top of page
  • Writer's pictureDaniel Meyerowitz-Katz

Open vs closed class actions: is the ALRC throwing the baby out with the bathwater?


The Australian Law Reform Commission's Report 134, entitled "Integrity, Fairness and Efficiency—An Inquiry into Class Action Proceedings and Third-Party Litigation Funders", has finally been released publicly. As I have noted before, I was involved in authoring two submissions to the inquiry into the litigation funding and class action systems (on behalf of Levitt Robinson and NSW Young Lawyers), and I am happy to say that both submissions were extensively cited in the final Report.

I have many thoughts on the Report, some positive and some negative. Time permitting, I will post them all on this blog in due course. For this post in particular, however, I am focussing on a discrete issue which I find quite concerning: that is, the ALRC's proposed amendment to s 33C(1) of the Federal Court of Australia Act 1976 (Cth) ("FCA Act"), which the Commission says was designed to ensure that all class actions are commenced as an "open class".

The Proposed Amendment

The recommended amendment is that the words “some or” should be removed from the phrase "some or all of them" in s 33C(1) of the FCA Act. Viewed in full context, the provision then becomes:

“Subject to this Part, where:

(a) 7 or more persons have claims against the same person; and

(b) the claims of all those persons are in respect of, or arise out of, the same, similar or related circumstances; and

(c) the claims of all those persons give rise to a substantial common issue of law or fact;

a proceeding may be commenced by one or more of those persons as representing all of them.”

The idea of requiring all class actions to be open class actions was proposed in the ALRC's original discussion paper in this inquiry. Professor Vince Morabito then published a very informative paper in response, which set out the different ways that "closed classes" have been used under Part IVA of the FCA Act. As Professor Morabito endeavoured to show, distinguishing between an "open class" action and a "closed class" action is not straightforward, and there are many legitimate ways in which the definition of a class can be restricted.

The ALRC addressed this article at [4.17]-[4.18]:

"In a 2018 article on open and closed classes, Professor Morabito noted that his data suggests ‘that determining whether all alleged victims of the impugned conduct have been included in the class action litigation is not as easy as one may think.’17 There have been two principal ways in which individuals who may have suffered loss have been excluded from participating in a class action:

  • through opt in devices such as a requirement to sign a funding agreement with a litigation funder or sign a solicitor’s costs agreement with the lead plaintiff’s solicitors, and

  • through drafting the statement of claim in a manner that defines the class narrowly (examples include: listing all the claimants individually, bringing an action on behalf of members of a particular association or trade union, or limiting the actions to claimants with a minimum thresold loss or minimum purchase volume of a product).

The recommendation that all class actions be open class and opt out is primarily directed at ensuring that the class action regime does not require potential group members to sign up with a lawyer or funder in order to participate. To the extent that a third-party litigation funder or law firm sought that outcome by defining the group narrowly, the ALRC considers that the Court has the necessary discretion to distinguish between a statement of claim that reasonably defines the class and one where the description of the class is crafted to require, as a practical matter, the signing up with a lawyer or funder to participate."

It seems to follow that the ALRC is not opposed to the use of the "closed" versus "open" class devices in the manner identified by Professor Morabito—that is, it should not be an issue that classes are sometimes defined narrowly by, for example, listing all of the class in a schedule (meaning that people with similar claims who are not in the schedule are excluded). The ALRC's goal in proposing the amendment is limited to preventing the class from being defined by reference to people having entered into costs agreements with particular solicitors or funding agreements with a particular funder.

However, as I read it, the proposed amendment goes well beyond the ALRC's stated goal. Further, as someone who has been involved in drafting several group definitions in representative actions, I am concerned that the amendment, as drafted, would be unworkable.

Effect of the Amendment

In order to fully appreciate the effect of the proposed amendment, it is necessary to understand the role of s 33C in the scheme of Part IVA. As stated by the Full Federal Court in Ethicon Sàrl v Gill [2018] FCAFC 137 at [7]:

‘Two “gateway” provisions contained in Part IVA of the Act were relevant to the commencement of this proceeding. The first was s 33C, which only allowed the applicants to commence if: (a) seven or more persons had claims against the respondents; (b) the claims of all of those persons were in respect of, or arose out of, the same, similar or related circumstances; and (c) the claims of all of those persons gave rise to a substantial common issue of law or fact. The second “gateway” provision, s 33H, required the applicants to specify common issues and identify or describe the group members in either “the application commencing the proceeding or in a document filed in support of such an application”. Section 33H operates to ensure that the Court can assess, at the outset, whether the specified pre-conditions for the commencement of a class action (contained in s 33C) were present.’

