FEDERAL COURT OF AUSTRALIA

Kusmanoff v Retail Employees Superannuation Pty Limited [2026] FCA 1728

File number:

VID 1107 of 2023

Judgment of:

BUTTON J

Date of judgment:

17 November 2026

Catchwords:

REPRESENTATIVE PROCEEDINGS – notices – registration process – application by Respondent for an

order that a notice be sent to group members foreshadowing

an intention to seek an order that only registered group

members can participate in any settlement – where

Applicant opposed the application – where Applicant’s

claims concern the automatic provision of income protection insurance to inactive superannuation fund members in particular circumstances – where the parties have no way of identifying which of the approximately 950,000 inactive members are group members – uncertainty as to group size, quantum, and settlement participation rates – where early resolution of the proceeding in the best interests of group members as a whole – where registration process would significantly enhance the prospect of settlement of the proceeding – application granted

Legislation:

Federal Court of Australia Act 1976 (Cth) ss 33J, 33X, 33ZF

Civil Procedure Act 2005 (NSW) s 175

Civil Proceedings Act 2011 (Qld) ss 103T, 103ZA

Supreme Court Act 1986 (Vic) ss 33ZF, 33ZG

Cases cited:

Alford v AMP Superannuation Ltd (No 2) [2024] FCA 423

Anderson-Vaughan v AAI Ltd (No 2) [2024] VSC 65

Blairgowrie Trading Ltd v Allco Finance Group Ltd (Receivers & Managers Appointed) (In Liq) [2015] FCA 811; (2015) 108 ACSR 1

BMW Australia Ltd v Brewster [2019] HCA 45; (2019) 269 CLR 574

Clarke v JB Hi-Fi Group Pty Ltd [2025] VSC 664

Earglow Pty Ltd v Newcrest Mining Ltd [2015] FCA 328; (2015) 230 FCR 469

Fox v Westpac Banking Corporation [2023] VSC 414

Green v Johnson and Johnson Pacific Pty Ltd (No 2) [2026] FCA 1094

J Wisbey & Associates Pty Ltd v UBS AG (No 2) [2024] FCA 147

Lendlease Corporation Ltd v Pallas [2025] HCA 19; (2025) 423 ALR 23

Merrett v State of Queensland (Queensland Police Service) (No 2) [2026] QSC 154

Nelson v Beach Energy Ltd [2025] VSC 339

P Dawson Nominees Pty Ltd v Brookfield Multiplex Ltd (No 2) [2010] FCA 176

Parkin v Boral Ltd [2022] FCAFC 47; (2022) 291 FCR 116

Petrucci v Toyota Finance Australia Ltd [2026] VSC 191

Preece v Aristocrat Leisure Ltd [2025] FCA 742

Regent Holdings Pty Ltd v State of Victoria [2012] VSCA 221; (2012) 36 VR 424

Division:

General Division

Registry:

Victoria

National Practice Area:

Commercial and Corporations

Sub-area:

Commercial Contracts, Banking, Finance and Insurance

Number of paragraphs:

99

Date of last submissions:

28 October 2026

Date of hearing:

16 October 2026

Counsel for the Applicant:

WAD Edwards KC with A Mobrici

Solicitor for the Applicant:

Shine Lawyers

Counsel for the Respondent:

C Caleo KC with A Batrouney

Solicitor for the Respondent:

Allens

ORDERS

VID 1107 of 2023

BETWEEN:

PAUL KUSMANOFF

Applicant

AND:

RETAIL EMPLOYEES SUPERANNUATION PTY LIMITED (ACN 001 987 739)

Respondent

order made by:

BUTTON J

DATE OF ORDER:

17 November 2026

THE COURT ORDERS THAT:

1.    The parties are to confer and provide proposed orders to chambers (including the notice to be issued to potential group members).

2.    If the parties cannot agree on the terms of the notice to be issued and the terms of orders giving effect to these reasons, they should notify chambers by no later than 4:30pm on 23 November 2026 so that further orders can be made for the parties to put their positions.

3.    The case management hearing on 20 November 2026 will be adjourned to 4 December 2026.

Note:    Entry of orders is dealt with in Rule 39.32 of the Federal Court Rules 2011.

REASONS FOR JUDGMENT

BUTTON J:

1    The Respondent (REST) has applied for orders that provide for the conduct of a pre-mediation registration process for group members. As there has been no opt-out process yet, the proposal is for a single notice to be issued which covers opt out and the registration process. The orders sought provide for the notice to inform group members that, upon any settlement reached before the first day of trial (and subject to Court approval), the parties will seek an order that would mean that group members who have not registered will not be permitted to participate in the settlement.

2    The Applicant opposes REST’s application.

3    The application needs to be considered in the context of the substantive proceeding. The Applicant alleges that REST breached its obligations (under general law, the Superannuation Industry (Supervision) Act 1993 (Cth) and pursuant to the Trust Deed) to some of its members by continuing to provide income protection insurance policies as a default, and charging the costs of that default insurance, to inactive members for whom that insurance was not of benefit.

4    The group on whose behalf the proceeding is advanced is defined as follows in the Statement of Claim (SoC) (omitting reference to deceased members, persons with family law arrangements and the exclusion of Commonwealth judges and registrars):

2.     The Applicant and the persons whom they represent (Group Members) are persons who:

2.1.     satisfy all of the following paragraphs:

(a)     were members of the Retail Employees Superannuation Trust (REST Fund) between 5 December 2008 and 30 June 2019;

(b)     were provided with income protection insurance as an automatic incident of their membership of the REST Fund;

(c)     were charged insurance premiums and/or fees in respect of such income protection insurance, which amounts were automatically deducted from their superannuation account balance in the REST Fund;

(d)     did not make or receive any contributions, rollovers or transfers, or receive any benefit payments for a period of more than 12 months; and

(e)     following the period of more than 12 months referred to in paragraph (d) above, continued to be charged the insurance premiums and/or fees referred to in paragraph (c) above;

…

AND

2.4    suffered loss or damage by the conduct pleaded herein; …

5    Members who were automatically insured, but who had not made or received any contributions, rollovers or transfers into their accounts for more than 12 months, are referred to in the SoC as Inactive Members: SoC [20].

6    Only a subset of Inactive Members are Group Members. It is only Inactive Members who have “suffered loss or damage by the conduct pleaded herein” that are Group Members. There are two categories of Inactive Members who are said to have suffered loss or damage (referred to in the SoC as the Non-Benefiting Inactive Members):

(a)    Dual-Fund Inactive Members; and

(b)    Non-Working Inactive Members.

7    Dual-Fund Inactive Members are Inactive Members “who were also members of other superannuation funds which provided those members with income protection insurance as an incident of membership”: SoC [21].

8    Non-Working Inactive Members are Inactive Members “who were likely to be unemployed and/or not receiving income during the 12 month period in which they had not made or received any contributions, rollovers or transfers into their REST Fund account”: SoC [22].

9    The Applicant alleges that the Group Members have suffered loss or damage due to, among other things, the reduction in the balance of their accounts through the deduction of insurance premiums, and the consequential reduction in the income generated and to be credited to their accounts: SoC [34], [38].

10    In support of its application, REST relies on an affidavit of its solicitor, Mr Malcolm Stephens, and an affidavit of a Senior Product Manager at REST, Mr Andrew Mortimer, both sworn 21 September 2026. In opposing the application, the Applicant relies on two affidavits of his solicitor, Mr Craig Richard Allsopp, sworn 8 and 16 October 2026.

The status of the proceeding, and the issue leading to the making of the registration application

11    The matter is set down for trial on 4 October 2027 on an estimate of four weeks. The Applicant’s evidence (lay and expert) is due to be completed by 18 December 2026, with the Respondent’s lay and expert evidence due by 9 April 2027, and the Applicant’s reply evidence due by 11 June 2027.

12    A mediation is to be held by 18 June 2027, on the basis that the mediation may continue after this date. The mediation will take place after the evidence is completed, but before the parties have incurred the costs of trial preparation.

13    REST can identify, from its records, the “Inactive Members”. Its records suggest there are approximately 950,000 Inactive Members. However, being an Inactive Member is only one component of the group member definition: Inactive Members must also be a Dual-Fund Inactive Member or a Non-Working Inactive Member in order to be a Group Member. REST does not know how many, or which, of its Inactive Members:

(a)    had other superannuation accounts with a fund other than REST through which they had income protection insurance; or

(b)    were unemployed and/or not receiving income.

