Do Mass Tort Negligence Class Actions Adequately Compensate Victims and Effectively Deter Tortfeasor Wrongdoing?

Abstract

The objectives of the class action procedural device and the tort of negligence overlap in that they both seek to compensate victims for their loss and, according to some, they both seek to deter tortfeasor wrongdoing. However, in practice, does the class action procedure facilitate the fulfilment of these dual aims? This article explains how the class action aids compensation and deterrence, but also addresses the significant issues that hinder the class action’s effectiveness in meeting its compensatory and deterrence aims in an Australian and New Zealand context. It concludes by laying the foundation for further research to improve the effectiveness of class actions or by possibly adopting supplementary or alternative remedial/regulatory procedures.

1. Introduction

Civil procedure mechanisms provide a framework for hearing and determining substantive legal disputes. Class actions, as a civil procedural mechanism, allow for individual claims with common issues to be aggregated and heard simultaneously against a wrongdoer. In this respect, class members may hope that their substantive legal rights are vindicated, that they recover compensation for any loss suffered and that tortfeasors (and potential tortfeasors) are deterred from further wrongdoing. However, in practice, does this occur? Do mass tort negligence class actions adequately compensate victims and effectively deter tortfeasor wrongdoing? This article begins a conversation in answer to this question. In short, while class actions can compensate claimants and deter wrongdoing to some extent, significant issues hinder their effectiveness. Ultimately, this begs the question as to whether there is a more effective way to remedy mass negligence harms through amending class action procedure, by adopting a new procedure or by combining the class action with other mechanisms.

This article begins by discussing the objectives of the tort of negligence and the class action mechanism. We discuss these objectives in an Australian and New Zealand class action context. We then turn to assessing compensation and deterrence in the class action context with a focus on three case studies: the PFAS class action, the DePuy Hip Implant class action and the Strathboss Kiwifruit Growers class action. We conclude by summarising our views to encourage further research in this area.

2. The Objectives of the Tort of Negligence

The law of torts imposes a framework for shifting responsibility for loss from one person or entity to another. It originated by shifting loss between parties when a tortfeasor engaged in certain actions regardless of their fault or negligence (referred to as torts of strict liability). However, as the doctrine developed, the courts, in many instances, began to impose a condition that the tortfeasor be at fault, whether intentionally or negligently, before shifting responsibility to the tortfeasor for the victim’s resulting losses (Todd 2023, p. 16; Sappideen et al. 2024, pp. 9–11).

In the tort of negligence, the law shifts responsibility when a tortfeasor negligently breaches a duty of care owed to the victim which causes the victim foreseeable loss. The reasons why the law of tort imposes responsibility in this manner, particularly in the case of negligence, have been subject to rigorous academic and jurisprudential debate (Todd 2023, Footnote 95; Sappideen et al. 2024, pp. 6–7). Academics have disagreed on the objectives of tort law, and on the degree of their importance. For example, some claim that the law of torts is to rectify losses and apportion moral responsibility for wrongdoing,1 while others claim it is for “utilitarian efficiency” to “maximize [sic] wealth or other social welfare goals”2 (Zipursky 2002, p. 386; Wright 1995, p. 160).

Notwithstanding these differences, from a functionalist perspective, and for the purpose of this article, there are two tort law objectives that dominate and are often cited in tort law theory. The two objectives are as follows: (1) to compensate victims for their losses (“compensation aim”); and (2) to “influence conduct, promote safety and deter wrongful behaviour”3 (“deterrence aim”) (Abel 1989, p. 791). Weinrib categorises these dual aims of tort law as “multiple-goal instrumentalism” (Weinrib 1989, pp. 502–4). Indeed, the tort of negligence has been concerned with compensation and deterrence since its inception. In Donoghue v Stevenson [1932] AC 562 at 580, Lord Atkin alluded to these dual aims when he stated the following:

The liability for negligence, whether you style it such or treat it as in other systems as a species of ‘culpa’, is no doubt based upon a general public sentiment of moral wrongdoing for which the offender must pay.

In relation to the compensation aim, the measure of damages is to put the victim back into the position they would have been in had the harm not occurred (Atkin 2023, p. 1485). In this respect, compensatory damages are divided into two sub-categories: special damages and general damages. Special damages compensate for loss which can be objectively calculated in monetary terms (Atkin 2023, p. 1490). General damages compensate for intangible loss which cannot be objectively calculated in monetary terms (for example, loss for mental distress, hurt or humiliation) (Atkin 2023, p. 1490).

In relation to the deterrence aim, some claim that imposing liability on a wrongdoer for the harm they have caused will change the future conduct of both the wrongdoer (specific deterrence) and on other potential wrongdoers (general deterrence), or “that tort can deter people and organisations from causing harm” (Landes and Posner 1987; Roisman et al. 2004; Popper 2011). Tort liability can be a mechanism through private law to change or avert behaviour, rather than relying on criminal law sanctions or regulatory law (Rooij and Brownlee 2021). Deterrence in tort can be examined from an efficiency and law and economics approach, but also by reference to fairness and justice, including preventing harm (Schwartz 1997).

The dual aims present admirable goals. However, in practice, does the law of negligence in fact adequately compensate victims for their losses or deter tortfeasor wrongdoing? Furthermore, where a wrong creates mass loss for numerous victims, does the class action mechanism assist in furthering these alleged dual aims of negligence in providing adequate compensation and deterring wrongdoing? This article beings to address the latter question. In this respect, we now address the purposes of the class action mechanism and how they overlap with the alleged dual objectives of the tort of negligence, being compensation and deterrence.

3. Australian Class Actions

Australia adopted a class action regime with the introduction of Part IVA into the Federal Court of Australia Act 1976 (Cth) (FCAA) in 1992. The regime was subsequently adopted in a number of Australian States, including Victoria and New South Wales. Court rule-based regimes also exist but are used much less (e.g., Federal Court Rules (Cth) r 9.21).

At a general level, the Australian class action aggregates or combines the claims of numerous persons or entities and then allows a representative to litigate those claims on behalf of the group. These broad requirements are given effect through the legislation. A class action may be commenced by a representative party on behalf of group members where there are seven or more group members who have a claim against the same person, and the claims of the group members are connected through “same, similar or related circumstances” and the existence of “a substantial common issue of law or fact” (FCAA ss 33C, 33D). The commencement requirements present a low threshold for starting a class action and have been described as “deliberately undemanding” (Dyczynski v Gibson (2020) 280 FCR 583, [331]).

A group member’s consent is not required for their inclusion in the class action, but the group member must be given an opportunity to “opt out” or exclude themselves from the class action (FCAA ss 33E, 33J, 33X(1)(a)). The opt out approach is important as it favours inclusion of claims without the need for a person to take action, or indeed, even be aware that a class action including their claim has commenced. However, the courts have permitted class action group definitions to be constructed so that only group members who have retained a specific lawyer or contracted with a litigation funder are part of a class action (Multiplex Funds Management Ltd. v P Dawson Nominees Pty Ltd. (2007) 164 FCR 275). If a group member falling within the defined class does not opt out then they are bound by the outcome of the proceedings (FCAA s 33ZB). The court must approve class action settlements (FCAA s 33V).

