Total and Permanent Disability (TPD) insurance is under sustainability pressure. Rising mental health claims now account for one in three TPD claims paid.[1] Insurance premiums have increased sharply. International reinsurers have begun to reconsider their exposure to the Australian TPD insurance market.
The Australian Prudential Regulation Authority (APRA) is concerned that, as the ultimate owners of group insurance product design, superannuation trustees’ narrow interpretation of the best financial interests duty (BFID) can hinder action when design change is needed.[2]
As we explain below, BFID, properly understood, gives trustees scope to act rather than a reason to hold back. What makes a decision defensible is a sound, well-documented process, a holistic view of the financial impact of design choices, consideration of evolving needs, and impartiality across cohorts of members.
The high bar for judicial review of trustees’ discretionary decisions
While the perceived risk of breaching BFID (or its predecessor, the best interests duty) is a common reason for caution, no contested court challenge to a trustee’s discretionary decision alleging breach of BFID under section 52(2)(c) of the Superannuation Industry (Supervision) Act 1993 (Cth) (SIS Act) has succeeded.[3] The biggest obstacle any challenger must overcome is the well-recognised high bar to challenging a trustee’s discretionary decisions, as set out in Karger v Paul: judicial review of a trustee’s exercise of discretion is confined to asking whether the discretion was exercised in good faith, upon real and genuine consideration, and for a proper purpose.[4]
The principle remains embedded in superannuation law, notwithstanding some qualifications. The High Court in Finch v Telstra Super Pty Ltd (Finch) held that decisions determining member benefit entitlements attract a heightened duty of inquiry - but crucially, the Court left open the question of how far the Karger v Paul principle should be qualified in the superannuation context.[5] Subsequent decisions have reinforced that the Karger v Paul principle continues to govern discretionary trustee decisions. In APRA v Kelaher, Justice Jagot distinguished Finch on the basis that the heightened inquiry obligation does not extend to day-to-day administrative decisions, such as whether to pursue a claim on behalf of the fund.[6] In Brady v NULIS Nominees (Australia) Ltd (No 4) (Brady (No 4)), Justice Markovic, accepting that the Karger v Paul principle (as qualified by Finch) continues to govern the review of discretionary trustee decisions, upheld a trustee’s decision to continue grandfathered commission arrangements, finding that the decision was made after a thorough process and was objectively justifiable as in members’ best interests.[7]
The tension between the positive statutory language in BFID and constraints on judicial review has been picked up by Justice Jackman in extrajudicial writing.[8] His Honour observes that the requirement for a trustee’s decision to be ‘reasonably justifiable’ should be treated as permitting a wide range of decisions a reasonable trustee could have made. Drawing on Lord Woolf MR’s observation in Equitable Life Assurance Society v Hyman[9] that there are ‘marked similarities’ between review of trustees’ discretions and judicial review of administrative action, Justice Jackman endorses the analogy with Wednesbury unreasonableness: the relevant question is whether a reasonable person in the position of the trustee could possibly have made the decision in question.
Focus on the quality of the decision-making process
While the precise boundaries of BFID remain unsettled,[10] cases shed light on what a defensible decision looks like.
BFID will be breached if the trustee’s subjective purpose is contrary to beneficiaries’ best financial interests, or if the decision is not reasonably justifiable as in their best financial interests, assessed objectively at the time it was made.[11] However, it:
- does not require the trustee to identify a single optimal course of action, as long as the decision is within the range of decisions reasonably open to the trustee – noting that there may be cases where an issue is so finely balanced that a decision either way can be regarded as objectively right;
- is not to be assessed with hindsight - the relevant question is whether the course of action adopted could reasonably be regarded as being in the interests of the beneficiaries at the time when it was made; and
- does not demand perfect decision-making - a decision that is reasonably justifiable is not a breach merely because more information could have been obtained or a better outcome was available.[12]
The Full Federal Court’s recent decision in Brady v NULIS Nominees (Australia) Ltd [2026] FCAFC 68 offers two practical lessons for trustees.
