Dwyer v Volkswagen Group Australia Pty Ltd

1. [1]

   GLEESON JA: This appeal concerns a representative proceeding brought by Prof Philip Dwyer as a representative plaintiff against Volkswagen Group Australia Pty Ltd (VW) claiming damages for himself and on behalf of group members in respect of the supply of Volkswagen motor vehicles which were alleged not to be of acceptable quality or, in some cases, merchantable quality, because the airbags fitted to the vehicles carried a risk of mis-deployment or rupture. Following delivery of his principal judgment on 18 June 2018 and a second judgment on 7 September 2021, the primary judge made final orders on 16 September 2021 determining the common questions in the proceedings adversely to the plaintiff and group members and dismissed the proceedings with costs. Prof Dwyer appeals against those orders and the answers to some of the common questions.

Nature of the case

1. [2]

   VW imported into Australia various models of Volkswagen vehicles manufactured by its German parent company, Volkswagen Aktiengesellschaft (VW AG). In October 2013, Prof Dwyer (the appellant) purchased a Volkswagen Passat from VW for approximately $40,000. Some 83,000 group members in the representative proceeding purchased various models of new and second-hand Volkswagen vehicles in Australia between 2007 and 2018 (the VW Vehicles). The VW Vehicles were fitted with driver side airbags manufactured by Takata Corporation or its related entities (Takata). Those airbags contained a propellant known as phase stabilised ammonium nitrate (PSAN).

2. [3]

   The appellant commenced a representative proceeding against VW under Pt 10 of the Civil Procedure Act 2005 (NSW). He sought damages for himself and on behalf of group members claiming that the VW Vehicles supplied from 1 January 2011, including his own, failed to comply with the consumer guarantee of acceptable quality in s 54 of the Australian Consumer Law (ACL) because the vehicles were not safe and free from defects, given the installation of the Takata airbag. He also claimed that the supply of the VW Vehicles to group members prior to 1 January 2011 was a breach of the statutory warranty of merchantable quality in s 74D the Trade Practices Act 1974 (Cth) (TPA). It was common ground that any distinction between the concepts of “merchantable” and “acceptable” quality was not material for the purpose of the proceedings.

3. [4]

   The questions for determination at the trial before Stevenson J were all those arising on the appellant’s claims and the questions common to the appellant’s and group members’ claims identified in consent orders made by Sackar J on 21 June 2019, as varied on 22 May 2020.

4. [5]

   The appellant’s case was that the use of PSAN as the propellant in the airbags created a risk of mis-deployment or rupture of the airbag, and that risk was present at the time he purchased his vehicle. He contended that:

5. [6]

   The appellant accepted at trial that the risk of mis-deployment or rupture of the airbags in the VW Vehicles is unquantifiable in that it cannot be predicted when and in what circumstances the PSAN in the airbags might degrade to a functionally significant degree. On his case, this is what made the VW Vehicles of unacceptable quality.

6. [7]

   The appellant and each group member had used their vehicle since purchase without incident. In the circumstances explained below, VW replaced the airbag in the appellant’s vehicle at no cost during a routine service in 2019. By the time of the trial (in May 2021), VW had replaced the airbags in each group member’s vehicle, subject to some exceptional cases: see [28] below.

7. [8]

   The primary judge rejected the appellant’s claim (Dwyer v Volkswagen Group Australia Pty Ltd t/as Volkswagen Australia [2021] NSWSC 715). In his principal judgment, his Honour concluded (at [14]) that the appellant had not established any link between the propensity of PSAN to degrade and any relevant, functionally significant propensity of the PSAN to degrade in the particular airbag installed in his vehicle so as to cause it to explode or malfunction. Therefore, the appellant had not established that his vehicle was not of acceptable quality when he purchased it.

8. [9]

   Against the possibility that this conclusion was wrong, his Honour went on to deal with two further issues. Assuming breach of the guarantee of acceptable quality, his Honour found that the defence relied on by VW under s 271(2)(a) of the ACL applied, because the assumed failure to comply with the consumer guarantee occurred “only because of” the act, default, or omission of another person, relevantly, Takata. His Honour also found that the appellant had not suffered any damage by reason of the installation of a Takata airbag in his vehicle, not least because VW had without charge replaced the Takata airbag with an airbag that is undoubtedly sound and, in any event, damage had not been proved by the appellant.

9. [10]

   In his second judgment, the primary judge dealt with the question of costs and the answers to the common questions (Dwyer v Volkswagen Group Australia Pty Ltd t/as Volkswagen Australia (No 2) [2021] NSWSC 1137).

10. [11]

    There are three parts to the appeal. To the extent that it relates to questions common to the claims of group members, the appeal is brought as a representative proceeding on behalf of group members: Civil Procedure Act, s 180(1)(a).

11. [12]

    The first concerns a challenge to the findings that the appellant had failed to establish that the VW Vehicles were not of “acceptable” quality, or in some cases of “merchantable” quality. The case on appeal is limited to the alleged risk of rupture of the airbags in the VW Vehicles; the appellant did not press his case below concerning the alleged risk of mis-deployment of the airbags.

12. [13]

    The second part concerns whether his Honour erred in his contingent finding that the exception in s 271(2)(a) of the ACL would apply to defeat the appellant’s claim for damages, assuming that the appellant’s vehicle was not of acceptable quality.

13. [14]

    The third part concerns whether his Honour erred in his contingent finding that the appellant had not suffered any damage and the adverse answers given to the claim for damages on behalf of group members.

14. [15]

    For the reasons that follow, the appeal should be dismissed.

Structure of these reasons

1. [16]

   The structure of these reasons is:

Factual background

1. [17]

   The primary facts are not in dispute. The following outline is largely taken from his Honour’s unchallenged summary in the principal judgment, supplemented by some additional matters which provide context for the appellant’s claim.

2. [18]

   The frontal driver side Takata airbags installed in Volkswagen vehicles contained four major subsystems: (a) an inflator; (b) a fabric bag that inflates; (c) a cover that matches the car’s interior and tears open in a controlled way when the bag is inflating; and (d) a sensor system that detects rapid deceleration that is consistent with a crash scenario. A diagrammatic representation of these components, given by Mr Robert Renz, the appellant’s expert, in Figure 2a of his report appears below.

3. [19]

   The airbag was designed to inflate upon a relevant impact by igniting the booster and then the main PSAN propellant to produce a controlled explosion that generated a hot gas (a “gas generant”) to fill the airbag with pressure within approximately 50 milliseconds of impact.