Australia’s class action regime has multiple objectives: access to justice, efficient resolution of disputes, avoidance of multiple suits and the risk of inconsistent findings, reduction of costs for the parties and the courts, and the deterrence of misconduct (Timbercorp Finance Pty Ltd. (in liq) v Collins (2016) 259 CLR 212, [43]; BMW Australia Ltd. v Brewster (2019) 269 CLR 574, [82], [97]; Legg 2022, pp. 34–36). Of central concern here is compensation and deterrence.

An important and related issue is the financing of class actions. When class actions were introduced into Australia, the legislature did not address how class actions would be funded or how costs would be addressed.

There are two main categories of costs in Australian litigation: the costs incurred by the representative party in bringing the proceedings (lawyer’s fees and disbursements) and the defendant’s costs. The former can be dealt with by the representative party paying those costs, through a conditional or no-win no-fee arrangement, or through litigation funding. Some of the costs of bringing the class action can usually be recovered from an opponent if the class action is successful. If the class action is unsuccessful, then the representative party (but not group members) is liable for the opponent’s costs (FCAA s 43(1A)). Litigation funders typically pay the lawyer who is acting for the class action and assumes liability for the defendant’s costs if the class action fails, but in return receive a percentage or share of the class action recovery (Legg 2021, pp. 222–23; Chen and Legg 2022, p. 955). The State of Victoria added to these options, with the adoption of group costs, orders that allow lawyers to recover a percentage of any recovery but require that the lawyer be liable for the defendant’s costs if the class action is unsuccessful (Legg 2021, pp. 231–34). The commission amount paid to funders can be adjusted by the court at the settlement approval stage if it considers the commission is disproportionate to overall class member recovery (BMW Australia Ltd. v Brewster 2019, 269 CLR 574, [3] and [232]; Waye et al. 2025 p. 140).

4. New Zealand Class Actions

There are no specific statutory-based civil procedure rules which govern the use of class actions in New Zealand. Instead, parties can aggregate litigation through various mechanisms with the most common being under the “representative rule” pursuant to High Court Rule 4.24 (HCR 4.24) (Chamberlain and Watson 2021, p. 307).

HCR 4.24 provides that “[o]ne or more persons may sue or be sued on behalf of, or for the benefit of, all persons with the same interest in the subject matter of a proceeding” by either the consent of all class members or by order of the court on application of a party. In practice, the courts now allow HCR 4.24 to facilitate class-wide actions. However, outside of HCR 4.24, the civil procedure rules applied in those actions are developed by the courts relying on their inherent powers on a case-by-case basis. In this respect, the New Zealand courts have been heavily influenced by developments occurring in class actions in the United States, Canada and Australia (Chamberlain and Watson 2021, pp. 307–16).

In New Zealand, the court will grant leave for a representative action to proceed on a class-wide basis provided (Saunders v Houghton (No 1) [2009] NZCA 610; [2010] 3 NZLR 331 at [13]):

(a)

The order does not confer a right of action or defence which would not have been available in a separate proceeding;

(b)

There is a shared interest in common by all class members; and

(c)

It is for the benefit of other class members that the plaintiff sues in a representative capacity on their behalf.

The “same interest” requirement means that the class members must have “a significant common interest in the resolution of any question of law or fact arising in the proceeding” (Cridge v Studorp Ltd. [2017 NZCA 376; (2017) 23 PRNZ 582 at [11]). The court can, on application, permit the action to proceed on an opt out basis (Southern Response Earthquake Services Limited v Ross [2020] NZSC 126; [2021] 1 NZLR 117; Chamberlain 2021, p. 142).

The plaintiffs’ and defendants’ legal costs of the proceeding are paid, usually on a scaled basis, by the unsuccessful party in the litigation. If the representative plaintiff is unsuccessful in the litigation, they bear the class members’ legal costs. However, in practice, a third-party litigation funder usually indemnifies the representative plaintiff and bears the plaintiff’s legal costs on “no win, no fee” basis. The court can, on application, order a common fund order so that any funder commission is borne by the class members—not just those who sign the litigation funding agreement (Simons v ANZ Bank Limited [2024] NZCA 330; [2024] 3 NZLR 485). The application can be made at an early stage of the proceeding. However, the court noted that “[c]ritical to this conclusion is that the court will closely scrutinise the CFO [common fund order] and approve any settlement” (Simons v ANZ Bank Limited [2024] NZCA 330; [2024] 3 NZLR 485, [135]).

Empirically, the number of representative actions filed on a class-wide basis has increased over the past three decades (Chamberlain and Watson 2021, p. 319; Waye et al. 2025, p. 143). In 2022, the New Zealand Law Commission (NZLC) completed a review of class actions and litigation funding in New Zealand. They recommended that New Zealand enact specific trans-substantive legislation to govern the use of class actions in New Zealand. However, to date, the government has not enacted the proposed legislation.

The NZLC’s final report notes that there are two primary objectives of class actions: improving access to justice and enabling economy and efficiency of litigation (NZLC 2022, p. 59). While not adopting deterrence as an objective, the NZLC noted that there was debate as to whether it should be or whether it is merely a byproduct of class actions. The NZLC concluded that compensation is the primary objective of class actions rather than influencing compliance with the law or deterring wrongful behaviour. Noting that academics hold differing views, the NZLC considers that deterrence should be an objective dealt with by regulators (NZLC 2022, pp. 57–59; Waye et al. 2025, p. 153).

5. Assessing Compensation

Mass tort class actions are prevalent in Australia. Morabito’s empirical study from April 2023 records that a total of 831 class actions were filed in the Federal and State Courts in Australia with 98 cases (11.7%) categorised as product liability claims, 80 cases (9.6%) categorised as mass tort claims (includes environmental claims) and one case (0.12%) is categorised as a data privacy claim (Morabito 2023, p. 37). Morabito states that categorising the claims is a “somewhat subjective process rather than an exact science” and that “… frequently class actions fall into more than one category” (Morabito 2023, pp. 15–16). For example, a product liability claim often contains statutory and tort-based causes of action. Nevertheless, one can extrapolate from the data that mass tort causes of action account for at least 21% of all class action litigation in Australia. Of these, 18 (18.3%) product liability claims are funded by a TPLF and 14 (17.5%) mass tort claims are funded by a TPLF (Morabito 2023, pp. 15–16).

