give a member this autonomy over their benefits; is workable, predictable and will limit opportunity for fraud on the member and the member’s family.

A dentist might be better than the cheapest guy with a drill


Proprietary company setups are not all the same. The $512 ASIC registration fee doesn’t get you a constitution for your company. Company constitutions vary and are on a quality spectrum and quality can count just like with any product.

Is a company constitution worth having?

A company set up without a constitution gets a one size fits call called the replaceable rules which gives a bare bones way for the company, and those involved in it, to operate. One size fits all can lead to a unintended outcomes. For instance an often unforeseen, easy to trip, requirement is to notify other directors of a conflict of interest between a director and the company. A properly tailored company constitution can modify conflict of interest rules away from the one size fits all to suit a company where only mum and dad are directors. Failure to do this can get weaponised like, say, when directors get divorced. And don’t think that this is the only reason why the replaceable rules may be a poor fit for your company.

Getting a capital structure of a company right

I do work sorting out situations made worse because companies are not understood by those setting them up. A company’s ideal capital structure is a big issue when a company is acting in its own right and not a trustee. Unless you understand the impact of s112-20 of the ITAA 1997 on the issue of shares in the company you’re a big chance to pay more capital gains tax that you might have when you sell or exit out of a company that has grown.

Company capital structure fails can lead to unnecessary loss of small business CGT concessions for small business which can amount to a big economic cost where a company ends up being a good business.

Getting “my” money out of a company

Shareholders try to get “their” money out of a company following a poorly executed lawyer free setup is another world of grief which can ironically bring in the lawyers, the ATO and expensive insolvency specialists.

A reckoning on death

Lots of problems don’t show up until a shareholder dies. This is often when the problem comes to my desk. It is sad when a family is tied in knots because their company establishment going way back was stuffed up. Any company setup, whichever way, might seem the same through times of smooth sailing. Why bother with the pesky paperwork at all? Wait, too, until the shareholders divorce, a fight amongst shareholders ensues or there is trading or tax trouble with the company: a sudden turn of interest, then, in the company’s capital, structure and records.

The “professional services” industry – escape for profit

Most of the non-legal providers on the internet are suss. They are derived from the offshore tax haven shell company “professional services” industry or use their business model. ICIJ media gives you an idea of their ethics https://cutt.ly/oUO7bvW and how they help their customers deal with local rules and commitments (not). Their model is to hide and escape from them.

Company constitutions, trust and SMSF deeds and partnership agreements are legal documents, and these providers are there to help you escape from having to get them from a lawyer charging a fee who is ethically obliged to professionally prepare them and whose work is covered by a professional indemnity/negligence insurance to protect you. And what about these rights? What a solicitor must tell you https://go.ly/P0jLU Worth having?

Their model is often something like this: we are not lawyers, so we give you escape from lawyers with this service. But we offer documents which are (based on) documents authored by a lawyer.

Reality check on unqualified legal practice

However you take this double-think pitch on the merits of avoiding lawyers, a reality is that the model is illegal: see the Federal Court case of Australian Competition & Consumer Commission v. Murray [2002] FCA 1252 https://jade.io/article/106192 to appreciate how documents supplied this way is from an unqualified legal practice source.

Ah! lawyers

There is a misconception that lawyers in this space are not worth the fees. I, for one, reckon my operation is lean and mean. And there are others like me. Sure my company and trust setup services cost a little more because my setups involve me thinking about and taking responsibility for what I am asked to do, and guiding clients on their setup choices based on what I know about them and thirty-five years’ experience of the ever changing traps – and that can’t be done by AI, yet.

What you get

So I can’t “compete” with a non-thinking service which gives you a company, trust or SMSF setup from a sausage cutter: documents all done and delivered instantaneously, with your credit card charged just as fast. But you get my drift: this blindingly impressive service just may be just too fast, hassle-free and brain-free. Look at the fine print (hello accountants) about who takes responsibility for loss if anything, including data inputs for which the inputter is made fully responsible, turn out not quite right.

So I agree. It just might be better to go to a dentist than the cheapest guy with a drill.

This post is actually from a post I made to another blog.  I think it’s worth another post on this blog even though it’s a more unruly and uncompromising than my usual posts here!

Should our SMSF have kept its Principal Employer?


Last month’s piece Lost SMSF trust deed replacement deeds – are they a scam? is my exposé of SMSF (self managed superannuation fund) trust deed variation techniques revealed as dodgy in the light of high Australian legal authority there set out.

So my exposé can be better appreciated and understood: this month I turn to some typical dilemmas faced by a SMSF trustee trying to update SMSF trust terms to:

  • keep them up to date with changing superannuation and tax laws; and
  • introduce capabilities so that opportunities presented by current regimes impacting superannuation funds can be effectively used.

To bring in the new, keep the old

One can see from my exposé that, to introduce new SMSF trust terms to a SMSF, a trustee needs to paradoxically keep the old.

Possibly no starker reminder of this are older SMSFs where the power of vary trust terms in the original trust deed (OTD) unconditionally requires the Principal Employer (or the “Employer” or the “Founding Employer”  – descriptions of this substantially similar role from the days of employer-sponsored superannuation vary) to initiate or consent to update trust terms of the SMSF.

My exposé further explains:

  • aside from in the narrowest of exceptions, a valid deed to vary SMSF trust terms requires a rigid adherence to the requirements of the power to vary trust terms contained in the OTD of the SMSF; and
  • an update or change to the power to vary in a SMSF OTD made on a misunderstanding that the power to vary allows amendment of the power to vary itself, when it doesn’t, is ineffective.

Invalid replacement of the power to vary


On that misunderstanding by a deed provider (unfortunately I can’t say deed lawyer here because, due to regulatory failings, SMSF legal documents with these errors are often supplied by non-lawyer outfits these days), the deed provider supplies a deed to vary SMSF trust terms by which the trustee purports to replace, among other trust terms, the power to vary in the OTD which power is replaced with the deed provider’s own contemporary take on an apt power to vary.

The SMSF trustee then considers the “replaced” power to vary which no longer requires the trustee to:

  • obtain the consent of the Principal Employer to vary trust terms; or
  • to take direction on the varied trust terms from the Principal Employer;

and decides that the redundant office of Principal Employer, no longer necessary with the evolution from employer-sponsored superannuation to self managed superannuation, can cease. The Principal Employer, say a company, is then de-registered and the office of Principal Employer under the SMSF lapses.

Marooned without a Principal Employer

As the “replaced” power to vary is of no effect this leaves the trustee unable to vary the SMSF trust terms further in future where there is no Principal Employer who can act under the power to vary from the OTD of the SMSF.

A question also arises whether the deed inserting the “replaced” power to vary also fails in its entirety where it contains an invalid replacement of the power to vary in the OTD. The answer to that question may vary case to case.

One can be more certain that deeds purporting to vary SMSF trust terms non-compliant with the power to vary in the OTD unconditionally requiring the consent etc. of the Principal Employer, will fail.

Other dated requirements in the power to vary

In retrospect many of the provisos which providers of SMSF OTDs included in powers to vary in SMSF OTDs seem unwise. Examples include provisos in powers to vary in OTDs that the trustee obtain the approval of:

  • the Commissioner of Taxation; or
  • the Insurance and Superannuation Commission;

to amendment of trust terms of the SMSF. These days the Commissioner of Taxation as the regulator of SMSFs is loathe to give such approval, which is not required by legislation, and the office of Insurance and Superannuation Commissioner no longer exists.

Unfortunately some old SMSF OTDs have these kinds of provisions and some way to deal with them needs to be worked out so that amendment compliant with the power to vary can take effect.