As already noted, it is only Inactive Members meeting one or both of those criteria who are Group Members.

14    It follows that REST does not know, and cannot find out from its own records, the number of Group Members, or the periods, within any 12-month period of inactivity, during which a member held another superannuation account with income protection insurance, or was unemployed or not receiving income. REST is not able, from its own records, to assess the quantum of the claims of Group Members.

15    For completeness, I note that REST has, on an ad hoc basis, been informed by some members of personal circumstances (eg that they had a separate superannuation account with income protection insurance, or that there had been a change in their employment status), but that has only occurred in respect of a very small number of Inactive Members, on an ad hoc basis, and does not provide a basis upon which to assess the number of Group Members, or the quantum of their claims.

The registration proposal in more detail

16    After the Applicant raised an objection that the timetable REST initially proposed for the pre-mediation registration process would divert resources from the preparation of his evidence, REST modified the timing of its proposal. Following those modifications, the Applicant did not raise any further objection based on the timing diverting resources from evidence preparation.

17    REST’s revised proposal provides for the following timetable:

(a)    23 April 2027 be fixed as the date for opt out and registration;

(b)    by 15 January 2027, REST would provide a distribution agent with a list of persons to receive the notice, including the person’s name and best available contact details (email or postal address, or both);

(c)    the notice would be issued by 29 January 2027, by email to those with email addresses, and by post to those with postal addresses only or where there has been an email bounce-back;

(d)    the notice would also be issued by the Applicant’s solicitors, Shine Lawyers (Shine), to those who have already registered interest with the solicitors or retained them, and would be displayed on Shine’s website;

(e)    the registration form could be completed online, or by emailing or posting a hard copy to Shine; and

(f)    by 21 May 2027, Shine would send a list of registered persons to REST’s solicitors.

18    Those receiving the notice would have approximately two and a half months in which to register (late January 2027 to 23 April 2027). The Applicant did not suggest that this was not a reasonable period.

19    The draft registration form prepared by REST asks some basic questions to identify whether the person is a Group Member. Relevantly, the questions include:

(a)    whether the person (or spouse or the relevant deceased person) was unemployed, on parental or other leave, or not receiving income for another reason, during any 12 month inactivity period;

(b)    if so, during what period(s) the person was unemployed or not receiving income; and

(c)    whether the person (or spouse or the relevant deceased person) had income protection insurance with another superannuation fund and, if so, during what period(s).

20    It is important to note that the Applicant accepts that a registration process will need to be carried out at some stage, if the matter settles; this is not a case in which it has been suggested that a settlement could be effected and distributed without a registration process.

The power to order a pre-mediation registration process and factors to be considered

21    Paragraph 1 of REST’s Interlocutory Application seeks an order fixing a date for opt out and registration pursuant to ss 33J and 33ZF of the Federal Court of Australia Act 1976 (Cth) (the FCA Act). Given s 33J is concerned with opt out, the application for a registration date (registration date) to be set is made under s 33ZF.

22    Paragraph 2 of the application seeks approval for the issue of a notice informing Group Members of their right to opt out, and of the intention — if the matter settles — to seek an order, the effect of which would be to provide that Group Members who have not registered by the registration date will remain Group Members but will not, without leave of the Court, be permitted to participate in the settlement. This aspect of REST’s application is advanced under s 33X of the FCA Act.

23    Following Lendlease Corporation Ltd v Pallas [2025] HCA 19; (2025) 423 ALR 23 (Lendlease), it is clear that the Court has the power under s 33X to authorise the issue of a notice that foreshadows an intention to apply for an order that group members who have neither registered nor opted out are bound by the settlement but cannot participate in the fruits of the settlement. That is the kind of process REST envisages.

24    As REST invokes s 33ZF in support of its application that the registration date be set, and because the effect of the orders REST seeks is to establish a court-ordered registration process, I proceed on the basis that the pre-mediation registration process should not be ordered unless it is supported by s 33ZF (and the suite of orders sought cannot be addressed only through the notice-approval power, s 33X). This is, in substance, the position that the Applicant advanced, whereas REST only addressed s 33X.

25    I accept the Applicant’s submission that s 33ZF must be engaged if a pre-mediation process is to be ordered, notwithstanding that the ordering of a pre-mediation process does not, by itself, carry with it any immediate consequences in terms of Group Members’ ability to participate in any future settlement. The conduct of a pre-mediation registration process involves significant expenditure and has an effect on the conduct of the proceeding in the lead-up to mediation and trial. To my mind, directing that such a process be engaged in goes beyond the exercise of a power to approve the issue of a notice (s 33X), the notice being part of the machinery by which a pre-mediation registration process is implemented.

26    Strangely, the role of s 33ZF in applications of this kind has not received much attention in the authorities following Parkin v Boral Ltd [2022] FCAFC 47; (2022) 291 FCR 116 (Boral). In Boral, the Full Court (Murphy and Lee JJ, Beach J agreeing) noted (at [140]) that the “Contradictor did not suggest before us that the first aspect, a regime allowing registration and the provision of information in advance of a mediation (prayers 5 to 8), is beyond an exercise of power pursuant to s 33ZF” (emphasis added). The High Court’s decision in Lendlease focused on s 33X and its NSW equivalent (being s 175 of the Civil Procedure Act 2005 (NSW)), confirming the power to issue a notice advising group members of consequences that may be sought in the future. The Applicant submitted that Lendlease “ought not be overread”, because in that case the parties agreed that the giving of the proposed notice was conducive to the settlement of the proceeding, and further submitted that “[n]othing that the High Court said in Lendlease, which is all about 33X(5), alters the fact that to engage it with a registration process endorsed by the court, you need to go through 33ZF”.

27    As is detailed below (see paragraph 38ff), there have been nine contested applications for orders for the conduct of a pre-mediation registration process since the Full Court’s decision in Boral. In the four of those cases conducted in this Court — J Wisbey & Associates Pty Ltd v UBS AG (No 2) [2024] FCA 147 (Wisbey), Alford v AMP Superannuation Ltd (No 2) [2024] FCA 423 (Alford), Preece v Aristocrat Leisure Ltd [2025] FCA 742 (Preece) and Green v Johnson and Johnson Pacific Pty Ltd (No 2) [2026] FCA 1094 (Green) — it is clear that the orders were sought, as they have been in this case, invoking both s 33X and s 33ZF. The four Victorian cases — Fox v Westpac Banking Corporation [2023] VSC 414 (Fox), Anderson-Vaughan v AAI Ltd (No 2) [2024] VSC 65 (Anderson-Vaughan), Clarke v JB Hi-Fi Group Pty Ltd [2025] VSC 664 (Clarke) and Petrucci v Toyota Finance Australia Ltd [2026] VSC 191 (Petrucci) — appear to have been advanced in reliance on s 33ZF and s 33ZG of the Supreme Court Act 1986 (Vic) (the Victorian Act) (the latter provision being found only in the Victorian Act). In the remaining case, Merrett v State of Queensland (Queensland Police Service) (No 2) [2026] QSC 154 (Merrett), being a Queensland case, as far as the reasons reveal (eg at [12]), the application was considered by reference to the equivalent of s 33X (being s 103T of the Civil Proceedings Act 2011 (Qld)), although the orders annexed to the judgment providing for registration were also made pursuant to the Queensland equivalent of s 33ZF (being s 103ZA).

28    Although the four cases in this Court involved applications under s 33ZF, as well as s 33X, the analysis referred to s 33X, without explicit consideration of s 33ZF (Wisbey at [63], [66], [68]–‍[70] (Beach J); Preece at [38]–[40] (Anderson J); and Green at [45]–[49] (Moshinsky J)). In Alford, Murphy J cited Wisbey (which referred to s 33X) but did not otherwise directly address the legislation.

29    I return, then, to s 33ZF and the test to be applied in determining whether to make orders providing for the conduct of a pre-mediation registration process, but note that the considerations discussed in the cases on pre-mediation registration processes — even where the focus was on s 33X — are relevant to the Court’s determination of whether the conduct of a pre-mediation registration process is appropriate or necessary to ensure that justice is done in the proceeding.