Most class actions settle (Morabito 2023, p. 26). Morabito records that 59 class actions in Australia have settled for AUD 50 million or more, equating to a total compensation of AUD 7,639,766,667 (Morabito 2025, p. 58). Five out of the top six settlement sums reached in Australian class actions are from mass tort claims alleging negligence. They are comprised of the following settlement sums: approximately AUD 494 million in the Kilmore East—Kinglake Black Saturday Bush Fire class action; AUD 440 million in the Queensland Floods class action; AUD 300 million in the Ethicon/Pelvic Mesh and the Marysville Black Saturday Bush Fire class actions respectively; and, AUD 250 million in the DePuy Hip Implant class action (Morabito 2023, p. 24). The fifth largest class action settlement occurred in the Uber settlement, which claimed the tort of conspiracy by unlawful means, and settled for AUD 271.8 million (Morabito 2025, p. 56).

On the face of it, these settlements prove that compensation is achieved in certain class actions. However, to measure their success, one needs to consider what was claimed as the total loss versus what the class members were awarded after deducting their legal and litigation funding expenses (Legg 2023, p. 48). One also needs to consider other options by which to compare class action outcomes. For example, is the comparison with individual litigation, or, for example, compensation one might receive under a government regulatory scheme such as the Accident Compensation Scheme operating in New Zealand. These are complex questions and present a far wider enquiry than can be accomplished in this article. However, for our purposes, we discuss anecdotally the compensation made in a few case studies and then conclude by discussing the practical issues that impede the performance of the class action’s compensatory function.

5.1. Compensation in Class Action Case Studies

In Australia, settlement sums are usually reached through what is termed the “global sum with individualised assessment approach” or the “process approach” (Gilsenan and Legg 2019, p. 16). The global sum approach is the most common settlement approach in mass tort class actions (Gilsenan and Legg 2019, p. 21). It is where the defendant pays a global sum to settle all claims and then an assessment occurs, usually through a claim’s administrator, of each individual’s loss. The first part of the enquiry places a cap on overall recovery for the class (Harvey et al. 2020, p. 74). The second part involves determining an individual’s eligibility, whether causation is established, what the loss is and whether the individual has reimbursement obligations with others. The process approach is where the “total settlement amount is not usually capped” but the process to obtain compensation requires a further adversarial process to determine eligibility and the amount of compensation in accordance with the terms of settlement (Gilsenan and Legg 2019, p. 24).

The court must approve all class action settlements (FCAA s 33V). The Federal Court of Australia stated in Australian Securities and Investments Commission v Richards [2013] FCAFC 89 at [7] that “[j]ustice will be satisfied where a settlement is ‘fair and reasonable having regard to the claims made by group members who will be bound by it”’. The factors the court considers in determining what is “fair and reasonable” usually include the complexity and duration of the litigation; the stage of the proceedings; the risks and prospects of success of establishing liability and damages; the risks of an appeal; and, the reasonableness of the settlement in light of the “best case” recovery and the attendant risks of litigation (Williams v FAI Home Security Pty Ltd. (No 4) [2000] FCA 1925; (2000) 180 ALR 459 at [19]; Modtech Engineering Pty Ltd. v GPT Management Holdings Ltd. [2013] FCA 626 at [11] and [13]; Webb v GetSwift Limited (No 7) [2023] FCA 90; 414 ALR 500 at [15]–[16]). The court focuses on the settlement amount compared to the value of the claim or loss suffered while considering various risks that may result in no or a reduced recovery. Further, the court’s role is to decide whether the proposed settlement is within the range of reasonable outcomes, rather than whether it is the best outcome which might have been won by better bargaining (Webb v GetSwift Limited (No 7) [2023] FCA 90; 414 ALR 500 at [15]–[16]). The amount to be paid in compensation is usually publicised through public settlement approval hearings and the court’s judgment approving the settlement.

In New Zealand, there are currently no statutory-based rules that require settlements to be approved by the courts. However, as a condition of granting opt out orders at the leave stage of the proceeding, the courts have recently ordered that any settlements will need court approval (Southern Response Earthquake Services Ltd. v Ross [2020] NZSC 126; [2021] 1 NZLR 117 at [101]; Simons & Ors v ANZ Bank Limited & Anor [2024] NZCA 330; [2024] 3 NZLR 485 at [46]). In practice, only a handful of representative actions have claimed negligence, and fewer of which we have access to important settlement terms. For example, in Re Strahl [2021] NZHC 3608, where negligence was claimed, the court issued a settlement approval judgment but redacted the settlement sums for reasons of confidentiality. However, one claim where settlement was obtained for a negligence cause of action although court approval was not required is in Attorney-General v Strathboss Kiwifruit Limited & Anor [2020] NZCA 98; [2020] 3 NZLR 247 discussed below.

(i)

PFAS Environmental Government Class Action

In 2020, the Federal Court of Australia approved the settlement of three class actions which involved claims against the Australian government for the Royal Australian Air Force’s use of firefighting foam containing per- and poly-fluoroalkyl substances (PFASs) that allegedly contaminated land (referred to as the Williamtown, Oakey and Katherine class actions).4 The class members claimed business and property loss but not personal injury loss. In approving the settlements, Justice Lee compared the proposed gross settlement sums with the “best possible recovery that might be able to be attained on behalf of group members”. In this respect, he stated (Smith v Commonwealth (No 2) [2021] FCA 837 at [64]):

In relation to the Williamtown class action, the settlement sum of AUD 86 million as a percentage of the total possible claim after costs is 97 percent. In relation to the Oakey class action, the settlement sum as a percentage of the best possible quantum recoverable and likely recoverable costs is 103.6 per cent. In relation to the Katherine class action, the Katherine settlement sum as a percentage of the best possible quantum recoverable and likely recoverable costs is 109.41 per cent.

Justice Lee commented that the gross recovery of AUD 212.5 million is “excellent” and “very high compared to comparable class actions” (Smith v Commonwealth (No 2) [2021] FCA 837 at [65]). However, he goes on to note that this amount does not take into account legal or litigation funding costs which were contested at the settlement hearing (Smith v Commonwealth (No 2) [2021] FCA 837 at [66]). The approved litigation funding costs amounted to AUD 53.125 million, the approved litigation funding expenses amounted to AUD 888,356, the approved legal costs amounted to approximately AUD 29.370 million and the approved additional payouts to the representative plaintiffs amounted to AUD 230,000 (Smith v Commonwealth (No 2) [2021] FCA 837). After these deductions, the amount distributed to class members was approximately AUD 128.887 million. The litigation funder, Omni Bridgeway, claims that “[t]he average distribution per household was around AUD 100,000, with households in the worst-affected areas receiving between AUD 150,000 and AUD 300,000”.5

Several class members were not pleased with the settlement. Legg and Prior assert that the “class action settlement had done little to change their [the class members’] circumstances, their property and the local environment” (Legg and Prior 2022, p. 11). Further, they state that “research on environmental litigation has observed, damages awarded as a result of the pollution of a river are not necessarily applied to reinstate the river to its former unpolluted state” (Preston 2008, citing Stone 1972, p. 8). In other words, the compensation received by the plaintiffs did not truly put them back in the position they would have been in but for the harm—which is the traditional measure of compensatory damages—but rather represents a monetary equivalent.