The right “applicable law”?

Powers to vary in SMSF OTDs frequently refer to an “applicable law”, or similar, broadly being the law that applied to SMSFs when the OTD was prepared. “Applicable law”, or whatever it may be, is usually defined in the OTD separately from the power to vary. When SMSF trust terms are generally updated, years later, the varied terms are understandably predicated on a different updated “applicable law”.

In my reckoning this means a deed varying SMSF trust terms probably needs to recognise and define two kinds of “applicable law” where compliance with “applicable law” is a proviso of the power to vary in the OTD:

  • firstly the statutes, regulations etc. that are apply to the SMSF under its updated terms; and
  • secondly the older laws prescribed as “applicable law” in the OTD, which may be redundant or repealed, which the trustee of the SMSF must nevertheless comply with to effectuate an update of trust terms in accordance with the power to vary in the OTD. The power to vary should then specifically refer to this second variety of “applicable law”. Restatement of these older laws can get complicated. For instance the Occupational Superannuation Standards Act 1987, which is often justifiably included as a component of “applicable law” in older superannuation OTDs, has been progressively renamed to the Superannuation Entities (Taxation) Act 1987,  the Superannuation (Excluded Funds) Taxation Act 1987 and the Superannuation (Self Managed Superannuation Funds) Taxation Act 1987.

An alternative view is that one stipulation of “applicable law” can suffice for the other on a reasonable interpretation of the OTD a court or tribunal may accept. That may be somewhat tenable if the OTD contains a interpretative provision contemplating amendments and re-enactments of statutes.

Still it is discomforting to rely on that interpretation of “applicable law” when the OTD specifically and restrictively defines what “applicable law” is and makes compliance with such “applicable law” a proviso to the power to vary. Adoption of multiple concepts of “applicable law” being:

  • one to support updated trust terms; and
  • the other to ground variations of the deed using the power to vary;

is a safer course in a deed to vary trust terms where “applicable law” is a proviso built into the power to vary in the OTD.


Proactive management of a SMSF with timely and effective amendment of SMSF trust terms to support that management can be a much more demanding and technical task then many will appreciate. It may pay for a SMSF trustee to carefully consider what the SMSF power to vary requirements in the OTD are, and what service the SMSF will be getting, rather than expecting that some plain vanilla SMSF deed amendment service is going to work.

Lost SMSF trust deed replacement deeds – are they a scam?

The writer has been reading about opportunity to replace lost trust deeds with a replacement deed from professional suppliers of replacement trust deeds, in SMSF Adviser and in other places. The writer is unconvinced that these replacement deeds are going to be legally effective particularly in relation to trust deeds to which the law in New South Wales applies.

Trust deeds lost in SA – Jowill Nominees Pty Ltd v. Cooper

On 2 July 2021 SMSF Adviser suggested that the South Australian case Jowill Nominees Pty Ltd v. Cooper [2021] SASC 76 provides an insight into issues a court will consider when a trust deed has been lost. This case concerned how trust rules of a trust governed by South Australian law can be varied by the SA Supreme Court on the application of the trustee pursuant to section 59C of the Trustee Act (SA) 1936. In the writer’s view this decision says nothing about variation of trust rules beyond the confine of a SA Supreme Court section 59C application.

Section 59C differs from the Trustee Acts to similar effect in other Australian jurisdictions including section 81 of the Trustee Act (NSW) 1925.

Regularity supports that there is a SMSF where its deed is lost

Where a trust, such as a self managed superannuation fund (SMSF), has been running for some time the trustee may be able to rely on the presumption of regularity to support the operation of the trust where the trust deed is lost.

The presumption of regularity is an evidentiary rule. It can apply where there is a gap in evidence about a prior act but where later acts and circumstances indicate likelihood that the prior act was performed. So in:

  • Sutherland v. Woods [2011] NSWSC 13 the NSW Supreme Court accepted that a SMSF trust deed and resolutions of a trustee of an active SMSF were signed on balance of probability although signed versions of these documents were missing from the evidence in the case; and
  • Re Thomson [2015] VSC 370 the Victorian Supreme Court treated a SMSF as operative in conformity with trust rules in a supposed later deed of variation even though an earlier deed of variation of the trust deed of the SMSF was lost and only an unexecuted version of the later deed of variation of the trust deed was available in evidence. Probabilities, and the surrounding facts such as the ongoing acceptance of the accounts of the SMSF based on the supposed later deed of variation, indicated likelihood that these deeds of variation had been completed and executed.

It is clear from the cases where the presumption of regularity is sought to be relied on that a court or tribunal will presume to aid a trustee unable to produce a missing deed only after an exhaustive search by the trustee for it:

He cannot presume in his own favour that things are rightly done if inquiry that he ought to make would tell him that they were wrongly done. 

Lord Simons in  Morris v. Kanssen  [1946] AC 459 at p. 475

Where a trustee of a trust, that has lost the trust deed of the trust, finds itself in dispute with the Commissioner of Taxation the presumption of regularity can counter the burden of proving the establishment of the trust on the trustee imposed by Part IVC of the Taxation Administration Act (C’th) 1953. See our post The burden of proof in a tax objection

The presumption of regularity is of procedural and not of substantive aid to establishing that a trust has been operating for some time in conformity with a valid and effective trust deed containing trust terms consistent with that operation where the trust deed cannot be produced. In the absence of evidence of the precise terms of a power of amendment, which is an exceptional power that can’t be presumed, the presumption of regularity, though, gives no substantial basis for amendment of trust terms to bring the terms of a SMSF trust deed back to terms that can be produced:

94. Variation of the terms of a trust (including by way of conferral of some new power on the trustee) is not something within the ordinary and natural province of a trustee. It is not something that it is “expedient” that a trustee should do; nor, fundamentally, is it something that is done “in the management or administration of” trust property. A trustee’s function is to take the trusts as it finds them and to administer them as they stand. The trustee is not concerned to question the terms of the trust or seek to improve them. I venture to say that, even where the trust instrument itself gives the trustee a power of variation, exercise of that power is not something that occurs “in the management or administration of” trust property. It occurs in order that the scheme of fiduciary administration of the property may somehow be reshaped.

Barrett JA in Re Dion Investments Pty. Ltd. [2014] NSWCA 367 at para 94

It follows that the presumption of regularity gives the trustee latitude to administer a trust on a presumed generic basis consistent with how that trust has been administered since inception where the trustee cannot produce the trust deed containing the trust terms. That presumption, though, would not ground alteration of trust terms where terms of a power of amendment which may not exist at all, cannot be specifically drawn on from the original trust instrument and complied with.

Law on amending lost trust deeds

How terms of a trust governed by the laws of New South Wales can be varied was considered by the Court of Appeal in Re Dion Investments Pty. Ltd. [2014] NSWCA 367. Re Dion Investments concerned an application to the Supreme Court to vary a trust deed of a trust by modernising its provisions for the benefit of the beneficiaries of the trust. In the writer’s view it is this Court of Appeal decision (by Barrett JA, whose decision Beazley P and Gleeson JA agreed with), not Jowill Nominees Pty Ltd v. Cooper, that gives insights into issues courts and tribunals, especially those in NSW, will consider when the effectiveness of instruments to amend trust terms:

  • where the trust deed of the trust has been lost and the power of amendment is not precisely known; or
  • in other circumstances where the variation to trust terms sought is not supported by, or are beyond, the power of amendment contained in the trust instrument such as in Re Dion Investments;

is to be considered.