- A decision is only as defensible as the evidence behind it. The trustee succeeded because its board papers and minutes showed months of genuine deliberation, including the board initially declining the proposal and asking for more information. Sound process, credible advice, robust deliberation and a contemporaneous record of the matters considered will be the best evidence a trustee has to show the decision was reasonably open to it.[13] In addition, evidence must be adduced to rebut the presumption in civil proceedings brought by a regulator in relation to a breach of the best financial interests covenant as the evidential burden of proof is borne by the superannuation trustee and once discharged the regulator bears the burden of proof on the balance of probabilities.[14]
- A decision must be judged in its full context. The Full Court rejected a challenge that impermissibly abstracted one feature of the arrangements – some members bearing the cost of commissions without receiving ongoing services – from the broader context: the risk that abruptly ending commissions would trigger adviser disengagement, member attrition and loss of scale, and derail the successor fund transfer.[15] For group insurance, this means that no single feature, such as premiums, can be assessed in isolation from the wider context of the design choice.
Understand the financial implications of design choices
Does a holistic judgment of this kind sit at odds with the requirement to act in members’ best financial interests? In our view, it does not. The word ‘financial’ was inserted in 2021 to clarify the existing best interests duty and to make clear that beneficiaries’ financial interests are the determinative consideration.[16] The amendment was confirmatory and reflects the position already recognised at common law that, where the purpose of a trust is to provide financial benefits, as with a fund providing retirement income, the best interests of beneficiaries are normally their best financial interests.[17]
Making financial interests determinative does not reduce the BFID assessment to comparing headline metrics. In the context of group life insurance, the financial implications of a design choice are manifold, and hard to reliably measure given the long-term and long-tail nature of life insurance. Trustees must instead get comfortable by looking beyond static headline numbers to what drives them, what may cause them to change, and how costs are likely to behave over time.
It follows that sound decision-making in beneficiaries’ best financial interests requires more than a simple premium rate comparison. Indeed, a low-cost option may prove less favourable financially once other factors are considered, including:
- Benefits relative to premium: The starting point is the value of cover obtained for the premium paid. A policy with a lower premium but narrower definitions or less generous benefit design may deliver worse financial outcomes for members. In practice, the scope of cover must be assessed when a lower premium is proposed.
- Quality of services: More restrictive claims processes or weaker insurer support may also erode the value members receive for their premiums. Trustees should find ways to translate these qualitative features into quantifiable measures so that their financial implications can be properly assessed.
- Stability and sustainability of premiums: Group insurance premiums are typically subject to adjustment mechanisms tied to claims experience. The availability of a premium freeze period is a legitimate consideration. However, a cheaper contract at inception (including during a rate guarantee period) may produce a higher total cost over the policy period as a result of the insurer exercising the premium adjustment right - particularly where the initial pricing was set aggressively to win the tender, the rate guarantee period is short or absent, and the trustee has limited contractual control over repricing.
- Costs of switching: Changing insurer is costly. Re-tendering, contract negotiation, system migration and integration, and administrative resources all erode premium savings, particularly if those savings prove unsustainable. The analysis focuses less on the cost itself than on the likelihood of a switch, the extent to which the trustee has preserved switching agility, and whether transition costs can be mitigated through contractual mechanisms such as insurer contribution to onboarding, data migration support, or run-off arrangements.
Know members and their needs
A trustee cannot credibly claim to have acted in members’ best financial interests without understanding who those members are and what they need. That understanding serves several purposes. It is a statutory requirement under section 52(7) of the SIS Act, which requires trustees to base their insurance strategy on the demographic composition of their members.[18] It is also a prudential expectation to understand the insurance needs of member[19] and a threshold step for discharging trustees’ Design and Distribution Obligations (for choice products only).[20] Beyond those requirements, it is a key means of evidencing that a decision is consistent with BFID. This is particularly so where the trustee can show genuine engagement with membership data and explain why specific product features are or are not appropriate for that membership.
But member needs are not static. Trustee decisions about insurance design should also be forward-looking. Demographic shifts - an ageing workforce, changing patterns of work, longer working lives - will change what members need from insurance. Society and medical treatment shifts – greater awareness of mental health and advances in medicine – will change the risk profile of insured members. How ‘future-proof’ a policy design is could be a compelling rationale supporting the trustee’s decision.