30    Section 33ZF of the FCA Act provides:

33ZF     General power of Court to make orders

(1)     In any proceeding (including an appeal) conducted under this Part [ie Pt IVA], the Court may, of its own motion or on application by a party or a group member, make any order the Court thinks appropriate or necessary to ensure that justice is done in the proceeding.

(2)    Subsection (1) does not limit the operation of section 22.

31    Section 33ZF confers on the Court a discretion to make orders, but that discretion is only available in respect of orders the Court “thinks appropriate or necessary to ensure that justice is done in the proceeding”.

32    While the power is broad, it is supplementary and the words of limitation must not be ignored: BMW Australia Ltd v Brewster [2019] HCA 45; (2019) 269 CLR 574 (Brewster) at [46] (Kiefel CJ, Bell and Keane JJ); Boral at [143]–[144] (Murphy and Lee JJ, Beach J agreeing). Those words of limitation mean that the s 33ZF power will not be enlivened where the order sought is “merely convenient or useful per se” but is not one that the Court thinks appropriate or necessary “to ensure that justice is done in the proceeding”: Earglow Pty Ltd v Newcrest Mining Ltd [2015] FCA 328; (2015) 230 FCR 469 (Earglow) at [33] (Beach J). However, the power is not to be read down by making implications or imposing limitations not found in the words of the provision: Brewster at [56] (Kiefel CJ, Bell and Keane JJ), citing the following observations of Wigney J in Blairgowrie Trading Ltd v Allco Finance Group Ltd (Receivers & Managers Appointed) (In Liq) [2015] FCA 811; (2015) 108 ACSR 1 at [98]:

A provision conferring a broad power on the Court should generally not be read down by making implications or imposing limitations which are not found in the express words: The Owners of the Ship “Shin Kobe Maru” v Empire Shipping Company Inc (1994) 181 CLR 404 at 421; Wong v Silkfield Pty Limited (1999) 199 CLR 255 at [11]. The only express limitation or requirement in s 33ZF is that the Court thinks the order is appropriate or necessary to ensure that justice is done in the proceeding. No other limitation should be read into the section.

33    The concept of ensuring that “justice is done in the proceeding” is not confined to the resolution of the substantive issues between the parties: Brewster at [48] (Kiefel CJ, Bell and Keane JJ). Interlocutory orders that are “apt to move the proceeding towards a just conclusion between the parties” are within the scope of s 33ZF: Brewster at [54] (Kiefel CJ, Bell and Keane JJ). The plurality in Brewster gave the following examples of the ways in which an order may connect with the statutory concept of ensuring that justice is done in the proceeding: the order contemplated may assist in determining an issue in dispute, preserve the subject matter, ensure the efficacy of any judgment, assist in the management of the proceeding to bring it to a resolution, or do justice between group members in relation to the costs of litigation: Brewster at [51] (Kiefel CJ, Bell and Keane JJ). In Lendlease, Gordon and Steward JJ observed that the class action legislative regime is directed at achieving fair and effective outcomes by settlement as well as by judgment, and a construction of its provisions facilitating the conduct and resolution of a class action, whether by settlement or judgment, is to be preferred: at [59], [90], [99].

34    It has long been recognised that, although open class actions typically proceed without the active involvement of group members — or proceed in that way at least for a time, unless and until group members need to come forward to press their individual claims — in some circumstances, group members may be required to take positive steps: Lendlease at [65] (Gordon and Steward JJ). Their Honours observed as follows in relation to the NSW equivalent of s 33ZF of the FCA Act (at [65], emphasis added):

Section 183 confers a broad power. As observed by Beach J in Earglow Pty Ltd v Newcrest Mining Ltd, this power can oblige group members to take positive steps in a proceeding, contrary to the general observation made by this Court in Mobil (set out above), such as the provision of discovery, the provision of particulars, the possible provision of a contribution towards security for costs, and the provision of particulars concerning group members’ identities to facilitate the service of subpoenas. In, for example, P Dawson Nominees Pty Ltd v Brookfield Multiplex Ltd (No 2), the Federal Court observed that it may be appropriate to make discovery orders binding on group members “in aid of mediation, for example where the parties face asymmetric information which may lead to an unfair settlement”. And in Regent Holdings Pty Ltd v Victoria, the Court of Appeal of the Supreme Court of Victoria also stated that ordering particulars and discovery can, in a given case, make sense in order to “provide the defendants with sufficient information to formulate rational settlement offers”.

35    As may be seen, Gordon and Steward JJ in Lendlease included the provision of information to defendants to provide them with sufficient information to formulate rational settlement offers, as among the examples of circumstances in which orders have been made requiring group members to take positive steps. Their Honours referred, with approval, to the decision of this Court in P Dawson Nominees Pty Ltd v Brookfield Multiplex Ltd (No 2) [2010] FCA 176 at [32] (Finkelstein J), and the decision of the Victorian Court of Appeal in Regent Holdings Pty Ltd v State of Victoria [2012] VSCA 221; (2012) 36 VR 424 (Regent Holdings) at [15] (Nettle, Redlich and Osborn JJA). In addition, in Earglow, Beach J observed that “[d]iscovery and particulars of group members’ claims may be accelerated in order to facilitate a mediation so that there can be an adequate appraisal of potential quantum ranges for the total class and the avoidance of a disproportionate level of information asymmetry”: at [49], see also [122].

36    I recognise that the Victorian cases must be read mindful that the class actions legislation in that State includes a provision not found in the federal, Queensland, or NSW, legislation: s 33ZG, which authorises orders under s 33ZF setting out a step group members must take in order to obtain a benefit. I also recognise that Regent Holdings (being the case to which Gordon and Steward JJ referred in the passage from Lendlease quoted above) involved a limited group that was, in effect, closed: see Regent Holdings at [14]–[15]. Nevertheless, the observations of the Victorian Court of Appeal concerning mediation are apposite and of general application. Their Honours, Nettle, Redlich, and Osborn JJA said (at [23], emphasis added):

Of course, mediation should be conducted without prejudice. But that does not mean that it should be conducted in ignorance. Court ordered mediation is not a game of bluff and bluster in which one party is free to mislead another to conclude that a claim is worth more than it is. It is designed to be an exercise in rational bargaining between relatively well-informed parties aimed at providing just compensation for worthy claims. The more accurate and complete the available information as to quantum, the more likely that rational settlements will be achieved.

37    Having regard to the foregoing, in my view, s 33ZF is available to support orders effecting a pre-mediation registration process, and the Court may be satisfied that such orders are “appropriate or necessary to ensure that justice is done in the proceeding”, if the orders sought will promote the early resolution of the issues in the litigation through mediation. I do not consider that the power is only enlivened if the prospects of mediation are improved to a degree that meets some stated threshold, whether expressed as a “material” or “significant” improvement, or by some other verbal formula. To the extent that the Applicant’s oral submissions suggested to the contrary, I reject them. Any such approach would read in a constraint not found in the language of s 33ZF.

38    There are a number of cases in which pre-mediation registration procedures have been ordered, and cases in which such orders have been refused. Since the decision in Boral, there have been nine contested applications for pre-mediation registration processes. Although the terminology of “soft class closure” has continued to be used in some of the cases — and by the parties in the present proceeding — to characterise applications of the present kind, that expression is best avoided as it is apt to introduce confusion. REST’s application does not provide for class closure.

39    Orders for pre-mediation registration were made in Fox, Anderson-Vaughan, Wisbey, Green and Merrett.

40    Orders were refused in Alford, Preece, Clarke, and Petrucci.

41    Although the Victorian class actions legislation differs inasmuch as the federal regime does not include s 33ZG, it is apparent that substantially the same considerations have been applied in both jurisdictions. The relevant considerations were identified by Beach J in Wisbey. His Honour’s summary has been referred to in the cases decided after Wisbey. The application in Wisbey was made under s 33J (regarding the opt-out component), s 33ZF (regarding the order fixing a date for registration) and s 33X (regarding approval of the notice). It may be seen that the application was accordingly structured in the same way as REST’s application.

42    In Wisbey, Beach J ordered, over the applicant’s objection, that a notice be distributed inviting group members to register. The notice was to state that, if a settlement were reached, an order would be sought at the settlement approval stage limiting participation in the settlement to registered group members. The group in Wisbey was defined by reference to transactions “arranged in Australia” with any dealer, which meant that the group included persons who had no trading relationship with any respondent and “the identity of a significant proportion of group members [was] not known or ascertainable by the parties”: at [4]–[6], [20], [118], [130].