Alteration of a trust by its founders

In the absence of a reserved power of amendment in a trust deed, can the trustee and the founders of a trust take action by a subsequent deed to vary an original trust deed (OTD)? The NSW Court of Appeal in Re Dion Investments indicates not. Barrett JA dispels this possibility where trusts and powers of the trust have been “defined” in an OTD:

41. Where an express trust is established in that way by a deed made between a settlor and the initial trustee to which the settled property is transferred, rights of the beneficiaries arise immediately the deed takes effect. The beneficiaries are not parties to the deed and, to the extent that it embodies covenants given by its parties to one another, the beneficiaries are strangers to those covenants and cannot sue at law for breach of them. The beneficiaries’ rights are equitable rights arising from the circumstance that the trustee has accepted the office of trustee and, therefore, the duties and obligations with respect to the trust property (and otherwise) that that office carries with it.

42. Any subsequent action of the settlor and the original trustee to vary the provisions of the deed made by them will not be effective to affect either the rights and interests of the beneficiaries or the duties, obligations and powers of the trustee. Those two parties have no ability to deprive the beneficiaries of those rights and interests or to vary either the terms of the trust that the trustee is bound to execute and uphold or the powers that are available to the trustee in order to do so. The terms of the trust have, in the eyes of equity, an existence that is independent of the provisions of the deed that define them.

Barrett JA in Re Dion Investments Pty. Ltd. [2014] NSWCA 367 at paras 41 to 42

Barrett JA then illustrates the point by this example:

43. Let it be assumed that on Monday the settlor and the trustee execute and deliver the trust deed (at which point the settled sum changes hands) and that on Tuesday they execute a deed revoking the original deed and stating that their rights and obligations are as if it had never existed. Unless some power of revocation of the trusts has been reserved, the subsequent action does not change the fact that the trustee holds the settled sum for the benefit of beneficiaries named in the original deed and upon the trusts stated in that deed. The covenants of a deed may be discharged or varied by another deed between the same parties (West v Blakeway (1841) 2 Man & G 751; 133 ER 940) but the equitable rights and interests of a beneficiary cannot be taken away or varied by anyone unless the terms of the trust itself (or statute) so allow.

Barrett JA in Re Dion Investments Pty. Ltd. [2014] NSWCA 367 at para 43

Alteration of a trust by all beneficiaries of a trust

SMSF Adviser and some SMSF deed suppliers express the view that persons who can compel the due administration of the trust can complete a replacement deed that varies and replaces a lost SMSF trust deed.

This view relies on a rule of equity from Saunders v. Vautier (1841) [1841] EWHC J82, 4 Beav 115, 49 ER 282. The rule is that where all of the beneficiaries of a trust are sui juris (of adult age and under no legal disability), the beneficiaries may require the trustee to transfer the trust property to them and terminate the trust. In Re Dion Investments, Barrett JA. recognises that this rule can entitle beneficiaries relying on the rule to require that the trustee hold the trust property on varied trusts:

but, if they do so require, the situation may in truth be one of resettlement upon new trusts rather than variation of the pre-existing trusts (and the trustee may not be compellable to accept and perform those new trusts: see CPT Custodian Pty Ltd v Commissioner of State Revenue [2005] HCA 53; 224 CLR 98 at [44]).

Barrett JA in Re Dion Investments Pty. Ltd. [2014] NSWCA 367 at para 46

For a trust that is a SMSF impediments to and implications of variation by the force of using the rule from Saunders v. Vautier are:

  • relatives and other dependants beyond the members of a SMSF, being all of the beneficiaries, must consent to using the rule from Saunders v. Vautier. Children, and others lacking legal capacity, who cannot consent to using the rule, are beneficiaries who can complicate use of the rule to vary a SMSF trust: Kafataris v. Deputy Commissioner of Taxation [2008] FCA 1454; and
  • if the beneficiaries do apply the rule from Saunders v. Vautier, resettlement of a SMSF trust on taking that action gives rise to:
    • CGT event E1 or E2 for each of the CGT assets of the SMSF under Part 3-1 of the Income Tax Assessment Act 1997. It follows that action taken by SMSF beneficiaries in reliance on the rule from Saunders v. Vautier will have comparable capital gains tax consequences to a transfer of all members’ benefits to a newly established SMSF; and
    • prospect that a new ABN and election to become a regulated superannuation fund for a new resettled SMSF will by required by the regulator.

Much reliance is placed by SMSF Adviser and by deed suppliers’ websites promoting replacement deed services on Re Bowmil Nominees Pty. Ltd. [2004] NSWSC 161. In Re Bowmil Nominees Pty. Ltd. . Hamilton J of the NSW Supreme Court, as a matter of expediency, allowed beneficiaries to vary a SMSF trust deed beyond limitations in the amendment power in the trust deed utilising the rule in Saunders v. Vautier on this basis:

20. Since it is appropriate that the trustee act upon the informed consent of beneficiaries who are sui juris and unnecessary applications to the Court for empowerment are not to be encouraged, I propose to adopt the course followed by Baragwanath J in the New Zealand case. I do not propose to make an order under s 81 of the TA empowering the making of the amendment, although I have expressed the view that the Court has power to do so and would be prepared to do so if it were necessary. Rather, I shall make an appropriate declaratory order to the effect that it is expedient that the proposed deed of amendment be entered into and that it will be appropriate for the trustee to act in accordance with it.

Re Bowmil Nominees Pty. Ltd. [2004] NSWSC 161 at para. 20

Update of trust terms by a court

The Court of Appeal in Re Dion Investments agreed with Young AJ, the primary judge, that post-1997 court decisions, including Re Bowmil Nominees Pty. Ltd., which relied on a misunderstanding of the extent of court power to vary trust deeds, particularly in relation to the statutory powers of a court to alter the terms of the trust viz. the aforementioned section 81 in NSW and section 59C in SA, which misunderstanding originated from this obiter dicta of Baragwanath J in Re Philips New Zealand Ltd [1997] 1 NZLR 93

The Court will not willingly construe a deed so as to stultify the ability of trustees, having proper consents, to amend a deed to bring it into line with changing conditions.

Re Philips New Zealand Ltd [1997] 1 NZLR 93 at page 99

were not correctly decided. Barrett JA said:

100. For these reasons, I share the opinion of the primary judge that the post-1997 decisions that have proceeded on the basis that variation of the terms of a trust is, of itself, a “transaction” within the contemplation of s 81(1) rest on an unsound foundation. The court is not empowered by the section to grant power to the trustee to amend the trust instrument or the terms of the trust. It may only grant specific powers related to the management and administration of the trust property, being powers that co-exist with (and, to the extent of any inconsistency, override) those conferred by the trust instrument or by law.

Barrett JA in Re Dion Investments Pty. Ltd. [2014] NSWCA 367 at para 100

In particular. the decision in Re Bowmil Nominees Pty. Ltd. and the other post-1997 decisions referred to in Re Dion Investments cannot be reconciled with the Court of Appeal decision in Re Dion Investments where Barrett JA found:

96. In such cases, however, the creation of what is, in terms, a power of the trustee to amend the trust instrument is a superfluous and meaningless step. When the court, acting under s 81(1), confers on a trustee power to undertake a particular dealing (or dealings of a particular kind), “it must be taken to have done it as though the power which is being put into operation had been inserted in the trust instrument as an overriding power”: Re Mair [1935] Ch 562 at 565 per Farwell J. The substantive power that the court gives comes into existence by virtue of the court’s order. It does not have its source in the terms of the trust. There is no addition to the content of the trust instrument. That content is supplemented and overridden “as though” some addition had been made to it. The terms of the trust are reshaped accordingly.