This is particularly so for mental health, where some claimants may benefit more from early intervention, staged support and continued engagement with treatment than from a single lump sum paid years after the onset of disability. Alternative models have emerged in group and retail markets to address these changing needs, including severity-based TPD definitions that assess impairment rather than occupational incapacity, continuous care structures that focus on ongoing support requirements, and staged benefit payments that maintain claimant engagement with recovery pathways. Where sustainability pressures have persisted and mental health claims have become an increasingly prominent driver of claims experience, consideration of these alternatives is a compelling indication of the trustee’s engagement in ‘real and genuine consideration’ of insurance offerings for members’ best financial interests.
Act impartially among classes of beneficiaries
Many view BFID as a statutory duty derived from the trustee’s foundational duty at common law, as set out by Megarry V-C in Cowan v Scargill.[21] The duty to act impartially, whether viewed as an element of BFID, a common law duty, or a statutory obligation to act fairly within and between classes of members that applies cumulatively with BFID,[22] is a central consideration in trustee decision-making.
The impartiality duty applies to decisions on group insurance. Group insurance is, however, a product built on pooling and redistributing risk and cost across the membership, where trade-offs are inherent in every design choice. A decision that advantages one cohort, whether through broader cover or lower premiums, may prove to have disadvantaged another. This tension can be resolved by a robust decision-making process that considers fairness across all member cohorts in the context of all of the terms of the policy. Trustees should be comforted that this decision must be judged at the time of entry into the group insurance policy (rather than in hindsight), in light of the inherent uncertainty of how the claims experience will evolve during the term of the policy.
That emphasis on process reflects the high bar to challenging a trustee’s discretionary decision. Courts have held that a decision that appears to prefer one interest over another does not, of itself, expose the decision to criticism, as long as the decision is made in good faith, upon real and genuine consideration, and for a proper purpose, such that the preference is the result of a proper exercise of the discretionary power.[23] Applying that process-focused inquiry, Justice Markovic in Brady (No 4) upheld the trustee’s decision to maintain existing commission arrangements for a defined cohort of members as fair notwithstanding unequal treatment with other members, because, viewed holistically, maintaining the status quo served the best interests of all members.[24]
In applying the duty to act impartially, two connected themes emerging from decided cases are of relevance to the design of group insurance products.
- Intergenerational fairness: The duty is owed to beneficiaries. APRA has expressed the view that imposing the entirety of a liability or loss on the cohort of members at the moment it crystallises may be particularly unfair where the circumstances giving rise to it arose when the membership differed substantially.[25] Similarly, it is questionable whether it is consistent with the duty to act impartially and the fairness covenants for current insured members to benefit from unsustainably low premiums at the expense of those who remain insured and later pay higher premiums, for example once an unsustainably low rate guarantee period ends, particularly where contemporaneous evidence before the trustee indicates that the initial pricing is unsustainable.
- Cohort-based analysis: A trustee acting impartially and fairly would avoid inappropriately benefiting one class or group at the expense of another. Balancing collective and individual interests should take into account actuarial advice and assumptions that inform the evaluation of cohorts sharing similar characteristics, and rely on the actuarial analysis to identify affected cohorts, quantify the detriment and weigh it against the realistic alternatives.[26] Such advice is among the strongest evidence that a decision is reasonably open to the trustee, when it appears to give rise to fairness concerns.
APRA’s 2023 review of group insurance sustainability gives a practical sense of where fairness concerns are likely to arise. The alarm bells ring for features whose cost is pooled across all insured members but whose benefit is concentrated in a few, or which expose the pool to high risk and large sums insured that have not been priced through underwriting.[27]
Where does this leave us?
While there is no checklist for BFID compliance - the duty is deliberately and appropriately expressed at a level of generality that allows it to adapt to changing circumstances and evolving needs – one thing is clear. BFID, properly understood, should not hinder trustee action. The legal framework gives trustees enough room to act, provided their decisions are supported by sound process, holistic assessment, a genuine understanding of members’ needs, and informed trade-offs.
In the context of group TPD, one of the factors that a trustee must consider is the sustainability of premiums. While not the only factor, it is an important one in a trustee’s consideration of BFID and the fairness covenants, particularly given the cost of transitioning group insurance between life insurers.
Against continuing deterioration in TPD claims experience, rising costs to members, and real questions about whether TPD will remain affordable and available, maintaining the status quo may become increasingly difficult to justify as being in members’ best financial interests.