43    The registration process proposed in Wisbey would enable the parties to obtain information they did not have and could not reliably ascertain, including the number of potential group members, the number who wished to participate in any settlement, their identity and whether they satisfied the group definition, the volume of their qualifying trading, and whether they had already received compensation: at [75].

44    In Wisbey, Beach J set out the principles to be applied in determining whether to make orders for registration and the issue of a notice of the kind considered in Lendlease, at [69]–[70] (emphasis added):

69     The issue for the Court when asked to make registration orders of this kind is thus not whether it has the power to make them, but rather whether it is an appropriate exercise of its power under s 33X. The relevant considerations in this context can include whether it is in the interests of group members as a whole to require registration before any prospective settlement is on the table, the point which the proceeding has reached, the attitude of the parties, the complexity and likely duration of the case, with protracted litigation and greater complexity increasing the interests of group members in avoiding litigation risk through achieving a settlement, whether group members have adequate notice of the change and reasonable time to decide whether to register, and whether an estimate of the size and number of claims can be made.

70     Whilst the significance of these considerations will vary with the circumstances of the case, a paramount factor is the extent to which a registration process is likely to improve the prospects of achieving a reasonable settlement.

45    Justice Beach’s description of the relevant considerations has subsequently been applied in a number of cases, including Alford (at [91]), Preece (at [40]), Clarke (at [21]), Petrucci (at [26]), Merrett (at [32]) and Green (at [49]), save that Murphy J in Alford (at [92]), and Anderson J in Preece (at [40]) considered the extent to which registration would improve the prospects of achieving a settlement at mediation to be an important (but not “paramount”) factor.

46    As Beach J observed in Wisbey, the significance of the considerations will vary depending on the circumstances of each case. Nevertheless, it is instructive to have regard to the factors that were important in the court ruling for, or against, a pre-mediation registration process in each of the nine cases referred to above. Taking them in chronological order (and noting that Fox was decided before Wisbey, and Anderson-Vaughan was heard before Wisbey was handed down, but delivered on the same day as Wisbey):

(1)    Fox concerned three class actions concerning “flex commissions” in motor vehicle finance. The groups in the three cases comprised approximately 403,000, 373,000 and 193,900 potential members. The defendants sought orders under ss 33ZF and 33ZG of the Victorian Act requiring group members to register in order to seek any benefit under any in-principle settlement reached at or in consequence of the mediations. In this way, the application being decided went beyond providing for a pre-mediation registration process, and the issue of a notice foreshadowing orders that would be sought in connection with settlement approval. Justice Nichols determined to make the orders sought. Her Honour’s reasons reveal the following relevant factors:

(a)    a registration process would be required if a settlement was reached, in any case: at [33];

(b)    the size of the class was large, but known — the difficulty lay in knowing how many group members would participate, and in quantifying individual group member claims, which were highly individualised in nature: at [100]–[104], see also [38], [46], [51], [93]ff;

(c)    records were available about likely class members so as to ensure that court-approved notices could be issued to each of them: at [69]ff;

(d)    it was a matter of speculation how much difference to registration rates it would make if a notice advised of an in-principle settlement, as compared with voluntary registration prior to mediation: at [87];

(e)    it had not been shown that, without an incentive to act in the form of “money on the table” following an in-principle settlement, there was a real prospect that group members would not be able to exercise their rights: at [88] and [111];

(f)    the attitude of the parties to the application is relevant, but “the position of the practitioners representing either party cannot be determinative in the sense of arrogating to the parties a decision that must remain one for the Court”: at [28]; and

(g)    there is no general rule that class closure orders prior to mediation are generally only made with the consent of the plaintiff, or when both parties concur: at [45].

(2)    Anderson-Vaughan concerned add-on insurance sold through motor vehicle dealers. There were 298,687 policies and approximately 200,000 unique group members, and approximately 77 per cent of the policies (around 230,000) fell within a “Non-Statutory Claim Period” cohort whose holders were required to prove unilateral mistake in order to overcome limitation defences: at [1], [35]–[38], [42]. The defendants sought orders for registration and soft class closure of a similar kind to those made in Fox, pursuant to ss 33ZF and 33ZG of the Victorian Act. The plaintiff opposed those orders being made. In deciding to make the orders sought, Delany J considered the following factors in particular:

(a)    the size and structure of the class meant that establishing the participation rate of each of the two cohorts was important: at [65];

(b)    knowing registration rates would enable the parties to assess the worst case for the defendants and the best case for the plaintiff and the group members, which was “likely to be a critical element in attempts to resolve the proceeding”: at [66];

(c)    the more “accurate and complete the available information as to quantum, the more likely that rational settlements will be achieved”: at [67], quoting Regent Holdings at [23] (Nettle, Redlich and Osborn JJA), see also [61], [68]–[69]; and

(d)    group members would have to take a positive step at some stage, if the proceeding settled or succeeded at trial, so the question about a registration process was when that should occur (not if it should occur): at [77].

(3)    The context in Wisbey has been set out above. In determining to make the orders sought, Beach J considered the following matters:

(a)    a mediation would be “significantly more likely to succeed when the parties can reasonably estimate the number of group members who will participate in any settlement, and the likely quantum of their claims”, and without that information, “it is difficult for the parties to agree on a figure, let alone be confident that any in-principle settlement will result in finality”: at [118];

(b)    there was a “clear inability” to make those estimates without registration: at [120]–[121];

(c)    although the applicant had proposed a “top down” method for estimations using publicly available trading data, and participation and recovery rates inferred from overseas proceedings, there were deficiencies in those information sources, making the applicant’s proposed approach “suboptimal”, leaving an “uncertain baseline of information” and lacking “the requisite level of detail to understand the claim size of the group”: at [82]–[85], [124]–[126];

(d)    the applicant’s suggestion that an expert could be briefed to make estimates with “unidentified appropriate adjustments on the basis of equally unidentified appropriate assumptions and inferences” did not address the respondents’ concerns about the limitations in the public data identified by the applicant: at [133]–[135];

(e)    a registration process would be required sooner or later, and the costs of a registration process would have to be borne eventually: at [182], [185];

(f)    a registration process would allow the parties to assess whether the costs being incurred are proportionate to the total quantum of group members’ claims: at [166];

(g)    the period covered by the orders would be confined to the mediation, and the Court would retain control over class closure and the capacity to grant leave to unregistered group members to participate, even after class closure orders are made: at [163]–[164], [180]; and

(h)    while such orders had generally been made on the joint application of the parties, on the facts of the case, it was appropriate to make the orders over the objection of the applicant: at [90]–[91].

(4)    Alford was decided shortly after Wisbey. The case was a superannuation class action alleging breaches of the trustee’s obligations said to result in higher fees and lower investment returns. The proceeding concerned 21 superannuation products over periods of up to 12 years and had approximately two million group members. The trustee respondents sought orders including a pre-mediation registration process. In dismissing the application, Murphy J considered the following:

(a)    the applicants were opposing the application on cogent grounds, and the respondents’ interests were inimical to group members’ interests such that caution was required, but it remained a matter for the Court to decide whether to make the orders sought: at [65]–[69], citing Fox at [28] (Nichols J);

(b)    the respondents’ evidence did not establish that their records were insufficient to enable them to compile a representative sample of group members’ claims (as distinct from a sample of only those who register) to participate efficiently and effectively in the mediation: at [70]ff;

(c)    the respondents’ own records were likely to be a better source of a representative sample of group members’ claims than a registration regime with a very low take-up: at [78];

(d)    the respondents’ application was a “cloak to disguise the fact the respondents wish[ed] to confine the class to just those group members who register”: at [4(b)], [82];

(e)    superannuation members typically have a low level of engagement, and the proposed registration regime was therefore likely to result in a very low rate of registration, and noted that this was significant to his decision on the application: at [19]–[20], [30]–[31];

(f)    the registration regime would be costly in a case where it was not the case that a registration process would be necessary at some point: at [84]–[87];

(g)    a registration process would divert the applicants’ legal team’s resources away from preparing for mediation and trial: at [46]–[49];

(h)    the relationship between group members and the trustee respondents was that of trustee and beneficiary, which pointed away from making class closure orders that would mean some beneficiaries would not benefit from a settlement without taking active steps: at [4(e)], [83]; and

(i)    with regard to Wisbey, the likelihood of a registration process improving the prospects of achieving a reasonable settlement was an “important consideration” (but disagreed with Beach J’s characterisation of it as “a paramount factor”): at [91]–[92].