97. Conferral of specific new powers pursuant to s 81(1) should not be by way of purported grant of authority to amend the trust instrument so that it provides for the new powers. Rather, the court’s order should directly confer (and be the sole and direct source of) the powers which then supplement and, as necessary, override the content of the trust instrument. And, of course, the only specific powers that can be conferred in that direct way are those that fall within the s 81(1) description concerned with management and administration of trust property.

Barrett JA in Re Dion Investments Pty. Ltd. [2014] NSWCA 367 at paras 96-97

A variation relying on a power of amendment in trust terms is not a variation of a trust deed but a variation of trust terms contained in a trust deed. Barrett JA explained this in Re Dion Investments:

44. It is, of course, commonplace to speak of the variation of a trust instrument as such when referring to what is, in truth, variation of the terms upon which trust property is held under the trusts created or evidenced by the instrument. A provision of a trust instrument that lays down procedures by which it may be varied is, of its nature, concerned with variation of the terms of the trust, not variation of the content of the instrument, although the fact that it is the instrument that sets out the terms of the trust does, in an imprecise way, make it sensible to speak of amendment of the instrument when the reference is in truth to amendment of the terms of the trust.

45. Where the trust instrument contains a provision allowing variation by a particular process, the situation is one in which the settlor, in declaring the trust and defining its terms, has specified that those terms are not immutable and that the original terms will be superseded by varied terms if the specified process of variation (entailing, in concept, a power of appointment or a power of revocation or both) is undertaken. The varied terms are in that way traceable to the settlor’s intention as communicated to the original trustee.

Barrett JA in Re Dion Investments Pty. Ltd. [2014] NSWCA 367 at paras 44-45

Significance of the power of amendment as expressed in an OTD

A power of amendment of a SMSF, or any other express trust, is a precise reflection of the settlor’s (founder’s) intention of conditions for amendment of the trust communicated in the trust terms in the OTD and supplies the only lawful way trust rules in a trust deed, otherwise immutable, can be amended aside from narrow exceptions:

  • where beneficiaries can invoke the rule in Saunders v. Vautier and, by doing so, resettle the SMSF on a new trust; or
  • by court order to vary trust terms or, in NSW, to allow dealings of a particular kind despite trust terms, in accordance with a state or territory Trustee Acts such as section 59C of the Trustee Act (SA) 1936 and section 81 of the Trustee Act (NSW) 1925;

as considered above.

Amendment practice

It follows that a power of amendment in an OTD of a trust:

  • needs to remain, as it was in the OTD, as a term of the trust unless the power of amendment itself can be amended, should that be possible and has so been amended; and
  • is best extracted, repeated and given prominence in a deed of variation which replaces the other trust terms of a trust so that trust terms are clear and traceable on an ongoing basis.

Extraction and repeat of a reserved power of amendment from an OTD is not always just a matter of extracting the paragraph or paragraphs in the OTD containing the power of amendment. In the writer’s experience powers of amendment in older SMSF OTDs are frequently premised on laws and practices that prevailed when the superannuation trust was established e.g. such as in the former Occupational Superannuation Standards Act (C’th) 1987 and practices relating to now redundant regimes of employer sponsored superannuation. To remain traceable to the settlor’s (founder’s) intention as communicated to the original trustee, conditions specified for amendment in a power of amendment based on laws and practices, even where those laws and practices have evolved or become redundant since establishment of the trust; need to be complied with and reflected cogently in the extraction and repeat of the power of amendment in a deed of variation, within reason, if the power of amendment is to remain as a trust term in an exercisable form in the deed of variation.

When can a power of amendment in an OTD itself be amended?

Amendment of the power of amendment itself may be possible but unlikely if the amendment provision in the OTD itself does not expressly permit it. In Jenkins v. Ellett [2007] QSC 154, Douglas J. stated:

The scope of powers of amendment of a trust deed is discussed in an illuminating fashion in Thomas on Powers (1st ed., 1998) at pp. 585-586, paras 14-31 to 14-32 in these terms:

“In all cases, the scope of the relevant power is determined by the construction of the words in which it is couched, in accordance with the surrounding context and also of such extrinsic evidence (if any) as may be properly admissible. A power of amendment or variation in a trust instrument ought not to be construed in a narrow or unreal way. It will have been created in order to provide flexibility, whether in relation to specific matters or more generally. Such a power ought, therefore, to be construed liberally so as to permit any amendment which is not prohibited by an express direction to the contrary or by some necessary implication, provided always that any such amendment does not derogate from the fundamental purposes for which the power was created ….It does not follow, of course, that the power of amendment itself can be amended in this way. Indeed, it is probably the case that there is an implied (albeit rebuttable) presumption, in the absence of an express direction to that effect, that a power of amendment (like any other kind of power) cannot be used to extend its own scope or amend its own terms. Moreover, a power of amendment is not likely to be held to extend to varying the trust in a way which would destroy its ‘substratum’. The underlying purpose for the furtherance of which the power was initially created or conferred will obviously be paramount.”

Jenkins v. Ellett [2007] QSC 154 Douglas J. at paragraph 15

One can see the parity between what was said in Jenkins v. Ellett and in Thomas on Powers and in paragraph 94 in Re Dion Investments Pty. Ltd., as set out above, about a trustee’s proper role not being concerned to question or improve trust terms. See the writer’s article Redoing the deed https://wp.me/P6T4vg-3x#rtd

Update of the power of amendment?

The writer sees confusion among SMSF deed suppliers over the difference between the OTD and the trust terms in the OTD and who consequently fall into the trap of treating the power to amend as updatable by the same power to amend.

So instead of relocating the power of amendment in the OTD to updated trust terms, suppliers simply replace that power with their own take on an apt power of amendment departing from Barrett JA’s dictum that it is not for the trustee, far less a variation deed supplier, to “question the terms of the trust or seek to improve them”. Following Re Dion Investments and Jenkins v. Ellett a replacement of a power of amendment that is not amendable is a deviation from the power of amendment prone to be:

  • beyond the power of:
    • the parties entrusted with the power of amendment; and
    • a court, even if an order of the court for the replacement power had been sought; and
  • thus void.

Later deeds of variation of SMSFs based on a deviation

As in Re Thomson trust deeds of SMSFs will likely be varied more than once so that trust terms (governing rules) can better reflect evolving law and practice with SMSFs. An unlawful replacement of a power of amendment which deviates from the power of amendment in the OTD of a SMSF lays a trap when a trustee seeks to make a further amendment to the trust terms of the SMSF: Based on the above authorities a further deed of variation reliant on the “updated” power of amendment in an earlier deed of variation, rather than the power of variation in the even earlier OTD of the SMSF, will fail and be void unless the updated power of amendment in the earlier deed of variation is in conformity with the power of amendment in the OTD.

So are replacement SMSF trust deeds a scam?

The writer suspects many SMSF deed suppliers who supply replacement SMSF deeds don’t understand or follow the implications of Re Dion Investments. As a considered NSW Court of Appeal decision Re Dion Investments is binding legal precedent that rejects the authority of first instance NSW Supreme Court decisions referred to and discussed by the Court of Appeal, including Re Bowmil Nominees Pty. Ltd., that rest on an “unsound foundation” .

It is unfortunate that these cases are still being used as spurious authority on the websites of SMSF deed suppliers in support of claims that lost SMSF deed replacement deeds are of greater efficacy as variations of a trust deed than courts and tribunals, especially NSW courts, will be prepared to accept or order following Re Dion Investments. The writer wouldn’t say these claims are a scam necessarily because, as this post shows, the present state of law is complicated, difficult and more restrictive than understood by courts in the post-1997 cases referred to in Re Dion Investments.