(ii)

DePuy Personal Injury Class Action

In 2011, representative plaintiffs filed a class action on behalf of approximately 5000 class members claiming that DePuy and Johnson & Johnson Medical Pty Ltd. (J&JMPL) negligently released ASR hip implant medical devices into the Australian market, failed to monitor and test the performance of the implants and failed to suspend or withdraw the implants when concerns about their performance arose. The class members claimed they suffered personal injury from the ASR implants. DePuy admitted that they owed a duty of care to class members but denied breach.

In 2016, the Federal Court of Australia approved a settlement in a class action against DePuy and J&JMPL (Stanford v DePuy International Ltd. (No 6) [2016] FCA 1452). Prior to settlement, the defendants paid AUD 136 million to class members or third parties on behalf of class members pursuant to a voluntary reimbursement program. These costs mostly consisted of treatment or related costs for revision surgery (Schimmel et al. 2021, p. 413). The class action trial on the common issues ran for 17 weeks before settlement was approved for AUS 250 million. The legal costs at the time of settlement were approximately AUD 36 million. The solicitors funded the proceedings on a “no win, no fee basis” (Morabito and Ekstein 2016, p. 81). There is no publicly available record of any commission they received. There was a mechanism for class members to receive a fast-track payment of AUD 55,000 if they so chose (at [64] and [140]). Of course, fast-track payments compensate class members quicker than receiving an individualised assessment; however, the trade-off is that they often recover less than they would otherwise (Legg 2018, p. 676). If class members elected to receive an assessment for their settlement amount, then any claim for non-economic loss had to be over AUD 40,000 (at [65]). As of October 2021, the court-approved administration costs for administering the settlement fund were just under AUD 16 million. The court noted that it was “an overrun of $10 million” (Stanford v DePuy International Pty Ltd. (No 8) [2024] FCA 35 at [84]).

In approving the settlement, Justice Wigney noted “[t]he trial was complex, hard-fought and difficult. There was little common ground” (at [28]). He also noted at [125] that

… despite a 17 week trial, and despite the applicants’ intuitively strong case, there remained considerable uncertainty and risk for the applicants and group members in terms of the outcome of the trial. And the stakes were very high indeed. To put it bluntly, the risk of the applicants failing completely could not be excluded. That would have meant no recovery by the applicants and group members at all, and the likelihood of an adverse costs order against the applicants for potentially many millions of dollars. Nor was it possible to exclude the risk that, even if successful on liability, the applicants might have obtained a materially lower award of damages than they contended for. That too may have had implications for the group members as a whole.

As of July 2019, approximately three years after settlement approval, 400 class members received compensation from the claim administrators. As of December 2020, approximately four years after settlement approval, the settlement administrators had paid out 1561 claims totalling AUD 139 million with 66 remaining (Schimmel et al. 2021, p. 413). As at April 2025, approximately nine years after settlement approval, the settlement administrators were still working through three complex class member claims.6 The time taken from settlement approval to distribution of funds brings into question the effectiveness of the compensation because of the time-value cost of money. In other words, compensation received in a timely fashion has more value from an inflationary perspective but also because it remedies the harm faster. This is particularly important in personal injury cases where claimants usually need funds urgently to pay medical treatment and rehabilitation costs.

In New Zealand, citizens who suffered personal injuries from the DePuy hip implant received compensation through a government regulatory scheme under the Accident Compensation Act 2001 (ACA). The amount they received has not been made publicly available. Some of those claimants sought to join proceedings in England and commenced proceedings in New Zealand for additional compensatory cover. However, both failed because of the statutory bar in suing for compensatory under section 317 of the ACA (Allan & Ors v Depuy International Systems [2015] EWHC 926 (QB); McGougan & Ors v Depuy International Ltd. & Anor [2018] NZCA 91). The claimants could file proceedings for exemplary damages against DePuy (section 319 of the ACA). However, this has not occurred given the low exemplary damages awards in New Zealand.

(iii)

Kiwifruit Environmental Government Class Action

In New Zealand, in 2009, the Ministry of Agriculture and Forestry (now called the Ministry of Primary Industries (MPI)) gave a permit to import kiwifruit pollen into New Zealand. Unfortunately, the shipment contained bacteria known as Psa3. The bacteria, which are lethal to kiwifruit orchards, spread in the Bay of Plenty and Northland. As a result, several kiwifruit growers suffered financial and emotional loss. In 2014, Strathboss Kiwifruit Limited (Strathboss) brought representative action proceedings on behalf of 200 kiwifruit growers against the Attorney General for NZD 450 million. The proceeding alleged negligence against the Attorney General.

The High Court found that the Attorney General owed a novel duty of care to the kiwifruit growers which it breached (Strathboss Kiwifruit Ltd. & Anor v Attorney-General [2018] NZHC 1559). The Court of Appeal overturned the High Court decision on the basis that MPI has statutory immunity preventing liability and, in any event, it was not “fair, just or reasonable” to impose a duty of care (Attorney General v Strathboss Kiwifruit Ltd. & Anor [2020] NZCA 98; [2020] 3 NZLR 247). Strathboss appealed to the Supreme Court and leave was granted (Strathboss Kiwifruit Ltd. & Anor v Attorney General [2020] NZSC 68).

In 2020, before the Supreme Court hearing, the Attorney General settled with the kiwifruit growers for a gross sum of NZD 40 million. It was reported that approximately NZD 15 million (or 40% of the total settlement) went to pay legal costs and litigation funding fees.7 The lawyer acting for Stathboss commented that there was “a mixture of happiness and disappointment among the plaintiff growers about the settlement” and that “our team was always aware that getting a court ruling would be hard in a Supreme Court”.8 While this case highlights the significant transaction costs of class actions, it also shows how the class action mechanism can be used to achieve a settlement when a legal claim bears risks.

5.2. Impediments to Class Action Compensatory Aims

As a starting point, there are inherent risks when a claimant’s access to justice and, in turn, chances of receiving compensation for loss, are contingent on proving the elements of a negligence cause of action. While, undeniably, the class action device assists claimants who may not have the ability or means to individually sue a well-heeled defendant, a large number of those claimants will not get relief because they lack the necessary evidence to prove the defendant owed a duty of care or that the defendant caused the loss (Morabito and Ekstein 2016, p. 62). Some accidents are merely “acts of God” and are not caused by any person, while others can be caused by several factors. There are numerous examples where class actions in Australia have been discontinued because of causation issues. For example, the class action for personal injury damages in Hogan v Sydney Water Corporation Limited was discontinued because the class could not substantiate a causation claim on a group basis (Federal Court of Australia, Proceeding No NG 1119, 1998). The class action in Philip Morris (Australia) Ltd. v Nixon (2000) 170 ALR 487 was discontinued because the plaintiff could not prove causation in relation to every defendant (Clark and Harris 2009, p. 91).