In other words, in order for a representative proceeding to be commenced, it must fulfil the requirements of s 33C. If it does not, the proceeding has not been validly commenced. Further, under s 33H, the originating process must identify the group members in a way that establishes that each of them is a person with a claim against the respondent who satisfies the criteria in s 33C. At [24] the Court in Ethicon Sàrl stated:

"when commencing the proceeding, it was necessary for the applicants to describe or otherwise identify the group members to whom the proceeding related. The group members were, as s 33A makes plain, “a member of a group of persons on whose behalf a representative proceeding has been commenced” (emphasis added). In identifying or describing those persons, it was not necessary to name them, nor to specify their number (see s 33H(2)), but it was necessary that the group membership be certain"

It follows that, in order to draft the group definition, the applicant’s lawyers must form a view as to which people could potentially comprise the group members, and draft a definition that identifies each of these people with sufficient particularity for the class action to have been competently commenced. As may be expected, different legal teams can and do form different views in relation to this issue. A recent example was the three competing class actions the subject of Impiombato v BHP Billiton Limited (No 2) [2018] FCA 2045. The Impiombato proceeding defined the group to include people who had purchased BHP shares during the period from 21 October 2013 until close of trade on 9 November 2015, the "Klemweb" proceeding had a period of 27 August 2014 to 9 November 2015, and the "LACERA" proceeding had a period of 8 August 2012 to opening of trade on 9 November 2015.

There are many reasons why a decision might be made to include or exclude a particular group of people. Generally, it will boil down to the strength of the claims based on the state of the applicants' knowledge at the time that the claim is filed. Other factors informing the decision may include the information available to the particular applicant, the applicants' resources, and its appetite for risk.

So in the three BHP proceedings, it is likely that the different date ranges were chosen based on the different knowledge that the lawyers had of what had occurred, their different case theories, and their different appetite for risk. The LACERA action, with the broader date range, was making a much more ambitious claim than the Klemweb action—which included many more group members, but which would have been substantially more difficult and costly to prove.

Under the proposed amendment, all claims would have to be commenced, no matter how difficult or remote. Bearing in mind that having a “claim” is not the same as being entitled to any relief: Ethicon Sàrl at [26], and considering the operation of s 33C as a "gateway" provision, this means that in order for the class action to be competently commenced, the applicant will have to include in the group all claims that: are against the same respondent; are in respect of, or arise out of, the same, similar or related circumstances; and give rise to a substantial common issue of law or fact—that is, irrespective of the relative merits of those claims and the potential difficulty or expense that prosecuting them would require. I emphasise again that if this is not done, the proceeding will not be validly constituted as a class action.

It is also important to note that, if the proceeding does not comply with s 33C when filed, "fixing it up later" may not be an option. This is because of the operation of s 33ZE of the FCA Act, which suspends the limitation period for group member claims when the class action is commenced. If the class action does not "pass through" the s 33C "gateway", then s 33ZE does not come into operation—meaning that the limitation period is not suspended for any of the group members. By the time the definition is amended to include the entire group as required by the new proposed s 33C, the limitation period may have expired.

The obvious way around this issue would be for plaintiff lawyers to start drafting the group definition as broadly and vaguely as possible, with the plan of whittling it down as the case progresses. There are obvious disadvantages to this practice, and it has substantial scope for increasing otherwise avoidable interlocutory applications about the validity of the group definition. Professor Morabito gives the vitamins cartel class action as an example where this seems to have occurred, and quotes Merkel J as saying at one interlocutory hearing:

"Every man, woman and child who has been in this country between 1992 and 1999; every person who is engaged in the food chain, the supply of every food and animal product; is a plaintiff, a group member. I can’t conceive of how that could ever be tried … there are some areas where there will be losses that are seriously claimed. By all means, they should be singled out, but not everyone in the food chain in this country … Is there any authority anywhere in the world that says that the class should be drawn so widely that [any person] who may have some kind of claim, no matter how small, no matter how nebulous and no matter how difficult to prove, ought to be included in the class?"

My reading of the ALRC's proposed amendment is that it would encourage precisely the kind of drafting that Merkel J was taking issue with in that excerpt.

If, as it seems from the Report, the ALRC’s actual goal is to prevent the group definition from being by reference to a funding agreement, then why not just introduce a sub-paragraph to that effect? Eg, “A person may not be excluded as a group member in a proceeding commenced under this Part only because that person has not agreed to particular funding terms or has not retained a particular solicitor.”

I personally do not agree that restricting class actions to closed classes only is a good idea in the first place, but if we accept the ALRC’s conclusion that it is a good idea, wouldn’t that be a much more effective way of achieving the ALRC’s goal?

426 views0 comments

Recent Posts

See All
bottom of page