4. [20]

   A schematic cross-sectional diagram of such an inflator contained in Figure 9 of an affidavit provided by Mr Andreas Schade, a technical expert employed by VW AG, appears below.

5. [21]

   The location and operation of the Takata airbags fitted in the VW Vehicles was described by his Honour at [29]-[33]:

6. [22]

   The appellant’s vehicle was manufactured by VW AG and fitted with a Takata driver side airbag which contained PSAN as the propellant: at [34]. Takata manufactured the airbag installed in the appellant’s vehicle pursuant to a contract with VW AG: [35]-[36]. The specifications given to Takata by VW AG did not specify the propellant to be used in the airbags: at [38].

7. [23]

   On 3 November 2015, the United States Department of Transportation National Highway Traffic Safety Administration published a Coordinated Remedy Order which imposed recall and remedy obligations on certain vehicle manufacturers (not Volkswagen) in respect of Takata airbags installed in those vehicles. That report noted:

8. [24]

   Notwithstanding the occurrence of such incidents, the appellant accepts that there has been no reported instance of injury from a rupture of a driver side Takata airbag installed in a Volkswagen vehicle.

9. [25]

   In February 2016, VW AG commenced an empirical analysis program for the purpose of investigating whether the PSAN gas used in the airbag inflators in Volkswagen vehicles constituted a safety risk. This involved the collection and testing of some 20,000 airbag inflators from the field from vehicles manufactured from 2005 onwards and from various climatic zones around the world: at [67]-[68]. The details of this testing program and analysis are summarised at [56]f below.

10. [26]

    VW AG’s product safety committee determined that there was no systemic safety risk associated with Takata smokeless driver inflator (SDI) airbag inflators installed in Volkswagen vehicles that would warrant their recall. The Court was also informed that no European regulator has required a recall of Volkswagens fitted with Takata airbags: at [75]. Nevertheless, VW AG and its related companies conducted recalls of vehicles fitted with Takata SDI airbag inflators in those countries where recalls were required by relevant government authorities, including in Australia.

11. [27]

    In February 2018, the Assistant Minister to the Commonwealth Treasurer issued a recall notice styled Consumer Goods (Motor Vehicles With Affected Takata Airbag Inflators and Specified Spare Parts) Recall Notice 2018 (Cth), which came into effect in March 2018: at [54]. The recall notice applied to a wide class of motor vehicles, including the VW Vehicles, and required the recall of “Affected Takata Airbag Inflators” including driver side airbags that used PSAN as a propellant to activate the airbag upon a collision: at [55]. The notice stated that it had been issued after a detailed investigation by the Australian Competition and Consumer Commission (ACCC) into possible risks involved in using vehicles containing Takata airbags manufactured by Takata Corporation of Japan: at [56].

12. [28]

    Following the recall notice, VW initiated a recall program. The VW Vehicles were recalled and the airbag inflators in them progressively replaced by the time of trial, subject to minor exceptions where VW had applied to the ACCC for an allowable exemption (for example, where the vehicle has been stolen, scrapped or where the owner has not been able to be contacted): at [64], [66]. As of 31 December 2020, the oldest VW Vehicle containing a Takata airbag was approximately 14 years old. All other relevant VW Vehicles were no more than 12 years old.

13. [29]

    The appellant’s case is that his vehicle was not as safe and free from defects as the hypothetical reasonable consumer would regard as acceptable because a safety device in the vehicle for the stated purpose of minimising the risk of injury upon collision – the airbag – itself carried a risk of causing injury or death which was not present in airbags that used other propellants.

14. [30]

    The appellant accepts that his vehicle is, by reason of replacement of the airbag, as valuable today as it would have been had it been supplied originally with non-defective airbags. The appellant’s complaint is that his vehicle, at the time of supply, was worth less than the price he paid for it by reason of the defective airbag installed in it. It is said that notwithstanding the replacement of the airbag some six years after purchase, the appellant has still suffered loss and damage.

The statutory provisions

1. [31]

   The appellant’s claim for damages for himself and on behalf of the group members who purchased their VW Vehicles from 1 January 2011, was based on a statutory cause of action conferred by s 271 of the ACL. The damages claim advanced on behalf of group members who purchased their VW Vehicles prior to 1 January 2011, was based on a statutory cause of action conferred by s 74D(1) of the TPA. The applicable statutory provisions are set out below.

2. [32]

   The Australian Consumer Law commenced on 1 January 2011 and is set out in Sch 2 of the Competition and Consumer Act 2010 (Cth). Section 54 provides a consumer guarantee of acceptable quality:

3. [33]

   “Supply”, in relation to goods, as defined in s 2(1), includes “supply (including re-supply) by way of sale, exchange, lease, hire or hire-purchase”. “Goods” as defined in s 2(1), includes “(a) ships, aircraft and other vehicles; … (f) second-hand goods; and (g) any component part of, or accessory, to goods”.

4. [34]

   There is no dispute that these provisions were satisfied in the present case, relevantly:

   1. (1)

      the appellant and at least some group members were “consumers” within the meaning of ACL, s 3(a) and (b);

   2. (2)

      the VW Vehicles are “goods” as the inclusive definition of that term in s 2(1) includes “vehicles”, which are expressly referred to in sub-par (a) of the definition (“ships, aircraft and other vehicles”); and

   3. (3)

      VW admitted on the pleadings that the VW Vehicles were supplied in trade or commerce, other than by way of sale by auction.

5. [35]

   Section 54(2) and (3) define the concept of “acceptable quality” for the purpose of the guarantee:

6. [36]

   The first part of this definition of “acceptable quality” contains a set of quality elements in s 54(2). The present case concerns whether the goods are free from defects and safe: s 54(2)(c) and (d). The second part of the definition contains a reasonable consumer test which applies an objective evaluation of the quality elements, in accordance with the expectations of a hypothetical reasonable consumer who is fully acquainted with the state and condition of the goods, including any hidden defects: s 54(2), taking into account the factors specified in the third part of the definition referred to in s 54(3), which include any other relevant circumstances relating to the supply of the goods: s 54(3)(e).

7. [37]

   The hypothetical question of whether the reasonable consumer would have regarded the quality of the goods as acceptable is to be determined at the time of supply. However, as Wheelahan J observed in Australian Competition and Consumer Commission v Jayco Corporation Pty Ltd [2020] FCA 1672 at [27]:

8. [38]

   Part 5-4 of the ACL (“Remedies relating to guarantees”) contains Div 1 (“Action against suppliers) and Div 2 (“Action for damages against manufacturers of goods”). Subdivision A of Div 1 concerns actions by a consumer against suppliers of goods if a consumer guarantee, including under s 54, is not complied with. The remedies available to a consumer depend on whether the failure to comply with the guarantee is a major failure or not: s 259(2) and (3).