(5)    Preece was a class action commenced on behalf of people who had purchased virtual currency in poker-machine-style mobile phone games, which were said to be offered contrary to the Interactive Gambling Act 2001 (Cth). Claims were advanced in restitution, for misleading or deceptive conduct and, in respect of a subset of vulnerable group members, for statutory unconscionable conduct. The respondents applied for orders for the distribution of pre-mediation registration notices to group members, foreshadowing the respondents’ intention to seek orders that only group members who registered may benefit from any future settlement. The applicant opposed the respondents’ application. In refusing to make the orders sought, Anderson J considered the following factors:

(a)    the respondents could, from their own records, reasonably estimate the aggregate value of group members’ claims for the purpose of participating in a productive mediation, and also had data that would allow them to estimate the value of individual claims: at [77]–[78];

(b)    uncertainty about quantum is a common feature of class actions and did not mean the parties could not have a productive mediation: at [79];

(c)    it was not necessary for the respondents to have more information about the vulnerable persons (who had unconscionable conduct claims) as the information sought was likely to be unreliable and those group members did not claim additional loss on account of their unconscionability claims: at [80];

(d)    the cost of the registration process would be high, and the cost was not warranted when assessing whether it was likely to aid settlement: at [81]; and

(e)    little weight was to be placed on the respondents’ view that, in the absence of registration, the matter was unlikely to be capable of settlement; such an assertion must be viewed against the respondents’ self-interest in minimising group member participation: at [82], citing Wisbey at [55] (Beach J).

(6)    Clarke concerned claims in respect of extended warranties sold by JB Hi-Fi to a class estimated at approximately eight million people over two claim periods. JB Hi-Fi sought orders for registration concurrently with opt out and for closure of the class until the commencement of the trial, pursuant to ss 33ZF and 33ZG of the Victorian Act. The plaintiff opposed these orders, and Delany J declined to make the orders sought, on the following grounds:

(a)    there was no satisfactory proposal for how group members would be notified, given contact details were only held for less than 40% of potential group members: at [68]–[79];

(b)    there was a prospect of conflict between group members if an order was made for soft class closure and there was a later opportunity for group members to register in advance of a settlement approval hearing (ie once there was “money on the table”), in addition to the settlement being diluted through late registrations: at [92]ff; and

(c)    the plaintiff’s opposition to the proposed orders and the cost of the registration process were also factors going against making the orders sought: at [103].

(7)    Petrucci concerned two proceedings against Toyota Finance Australia relating to flex commissions and add-on insurance in connection with the sale of motor vehicles. The classes comprised approximately 519,900 and 163,890 primary borrowers (approximately 130,000 of whom were members of both proceedings), and contact information was held for 99 per cent of group members. Justice M Osborne declined to make the orders sought by the defendants (opt out, registration and soft class closure pursuant to ss 33ZF and 33ZG of the Victorian Act), on the following grounds:

(a)    the proceeding was at an early stage, and registration at that stage may well result in an artificially low participation rate, which would make it problematic for the plaintiffs’ solicitors to recommend any settlement: at [31]–[32];

(b)    real caution is required when an application is opposed by the plaintiff: at [32]–‍[33];

(c)    the costs of the registration process were high and, without a settlement offer on the table, it could not be assumed that the costs were proportional: at [24], [35];

(d)    a registration process might not ultimately be required as the defendants likely had bank account details for many group members: at [37];

(e)    uncertainty about quantum is not uncommon in class actions, and there was much that was known, including the size of each class, the dates of the loans and the prima facie loss, which sophisticated practitioners should be able to use to make reasonable estimates: at [38];

(f)    while not impugning the defendants’ motives, the court should be alive to the advantage to the defendants in minimising group member participation: at [39]–[40]; and

(g)    a meaningful mediation could occur without registration: at [41].

(8)    Green concerned an application by the respondent for an order under ss 33X(1)(a), 33X(5) and 33ZF(1) of the FCA Act that a notice be sent to group members foreshadowing an intention to seek an order that only registered group members can participate in any settlement. The group in Green comprised persons who purchased certain cold and flu products manufactured by the respondent, which were said to be ineffective. The matters to which Moshinsky J had regard in determining to make the orders sought included the following:

(a)    the class was large — at least eight million people — who would have very small claims (given the price of the cold and flu products): at [62(a)–(b)];

(b)    a registration process would be required at some point (assuming settlement or judgment), and the cost was not an impediment: at [62(c)], [71]–[72];

(c)    the need to provide some kind of verification of their purchases was likely to be difficult for many group members, and there was “great uncertainty as to the number of Group Members who would seek to participate in any settlement”: at [62(d)], [63] (emphasis in original);

(d)    while the applicant was confident that group member losses could be meaningfully estimated through discovery and expert evidence, that did not address the “critical issue, namely the participation rate”, and an assertion that informed assumptions could be made about that matter was not supported by any analysis of how participation rates could be estimated: at [64] (emphasis in original);

(e)    without information about the likely participation rate, it would be very difficult to settle the proceeding, whereas a registration process would provide information as to the likely participation rate and therefore significantly enhance the prospect of settlement of the proceeding, as the parties will be able to “put a realistic value on the size of the claim”: at [65]–[66];

(f)    a registration process would not preclude the Court allowing some late registrants to participate, and the possibility of late registrants coming forward, even in large numbers, could be accommodated by the parties being able to make appropriate allowances such that the registration process would still provide a good indication of participation: at [67];

(g)    the “money on the table” point was not persuasive because the size of the claims of group members was likely to be so small that, even following a settlement, they may not be sufficiently incentivised to register: at [68];

(h)    the Court is not bound to make the orders foreshadowed in the notice, which is “a significant measure of protection for group members”: at [73]; and

(i)    any settlement reached without information about the likely participation rate could be unfair to the interests of group members as it could presuppose a lower rate of participation, only for the actual sums to be received by each group member diminishing following actual registration: at [75].

(9)    Merrett was a class action brought on behalf of current and former employees of the Queensland Police Service arising from COVID-19 vaccination directions. The class was confined, with a known upper bound of 18,987 employees, divided into three sub-groups (the “Vaccine Refusal”, “Vaccine Coercion” and “Refused Medical Exemption” groups): at [50], [52]. In ordering a pre-mediation registration process over the opposition of the plaintiffs (who contended that it was too early in the proceeding), Williams J had regard to the following:

(a)    the most important consideration was the extent to which registration was likely to improve the prospects of a reasonable settlement: at [33], citing Wisbey at [70] (Beach J);

(b)    there was no reliable information available to the parties about the size of the Vaccine Coercion group, which, absent registration, was “at best a ‘guess-‍timate’” — the parties did not know whether the group members in that sub-group were in the “tens, the hundreds, the thousands or towards the ‘upper bound’ of 18,987”: at [75]–[77];

(c)    a registration process that would enable the parties to know how many group members would participate was likely to increase the prospects of achieving a reasonable settlement: at [79]–[80];

(d)    adequate notice could be given to group members via email and post (for existing and former employees, respectively), combined with newspaper advertisements as a safeguard for any out-of-date mail addresses: at [81]–[82], [86]; and

(e)    the factor of the attitude of the parties was neutral, with one favouring it and one opposing; opposition by a party (here, the plaintiffs) was not determinative: at [86].

47    From this survey, several points of commonality emerge.

48    The ability of the parties to make a reasonable estimate of the number and value of the claims based on the respondent’s records is a key consideration. Where the available records were regarded as adequate for that purpose, pre-mediation registration orders have been refused: eg Alford at [17] (Murphy J); Preece at [71]–[79] (Anderson J); Petrucci at [38] (M Osborne J). Conversely, where the size of the group, or the number likely to participate, could not be reasonably estimated from available records, that has supported the making of orders effecting a pre-mediation registration process: eg Fox at [101]ff (Nichols J); Anderson-Vaughan at [83] (Delany J); Wisbey at [117]ff (Beach J); Green at [63]ff (Moshinsky J); and Merrett at [77]ff (Williams J). In Clarke, Delany J accepted that a registration process would provide certainty for participation rates and thus assist with mediation, but nevertheless refused to approve a pre-mediation registration process on the basis that no adequate notice could be given to group members, such that it would not provide the parties with a reasonable estimate of their actual maximum exposure for the purposes of mediation: at [84], [98].