Council of Australian Life Insurers, ‘Mental Ill Health Is Straining Australia’s Safety Net’ (Media Release, 11 July 2025).
APRA Executive Director Jane Magill, ‘TPD sustainability: a product made for one world tested in another’ (Speech, All Actuaries Summit, 27 May 2026).
Breaches of section 52(2)(c) found in Steer v AMP Life Ltd [2021] SADC 109 and Notaras v Notaras [2012] NSWSC 947 did not concern discretionary decisions: the former arose from a failure to comply with a statutory notice obligation, and the latter from a trustee’s misappropriation of fund monies. The only other BFID-related proceeding resulting in declarations of contravention is Australian Securities and Investments Commission v AustralianSuper Pty Ltd [2025] FCA 102 (Hespe J). That proceeding was not contested and was resolved on an agreed basis.
Karger v Paul [1984] VR 161, 163 (McGarvie J).
Finch v Telstra Super Pty Ltd (2010) 242 CLR 254 [64]-[66] (French CJ, Gummow, Heydon, Crennan and Bell JJ) (‘Finch’).
Australian Prudential Regulation Authority v Kelaher [2019] FCA 1521 [154] (Jagot J) (‘Kelaher’).
Brady v NULIS Nominees (Australia) Ltd (No 4) [2024] FCA 1374 [484], [661] (Markovic J) (‘Brady (No 4)’), affirmed in Brady v NULIS Nominees (Australia) Ltd [2026] FCAFC 68 (Lee, Hespe and Button JJ) (‘Brady Appeal’).
Justice Jackman, ‘Superannuation Trustees’ Duty to Make Money for Their Beneficiaries’ (Speech, Law Council of Australia Superannuation Lawyers’ Conference, Canberra, 28 March 2025).
Equitable Life Assurance Society v Hyman [2002] 1 AC 408 [17].
In Brady Appeal, the appellant’s initial ground challenging the application of the Karger v Paul standard of review was not pressed. The Full Court dismissed the appeal on the evidentiary record, noting that the judgment ‘does not provide the occasion to wade into the question of the appropriate standard of review’: at [100] (Lee J).
Kelaher [63] (Jagot J).
Kelaher [53], [55], [58], [64]-[65] and [308] (Jagot J); see also Brady (No 4) [658], [661] (Markovic J); Brady Appeal [24], [97] (Lee J).
Brady Appeal [104]-[109] (Lee J); see also, in the group insurance context, Manglicmot v Commonwealth Bank Officers Superannuation Corporation Pty Ltd [2011] NSWCA 204 [54]-[64], [90] (Giles JA, Young and Whealy JJA agreeing).
SIS Act, section 220A.
Brady Appeal [102]-[103] (Lee J).
Explanatory Memorandum, Treasury Laws Amendment (Your Future, Your Super) Bill 2021 (Cth), [3.9]-[3.10]; see also ASIC v AustralianSuper Pty Ltd [2025] FCA 102 [124] (Hespe J).
Kelaher [65] (Jagot J), citing Cowan v Scargill [1985] Ch 270, 287, 289.
SIS Act, section 52(7)(a)(iii).
APRA, Prudential Practice Guide SPG 250 Insurance in Superannuation; see also APRA, ‘Sustainability of life insurance in superannuation’ (Web page, December 2023).
Corporations Act 2001 (Cth), section 994B.
Cowan v Scargill [1985] Ch 270, 287.
SIS Act, section 52(2)(e)–(f). In Brady (No 4) [711]-[714], Markovic J considered fairness between members under the fairness covenants in sections 52(2)(e) and (f).
Edge v Pensions Ombudsman [2000] Ch 602, 627.
Brady (No 4) [716]–[717], [731]–[733] (Markovic J).
See APRA’s submissions in Application by LGSS Pty Ltd atf Local Government Super [2021] NSWSC 1613 [146] (Ward CJ in Eq) and Application by Maritime Super Pty Ltd atf Maritime Super [2021] NSWSC 1614 [162]-[163] (Ward CJ in Eq).
See Re KCA Super Pty Ltd (No 2) [2011] NSWSC 1301 (Brereton J).
APRA, ‘Sustainability of life insurance in superannuation’ (Web page, December 2023).