9. [39]

   The present case involves Div 2 of Pt 5-4 which concerns action by an affected person against the manufacturer of goods if the consumer guarantee under s 54 is not complied with.

10. [40]

    Section 271(1) is the remedial provision which entitles consumers to claim damages against a manufacturer for breach of the consumer guarantee in s 54:

11. [41]

    An “affected person” who may claim damages under s 272(1) is defined by s 2(1) as including: (a) a consumer who acquires the goods; (b) a person who acquires the goods from the consumer (other than for the purpose of resupply); or (c) a person who derives title to the goods through or under the consumer.

12. [42]

    VW admitted on the pleadings that it is a manufacturer of the VW Vehicles by operation of ACL, s 7(1)(e) which provides that a manufacturer includes “a person who imports goods into Australia if, (i) the person is not the manufacturer of the goods, and (ii) at the time of importation, the manufacturer of the goods did not have a place of business in Australia”.

13. [43]

    The entitlement to claim damages is qualified by s 271(2) which provides that s 271(1) does not apply if the consumer guarantee is not complied with “only because of” one of three specified circumstances:

14. [44]

    Section 272 specifies the type of the damages recoverable against a manufacturer of goods for breach of a consumer guarantee:

15. [45]

    Prior to 1 January 2011, Part V of the TPA (“Consumer Protection”) included Div 2 (“Conditions and Warranties in Consumer Transactions”) and Div 2A (“Actions against Manufacturers and Importers”). Division 2A contained s 74D which conferred on consumers a direct right of action against a corporate manufacturer of goods which were not of merchantable quality:

16. [46]

    As will be apparent from s 74D(2), the entitlement to claim damages under s 74D(1) is qualified by s 74D(2) which relevantly provides that s 74D(1) does not apply if the warranty of merchantable guarantee is not complied with “by reason of” one of two specified circumstances, occurring after the goods have left the control of the corporate manufacturer.

17. [47]

    Again, it is not in dispute that at least some of the group members who acquired a VW Vehicle prior to 1 January 2011 satisfied s 74D(1), being a consumer within the definition of that term in the TPA, s 4B, who acquired a “good” being VW Vehicles and that VW is a deemed “manufacturer” by reason of the operation of the extended definition of “manufacturer” in TPA, s 74A(4) which was in similar terms to the definition of manufacturer in ACL, s 7(1)(e): see [42] above.

The common questions

1. [48]

   The questions for determination at the trial were all those arising on the appellant’s claims and the questions common to the appellant’s and group members’ claims as identified in Annexure A to consent orders made by Sackar J on 21 June 2019 (following a judgment delivered by Sackar J on 11 December 2018: Takata Airbags Class Action – Common Questions [2018] NSWSC 1868), together with an additional common question, Q20A, identified in the consent orders made by Sackar J on 22 May 2020.

2. [49]

   It is not necessary to reproduce in full the common questions. To the extent that the appellant challenges the answers to some of the questions, those questions and answers given by his Honour are set out in the Schedule to these reasons.

Standard of appellate review

1. [50]

   The appeal to this Court is an appeal by way of rehearing under s 75A of the Supreme Court Act 1970 (NSW). Before this Court can interfere with the findings made by the primary judge, error must be established: Scenic Tours Pty Ltd v Moore [2018] NSWCA 238; (2018) 361 ALR 456 (Scenic Tours) at [255], citing Minister for Immigration and Border Protection v SZVFW (2018) 264 CLR 541; [2018] HCA 30 at [30] (Gageler J).

2. [51]

   Neither party directed submissions to the standard of appellate review which applies to the challenge to the finding that there was no breach of the guarantee of acceptable quality in ACL, s 54. Although the determination of this question involved the exercise of an (objective) evaluative judgment, the test for establishing error with respect to whether the appellant’s vehicle was of acceptable quality is the general “correctness” standard of review in Warren v Coombes (1979) 142 CLR 531; [1979] HCA 9, rather than the “deferential standard” which tolerates a range of outcomes.

3. [52]

   That is because the question of whether the appellant’s vehicle was of acceptable quality must be answered in accordance with the expectations of a hypothetical reasonable consumer applying the express statutory criterion in ACL, s 54(2) and (3), and calls for a decision where there is one correct answer, or adopting the language of Gageler J in SZVFW at [49], a “unique outcome”; see also Scenic Tours at [254]; Sun v Chapman [2022] NSWCA 132 at [8]-[13], [114]-[117] and [189].

4. [53]

   Further, whilst due respect is to be paid to the conclusions reached by the primary judge, this Court must make its own judgment as to the proper inferences to be drawn from the established facts and give effect to its own conclusions if they differ from those of the primary judge: Scenic Tours at [254], citing SZVFW at [18] (Kiefel CJ), [27], [48]-[50] (Gageler J). Although his Honour heard evidence from the appellant, neither party submitted that the appellant’s evidence informed his Honour’ evaluative judgment in a manner that justifies this Court in giving his Honour’s assessment a significant degree of respect for that reason: Scenic Tours at [255], citing SZVFW at [33] (Gageler J).

5. [54]

   At trial, the appellant’s case on liability was that the VW Vehicles necessarily carried a risk of airbag mis-deployment or rupture through PSAN degradation in the presence of moisture and temperature changes. The appellant characterised the risk as “unquantified and unquantifiable”. His case was that:

6. [55]

   The appellant sought to prove his case on liability by reference to the empirical analysis program conducted by VW AG and expert evidence from Mr Renz. It is of assistance first to refer to this evidence and his Honour’s findings before summarising the findings on acceptable quality.

7. [56]

   Mr Schade was a technical expert employed by the Product Analysis Department within the Technical Development Department of VW AG. He had been involved in the design and implementation of the empirical analysis program and the analysis of its results since its inception in February 2016.