49    As may be seen from the survey of the cases above, most of those cases involved circumstances in which the group size could be reasonably estimated; it was the need for the respondent to be able to know something about likely participation rates that drove the approval of a pre-mediation registration process. REST is not even at that starting point. In this proceeding, the issue is much more far-reaching than not being able to assess participation rates against a known group member cohort.

50    In many of the cases referred to above, the ability of a respondent to assess its exposure having regard to participation rates was seen as contributing to the prospect of a rational settlement being achieved. The position is even more pronounced when not even the group size is known, and cannot be assessed by the respondent. In this respect, the uncertainty concerning quantum in the present case goes well beyond the usual levels of uncertainty in class actions, which has been noted in some of the cases in which pre-mediation registration processes have been declined: eg Preece at [79] (Anderson J).

51    Whether registration will be required at some point in any event has also been significant. The inevitability of a registration process at some point supported the orders made in Fox, Anderson-Vaughan, Wisbey and Green. The prospect that no registration process might ultimately be required told against the orders in Alford where (at that time) Murphy J considered that most group members could be paid through the trustee’s records or the Australian Taxation Office (ATO) trustee voluntary payment mechanism (at [51]ff), and in Petrucci, where direct payment to group members’ bank accounts was a possibility (at [37] (M Osborne J)).

52    The cost of a registration process is significant, but the cost is of less concern if a registration process will be required at some point in any event (eg Wisbey at [181]–[182], [185] (Beach J); Green at [72] (Moshinsky J)).

53    While uncertainty regarding quantum is a common feature of class actions and is ordinarily manageable (eg Preece at [79] (Anderson J); Petrucci at [38] (M Osborne J)), uncertainty about participation rates, particularly in very large classes, can present a real impediment to a reasonable settlement being reached (eg Fox at [101]ff (Nichols J); Anderson-Vaughan at [66] (Delany J); Green at [63]ff (Moshinsky J)).

54    The capacity to give effective notice to group members of the registration process is relevant (eg Fox at [69]–[70] (Nichols J); Anderson-Vaughan at [76] (Delany J); Merrett at [81]–[82] (Williams J), where the availability of contact details meant that personal notice could be given to substantially all group members, cf Clarke where contact details were not held for more than 60% of potential group members, and there was no developed proposal for distribution by other means: at [84] (Delany J)).

55    The attitude of the parties is relevant, and caution is required when an applicant opposes the orders sought. However, an applicant’s opposition to a registration process is not determinative and there is no general rule that such orders are made only by consent: Fox at [28]–[30] (Nichols J); Wisbey at [90]–[91] (Beach J); Alford at [69] (Murphy J); Green at [60] (Moshinsky J); Merrett at [86] (Williams J). Conversely, a respondent’s assertion that the proceeding cannot settle without registration carries little weight, and the court should be alive to the self-interest of respondents where they may seek orders that advantage them through minimising group participation: Wisbey at [55] (Beach J); Preece at [82] (Anderson J); Petrucci at [39]–[41] (M Osborne J).

56    The Applicant submitted in his written submissions that Green was wrongly decided and should not be followed insofar as Moshinsky J did not afford “substantially greater weight” to the applicant’s views as to what was in group members’ interests. I note that, after the hearing of REST’s application, an application for leave to appeal in Green was filed by the applicant in that proceeding.

57    The additional benefit of a registration process enabling the parties to ensure that legal costs were commensurate with the claim size — a matter raised by Beach J in Wisbey (at [165]ff) — was also noted by Moshinsky J in Green at [51], [76], and by Williams J in Merrett at [77].

Why a registration process should be ordered in this case

58    In many of the cases referred to above in which a pre-mediation registration process was ordered, the ability of a respondent to assess its exposure having regard to participation rates was seen as contributing to the prospect of a rational settlement being achieved. Many of those cases were proceedings in which the group size could be reasonably estimated, and it was the need for the respondent to be able to know something about likely participation rates that drove the approval of a pre-mediation registration process.

59    In the present case, REST is not even at that starting point of being able to assess the size of the group. In this respect, the uncertainty concerning quantum in the present case goes well beyond the usual levels of uncertainty in class actions, which has been noted in some of the cases in which pre-mediation registration processes have been declined.

60    The critical issue in the present case is that REST says it does not know, and cannot from its own records find out (or estimate), how many Group Members there are. In this respect, the case is an unusual one.

61    The Applicant disputes that, without a registration process, REST would be going into a mediation ignorant as to the size of the group and the size of the group’s financial claims. While pointing to some limitations in REST’s evidence on what its records reveal, the Applicant did not seriously contend that the size of the group or the scale of the financial claims of the group could be assessed, or estimated, by REST based on its records. In this respect, the present case is very different from Alford — on which the Applicant relied heavily — in which Murphy J considered that the respondents in that case could estimate their exposure using their own records.

62    The Applicant contended that no pre-mediation registration process should be undertaken for reasons which included a contention that a reasonably reliable estimate of group size and financial exposure could be undertaken with the assistance of experts. Mr Allsopp’s first affidavit was sworn on 8 October 2026. In it, he said:

[39]    On 27 August 2026, and for the purposes of this proceeding, the Applicant engaged Mr Nicholas Florentzos, an actuary and consultant. Mr Florentzos has considered information currently available to Shine Lawyers, information that is publicly available and information which REST says that it holds, and has provided a view that a realistic estimate of the number of group members may be achieved. …

[42]    The Applicant has engaged Ms Dawna Wright of FTI, a forensic accountant, for the purpose of estimating quantum. Ms Wright has advised that she can undertake a realistic estimate of quantum on the material available, including the estimate of group members to be undertaken by Mr Florentzos. …

Mr Allsopp exhibited an email to Shine from Mr Florentzos of 7 October 2026, and an email from Ms Wright of 8 October 2026, to the affidavit he swore that day.

63    In his email, Mr Florentzos said:

I refer to your letter to me dated 7 October 2026.

1.     Question 1 of your letter asked for my opinion as to whether it is possible with a reasonable degree of accuracy, to estimate the numbers of each of, during the Relevant Period:

i)     Dual-Fund Inactive Members;

ii)     Non-Working Inactive Members; and

iii)     Non-Benefiting Inactive Members.

2.     In my opinion, it is possible with a reasonable degree of accuracy to determine an estimate of the Dual-Fund Inactive Members, the Non-Working Inactive Members and the Non-Benefiting Inactive Members during the Relevant Period.

3.     In my opinion, this estimate can be obtained by:

(a)     Utilising the materials, data and documentation provided in your letter; and

(b)     Obtaining information available in the public domain. Recent and historical information can be obtained from sources including: …

64    In that email, Mr Florentzos went on to list 13 publicly available sources of information (including the Australian Prudential Regulation Authority, the ATO, the Australian Securities and Investments Commission, the Productivity Commission, the Fair Work Ombudsman, as well as the product disclosure statements and annual reports of various superannuation funds). Mr Florentzos said that the estimate could be prepared by “[s]electing the relevant items from (a) and (b) to then create a methodology, calculate and make the determination”.

65    It is notable that Mr Florentzos did not explain, nor does he appear to have identified, what information is contained in the various sources that would enable him to estimate the number of members in the categories he identified. Nor did he explain, even very broadly, what methodology he might apply. Mr Florentzos was asked by Shine to identify the methodology he would use and the information he would use in the second question posed by Shine, but Mr Florentzos said in his email that that question would be answered “in due course”.

66    By pointing to Mr Florentzos’s failure to identify the information he would use and what his methodology would involve, I intend no criticism of Mr Florentzos. He was only asked by Shine the same day (7 October 2026) to give his opinion. Clearly his response was prepared under time pressure. The letter Shine sent to Mr Florentzos was only put into evidence by the Applicant after REST objected to the admissibility of the emails from Mr Florentzos and Ms Wright and to the parts of Mr Allsopp’s first affidavit deposing to their respective views.