In practice, most plaintiffs do not care about the legal technicalities of a negligence cause of action—they just want recovery for their loss. In this respect, for defendants without insurance, the plaintiffs run the very real risk of not being compensated at all. In other words, compensation is contingent on the defendant’s financial means. For example, after litigation started for the mass tort asbestos harms in Australia, the “largest asbestos manufacturers in Australia … rearranged their assets to limit their asbestos liabilities” (Spender 2003, p. 253). Conversely, class member recovery is also contingent on being able to fund class action litigation in the first place. In an Australian class action concerning the drug Roaccutane, solicitors representing the class on a “no win, no fee” basis discontinued it “following the realisation … that they lacked the resources to run a proceeding of [that] magnitude” (Morabito and Ekstein 2016, p. 80).

When a negligence claim is successful on the issue of liability, in practice, a large portion of claimants are not fully indemnified for their loss in the class action for several reasons. First, there are federal and state statutory caps on the damages victims can claim for personal injury loss (Gilsenan and Legg 2019, pp. 22–23). The statutory cap reforms were “driven by commercial interests of insurers who were concerned about the quantum of personal injury damages payments and their impact on the premium pool for liability insurance” (Harvey et al. 2020, p. 48). In the Federal Courts, the Competition and Consumer Act 2010 (Cth) contains an eligibility threshold for non-economic loss and caps damages for non-economic loss, economic loss for income and gratuitous care. It also provides that exemplary damages cannot be awarded for personal injury claims. Harvey et al. state that the impact of these reforms is that mass tort class actions where victims seek more modest damages sums for personal injury have declined (Harvey et al. 2020, p. 71; Schimmel et al. 2021, p. 404). This significantly impacts access to justice for those who suffer from less grave accidents. They state (Harvey et al. 2020, p. 71)

For example, the early years of the class actions regime (that is, in the decade between 1992 and the early 2000s) saw a slew of product liability class actions arising from contaminated food such as Kraft peanut butter, pizza, pork rolls at a Vietnamese restaurant, Spanish Mackerel and oysters. These types of claims have now disappeared. Similarly, claims for relatively modest injuries arising from adverse effects of pharmaceutical products are no longer pursued; an example of one of the pre-tort reform pharmaceutical claims is the class action relating to adverse effects of Travacalm.

It is also important to note that statutory caps can vary depending on jurisdiction. In Gill v Ethicon Sàrl (No 10) [2023] FCA 229 at [123]–[130], Justice Lee discussed how the amount of compensation that the lead representative plaintiff was entitled to for non-economic loss from a negligence claim would differ depending on whether the wrong occurred in the Australian state of New South Wales or Western Australia. This disparity raises significant concerns as to the fairness of damages awards in federal class actions that claim negligence. From a compensatory perspective, a class member’s compensation may be considered arbitrary if it is based solely on where they suffered the wrong as opposed to remedying the harm caused.

Second, the loss claimed for each plaintiff is often so varied, particularly in personal injury claims, that the use of the class action procedure can be antithetical (Clark and Harris 2009, p. 90). Inaccuracies can occur in assessing damages awards with some class members receiving too much while others too little (Woodhouse et al. 1967, p. 52; Legg 2016, pp. 94–95).9 This is particularly acute in class actions because most settle and the class member is often not actively engaged in the negotiation process. Settlement sums are reached either through the “global sum with individualised assessment approach” or the “process approach” (Gilsenan and Legg 2019, p. 16).

In relation to the global sum approach, the time and costs associated with the claim administrator’s enquiry eats into the overall funds available for distribution to class members. In Australia, examples of this type of settlement have occurred in the Kingslake Black Saturday Bush Fire class action (Matthews v AusNet Electricity Services Pty Ltd. [2014] VSC 663 at [420], the DePuy Hip Implant class action (Stanford v DePuy International Ltd. (No 6) [2016] FCA 1452) and the Ethicon/Pelvic Mesh class action (Gill v Ethicon Sàrl (No 10) [2023] FCA 229). As Barnes and Macdonald state about the bush fire class actions in Australia (Barnes and McDonald 2021, p. 41; Waye 2018, p. 122),

All these actions were settled on a percentage of loss basis, meaning the defendant would pay each group member the portion of its losses specified in the settlement agreement. For such a payment to be made, each group member’s losses was assessed individually. … The time required to undertake this negotiation meant that the last payment occurred nine years after the settlement was first reached.

In relation to the process approach, the process to obtain compensation requires a further adversarial process to determine eligibility and the amount of compensation in accordance with the terms of settlement. Gilsenan and Legg cite an example of this being the LCS^® Duofix™ Knee Replacement class action in Australia (Barnes and McDonald 2021, pp. 30–31). They state “[c]oncerns with cost and delay are embedded in the application of substantive fairness” and Barnes and McDonald 2021, pp. 28, 30–31):

[The process approach] may be costlier and more time-consuming on a per-claim basis due to the adversarial approach. A defendant is incentivised to carefully consider each claim and contest those that it believes are not eligible, as the more ineligible the claims, the lower will be the compensation amounts payable. The LCS^® Duofix™ Femoral Components class action took over three years to deal with about 400 claims. In contrast, the much larger Kilmore East Kinglake bushfire SDS, with 1481 I-D claims and 9174 economic loss and property damage (“ELPD”) claims, took about two years and five months.

Third, class actions have high transaction costs (Legg 2016, p. 100).10 The damages the claimant receives are usually reduced by payments to interested third parties including plaintiff lawyers, third-party litigation funders, settlement fund administrators and disbursements (for example, to expert witnesses and court filing costs). Spender states (Spender 2010, p. 45)

Although we may decry the remedial distortions and compromises that lay in the wake of class actions, our opposition must be linked to sub-optimal outcomes. Suboptimal outcomes might be seen in poor allocations which result in exploitation of passive right holders, or windfall gains to middlemen, or unprincipled remedial doctrine.

In the Ethicon/Pelvic Mesh class action, the plaintiffs’ lawyers and the TPLF claimed a combined cost of over AUD 100 million out of an AUD 300 million settlement. This is before any costs for claims administration are deducted. The Federal Court of Australia has approved AUD 83 million to pay costs to date (Gill v Ethicon Sàrl (No 10) [2023] FCA 229; Gill v Ethicon Sàrl (No 11) [2023] FCA 229; Gill v Ethicon Sàrl (No 12) [2023] FCA 902).

Last, the damages that are paid on judgment or settlement are usually in the form of a lump sum payment which cannot be adjusted later because the assessments are made with finality (Woodhouse et al. 1967, p. 60). The claimant may suffer “financial pressures of present living” which may mean funds are insufficient to cover long-term medical and rehabilitation costs (Woodhouse et al. 1967, p. 77). Final settlement sums can also impact future claimants whose injuries are not yet known and their ability to recover if they are included in the class definition (Clark and Harris 2009, p. 91). This is particularly so with the global sum approach.