8. [57]

   As indicated, some 20,000 airbag inflators were collected from the field from Volkswagen vehicles manufactured from 2005 onwards and from various climatic zones around the world. As of 2020, the airbags retrieved from the field represented vehicles of up to 15 years in age. The airbag inflators were subjected to several tests: (i) physical examination of the inflated module; (ii) investigation of the chemical and physical properties of the propellant; (iii) computerisation tomography scans (CT scans); (iv) deployment tests and tank tests; and (v) closed vessel analysis of pressure/time curves. Mr Schade gave evidence that there were approximately 7,000 deployment tests, 5,000 tank tests (in fact 5,423) and 7,000 closed vessel tests (in fact 7,492).

9. [58]

   Competing expert evidence was given as to the risk of mis-deployment or rupture of the airbags in the VW Vehicles. Mr Renz was called for the appellant; he was a chemist with mechanical engineering technology expertise and much experience working on automatic airbag ignitors, inflators and micro gas generators. Prof Thomas Klapötke was called by VW; he was a Professor of Chemistry at Ludwig Maximilian University in Munich, Germany.

10. [59]

    Addressing the evidence of the real-world experience of airbags in Volkswagen vehicles, his Honour made the following general findings:

11. [60]

    Turning to the specific testing as part of the empirical testing program, the evidence and his Honour’s findings may be summarised as follows.

12. [61]

    Mr Schade gave evidence that the physical examinations of the retrieved airbags involved visual examination to identify signs of changes in inflator behaviour and internal examination by deconstructing the inflator into constituent parts. The CT scans involved a non-invasive examination of the inflator and used two-dimensional, cross-sectional images, which can be combined to produce a three-dimensional image of the inflator.

13. [62]

    His Honour found that the physical examinations and the CT scans showed differences between airbag inflators fitted to Volkswagen vehicles and those fitted to other vehicles: at [80]. As his Honour said (at [81]):

14. [63]

    After noting at [82] that Mr Schade gave unchallenged evidence that as a result of this testing, the airbag inflators fitted to the Volkswagen Vehicles did not show any functionally relevant anomalies and no material manufacturing faults that would lead to rupture, his Honour found at [83] that this provided support for the submission made on behalf of VW that airbag inflators for Volkswagen vehicles are different from the inflators in other vehicles in ways that meaningfully affect their performance.

15. [64]

    Mr Schade explained that the “closed vessel” testing involved the removal of the PSAN propellant from retrieved inflators, a sample of which was then placed in a reinforced closed vessel and ignited. The burn rate of the propellant and the pressure generated by the burning of the propellant (which produces the gas which inflates the airbag cushion) was calculated. The closed vessel allowed the measurement of much higher pressures generated by the propellant, once ignited, than pressures which are reached within an inflator during a typical deployment of an airbag. The burn rate of the propellant is derived from the pressure over time curves that are produced in the closed vessel test. The derived burn rate is then compared to the normal burn rate and a predetermined “critical” integrated burn rate which is the burn rate that produces the maximum pressure that the inflator housing is designed to withstand. If the derived burn rate indicated exceeds the desired inflator housing pressure, then inflator rupture can occur. A burn rate that produces the pressure at levels at which the housing can rupture is known as the “critical burn rate”.

16. [65]

    As of 1 April 2021, there were 7,493 closed vessel tests. The results of the closed vessel testing were shown in Figure 21 to Mr Schade’s affidavit, which was reproduced as an attachment to the principal judgment. Mr Schade gave evidence that the closed vessel analysis confirmed that the inflators from Volkswagen vehicles did not reveal any critical anomalies in the burn rate of the inflator. He explained that: (i) the normal or expected integrated burn rate for PSAN was between 1,700 to 1,800 MPa*mm/s, (ii) the pressure/time curves plotted in pink and purple in Figure 21 indicated that certain older SDI field inflators retrieved from Volkswagen vehicles from hot and humid climates (Zone 1 and 2) exhibited slightly elevated burn rates, (iii) the highest recorded integrated burn rate for the SDI inflators returned from Volkswagen vehicles was measured by the closed vessel test to be 2051 MPa*mms, and (iv) the pressure/time data results demonstrate that the SDI inflators installed in the Volkswagen vehicles exhibited integrated burn rates in the non-critical range.

17. [66]

    His Honour observed that the closed vessel testing involved subjecting the PSAN in the airbag inflators retrieved from the field to pressure in a reinforced closed vessel in order to determine whether its integrated burn rate or IBR measured in megapascals per millimetre per second, exceeded a “normal” or “expected” rate of between 1,700 and 1,850 MPa*mm/s or a critical rupture rate determined by Takata to be 3,000 MPa*mm/s: at [84], [91], [97]. His Honour found that the evidence established that the airbags are designed with an industry standard 150 per cent safety margin above the normal or expected burn rate, which would imply a rupture pressure of 2,775 MPa*mm/s: at [103], [137]-[138]. His Honour noted that the two measures were broadly consistent with one another: at [138]. The evidence was that the highest test result observed was 2,051 MPa*mm/s: at [94]. His Honour accepted that this was well below the critical rupture rate, whether that figure was approximately 2,775 MPa*mm/s as suggested by Mr Renz, or 3,000 MPa*mm/s calculated by Takata: at [102], [103], [140].

18. [67]

    Two further findings in relation to the closed vessel tests should be noted: (i) the results as recorded in Mr Schade’s Figure 21 showed the PSAN was subjected to pressures up to 90 MPa, being much higher than would normally be experienced in the field as the airbags are designed so that when pressure reaches 40 to 50 MPa vent holes open and the airbag inflates: at [90], and (ii) Prof Klapötke agreed that the experiments conducted by Volkswagen show a progressive deterioration in the PSAN that exceeds the normal or expected parameters set by Volkswagen by about 8 per cent, for the older inflators, not for the new ones, and agreed that said that this was “statistically relevant, though it’s not dangerous”: at [93].

19. [68]

    The “tank” tests involved the inflator as a unit being placed in a sealed tank and sensors measured the internal and external pressure during the firing of gas generators: at [104]. Mr Schade gave evidence that the burn rate of the propellant is inferred from the tank pressure observed during the tank test. High tank pressure indicates excessively rapid burn-off behaviour.

20. [69]

    Prof Klapötke gave evidence that tank pressure and inflator chamber pressure provide additional information on the propellant burning behaviour inside the inflator because it measures the pressure which occurs inside the inflator housing and in the tank upon deployment of the inflator being tested. He said that, as a result, the performance of the propellant as it burns within the housing is measured, as opposed to the burn rate of the propellant alone, as is measured in closed vessel testing.