67    Nevertheless, the point remains that beyond a statement of confidence by an expert that he could, by some means and using unidentified data points, provide an estimate of the three categories of members, there is no evidence refuting REST’s position that it cannot reasonably estimate the number of Group Members. Given the complete lack of any substantiating detail or information in Mr Florentzos’s email, I give it little weight and accept REST’s position.

68    Like Mr Florentzos, Ms Wright was only asked to give her opinion the day before she responded (7 October 2026). However, her email was less controversial as she said she could — using figures as to the total size of each cohort as a base assumption (presumably to be provided by Mr Florentzos), and using other data and information sources — prepare an estimate of the total quantum of loss suffered by Non-‍Benefiting Inactive Members with a reasonable degree of accuracy.

69    The next matter of significance raised by the Applicant in opposing the orders sought was a contention that registration rates in any pre-mediation registration process would be low and not representative of the true extent of Group Members who would wish to participate in a settlement. That conclusion is based on several contentions:

(a)    members of superannuation and financial services class actions will generally have less “inherent motivation” to engage than members of some other types of class actions, such as personal injury, consumer and shareholder class actions, and the problem of disengagement in superannuation is pronounced for “millennials”, who would constitute a portion of REST’s membership;

(b)    group members are more motivated to register when a settlement has been achieved, and in response to settlement distribution notices;

(c)    the extent to which insurance group members are likely to engage with class action proceedings “further increases significantly [in addition to the increase after settlement] when settlement notices are accompanied by an extensive media campaign”; and

(d)    Shine’s data from the Jessica Amy Challenor v QSuper Board (VID 691 of 2021) (QSuper) and Gregory John Lenthall & Anor v Westpac Banking Corporation & Anor (NSD 1812 of 2017) (Westpac) class action registration processes shows that engagement by superannuation and insurance group members significantly increases after settlement and confirms Mr Allsopp’s asserted “experience that the number of group members who might register prior to settlement is unlikely to reflect (and, in fact, is likely to substantially understate) the number of group members who ultimately register”.

70    In large part, these contentions are stated by Mr Allsopp in his first affidavit based on his “opinion and experience”. That incantation, or variants thereof, occurs almost 50 times in Mr Allsopp’s 86-paragraph affidavit. While the experiences and views of a practitioner who has worked in class actions for many years are relevant, they are more useful when the opinion or fact being stated is tied to the practitioner’s experience. Mr Allsopp’s first affidavit is replete with such assertions of fact and statements of opinion, but they are not grounded in his actual experience in any way that is exposed to the reader of his affidavit. By that I mean that Mr Allsopp’s affidavit does not descend to explain what it is about his experience that has led him to the views he has; rather, the bare “in my opinion and experience” incantation (and variants thereof) is repeated.

71    Nevertheless, the point in paragraph 69(a) above may be accepted as according with common sense, at least so far as the comparison with personal injury cases is concerned; the proposition that group members in consumer class actions have more “inherent motivation” to engage is more debateable, and likely highly variable depending on the nature of the product or service in question (eg the purchase of a car that turns out to be defective, versus the purchase of a packet of cold and flu medication).

72    I also accept that potential group members who receive a notice advising of a settlement having been achieved, subject to court approval, will have a higher degree of motivation to register than potential group members who merely receive an invitation to do so, with no reference to any potential adverse consequences. However, the extent to which there will be a gap between registration rates when a settlement has been achieved, as compared to registration rates when a clear adverse consequence has been identified — relevantly, being exposed to the possibility of not being able to participate in any potential future settlement, if the orders that are foreshadowed by the notice are made — is speculative. Contrary to the Applicant’s position, I do not consider that the Westpac or QSuper cases support the contention that the gap has been established, or should be expected to be material.

73    Before turning to the QSuper and Westpac data, I note that the point referred to in paragraph 69(c) above does not weigh against a pre-mediation registration process in this case because Shine’s costings for a pre-mediation registration process anticipate that there would be a significant social media, digital advertising and radio advertising campaign.

74    As recounted by Mr Allsopp:

(1)    In QSuper, registration was not ordered until after a settlement had been reached. Only 1,000 registrations were received between April and May 2023, following the issue of opt-out notices, whereas more than 90,000 registrations were received between December 2025 (when settlement notices started being issued) and February 2026.

(2)    In Westpac, the opt-out notice encouraged, but did not require, registration. The opt-out notice in Westpac also did not identify any adverse consequence to group members of not registering, save that it stated that the funder may not continue to fund the proceeding if a sufficient number of registrations was not received. The data presented by Mr Allsopp showed that about 1,700 registrations were received following the issue of the opt-out notice, about 10,000 registrations were received after settlement notices were issued, and a further 15,000 registrations were received after another round of settlement notices were issued, this time accompanied by an extensive media campaign.

75    I do not consider the data concerning the registration patterns in QSuper or Westpac to be of much relevance. What occurred in those cases does not support the weight that the Applicant seeks to place on them; still less does what occurred provide support for Mr Allsopp’s opinion that a pre-settlement registration process will generate low levels of registration, which low levels are not representative of the proportion of Group Members who would ultimately want to participate.

76    In QSuper, no call to action for group members to register was issued before the settlement notices went out. The contrast between registrations before, and after, the settlement notices were issued in QSuper reveals nothing about what could be expected in a pre-mediation registration process which is coupled with a notice stating the effect of the orders that will be sought on settlement.

77    What occurred in Westpac also does not provide any guide as to what could be expected to occur if a registration process is conducted in the present case. As noted, the low level of registrations following the issue of the opt-out notice occurred in a context in which the notice did go beyond encouraging voluntary registration, by noting that the funding of the proceeding may be in jeopardy. What the data in Westpac do suggest, however, is that publicity and advertising may be important in getting group members to engage; in that case, the fact of there being “money on the table” when the first round of settlement notices went out was not enough to draw out the full extent of group members ultimately wishing to participate.

78    In supplementary submissions, the Applicant identified additional data about the take-up of registration processes in some other cases. In the main, the data were of limited assistance because (variously) the initial registration process was not undertaken pursuant to a process which included a notice warning of the potential consequences of a failure to register, or the initial registration scheme only covered a limited period, which had expired.

79    I accept that a pre-mediation registration process, even one accompanied by a media campaign, cannot be expected to flush out every Group Member who may ultimately want to participate in a possible settlement. But that does not mean that the dire consequences Mr Allsopp describes in his affidavit must follow. In particular, with a baseline of Group Member participation established through a pre-mediation registration process, the parties can approach a mediation making allowances for the prospect that some further Group Members may seek to participate and orders proposed for a settlement approval may need to include, as they often do, provision for at least some late registrants.

80    The fact that such allowances may not be scientific or mathematically precise may be accepted, but that does not diminish the utility of having a clear baseline of participation established by a pre-mediation registration process. That is particularly so when one recalls that the alternative is for the parties to go to mediation without even knowing how many Group Members there are. Establishing a firm baseline of participating Group Members will allow the parties to approach the mediation on an informed basis, which I am satisfied will enhance the prospects of the parties reaching a rational and supportable settlement. It should be recalled that mediations are supposed to involve “rational bargaining between relatively well-informed parties aimed at providing just compensation for worthy claims”; they are not supposed to be “conducted in ignorance”, or to constitute a “game of bluff and bluster”: Regent Holdings at [23] (Nettle, Redlich and Osborn JJA).

81    Another factor of concern is the cost of conducting a pre-mediation registration process. Shine prepared estimates that included the cost of responding to queries as well as the cost of an advertising blitz on social media and radio, as well as digital advertising. The costs ranged from $547,860 to $1,146,598, depending on the rate of contact from Group Members (or potential Group Members). While it was not entirely clear from Mr Allsopp’s affidavit, it would appear that these estimates do not include the raw costs of distributing the registration notices to Group Members. Mr Allsopp deposed elsewhere in his affidavit to his belief that — based on his experience and Shine’s review of several quotations from mailing houses — the cost of distributing opt-out notices to Group Members was at least around $670,000, but that is a cost that will need to be incurred even if the application is refused, and only a bare opt-out notice is distributed.