5.3. Empirical Comparative Research Needed

There is a dire need for more empirical research on the compensation outcomes of class actions compared to regulatory measures. Legislatures and rules committees should be encouraged to enact legislation and regulations which require decision-making bodies to publicly release comprehensive data on compensation amounts and their deductions. In Australia, the courts are required to approve class action settlements which means that compensation amounts and their deductions are publicly available. This information may not be available in other jurisdictions. From the empirical studies that are available, Legg studied class actions, alternative dispute resolution processes and a regulatory process that “compensated financial consumers arising from the Storm Financial collapse in Australia” (Waye 2018, pp. 128–29). He found that after regulatory proceedings, the “clients’ average recovery rate in ADR processes was 72 per cent of losses whereas class actions recoveries varied from 29 to 55 percent of loss” (Waye 2018, pp. 128–29).

Macleod and Hodges in Redress Schemes for Personal Injuries acknowledge that comparing compensation amounts is complicated (Macleod and Hodges 2017, p. 642). However, on reviewing regulatory schemes around the globe, they conclude (Macleod and Hodges 2017, p. 628)

… the administrative costs of all compensation schemes are vastly lower than the tort litigation system. Tort litigation usually has legal fees of around 55% of total compensation value awarded, often higher. Although administrative schemes are variable in their administrative costs, as a percentage of redress awarded they are hugely more efficient than tort.

Chamberlain, in comparing outcomes in the Ethicon Pelvic Mesh class action in Australia and the regulatory response to pelvic mesh claims in New Zealand, claims (McPherson 2023; Accident Compensation Corporation 2018)

… ACC reports showed that between 1 June 2005, and June … [30, 2018], the ACC paid out 439 claims [which incurred costs] for … [Pelvic Organ Prolapse (POP) or Stress Urinary Incontinence (SUI) mesh claims].

The 439 women were paid a total of approximately NZD 10.1 million (AUD 9.41 million), equating to a rough average of approximately NZD 23,000 (AUD 21,500) each, although payouts vary depending on individual circumstances.

While the calculations are very approximate, as are average payout estimates in Australia’s Johnson & Johnson class action [a crude average, dividing the settlement minus estimated costs between 11,000 women, suggests a payout of just under AUD 20,000], the figures appear to show a comparable level of compensation …

Further, the time between the decision date and the latest payout for the 439 victims is shown in

Table 1. Time taken between ACC decision date and latest payout to ACC pelvic mesh victims.

Years	Number of Victims	Percentage of Victims
Under 1	260	59%
1–2	78	18%
2–3	32	7%
3–4	24	5%
4–5	11	3%
Greater than 1	34	8%

(Accident Compensation Corporation 2018, p. 58):

However, the Ethicon Pelvic Mesh average payout per victim will likely be less than Australia’s overall average settlement payout in product liability class actions. In 2021, the average compensation of AUD 28,106 for over 11,370 claimants [was paid] in settled product liability class actions with AUD 55,506 and AUD 48,233 being the average payouts for medical device and pharmaceutical class actions respectively (Schimmel et al. 2021, p. 143).

The issue is one of consistency, or lack thereof, and the litigation risk for class members depending on which class action they are involved with.

6. Assessing Deterrence

As explained above, the class action also seeks to achieve deterrence, which aligns with tort law’s purpose. The Australian class action is a procedural mechanism that facilitates claims in tort through its representative and aggregative nature. The combination of (1) the pooling of resources or sharing of costs, (2) economies of scale and (3) the overcoming of the collective action problem makes claiming possible. Class actions change the traditional cost/benefit analysis as the aggregation of claims makes a claim that may be uneconomic to pursue alone feasible when combined with other claims. Costs can be shared with others who are similarly situated. The class action allows for the cost of bringing the action to be spread across many claimants, which may also give rise to economies of scale because as the number of group members increases, costs increase at a diminishing rate.

The collective action problem refers to the difficulty or problems that arise in group decision making. Mass torts tend to give rise to large, dispersed and disorganised groups of claimants. The benefits to the group, namely recovery of damages, exceed the aggregate costs to the individual members of the group, making action desirable. However, because benefits are dispersed among the group, an individual may be unwilling to incur the cost of action alone because individual costs would exceed their individual benefits. The class action allows the benefit to be pursued by sharing the costs. Moreover, without the class action, there would be additional costs of coordination in seeking to identify other similarly situated individuals. The opt out approach removes the need for this at commencement (Chen and Legg 2023).

Deterrence should come from the imposition of a range of costs—direct and indirect, financial and non-financial. The payment of compensatory damages, the costs of litigation and reputational harm associated with a successful class action would seem to intuitively operate as a potential source of deterrence in relation to mass torts. From a specific deterrence viewpoint, the rational tortfeasor should prefer to avoid these costs and should therefore be incentivised not to break the law. The extensive media coverage that class actions attract, typically because of the number of people affected and the quantum at stake, should promote not just specific deterrence but also general deterrence as the underlying conduct and the resolution of the class action is communicated widely, including to other similarly situated entities (Weinstein 1995, p. 6; Legg 2021, pp. 151, 161).

However, the effectiveness of class actions in achieving deterrence is difficult to measure. Empirical examinations of deterrence show that “deterrence matters a lot” but the failures of deterrence are also clearly visible. Moreover, the inability to measure deterrence also arises because effective deterrence involves proving a negative, that is, that a person or entity refrained from misconduct because of the fear of sanctions (Kennedy 2009, p. 40; Nagin 2018, p. 181).

To assess deterrence in this article, we take the view that deterrence is achieved when the class action imposes sufficient costs that the original misconduct is no longer profitable or desirable. The difficulty with this approach is that some class action costs, such as to reputation, are difficult to measure. Equally, the profit or savings made from the misconduct by the tortfeasor are often not known.

6.1. Deterrence in Class Action Case Studies

The class actions against the Australian Government in relation to the use of PFASs by the Defence force in firefighting foam provide an example where compensation was sought in relation to property or business losses and paid through a settlement that equated to roughly 100% of full recovery. But did it achieve deterrence?

PFASs had been used in firefighting foam since at least the 1980s because of its effectiveness in fighting liquid fuel fires. Defence had been warned in 1987 and 1991 that PFASs should be handled as a toxic waste and had “the potential to cause contamination of the environment” (Knaus 2017).

In 2004, Defence started phasing out its legacy firefighting foams containing PFASs as active ingredients and switched to “Ansulite”, which only contains trace elements of PFASs and is only used in emergencies or controlled environments to test equipment. Defence is currently undertaking a review of alternative firefighting foams that meet capability requirements as well as environmental protection requirements (Aylward 2018). In 2015, Defence’s national PFAS Investigation and Management Program was established to manage risks associated with PFAS contamination on and around Defence bases associated with the historic use of firefighting foams, including conducting remediation (Varghese 2024). Defence also established a procedure for an individual or business that has suffered loss or damage due to Defence activities involving PFASs to submit a legal claim (Australian Government n.d.).