21. [70]

    As of 1 April 2021, 5,423 inflators were subjected to the tank tests. There was a single incident of rupture of an airbag inflator in a tank test. Mr Renz accepted in cross-examination that the single incident was an anomaly as compared to the rest of the data. Mr Schade described this incident as an “outlier”. He said that “it was not something that our data showed at the time in 2019, or its shows today”, “it is so far out that it is an outlier by the data”, that the assumed burn rate for this test was well above 3,000 (MPa*mm/s) and, “we see that all the other burn rates that we have are far away from that”. He proffered the view that “this one outlier can be due to manufacturing issues at the time of its production”.

22. [71]

    After observing that little attention was paid to the tank tests in final submissions, save that it was emphasised on behalf of the appellant that during one of these tests, one inflator ruptured (at [105]), his Honour made the following findings:

23. [72]

    In their joint report, Mr Renz and Prof Klapötke agreed that the empirical testing conducted by VW AG was done using sound scientific and acceptable practices. They also agreed that the empirical testing shows an increase in the integrated burn rate or IBR, as the inflators age, with Prof Klapötke qualifying his agreement as follows:

24. [73]

    After noting at [120] that PSAN absorbs moisture from the air, his Honour made the following findings with respect to PSAN’s propensity to degrade:

The primary judge’s reasons on acceptable quality

1. [74]

   There is no dispute as to his Honour’s concise statement of the concept of “acceptable quality” in s 54. It is convenient to reproduce the reasons at [21]-[24] in full (citations omitted):

2. [75]

   His Honour found that the results of the empirical analysis program could not lead to a conclusion that the tested airbags were unsafe: at [110]. On the contrary, he found that the results suggested that the airbag inflators installed in the VW vehicles were safe and that increased integrated burn rates revealed by the tests were unlikely to translate to aggressive or unsafe deployment of the airbags: at [111].

3. [76]

   It is not necessary to summarise all of his Honour’s detailed findings at [137]-[146] on the propensity or risk of explosion case, as much of these reasons have already been referred to at [66] above. It is sufficient to refer to the following further findings:

4. [77]

   It is not necessary to refer to the detailed findings at [147]-[151] in relation to the appellant’s “risk of mis-deployment” case, since the appellant did not press this case on appeal. His Honour summarised conclusion on the “risk of mis-deployment” case at [152]-[153]:

5. [78]

   The appellant’s case on acceptable quality was that the state and condition of the goods must be assessed in light of six facts, which his Honour set out at [155]:

6. [79]

   His Honour concluded at [159] that he was not persuaded that the appellant’s vehicle was not of acceptable quality, nor persuaded that VW had failed to comply with the guarantee of acceptable quality in s 54(1). He summarised his reasons at [156]-[158]:

Grounds 1-10

1. [80]

   Grounds 1-10 contend:

2. [81]

   Grounds 1-4, 6 and 8 contend in various ways that his Honour ought to have found that the appellant’s vehicle was not of acceptable quality. It is common ground that the outcome of the challenge to the finding on acceptable quality governs the question of merchantable quality. Ground 5 challenges the findings in relation to the significance of the empirical testing program. Ground 7 challenges the specific finding in relation to the single instance of rupture of the airbag inflator in the tank tests. Ground 9 challenges the finding concerning the nature of motor vehicles. Ground 10 challenges the answers to some common questions and depends on the outcome of the other grounds.

3. [82]

   As indicated, the references in grounds 4, 7 and 8 to the appellant’s case based on the risk of mis-deployment can be ignored, as this case was not pressed on appeal.

4. [83]

   The appellant’s submissions addressed grounds 1-10 compendiously under three headings: Volkswagen’s testing, Expert evidence, and Unacceptable quality. Not all the grounds were the subject of submissions. It is said that sub-pars (e) and (f) of [155] of the principal judgment (set out at [78] above), encapsulates the way the appellant puts his case on appeal.

5. [84]

   It is convenient to address the grounds as advanced in the appellant’s submissions in the following order: (1) the closed vessel testing, (2) the single tank test result, (3) the empirical testing program, (4) the nature of motor vehicles, and (5) whether the VW Vehicles were of unacceptable quality.

The closed vessel testing

1. [85]

   It is common ground that the closed vessel test involves observing the PSAN burning in conditions that do not exist when an airbag is ignited in the field. Airbag inflators are designed so that once the propellant is ignited and starts generating gas, the pressure within the inflator will start to stabilise once it reaches a certain point, approximately 45-50 MPa, as gas is released from the vent openings. By contrast, in a closed vessel test the propellant continues to burn to conditions where the pressure significantly exceeds 50 MPa. The closed vessel tests subjected the PSAN in the inflators to pressures of up to 90 MPa, which his Honour found was much higher than would normally be experienced in the field as airbags are designed so that when pressure reaches 40-50 MPa vent holes open and the airbag inflates: at [90].

2. [86]

   About 8 per cent of the closed vessel test results exceeded the “normal” or “expected” range of the integrated burn rate. The appellant referred to the “normal” or “expected” range as the “accepted” range as determined by Takata and adopted by Volkswagen. VW submits that it does not follow that a test result above this level was unacceptable or to be deemed dangerous. It said that the relevant comparator was the critical rupture rate, which as his Honour observed at [100], was the rate at which there was the possibility of rupture of an airbag on inflation.

3. [87]

   The appellant made two related complaints in relation to his Honour’s findings in relation to the closed vessel testing. The first is that although his Honour recognised that the testing showed that the integrated burn rate of PSAN from retrieved airbag inflators exceeded Takata’s “normal” or “expected” range of between 1,700 and 1,850 MPa*mm/s in 8 per cent of results, his Honour focused on the fact that all the results were substantially below the “critical rupture point” of 3,000 MPa*mm/s as determined by Takata: at [100]-[103]. The second is that his Honour focused on the margin between the highest observed closed vessel test result of 2,051 MPa*mm/s and either the 3,000 MPa*mm/s critical rate calculated by Takata, or the 2,775 MPa*mm/s rate implied by applying a 150 per cent safety margin to the expected range: at [138]-[139].

4. [88]

   It is said that (i) the true significance of the closed vessel testing was that it showed that the PSAN in the airbag inflators fitted to the VW Vehicles had the propensity to degrade in the presence of moisture and temperature change, and (ii) the closed vessel testing showed that degradation of PSAN, which is a progressive and irreversible process, was occurring in Volkswagen vehicles.

5. [89]

   In support of these submissions, the appellant selectively quoted from the evidence of Prof Klapötke who accepted that the 8 per cent result was statistically significant and showed a progressive deterioration in the PSAN. This omitted reference to Prof Klapötke’s full answer, that although 8 per cent is statistically relevant, “it’s not dangerous”, as his Honour recorded in the passage of cross-examination extracted at [93].