82    I accept that conducting a pre-mediation registration process will be expensive. However, the Applicant accepts in this case that a registration process will need to be conducted at some point, if the matter settles or succeeds at trial. In this respect, the case is unlike Alford and Petrucci, where pre-mediation registration regimes were rejected in cases where the court was not satisfied that a registration process would be necessary at a later point in any event. Rather, the present case is more like Wisbey and Green, where the fact that a registration process would be required at some point meant that the costs of the process were to be considered in that light.

83    I do not consider that the pre-mediation registration process proposed by REST should be approached on the basis of any assumption that there will necessarily need to be a further registration process after any successful settlement. For example, the registration and related notice proposed by REST follows the same structure as that in Wisbey. The information in the proposed notice regarding the consequences of the “do nothing” option is in substantially the same terms as the notice in Wisbey. As REST noted in its supplementary submissions, there was no indication in Wisbey when the orders for registration and opt out were made that there would be a further opportunity to register after the relevant deadline and, after Wisbey settled, the notice distributed to group members did not afford an additional opportunity to register, but noted that it would be a matter for the Court whether to allow group members who did not register by the deadline to be treated as registered group members for the purpose of participating in the settlement.

84    I also do not consider that a notice combining opt out and registration will necessarily be confusing to Group Members, that being another concern of Mr Allsopp’s. Care is always required to ensure that notices to group members are comprehensible. I see no inherent difficulty in a notice explaining to Group Members that, if they do not wish to opt out, they can register to participate in any future settlement, as well as identifying the potential consequences if they do not register.

85    It should be understood that ordering that there be a pre-mediation registration process, with a notice of the kind REST proposes, does not mean that Group Members who do not register are not Group Members; on the contrary, they would still be Group Members.

86    Whether Group Members who did not register are to be precluded from participating in the settlement, while still being bound by the settlement, is a matter that will need to be determined if a settlement is reached and the settlement is presented for approval by the Court on that basis. A pre-mediation registration process does not determine whether an order of that kind will be made when a settlement approval application is presented. It should also be borne in mind that settlement approval applications are often presented seeking an order that permits some late registrants to participate, or even orders granting the settlement administrator discretion to allow late registrants to participate if they satisfy the administrator that there is a good reason why they did not register previously.

87    Whether the extent of such late registrations causes a concern regarding whether the settlement sum is reasonable would depend on the number of late registrations and on factors likely to be covered in a supporting opinion of counsel. Those factors include the thinking of an applicant’s legal team regarding the reasonableness of the settlement sum offered at mediation in light of the baseline of registered group members and the legal team’s assessment of the likelihood and extent of any cohort seeking to register late. For the reasons already set out, I do not consider that it should be assumed that the number of unregistered group members seeking to participate in a settlement will necessarily be so large as to dilute a settlement sum to the extent that it cannot be supported as fair and reasonable, or that reasonable allowances cannot be made in mediation for this issue.

88    I consider that the prospects of the proceeding being settled at mediation will be improved if the parties are able to approach the mediation with the benefit of the information about REST’s potential exposure that will be generated from a pre-mediation registration process. The authorities recognise that the parties having a realistic idea of a respondent’s exposure enhances the prospects of settlement: eg Anderson-Vaughan at [84] (Delany J); Wisbey at [127] (Beach J); Merrett at [79]–[80] (Williams J).

89    I consider it in the interests of Group Members as a whole that steps are taken that maximise the prospects of the proceeding being brought to a successful conclusion by settlement at mediation, as compared with proceeding with a time-consuming and expensive trial. In that regard, it is also relevant to note that this is not a case in which an initial trial could be expected to deliver a platform for resolution of the claims of the group as a whole. Without a registration process, an initial trial will not shed any light on the extent of group membership, or the financial claims of Group Members. This is acknowledged by the Applicant, who states in his pleading that he presently does not know particulars of Group Members or their loss or damage, and such particulars will be provided “following the initial trial at such time as it is necessary to plead and particularise Group Member claims”: SoC [37]–[38].

90    This is an appropriate point in the proceeding for a pre-mediation registration process to take place. Dates for mediation and trial have been set, and the pre-mediation registration process can be accommodated without disrupting the progress of the proceeding towards trial. REST’s proposal allows ample time for recipients of the notice to read and consider it, and to complete the registration process.

91    One matter that the parties did not address in their submissions, however, was the appropriateness of REST’s proposal that the registration process specified in its proposed notice apply to any settlement reached prior to the commencement of the initial trial of the proceeding (or such other time as agreed by the parties and/or ordered by the Court). In this regard, I note the observations of Nichols J in Nelson v Beach Energy Ltd [2025] VSC 339 (Beach Energy) at [20], considering the period prior to the commencement of the initial trial that would provide an incentive to the parties to resolve the proceeding before incurring the significant costs of final trial preparation.

92    As a final point, while I note that cynicism regarding the respondents’ motives was expressed in Alford and Petrucci, in any particular case, the merits of a pre-mediation registration process proposal should be considered on the facts of the case. As I have noted, REST in the present case is in a situation where even the group size is unknown, and I do not ascribe its application for a pre-mediation registration process to the cynical motives noted in some other cases. Further, and in any event, the observations of Gordon and Steward JJ in Lendlease are apposite and dispel the assumption that steps in the interests of a respondent are necessarily inimical to the interests of group members. Their Honours said (at [99], emphasis added):

The giving of such a notice, which is merely informative of Lendlease’s objectives, itself creates no detriment, and perhaps only enhances the possibility of settlement. In that respect, the observation of the Full Federal Court in Parkin cannot be ignored: court-approved settlements of securities class actions are “by an overwhelming margin” the most common way of resolving group members’ claims. Resolution of claims by settlement is fundamental to both Pt 10 of the CP Act and Pt IVA of the FCA Act. It is of no moment if the giving of the notice also promotes the commercial interests of Lendlease. If settlement is achieved, it would promote the interests of both Lendlease and group members.

The relevant settlement window to be specified in the notice and orders

93    REST’s application proposes that the Court make an order approving the giving of a notice that states that, upon any settlement being reached, an order will be sought in terms that would (subject to the grant of leave) exclude persons who have not registered by the registration date from participating in a settlement agreed “before the first date of trial listed in accordance with order 1 of the Orders made in this proceeding on 1 June 2026, or such other time as agreed by the parties and/or ordered by the Court”.

94    The mediation is due to commence by 18 June 2027, but may — as is not uncommon in class actions and other substantial proceedings — occur over a longer period; a mediation in such a case is not necessarily confined to one or more specified days. The order foreshadowed by REST’s application would apply to any settlement reached before the first day of trial. This presents a substantial settlement window in respect of which the foreshadowed order might be sought.

95    The Applicant did not address submissions to the discrete issue of the settlement window (presumably because he opposed the entire application).

96    I have two concerns with the order proposed by REST. First, potential Group Members should be provided with information that is as firm as possible; the potential for a “moving feast” should be avoided to the extent possible. The orders should not fix on a trial date specified in a particular order, as there is the potential for as-yet-unforeseen circumstances to result in that date changing. Secondly, and more substantively, it is likely to be preferable to specify a period of time that allows for the reality that a mediation may take place over some days, or even weeks, but not specifying a period that runs right up to the day before the trial commences. The observations of Nichols J in Beach Energy at [20] are apposite (albeit in a class closure context).

97    Subject to any further submissions the parties may make, my present inclination is that it would be preferable for the notice, and the corresponding order, either to refer to a period that ends when the mediator formally terminates the mediation, or ends one month prior to the commencement of the trial. The latter option would allow for some period of negotiation outside the formal mediation process, but would still close the settlement window sufficiently before trial so that the incentive remains to settle before heavy trial preparation costs are incurred.

Conclusion

98    The Applicant indicated that, if the substantive application were determined adversely to his position, he would want the opportunity to consider the specific terms of the notice REST proposes be issued, as well as the form of its proposed distribution orders. As I have determined that the pre-mediation registration process should be conducted, orders will be made directing the parties to confer on the terms of orders to give effect to these reasons, including the terms of the notice to be issued to potential Group Members and the precise method of distribution.

99    The parties should also confer regarding the settlement window to be referred to in the notice and corresponding order and, if necessary, make further submissions on this matter.

I certify that the preceding ninety-nine (99) numbered paragraphs are a true copy of the Reasons for Judgment of the Honourable Justice Button.

Associate:

Dated:    17 November 2026