On one view, the above description of events shows that Defence had already been deterred from using PFASs in an unsafe manner that could cause environmental harm, including reducing the value of land. On the other, the slow and secretive approach of the government to addressing the risks of PFAS contamination required public examination and the finding of fault through litigation. A government inquiry into the management of PFAS contamination in and around Defence bases noted that “the Williamtown community had ‘felt compelled’ to commence proceedings against Defence in November 2016 due to ‘over a year of inaction from the Commonwealth in adequately responding to or attempting to remediate the damage cause by the contamination’” (Joint Standing Committee on Foreign Affairs, Defence and Trade 2018, p. 91). The class action settled before there could be any findings of misconduct (Legg and Prior 2022, p. 9). However, as Australian class action settlements can only take effect with court approval, the settlement hearing and judgment publicised Defence’s conduct and the payment of substantial compensation.

Nevertheless, the class action did not seek compensation for the health effects that PFASs may have had on the persons living in Williamtown, Oakey or Katherine (Smith v Commonwealth of Australia (No 2) [2020] FCA 837 at [68]–[69]). This may be because the requirement to prove causation in terms of personal injury was regarded as more difficult than proving causation in relation to property loss (Legg and Prior 2022, pp. 3, 6). If the tortfeasor does not bear the full cost or impact of misconduct, then deterrence may be ineffective as they are not incentivised to avoid that cost. However, where the alleged wrongdoer is the government, and not a profit-seeking entity, the calculations may not be as financially driven. The government and its departments are usually concerned with acting in the interests of citizens, lest they lose support or even undermine trust in government. The court also appointed an epidemiological expert to prepare a report on the toxicology of PFASs and address whether PFASs are “causative or potentially causative of adverse human health effects”. While the issue was not decided, the report was obtained by the media who promoted its findings that “the existing body of evidence suggests that it is eminently possible … [PFAS] has negative health outcomes”, which was contrary to the findings of the Australian government’s expert health panel, appointed to investigate the effects of PFASs in 2018 (Fellner 2020).

The class action provided a vehicle to bring the government’s actions to the public’s attention and incentivise the government to address the concerns. In other jurisdictions, the experience has been similar, in that it was litigation brought by community members “that forced the regulators and the government to begin to act” (Australian Senate Select Committee on PFAS 2025, p. 6).

The class action’s ability to facilitate a claim that draws attention to conduct and, as a result, has a deterrent effect, can even operate when the class action is unsuccessful. In Sharma v Minister for the Environment, the court at first instance held that the Minister owed a duty of care to Australian children to avoid climate-related harm, but on appeal, the holding was overturned (Sharma v Minister for the Environment [2021] FCA 560 overturned by Minister for the Environment v Sharma [2022] FCAFC 35 which utilised Federal Court Rule 9.21 rather than Pt IVA). However, the underlying finding at first instance that climate change was occurring with adverse effects for Australia was followed in subsequent litigation (O’Donnell v Commonwealth of Australia [2023] FCA 1227, [41]). Thus, while the specific claim in Sharma failed, the court’s judgment included a factual finding that can be employed to assist in deterring activities that cause climate change.

6.2. Profitable Misconduct

Between late 2003 and December 2009, J&JMPL imported into Australia medical devices, referred to generally as “ASR implants”, manufactured by DePuy International Ltd. for use in hip replacement or resurfacing surgery. These metal-on-metal medical devices were implanted in nearly 100,000 patients worldwide, including 5000 patients in Australia. However, these implants caused complications, and forced 37.1% of patients to undergo revision surgeries within 7 years to remove or replace malfunctioning implants (Wienroth et al. 2014; Stanford v DePuy International Ltd. (No 6) [2016] FCA 1452 at [30]). As explained above, in 2011, a class action was brought against DePuy and J&JMPL. The plaintiffs alleged contraventions of the Australian Consumer Law and negligence in designing, manufacturing and supplying the ASR implants. In 2016, after a trial on the common issues, but before judgment, the matter settled for AUD 250 million (Stanford v DePuy International Ltd. (No 6) [2016] FCA 1452).

The Federal Court approved the AUD 250 million settlement as “fair and reasonable”. However, from a deterrence perspective, the settlement sum appears to be substantially lower than J&J/DePuy’s revenue from the hip implants. From the launch of the hip implants in 2003 to the worldwide recall in 2010, DePuy’s sales nearly doubled from USD 3 billion to USD 5.6 billion with a significant source of growth being sales outside of the United States (Luong (2024) citing (Johnson and Johnson 2004, p. 31; Johnson and Johnson 2011, p. 33)). Moreover, in 2007, DePuy posted USD 4.6 billion in sales and gained the top market share position for replacement hips in the United States and the no. 2 position globally (Kamp 2008).

For the payment of compensation to put the victim back into the position they would have been in had the harm not occurred, legal costs and reputational harm may together be insufficient to deter when the profits from the misconduct are sufficiently large. Even with the aggregation of claims through the class action, the focus on harm and compensation may result in insufficient costs. The class action may aid in deterrence, but more is needed, either through additional damages (exemplary or punitive damages being one option: Lens 2014) or through additional avenues for deterrence, such as regulatory suits. In Australia, regulators such as the Australian Competition and Consumer Commission (ACCC) and Therapeutic Goods Administration (TGA) have a range of enforcement tools, including civil penalty proceedings, that may be employed (Australian Competition and Consumer Commission 2025; Australian Government—Therapeutic Goods Administration 2025).

6.3. Repeat Tortfeasors

The DePuy class action was not the only litigation faced by J&J. Outside of Australia, there have been legal proceedings involving children’s Tylenol, Rolaids, talc-based baby powder and opioids (Johnson and Johnson 2025, pp. 93–100). In 2012, in Australia, a class action was brought against J&JMPL and its subsidiary, Ethicon, for pelvic mesh devices which were intended to alleviate stress urinary incontinence or pelvic organ prolapse. Justice Katzmann found J&JMPL and Ethicon liable in negligence for failing to conduct adequate clinical evaluations and issue adequate warnings of the risks of the pleaded complications as well as contraventions of the Australian Consumer Law (Gill v Ethicon Sàrl No 5 [2019] FCA 1905). As explained above, the matter settled for AUD 300 million.

The New York Times in reporting on litigation brought against J&J asked if the company was facing systematic issues in how it operated, putting profit before people, or whether, as a multinational conglomerate with more than 250 operating companies in 60 countries, there would be missteps in product development and sales (Singer 2010). From a deterrence perspective, repeated tortious misconduct by the same entity suggests that specific deterrence is ineffective. Despite multi-million dollar compensation payments (billions in the United States), the company appears to continue to operate in a manner that involves law-breaking. There is a continuing undervaluing of human harm that may become part of “business as usual”. Other corporations also see this conduct and the insufficient ramifications it attracts with the result of negative general deterrence, i.e., an incentive to break the law as others have found it profitable to do so.