6. [90]

   The appellant also selectively quoted from the evidence of Mr Schade. It is said that he accepted that the closed vessel test results showed “that our propellant does degrade. It changes compared to what it was when it was produced”. This omitted reference to Mr Schade’s full answer, “[i]t has a slight change”, and his explanation in relation to Figure 21 concerning the closed vessel test results that the pressure rate of 45 MPa to 50 MPa is the “designed pressure” in the field (or any tank test), so that the “important” results are the IBR to the “left of 45 [MPa]” in Figure 21.

7. [91]

   There was no evidence that an integrated burn rate in excess of the “normal” or “expected” range indicates an adverse impact on function. At trial, the appellant accepted that the observed integrated burn rate for the closed vessel tests at pressure levels of 45 or 50 MPa:

8. [92]

   The appellant’s submissions ignore the critical point which his Honour recognised at [101], when addressing the implications of the integrated burn rates being above Takata’s normal or expected range: none of the closed vessel tests approached the critical rupture rate or established an adverse impact on the function of the airbag. As his Honour found, the results of the closed vessel testing did not show that any degradation in PSAN would occur to such an extent or within such a timeframe as to affect the performance of the airbag inflators in a manner that had practical consequences for the vehicles’ owners: at [140], [142], [145].

9. [93]

   There is a further point. Mr Schade explained that the IBR is calculated using the data about the rate at which the propellant burned when the pressure within the closed vessel was between 10 MPa and 65 MPa. He said that the increased pressure in the closed vessel has an impact on the burn rate of the propellant, and the burn rate will accelerate even more as the pressure increases beyond 50 MPa. This evidence was unchallenged. VW submits that given the nature of closed vessel testing, the observed burn rate in the closed vessel testing at pressure levels above 50 MPa is higher than that which would be generated if the same PSAN was ignited in real world conditions. The appellant said that “whilst the actual pressure at any point in time in an inflator in the real world may be lower than the integrated burn rate, that fact does not support a contention that there is a greater gap between actual pressure and rupture point than that suggested by comparison of the integrated burn rates” (emphasis added).

10. [94]

    The appellant misstates and misunderstands VW’s submission. The point is not one of comparison of the actual pressure in an inflator in the real world and the integrated burn rate, rather, the point is that the actual pressure in the closed vessel tests above 50 MPa is higher than would be experienced in the real world and this has an effect on the observed burn rates above 50 MPa*mm/s. Given Mr Schade’s unchallenged evidence that the burn rate will accelerate even more as the pressure in the closed vessel increases beyond 50 MPa, VW is correct to submit that even the highest observed integrated burn rate of 2,051 MPa*mm/s is not proof the if PSAN in an airbag of equivalent age and conditions were ignited in the real world, it would reach that burn rate. The burn rate would in fact be lower than the rate observed in the closed vessel testing.

11. [95]

    It is said that two circumstances demonstrate error in his Honour’s finding that the closed vessel tests did not suggest that the critical rupture rate was reached. Each is considered below.

12. [96]

    The first matter is Takata’s investigations unrelated to Volkswagen vehicles which suggested that the minimum field exposure for problems with the airbags was seven years and the majority of field ruptures occurred after ten years or more. It is said that Volkswagen's closed vessel testing was only capturing the early part of the likely period of heightened risk because this testing involved vehicles that were between 5 and 12 years old. This submission was based on Figure 23 in Mr Schade’s affidavit but ignored that the summaries and tables in Mr Schade’s affidavit are effectively points in time.

13. [97]

    As Mr Schade explained, the testing was ongoing and the database maintained by VW AG was constantly being updated (monthly as of January 2021) as more tests were conducted. The evidence at trial captured all relevant tests, including closed vessel tests, as of 1 April 2021; this is evident from the content of the electronic spreadsheets that were provided by VW’s solicitors to the appellant’s solicitors on 7 April 2021. Documents extracted from those electronic spreadsheets made good VW’s submission that the closed vessel tests included vehicles manufactured in 2006.

14. [98]

    Nor can the results obtained by Takata for other manufacturers’ vehicles be translated to Volkswagen vehicles. The evidence was that there had been no such rupture in a Volkswagen vehicle anywhere in the world. As his Honour noted at [51], 50,000 of the 440,000 Volkswagen vehicles involved in accidents were nine years or older, and thus of an age where any problem with the airbags would likely materialise. Further, there is an unchallenged finding at [74] that the airbag inflators for Volkswagen vehicles were different from the inflators in other vehicles in ways that might meaningfully affect their performance.

15. [99]

    The second matter is that some one third of Volkswagen vehicles in Australia were still on the road at 15 years. It is said that the VW AG testing does not establish what happens after that. But this ignores several important matters.

16. [100]

    First, the appellant had the onus of proof that the airbags in the VW Vehicles were unsafe. Mr Renz accepted that he could not say that degradation of PSAN would reach a point of functional significance at any identifiable point in time, let alone within the service life of any of the VW Vehicles. When asked about the 8 per cent of the closed vessel test results which exceeded the “expected” integrated burn rate and the highest test result of 2,051 MPa*mm/s which exceeded the “expected” integrated burn rate by about either 11 per cent, or 15 per cent on Mr Renz’s implied limit, Mr Renz agreed that he did not have any data to indicate that that level of exceedance will affect the process of the airbag deployment in a way that impacts on its safe operation.

17. [101]

    Second, the closed vessel testing did not disclose unacceptable results, even for vehicles up to 15 years old. The testing established that the degree of degradation of PSAN in the 8 per cent of observed cases did not interfere with their safe and effective operation. As his Honour found, none of the results came anywhere near Takata’s critical rupture point: at [102]. The results of the testing were inconsistent with the appellant’s case that PSAN degradation is inherently dangerous.

18. [102]

    Third, nor did the closed vessel testing establish the appellant’s contention that “at some stage in the life of the Volkswagen vehicles, the degradation would reach the point where rupture would occur” (emphasis added). The appellant did not demonstrate that there is degradation to such an extent as to have an adverse impact on functionality of the airbags within any meaningful timeframe. The submission about the situation getting worse at some time in the future was merely speculation.

19. [103]

    There is no basis to infer from the results of the closed vessel testing that the critical rupture point would have been reached within the lifespan of the VW Vehicles. As his Honour found at [140], the appellant did not establish any connection between the possible degradation in the PSAN and the airbag performance in Volkswagens.