The current law appears to be that if a trustee wants to vary a SMSF trust deed, which is “not something within the ordinary and natural province of a trustee” especially in NSW, the parties given power to amend under a power of amendment must locate, have and rely on that power in or derived from the OTD to successfully amend terms of a SMSF trust without resettling it.

Other solutions, aside from supreme court applications allowed under:

  • section 81 of the Trustee Act (NSW) 1925, as pursued in Re Dion Investments
  • section 59C of the Trustee Act (SA) 1936, as pursued in Jowill Nominees Pty Ltd v. Cooper; or
  • comparable legislation in other Australian states and territories;

which are expensive litigation, are unlikely to be legally effective.

It follows that every effort should be made to find trust terms in an OTD so that the power of amendment in the deed will be carefully complied with when an amendment of a trust deed is to be undertaken. That includes where there have been earlier deeds of variation of the trust terms of a SMSF whose validity also rests on, and must be derived from the reserved amendment power defined in the OTD.


The author acknowledges the articles:

  • A matter of trusts – Presumption of regularity to the rescue? Milton Louca and Phil Broderick, Taxation in Australia March 2018 at page 436
  • The powers of a Court to vary the terms of a trust A consideration of in Re Dion Investments Pty. Ltd. (2014) 87 NSWLR 753 A paper presented to the Society of Trust and Estates Practitioners – NSW Branch Wednesday 21 October 2015 by Denis Barlin of counsel (who appeared as counsel for the section 81 applicant in the case)

that were useful in preparing this post and which contain greater detail on the issues discussed. The author also expresses his gratitude that these articles have been made available openly online.

Etmekdjian – the disqualified are out of the SMSF system

You are leaving the SMSF sector

In an Administrative Appeals Tribunal decision this month in Etmekdjian v. Commissioner of Taxation  [2020] AATA 3821 (1 October 2020), the AAT refused to extend a waiver of disqualification to the applicant so the applicant could manage his own self managed superannuation fund (SMSF). The applicant had been disqualified under Part 15 of the Superannuation Industry (Supervision) Act 1993 (SIS Act).

Automatic disqualification and its waiver

There are a thin 14 days following disqualification to apply for a waiver of the disqualification once a person is disqualified: section 126B. In this case the applicant had been disqualified automatically under sub-section 120(3) of the SIS Act on conviction, by a NSW Local Court, for Commonwealth Criminal Code offences. The offences were for dishonestly backdating employee share scheme elections under former section 139E of the Income Tax Assessment Act 1936.

The applicant was outside of the 14 days allowed to seek the waiver so the applicant sought an extension of time to do so from the AAT.

AAT decision

The AAT refused:

  • to accept that the applicant’s unsuccessful appeal to the District Court against the conviction stayed the conviction by the Local Court and thus the date of automatic disqualification for section 120 purposes; and
  • to allow the extension of time as there was an absence of exceptional circumstances explaining the failure to lodge the application for waiver against disqualification within 14 days.

Context of the AAT decision and Part 15 disqualifications

Deputy President Bernard McCabe observed at the outset in paragraph 1 of the AAT decision:

Managing a superannuation fund – even a small, self-managed fund – is a big responsibility. There is a public interest in managing these funds properly given the tax advantages they enjoy. To that end, the Superannuation Industry (Supervision) Act 1993 (Cth) (SISA) establishes rules designed to ensure prudent management. Part 15 of SISA includes rules regarding disqualified persons. A disqualified person may not be a trustee, investment manager or custodian of a superannuation entity, or be a responsible officer (such as a director) of a corporation that performs those roles: s 126K. A person can be disqualified by the Commissioner of Taxation (where the Commissioner is the regulator) or the Federal Court (where the Australian Prudential Regulatory Authority is the regulator) on a variety of grounds, but a person will be automatically disqualified in circumstances set out in s 120. .…

[2020] AATA 3821 at paragraph 1

The AAT found that it only had power to extend the 14 day period strictly when exceptional circumstances warranted that extension. Deputy President McCabe concluded:

….The law requires that I identify exceptional circumstances that prevented the applicant from complying with the 14 day time limit. It is not enough to establish the applicant had a good excuse, or that non-compliance does not result in any harm, or that the applicant has a good case in relation to the substantive issue. This is not a standard ‘extension of time’ case.

21. The applicant has failed to identify any ‘exceptional circumstances’ that prevented him from making the application within the time frame contemplated in s 126B(3). In those circumstances, the decision under review must be affirmed.

[2020] AATA 3821 at paragraphs 20 and 21

The AAT’s strict approach is no surprise when it comes to the disqualification rules in Part 15. Part 15 reflects a low tolerance approach in the SIS Act to persons designated unfit to manage a superannuation fund.

Difficulties – SMSFs with a disqualified trustee/director/member

The author has seen Part 15 disqualifications happen on bankruptcy by operation of sub-section 120(1) of the SIS Act.

A person disqualified on bankruptcy, or any other disqualified person such as Mr. Etmekdjian, can’t be a trustee, a director of the trustee or a member of a SMSF. A fund with a disqualified participant falls off the register of superannuation funds as a SMSF regulated by the Commissioner of Taxation. The fund becomes (at least notionally) regulated by the Australian Prudential Regulatory Authority (APRA) instead.

So the Australian Taxation Office won’t and can’t assist once the fund is no longer a SMSF.

Unless a deactivated SMSF, on a participant becoming Part 15 disqualified, can nimbly:

  • convert to an APRA regulated fund; and
  • appoint an approved trustee;

based on a power in the trust deed of the fund to do so, or disqualified persons promptly vacate the fund to prevent deactivation, the fund reverts to an administrative no man’s land. Even arranging a roll out of benefits to another fund is fraught following deactivation. The fund won’t be practically manageable or administrable.

Time for a SMSF deed upgrade?

Lack of capability in a SMSF trust deed to convert a SMSF to an APRA fund is one of a number of indicators that SMSF trust deed may need of an upgrade to comply with today’s SIS Act requirements.

Controlling who gets death benefits from a SMSF

A widower nominated his son and daughter to take equal shares of his superannuation benefits on his death on a basis not binding on the trustee. The daughter, who was the surviving trustee of her father’s self managed superannuation fund (SMSF) after his death, appointed her husband as the new co-trustee and excluded the son from control of the SMSF. The daughter refused to pay the son the equal share of death benefits based on the father’s non-binding death benefit nomination (DBN). The son unsuccessfully challenged the daughter’s conduct in the NSW Supreme Court: Katz v. Grossman [2005] NSWSC 934.

Dilemma – the SMSF trustee’s control over where death benefits go

Katz v. Grossman reveals a dilemma with SMSFs: whoever survives a member as trustee of a SMSF generally has significant autonomy as to whom death benefits of a deceased member will be paid to by default unless the member:

  • has taken effective steps to ensure the DBN is a valid binding DBN (BDBN) to bind the trustee to pay his or her benefits to:
    • dependant/s nominated by the member; or
    • the member’s estate by nominating the member’s legal personal representative (LPR); or
  • the member has made his or her pension reversionary to their chosen dependant: although a reversionary pensioner generally cannot be an adult child as only death benefits dependants contemplated by section 302-195 of the Income Tax Assessment Act 1997 can receive pension, including reversionary pension, death benefits.


The challenge of directing death benefits to dependants

Member control of superannuation is all well and good but selection of dependants to receive death benefits, either by member’s DBN or by the trustee of the fund, is fraught and is just as prone to dispute between disgruntled family beneficiaries as disputes over wills (Wills) and deceased estates are.