Again, the question arises as to how sufficient deterrence can be achieved. As raised above, putting a higher value on human life and suffering is required, which needs to be reflected in the calculation of loss or damages imposed. In addition, regulatory suits, such as pursuing criminal charges against high-ranking individuals, may be needed. On one view, the class action is just not enough. But, arguably, it is also necessary. As the New Yorker reported in relation to the talc litigation in the United States, “the only people who have ever held Johnson & Johnson responsible for its actions are those who have served on juries” (Cep 2022).

6.4. Transaction Costs and Deterrence

In both the PFAS class action and the Kiwifruit growers class action, roughly 40% of the total settlement went to pay legal costs and litigation funding fees. While these costs may undermine full compensation, as explained above, they do not undermine deterrence (Legg 2022, p. 154). This is because deterrence focuses on the costs imposed on the wrongdoer, regardless of what causes those costs. Indeed, costs incurred by the wrongdoer in defending litigation, such as legal fees and lost management time, may also act as a deterrent. An example of the range of potential costs imposed by alleged misconduct is captured by the J&J 2024 Annual Report which notes (Johnson and Johnson 2025, p. 11)

Product concerns, whether raised internally or by litigants, regulators or consumer advocates, and whether or not based on scientific evidence, can result in safety alerts, product recalls, governmental investigations, regulatory action on the part of the U.S. FDA (or its counterpart in other countries), private claims and lawsuits, payment of fines and settlements, declining sales and reputational damage. These circumstances can also result in damage to brand image, brand equity and consumer trust in the company’s products.

The class action typically magnifies transaction costs. The grouping of claims increases the amount of damages claimed, which in turn can increase the resources devoted to the defence of proceedings. Similarly, the high-stakes nature of mass torts can have a significant reputational impact. However, the quantum of a defendant’s legal costs, although measurable, is typically not made public. Costs for diverted or distracted management, who must deal with the class action rather than their usual responsibilities, are difficult to quantify and not publicly available.

6.5. Insurance

Insurance is generally defined as a contract by which one party (the insurer) undertakes to indemnify another party (the insured) against risk of loss, damage or liability arising from the occurrence of some specified event (Legg et al. 2024, p. 288). In the mass tort scenario, businesses may obtain Product Liability Insurance to protect against claims of negligence causing personal injury or property damage that a third party suffers because of the business’s product (Aon 2025). The presence of insurance cannot easily be determined, and some large organisations may self-insure.

While insurance may allow for risk-spreading and the availability of funds for compensation, it may also undermine deterrence because the wrongdoer is relieved from making payment. Instead, the insurance provider will meet the payment (Baker and Griffith 2010). This may be considered through the concept of moral hazard—if a person ceases to be liable for the costs of their actions, then their incentive to minimise the costs of those actions may be lost (Legg et al. 2024, p. 303). As John Fleming observed, “[t]he deterrent function of the law of torts was severely, perhaps fatally, undermined by the advent of liability insurance” (Fleming 1967, p. 823).

If companies and their decision makers are insulated from the costs of tortious misconduct, then those costs may not impact their decision making. However, insurance requires the payment of a premium and usually includes clauses specifying an excess/deductible or retention, i.e., an amount that must be paid by the company before the insurer is liable to pay. Entities that engage in wrongdoing will need to claim on their insurance more frequently and for larger payments, which in turn can see premiums rise (or insurance become unavailable). Similarly, an excess/deductible or retention can be set at a level to create an incentive for the entity to act prudently and take steps to avoid misconduct. The insurer and their pricing of insurance vis-à-vis risk of misconduct, if accurate, can facilitate deterrence. Insurance products may be better seen as filtering deterrence rather than removing it (Legg 2022, pp. 125, 163). The impact of class actions on insurance is illustrated by Australian shareholder class actions where the rise in the number and quantum of claims saw insurers respond by increasing premiums and excesses/retention amounts (CIMIC Group Limited v AIG Group Limited [2022] NSWSC 999, [336] (premium tripled and retention clause of AUD 1 m added); Nuix Limited v Berkshire Hathaway Specialty Insurance Company [2025] FCA 1002 (retention clause set at AUD 10 million)).

Further, the concern about the loss of deterrence has also been responded to through the argument that informal or non-monetary sanctions, such as reputational harm, may prevent dispersion because they “transcend the commercial banality of money” (Fisse and Braithwaite 1983, p. 82). The class action’s aggregation of claims and attraction of publicity impose costs that insurance cannot meet. However, the overall effectiveness of publicity or reputational harm and its deterrent effect on tortfeasors is difficult to measure.

7. Conclusions

The class action procedure can compensate tort victims for their losses and deter tortfeasor (and potential tortfeasor) wrongdoing. A good example of the effectiveness of the class action procedure can be seen in the PFAS class action case study discussed in this article. However, in most class actions, their effectiveness in meeting their compensatory and deterrence aims is severely hampered.

In relation to compensation, the class action often causes significant time delay and third-party costs before compensation is paid to class members which negates their effectiveness. For example, in both the PFAS and the Kiwifruit class action, approximately 40% of the settlement sums went to pay third-party costs. In the Kiwifruit class action, the growers did not receive compensation until six years after commencing proceedings. In the DePuy class action, most class members did not receive compensation until at least eight years after commencing the proceeding. In this respect, the type of harm suffered will likely impact the effectiveness of the compensation received. In situations where personal injury is suffered, such as in the DePuy class action, receiving compensation in a timely manner is even more crucial because of the often-urgent nature of the victim’s medical and rehabilitation costs if insurance or government health benefits are unavailable. Last, the defendant’s solvency, statutory caps, administrator settlement costs, risks associated with individualised assessments and the lump sum nature of compensation all impinge on the effectiveness of compensation in the class action context.

In relation to deterrence, it is extremely hard to measure whether class actions in fact deter, or sufficiently deter, tortfeasor wrongdoing. Intuitively, deterrence should influence tortfeasor behaviour by imposing a range of costs—direct and indirect, financial and non-financial. In this respect, the PFAS class action suggests that the Australian government started acting on PFAS concerns in a timely manner once litigation was commenced. However, profiting from misconduct and insurance coverage can dull the deterrence effect of class actions. We see these concerns at play, particularly when there are repeat class action tortfeasors, such as J&J, who make extremely large profits from their products which cause injury. Whether the class actions deter or are merely seen as a “cost of doing business” is unknown.

However, we can conclude that class actions are necessary but often not sufficient. Without the class action there may be no compensation, nor the costs and publicity that aid in deterrence. But by themselves, class actions often do not provide adequate compensation, especially due to costs and delay, or deterrence. This leaves one questioning whether reforms are needed. The class action procedure could be amended, or oversight improved, to better meet these aims. New procedures may need to be adopted to supplement or replace the class action in specific contexts. Alternatively, the class action may be seen as one element of a larger compensatory and regulatory regime that operates in conjunction with other sources of compensation and deterrence, such as government regulation, to be applied together or as alternatives to meet the needs of a specific situation.