20. [104]

    The challenge in grounds 2 and 3 to the findings in relation to the closed vessel testing should be rejected.

The single tank test result

1. [105]

   Ground 7 contends that his Honour erred by disregarding the single instance of rupture of the inflator that occurred in the tank tests conducted on some 5,000 inflators as part of the empirical testing program. The airbag inflator had been retrieved from a 2007 Volkswagen Polo located in Queensland, being XTLU00902851.

2. [106]

   Contrary to the appellant’s submission, his Honour did not disregard this result. It was expressly considered at [105]-[107], together with the evidence given by Mr Renz, who conceded in the passage of his cross-examination extracted by his Honour at [107], that the single result was an anomaly.

3. [107]

   It is said that the one ruptured inflator during tank testing demonstrates that there was in fact a breach of the guarantee of acceptable quality. Being a destructive test, by the time of trial all that was available was the data. This was reproduced in Mr Renz’s report in reply at par [82] in a graph which is set out below:

4. [108]

   In his report, Mr Renz attributed significance to the initial rise in pressure in the airbag inflator over the first two milliseconds, followed by a brief pause and then approximately 10 millisecond exponential rise in internal pressure, before bursting. Mr Renz said that this reflected the shape of the curve produced by mapping pressure over time in reports which concerned vehicles other than Volkswagen vehicles: the Orbital-ATK report of September 2016 and the first report of Exponent Inc of July 2016. Based on the similarities between the pattern of pressure seen in relation to the inflator retrieved from the 2007 Polo and his views as to the propensity of PSAN to degrade, Mr Renz considered that this rupture was due to the degradation of PSAN-based propellant in the inflator.

5. [109]

   Several observations can be made in relation to Figure 2 in the Renz report. The unlabelled X axis is time in milliseconds. The Y axis is labelled “Pressure” but does not give a unit, however, by comparing the graph in Figure 2 with that in the Orbital-ATK report, to which Mr Renz referred, the Y axis seems to be in atmospheres. Normal atmospheric pressure is about 101,325 pascals (Pa) or 101.325 kilopascals (kPa). By comparing the graph in Figure 6 in the Orbital-ATK report with the graph in Figure 2 prepared by Mr Renz, it would seem that the peak on Mr Renz’s graph at “900”, would, if it is measured in atmospheres, correspond to 90,000 kPa, which is consistent with the peak “runaway pressure” on airbags with degraded PSAN shown in the Orbital-ATK report.

6. [110]

   No specific submission was made at trial based on the graph in Figure 2 of the Renz report. There is, however, a deal of potential inaccuracy in Mr Renz’s graph. It shows that the pressure in the inflator (grey line) increased dramatically before the pressure in the tank (orange line) increased at all. This was not explored in the evidence or in submissions. (Given the nature of the tank test, a possible explanation might be the small delay between the rise in pressure observed in the inflator after ignition and the rise in external pressure observed in the tank, once the vent holes in the inflator open.) It also shows negative pressure for the inflator (grey line). The appellant did not address this matter, despite it being drawn to counsel’s attention in oral argument in this Court. It also refers to the “Inflator pressure” (grey line) as “Mpa”, presumably, megapascals, when the Y axis seems to be in kilopascals.

7. [111]

   It is said that Mr Renz was not cross-examined on his opinion in his report in reply (at par [84]) that this rupture was due to the degradation of PSAN-based propellant in the inflator. That misstates the position. Mr Renz was cross-examined, assuming the PSAN had degraded, as to what might have caused the PSAN to have degraded to the point of causing rupture in the particular inflator. He accepted that a production fault affecting that particular inflator that led to moisture ingress is “possible” but said that probability was “not very high”. Upon further cross-examination, Mr Renz ultimately accepted that the single instance of rupture in the tank tests was an “anomaly”. As his Honour noted at [105], little attention was paid to the tank tests in final submissions, save for Mr Renz’s acceptance that the single result was anomalous.

8. [112]

   The appellant sought to diminish Mr Renz’s answer as being “ambiguous”. But there was no ambiguity in the proposition accepted by Mr Renz in cross-examination. Having agreed with the cross-examiner that there was a “large body of data, including a large number of airbags from 2007 that have been tested and shown burn rates nowhere near the rupture level”, Mr Renz accepted the proposition that “[v]iewed in that light, this single result you would view as an anomaly as compared to the rest of the data”. The evidence did not stop there. Mr Schade gave evidence that the single instance of rupture in the tank tests was an “outlier”.

9. [113]

   An outlier is something which is not regarded as probative of the conclusion sought to be drawn from a statistical analysis because it is to be inferred that the data is corrupted or distorted in some way. Although challenged in cross-examination, Mr Schade maintained his view that it was correct to exclude this result because it “doesn’t fit into the data. It is an outlier statistic”. There was no error by his Honour in not finding that the single instance of rupture in the tank test demonstrated that the asserted vulnerability of the PSAN propellant to degradation could lead to rupture of the airbag inflator (grounds 6 and 7).

10. [114]

    It is said that the tank testing disclosed a 1:5,000 chance of failure and that his Honour’s reference to the entire empirical analysis program in which some 20,000 airbags were tested reveals a misunderstanding of the evidence. This misstates his Honour’s finding at [106]. His Honour did not say that the testing disclosed a 1:20,000 chance of failure. He merely observed that there was only one such occurrence of failure out of some 20,000 airbags that were tested in various ways.

11. [115]

    It is also said that whether the correct ratio is 1:5,000 or 1:20,000, it must be unacceptable from the standpoint of the reasonable consumer. That ignores that a single failure observed in a sample does not necessarily denote the probability of failure in the population. As VW correctly submits, to determine a probability, it would be necessary to at least generate a statistically significant number of failures and then consider that number against the total number of failures. Contrary to the appellant’s submissions, the single instance of rupture in the tank tests could not be regarded as probative of the conclusion sought to be drawn by the appellant from a statistical analysis.

12. [116]

    It is said that “the only test in which a rupture could occur was a tank test”. To the extent that this submission implied that the tank test was the only test indicating the likelihood of rupture, that is incorrect. The closed vessel testing involved measuring the integrated burn rate for PSAN in various igniters retrieved from the field and comparing the burn rates to the critical rupture pressure. None of some 7,000 tests even approached that pressure level.