With superannuation funds generally, and especially SMSFs, it is a challenge for a member to:

  • maintain an up to date expression of where he or she wants his or her benefits to go on his or her death; and
  • to effect those wishes by way of a DBN.

In many cases this will be inconsequential such as where a surviving spouse of a deceased member is the surviving trustee, or controls the trustee, and is the obvious dependant of the member to take death benefits. But where dependants are next generation, or where a member has a blended family, surviving trustee decisions to pay death benefits of the deceased may not align with the deceased member’s wishes or their DBN especially where trusteeship of the SMSF passes into unexpected and unprofessional trustee hands on their demise.

Section 17A of the Superannuation Industry (Supervision) Act 1993 (the SIS Act) limits who can be or control a trustee of a SMSF to:

  • the members of the SMSF; or
  • their enduring attorneys;

unless the fund is a single member fund and, in any case, trustees of a SMSF must be unremunerated in their role as trustee: paragraph 17A(2)(c) of the SIS Act. Member controlled superannuation by a SMSF can be a control vacuum isolated from professional trustee expertise following the death of a SMSF member unless a professional is involved in the limited ways possible under sections 17A and 17B.

Does a SMSF member need to control where their death benefits go?

Is it desirable that the member controls where he or she wants his or her benefits to go in any case? Superannuation is explicitly to provide for a member’s dependants when the member dies. The SIS Act defines a dependant:

“dependant”, in relation to a person, includes the spouse of the person, any child of the person and any person with whom the person has an interdependency relationship.

Section 10 of the SIS Act

A deceased member may nominate a dependant by a DBN that is not the dependant of the member most truly dependent or most in need of the member’s death benefits. That is why it is doubly desirable to have a competent and trustworthy person succeed the member as trustee who will genuinely assess these needs. It is on the assumption that such a trustee will survive the member as SMSF trustee that superannuation fund governing rules (SFGRs) generally give the surviving trustee an open discretion to select the dependant of the deceased member to take the member’s benefits unless one of the Exceptions applies.

So even with the guidance of a non-binding DBN (NDBN), which expresses a deceased’s wishes as to whom his or her benefits are to be paid, SFGRs, the SIS Act and trust law typically give a superannuation trustee a power to pay benefits to dependants the trustee chooses in the trustee’s discretion contrary to and despite a NDBN as occurred in Katz v. Grossman.

The binding death benefit nomination

To immunise a DBN from a wrong choice of trustee, who may select a dependant in their discretion at odds with the member’s wishes, a member can use a BDBN. A widow or widower in circumstances similar to Katz v. Grossman can prevent override of their wishes as to who is to be their superannuation dependant to take their death benefits by force of a BDBN to bind the trustee to pay to that dependant.

The BDBN obstacle course

A SMSF member seeking to impose a BDBN to control his or her superannuation needs to be sure it will take effect. There are numerous contingencies. Consider these:

  • is the capability in SFGRs allowing BDBNs effective and does it have integrity? Does the member appreciate that BDBN forms and arrangements differ from trust deed to trust deed? Not all BDBN arrangements in trust deeds are rigorous;
  • will the BDBN be validly completed? (Wareham v. Marsella discussed below is an instance of invalid completion of a BDBN);
  • if the BDBN is stated to be non-lapsing will it take effect as non-lapsing? That is: will the BDBN continue to bind the trustee more than three years after it is made? In the recent case Re SB; Ex parte AC [2020] QSC 139 a non-lapsing BDBN was accepted by the Supreme Court of Queensland. Non-lapsing BDBNs are understood to be feasible for SMSFs following:
    • Donovan v. Donovan [2009] QSC 26; and
    • Self Managed Superannuation Funds Determination SMSFD 2008/3 Self Managed Superannuation Funds: is there any restriction in the Superannuation Industry (Supervision) legislation on a self managed superannuation fund trustee accepting from a member a binding nomination of the recipients of any benefits payable in the event of the member’s death?
  • what if the member marries, divorces or commences a reversionary pension after making a BDBN?
  • will the BDBN have a fraud risk? Who needs to witness completion of a BDBN form by a member under the SFGRs? Depending on arrangements and the regime in the SFGRs for safe custody and verification of a BDBN, is there a risk that a BDBN may be altered by a dishonest successor trustee or a trustee in a conflict of interest with other dependants or “lost” so the BDBN won’t take effect as intended by the member? and
  • what if the SFGRs are subsequently amended so that a BDBN made under former SFGRs of a SMSF no longer comply with the later SFGRs?

So a member looking to rely on a BDBN to direct who will take their superannuation faces a veritable obstacle course turning on:

  • the SFGRs in the trust deed of the fund;
  • the member’s domestic circumstances; and
  • the security integrity of the BDBN arrangements;

in his or her quest to have a BDBN complied with by the trustee of the SMSF when the member is no longer around.

Better for a BDBN to be in a member’s Will?

Although a payment of death benefits is not a testamentary disposition:  McFadden v Public Trustee for Victoria [1981] 1 NSWLR 15, it is desirable that a BDBN should be set out in, or, in the least, kept with, the Will of a SMSF member to avoid some of the above contingencies.

Generally speaking Wills are:

  • subject to strict witnessing and other evidentiary requirements under state laws which reduce the prospect of fraud. By inclusion of a BDBN in a Will the BDBN can attract the same protections; and
  • revoked on marriage and, depending on state laws, altered by divorce. A dependant nominated in a BDBN may pre-decease the member. On any of these events BDBNs are ideally revisited and, in that context, a non-lapsing BDBN is especially fraught after a situation where a long-dated BDBN should have been updated to reflect changes in a member’s domestic situation. If a BDBN is in a Will there is a greater likelihood that desirable update of a BDBN will not be overlooked.

There is also the advantage of consolidated consideration and expression of the member’s wishes for his or her property and financial resources substantially in a single document. Superannuation death benefits of deceased superannuation members now frequently exceed deceased estate property governed by their Wills in value.

To include a BDBN in a Will there needs to be a basis or regime for making a BDBN in a member’s Will in the SFGRs (in the trust deed) of a SMSF. Ordinarily SFGRs/SMSF trust deeds do not provide for BDBNs to be set out in a Will and instead require the BDBN to be in a discrete BDBN form.

When there is no BDBN


  1. a BDBN fails;
  2. there is a NDBN but no BDBN; or
  3. no DBN at all;

what assurance does a member have that a trustee will act in the interests of and fairly to the prospective dependants of the member?

There is initially the issue with the first and third cases that there is no satisfactory expression of what the member wishes. This situation arose in the recent Victorian Supreme Court Appeal decision in Wareham v. Marsella [2020] VSCA 92.

Wareham v. Marsella

In Wareham v. Marsella the dependants of the deceased member of a SMSF included:

  1. the deceased’s daughter from her earlier marriage, Mrs. Wareham; and
  2. her husband of 32 years up to her death, Mr Marsella.

The deceased had made a BDBN in favour of her grandchildren at the inception of the SMSF but her grandchildren were not her superannuation dependants (see the definition in section 10 of the SIS Act above) so the BDBN was invalid.

Mrs. Wareham was the deceased member’s surviving trustee. Relations between her and the husband, Mr. Marsella, were strained. Mrs. Wareham appointed her husband Mr. Wareham as co-trustee. The trustees paid all of the deceased’s SMSF death benefits to Mrs. Wareham wholly excluding Mr. Marsella.

At first instance McMillan J. held that Mr and Mrs Wareham had exercised their discretion as trustees without giving real and genuine consideration to the interests of the dependants of the SMSF and:

  • set aside the exercise of their trustees’ discretion to favour themselves; and
  • removed Mr and Mrs Wareham as trustees of the SMSF.

This result was upheld on appeal to the Court of Appeal.

The court confirmed the wide autonomy the trustees of the SMSF had to select a dependant to take death benefits:

Apart from cases where trustees disclose their reasons, the exercise of an absolute and unfettered discretion is examinable only as to good faith, real and genuine consideration and absence of ulterior purpose, and not as to the method and manner of its exercise.

from Karger v Paul [1984] VicRP 13

Mr and Mrs Wareham did not give reasons for their decision to distribute all of the death benefits to Mrs. Wareham which meant that Mr. Marsella needed to establish:

  • bad faith;
  • lack of real and genuine consideration by; and/or
  • an ulterior purpose of

the trustees in making the decision (the Challengeable Grounds).  These are all matters that are challenging to prove before a court particularly where there are no expressed reasons of the trustees for making the decision. However in this exceptional case the Supreme Court could focus on:

  • the erroneous response by the trustees’ lawyers in correspondence with Mr. Marsella over his claim to participate as a dependant. From that it could be shown that the trustees misconceived their obligation to give Mr. Marsella’s claim a real and genuine consideration. For instance, the trustees’ lawyers had asserted in the correspondence that “Mr. Marsella was not a beneficiary or dependant and had no interest in the fund”; and
  • the bad faith of the trustees. The court observed that “the decision to pay no part of the death benefit to the deceased’s husband of more than 30 years was, at least, remarkable” and the “grotesquely unreasonable” nature of the decision to exclude him was enough to establish bad faith. As there was actual conflict between Mrs. Wareham, a trustee, and Mr. Marsella the court observed that it may remove a trustee in its discretion.

Balanced against that was:

  • the trustees’ resolution to pay Mrs. Wareham which did not reveal errors that establish any of Challengeable Grounds;
  • the power of the trustees to pay death benefits to a dependant who is a trustee despite the conflict of interest; and
  • the trustees did not give evidence and so where not examined about their consideration of Mr. Marsella’s claims as a dependant;

An exceptional case

So even though Mr. Marsella was successful the case was appealed and hard fought. The trustees’ lawyer’s unlikely lapses in the correspondence and the extreme outcome and treatment of a husband of more than 30 years were vital to the result especially where SFGRs expressly permit a trustee to favour themselves in death benefits discretionary decisions despite their conflict of interest with other potential dependants that could receive those death benefits.

Where a trustee is more cautious and opaque in the course of:

  • their decision to pay death benefits to their own benefit, despite their conflict of interest; and
  • in related correspondence;

the trustee will reveal little which will give a disgruntled dependant Challengeable Grounds to challenge the trustee’s exercise of discretion.

In that respect Katz v. Grossman more likely reflects the reality facing most disgruntled family members who miss out on death benefits, especially those who are not a spouse or in an interdependency relationship. In Katz v. Grossman Mr. Katz may not have been in a position to establish the Challengeable Grounds even though:

  • his father had nominated him on a non-binding basis to take an equal share of his death benefits; and
  • his sister and her husband as trustees of the SMSF instead distributed all of the death benefits of the father to his sister.


It follows that the authority of Wareham v. Marsella may only assist a spouse or interdependency dependant highly deserving of death benefits as dependants in compelling cases where:

  • a SMSF’s trustee makes identifiable error in the process of discretionary decision-making to pay death benefits to himself or herself despite their conflict of interest with the deceased’s spouse or interdependency dependant; and
  • where that spouse or interdependency dependant can endure legal action to challenge the decision based on the Challengeable Grounds.

Otherwise SMSF members need to ensure the right person or people are their successor trustee of their SMSF if a Katz v. Grossman or other situation where a dependant favoured by the member misses out on death benefits is to be avoided. A valid or current expression of wishes either in a NDBN or better:

  • in a BDBN, rigorously backed by SFGRs, included in or with the Will of the member to ensure its integrity, reducing the likelihood that a payment of death benefits will be made to the exclusion of a desired or deserving family member especially where, due to the confines of the SIS Act, the member can’t be confident that their successor SMSF trustee won’t use the opportunity to favour their own benefit; or
  • where a member foresees that their dependants will be in potential conflict, in next generation or blended family circumstances, by taking steps in accordance with the SFGRs to remove trustee autonomy to make the death benefits payment decision and to instead mandate that death benefits are to be paid to the LPR of the member in compliance with core purposes of superannuation which allow payment of death benefits to the LPR under section 62 of the SIS Act. In that case the member can set out how death benefits are to be left in his or her Will.

Does a SMSF that holds only life insurance satisfy the sole purpose test?


I was recently asked if a SMSF whose only assets are an insurance policy and cash topped up by contributions used to meet premiums on the policy would comply with the sole purpose requirement in the Superannuation Industry (Supervision) Act 1993.

It depends. It will depend on the terms of the life policy and significantly on the age of the member:

The sole purpose test in sub-section 62(1) is structured as follows:

(1) Each trustee of a regulated superannuation fund must ensure that the fund is maintained solely:

(a) for one or more of the following purposes (the core purposes ):

… or

(b) for one or more of the core purposes and for one or more of the following purposes (the ancillary purposes ):


In other words a fund maintained for any of the listed core purposes complies with the sole purpose test. A fund maintained for an ancillary purpose or purposes also complies with the sole purpose test so long as it is also maintained for a core purpose or purposes.

The life insurance policy in question would need to be considered. Will the policy pay out insurance proceeds to the trustee of the fund on the death of the member to be used by the trustee to pay death benefits?

Pursuing a core purpose

If the policy would indemnify the trustee on the death of the member before the member:

  • ceases gainful employment; or
  • reaches age 65;

the trustee of the fund would appear to pursue core purposes either in sub-paragraph 62(1)(iv) or (iv) which are:

(iv)  the provision of benefits in respect of each member of the fund on or after the member’s death, if:

(A)  the death occurred before the member’s retirement from any business, trade, profession, vocation, calling, occupation or employment in which the member was engaged; and

(B)  the benefits are provided to the member’s legal personal representative, to any or all of the member’s dependants, or to both;

(v)  the provision of benefits in respect of each member of the fund on or after the member’s death, if:

(A)  the death occurred before the member attained the age (65) prescribed for the purposes of subparagraph (ii); and

(B)  the benefits are provided to the member’s legal personal representative, to any or all of the member’s dependants, or to both;

by taking out life cover to fund these death benefits.

Pursuing an ancillary purpose

If the life cover in the policy:

  • is in respect of the life of a member who is over age 65 and who has ceased gainful employment; or
  • only extends cover following both of those events;

then the fund is only maintained for ancillary purposes in sub-paragraph 62(2)(iii) and (iv) which are:

(iii) the provision of benefits in respect of each member of the fund on or after the member’s death, if:

(A) the death occurred after the member’s retirement from any business, trade, profession, vocation, calling, occupation or employment in which the member was engaged (whether the member’s retirement occurred before, or occurred after, the member joined the fund); and 

(B) the benefits are provided to the member’s legal personal representative, to any or all of the member’s dependants, or to both;

(iv) the provision of benefits in respect of each member of the fund on or after the member’s death, if: 

(A) the death occurred after the member attained the age prescribed for the purposes of subparagraph (a)(ii); and 

(B) the benefits are provided to the member’s legal personal representative, to any or all of the member’s dependants, or to both;

and the fund is not maintained for core purposes and so the fund does not comply with the sole purpose requirement in sub-section 62(1).


To reiterate: where the member has both:

  • reached the age of 65; and
  • ceased gainful employment;

or the policy doesn’t fund death benefits before either case then core purposes in sub-paragraph 62(1)(iv) and (iv) don’t apply to life cover taken out by the fund and the sole purpose test in section 62 could be breached for failure to pursue a core purpose in addition to an ancillary purpose or purposes.

Income from private company investments – the tax scourge of SMSFs


A self managed superannuation fund (SMSF) is generally a low tax entity, particularly when in pension phase where a nil rate can apply and a low 15% rate can apply when not. Still the taxable income of a complying superannuation fund (SF) can be split into a non-arm’s length component and a low tax component under section 295-545 of the Income Tax Assessment Act (ITAA) 1997. The non-arm’s length component is taxed at the highest individual marginal rate which is 45% in the 2019-20 income year.

Non-arm’s length income

The non-arm’s length component for an income year is the complying SF’s “non-arm’s length income” (NALI) for that year less any deductions to the extent that they are attributable to that income.

NALI picked up on audit – even higher tax

The recent case in GYBW v. Commissioner of Taxation [2019] AATA 4262 (GYBW) is a cogent reminder of how NALI taxed at the highest marginal rate can arise in a SMSF. In GYBW a tax shortfall arose as the NALI not returned by the SMSF was detected in an audit by the Commissioner of Taxation. Hence even higher taxes applied including shortfall interest and penalties. There was a reduction in penalties on appeal to the AAT from “reckless” to “failure to take reasonable care” level.


Section 295-550 is one of a number of superannuation rules designed to protect the integrity of the low tax complying SF regime by combatting income shifting arrangements where income, that might be taxed elsewhere to another type of taxpayer at higher rates, is non-commercially shifted to a complying SF that attracts a low rate of tax.

Section 295-550 is directed at non-arm’s length dealings where complying SFs (and other superannuation entities) earn income from an arrangement which exceeds the income that the complying SF might have been expected to derive from the arrangement if the parties to the arrangement had been dealing with each other at arm’s length.

Where section 295-550 is enlivened all of the income from the arrangement is NALI taxed at the highest rate.

Private companies dividends prone to be NALI

At the forefront of NALI is dividend income from investment by complying SFs in private companies.

In GYBW Senior Member McCabe identified an objective test in sub-section 295-550(2) which looks at a question of fact: is a dividend paid by a private company to a complying SF consistent with an arm’s length dealing? A private company dividend paid to a SMSF is NALI to the SMSF if it is not. This objective test replaced the former provisions in Part IX of the ITAA 1936 under which private company dividends were treated as special income (the forerunner to NALI) as a matter of course. That is, unless the Commissioner exercised a discretion that it was unreasonable to treat the private company dividend as special income where the Commissioner became satisfied that the income was earned at arm’s length.

Sub-section 295-550(3) sets out factors to be considered in applying the objective test.

The facts and findings in GYBW

In GYBW, the SMSF was the SMSF of a partner in an accounting practice with the pseudonym D. His client and connection pseudonym K had volatile and valuable business interests which could earn significant income from Department of Defence contracts.

D retired from his accounting practice to become the chief financial officer of the B Group.

The various partnership and corporate dealings of K are complex and supporting evidence of them before the AAT was “difficult” and incomplete. The AAT did not accept:

  • that the evidence, though involving non-related parties D, K, K’s trust and the other partners and former partners of K; and
  • that legal advice received before the SMSF invested in B Holdings;

supported a finding that the shares in pseudonym B Holdings acquired by D’s SMSF were acquired on terms where dividends would be earned from the shares consistently with an arm’s length dealing.

Senior Member McCabe observed how parties at arm’s length from each other can engage in an non-arm’s length dealing just as non-arm’s length parties can engage in an arm’s length dealing. For instance, in the latter case, a family member of the seller acquiring stock exchange listed shares of the seller on a stock exchange. Section 295-550 is directed to the dealing viz. how the SMSF came to earn the private company dividends it earned, not to the relationship of the parties to the arrangement. The AAT was therefore sceptical about the acquisition by D’s SMSF of ordinary shares in B Holdings on its formation for a nominal sum where B Holdings was also able to obtain and exploit K’s business interests a day later which D contended had negligible value then.

That AAT observed that “Fortune shined on the business” of B Holdings and B Holdings earned more than $10 million over four years which likely explains why it was picked up for an audit by the Commissioner.

Darrelen applicable

After looking at the Explanatory Memorandum with which section 295-550 was introduced Senior Member McCabe concluded that the purpose of the section did not change nor was there any change to the factors to which regard was to be had. Therefore the Full Federal Court decision in Darrelen Pty Ltd v Federal Commissioner of Taxation (2010) 183 FCR 237, which concerned the former provisions in Part IX of the ITAA 1936, remained authoritative in Senior Member McCabe’s view. In Darrelen the court had held that dividends paid by a private company were special income. In the case the SMSF had acquired its four shares in that company for a cost far less than their market value in an earlier year of income notwithstanding that the same dividend amount was paid on all 100 shares in the income year it was paid.

The cost to the SMSF of the shares on which the dividend was paid

The cost to the SMSF of the shares on which dividends were paid is a specific factor that can be taken into account under paragraph 295-550(3)(b) in determining whether their payment is consistent with an arm’s length dealing. In applying the objective test Senior Member McCabe referred to Commissioner of Succession Duties (SA) v Executor Trustee and Agency Co of South Australia Ltd (Clifford’s Case) where the High Court set out its views on how to value shares in a company:

The main items to be taken into account in estimating the value of shares are the earning power of the company and the value of the capital assets in which the shareholder’s money is invested. But a prudent purchaser does not buy shares in a company which is a going concern with a view to winding it up, so that the more important item is the determination of the probable profit which the company may be reasonably expected to make in the future, because dividends can only be paid out of profits and a prudent purchaser would be interested mainly in the future dividends which he could reasonably expect to receive on his investment. Further, a prudent purchaser would reasonably expect to receive dividends which would be commensurate with the risk, so that the more speculative the class of business in which the company is engaged the greater the rate of dividend he would reasonably require. In order to estimate the probable future profits of a company it is necessary to examine its past history, particularly the accounts of those years which are most likely to afford a guide for this purpose. In order to estimate the rate of dividend that a prudent purchaser could reasonably require on his investment it is necessary to examine the nature of the business and the risks involved and to seek the evidence of business men, particularly members of the stock exchange and experienced accountants, who can testify to the appropriate rate from the prices paid for shares in companies carrying on a similar business listed on the stock exchange or from private sales of shares in such companies or from their general business experience.

[1947] HCA 10; (1947) 74 CLR 358 at p.362

and with the benefit of hindsight, and omissions in the evidence supporting D’s SMSF’s case about how the SMSF and B Holdings came to benefit in K’s business interests, the AAT found that dividends were not consistent with arm’s length dealing as they arose from shares acquired for less than their value so evaluated. The AAT found that the dividends received by D’s SMSF from B Holdings were NALI.

NALI rules extended to expenses

The NALI rules have extended to losses, outgoing or expenditures that are less than expected to the complying SF by the Treasury Laws Amendment (2018 Superannuation Measures No. 1) Act 2019 in Schedule 2.


Unless GYBW is overturned on appeal SMSF investment in a private company of a related party or in a private company of the connections of the SMSF seem destined for NALI high tax treatment. So SMSFs should be wary of investment in private companies generally: SMSF investment in a private company carries the suspicion that the investment is an opportunity to shift income from a higher taxed entity to a concessionally taxed SMSF.

It follows that the trustee of a SMSF looking to sustain concessional tax treatment needs to adequately document its dealings with and investment in private companies so the arm’s length character of the investment can be verified and, where need be, independent valuation supporting consistency with arm’s length dealing should be sought.