13. [117]

    It is said that the manufacturing defect explanation suggested by Mr Schade for the single instance of rupture occurring in the tank test “proves the appellant’s case”. According to the submission, unless VW has a perfect manufacturing process that eliminates such flaws, the lack of acceptable quality flows from the inevitable vulnerability to such flaws in a mass-produced product when compared to alternatives. But no finding was made by his Honour concerning any manufacturing defect in the single inflator that ruptured in the tank test.

14. [118]

    Given the state of the evidence, that is unsurprising. Mr Schade’s suggestion of a manufacturing issue was no more than that, a possibility. Mr Renz said that the probability of a production fault was not very high. He also acknowledged that, being a destructive test, analysis of what caused the PSAN in this particular inflator to degrade was “impossible”.

15. [119]

    Further, as VW submits, even if a different defect at the time of manufacture allowed moisture to enter the airbag inflator that ruptured, that does not establish that all airbags that use PSAN have the alleged propensity or risk of rupture. The appellant did not suggest, nor did the evidence establish, that a different defect was present in all VW Vehicles. Ground 7 should be rejected.

Reliance on the VW AG empirical testing program

1. [120]

   Ground 5 contends that his Honour erred in finding that the results of the empirical testing program by VW AG (i) suggest that the airbags are safe, (ii) suggest that there is no reason to predict that the airbag inflators installed in the VW Vehicles would become critical or dangerous, and (iii) could not lead to a conclusion that the airbags were unsafe. This ground can be dealt with relatively briefly, given that the matters relied upon in support of this ground have been largely addressed above.

2. [121]

   There is no challenge to his Honour’s finding at [78] that there is nothing in the evidence to suggest that the empirical analysis program was not a genuine attempt by VW AG to ascertain whether there was any problem with Takata airbags installed in Volkswagens.

3. [122]

   Nor is there any challenge to the methodology adopted by VW AG. As his Honour found at [70], Mr Renz, the appellant’s expert, agreed that the program involved ample testing across different climate zones and of inflators of different ages.

4. [123]

   It is said that the testing was confined to airbags and PSAN retrieved from vehicles that were no more than 12 years old, and that the testing could not predict, nor did it even purport to predict, whether, and if so when, the airbags will become critically dangerous. This submission has been addressed above. As his Honour found at [68], the empirical testing program involved airbags retrieved from Volkswagen vehicles from 2005 onwards. Contrary to the appellant’s submission, the closed vessel tests captured vehicle up to 15 years old: see [96]-[97] above.

5. [124]

   The balance of the submission ignored the matters referred to at [100]-[103] above. The fact that the closed vessel tests recorded that, for vehicles up to 15 years old, the burn rate observed in the closed vessel testing never approached the critical rupture rate, was consistent with the real-world experience. There had been no such rupture in a Volkswagen vehicle anywhere in the world and no meaningful degradation had been found in testing on inflators from Volkswagen vehicles up to 15 years old.

6. [125]

   It is not in issue that Prof Klapötke did not attempt to model the risk of when the PSAN might become unsafe. So much is accepted by VW. But contrary to the appellant’s contention that the predictive exercise of when PSAN might become critical or dangerous in “unknowable”, his Honour accepted Prof Klapötke’s evidence that whilst the timing and speed of the degradation process with respect to PSAN is “unknowable on the basis of theoretical model of prediction”, it is possible to model when the PSAN might become unsafe: at [134]. As his Honour found at [145], the appellant has not sought to demonstrate that this risk could eventuate within any meaningful timeframe.

7. [126]

   It is said that his Honour erred in referring to the empirical testing as “fierce”. This is a misreading of his Honour’s reasons. His Honour was simply making the point that the testing conducted on over 20,000 airbags, involving both observations from the actual tests and the trend observable across those tests, put VW AG in a position to conclude that there were no systemic safety risks in relation to the VW Vehicles. His Honour did not find that the empirical testing was determinative. Rather, he rejected the appellant’s submission that the empirical analysis program could lead to a conclusion that the tested airbags were unsafe: at [110].

8. [127]

   There was no error in finding that the empirical testing suggests, very strongly, that there is no reason to predict that the airbag inflators actually installed in Volkswagen vehicles will become critical or dangerous: at [77], [136]. Ground 5 should be rejected.

The nature of motor vehicles

1. [128]

   Ground 9 contends that his Honour erred at [158] by considering it relevant to the analysis of acceptable quality under s 54 “that motor vehicles are complicated pieces of machinery that may develop problems, even going to the safety of the vehicle, that may require rectification by the manufacturer during the vehicle’s lifetime”.

2. [129]

   In writing, the appellant acknowledged that his Honour’s observation was “self-evidently correct”. In oral argument, the appellant accepted that the observation does not seem to have impacted his Honour’s decision on liability. To the extent that the appellant said that it is not apparent what his Honour’s observation was intended to suggest, this is immaterial.

3. [130]

   Given that no error has been identified, ground 9 should be rejected.

Unacceptable quality

1. [131]

   The contention that the appellant’s vehicle was not of acceptable quality was put in two ways. In chief, the appellant said that his vehicle was not safe and free from defects as the hypothetical reasonable consumer would regard as acceptable because his vehicle was manufactured with a particular physical feature, PSAN propellant in the airbag inflator, that is responsible for a risk of rupture (because PSAN is subject to progressive degradation over time in the presence of temperature changes and moisture), which in turn translates into “increased” risk of death or injury.

2. [132]

   In reply, the appellant said that it is not necessary to establish that the good has a “relevant quality or characteristic” that makes the good of unacceptable quality, rather, the “relevant characteristic” can be “an unquantifiable risk of serious harm”, at least where (i) the risk is not fanciful, (ii) arises from the application of physics and common sense in respect of a physical-chemical process which is shown to occur in real world conditions, and (iii) has not been excluded for the realistic life of the good or otherwise dealt with by appropriate disclosure and/or safeguard measures.

3. [133]

   The different way in which the argument was put in reply reflected the evidentiary difficulties confronting the argument in chief. Each argument is addressed below.

4. [134]

   On the first way the unacceptable quality argument is put, the appellant characterised the PSAN propellant as a latent defect in the airbag inflators.

5. [135]

   The appellant does not challenge the finding at [156], insofar as it concerns the appellant’s vehicle, that the hypothetical reasonable consumer would "also be acquainted with the fact that there was no reason to think that probably, 'those outcomes' [(ie rupture or mis-deployment)] will happen at some stage in any Volkswagen and thus in the [appellant’s] vehicle". The appellant accepts that “it cannot be said that it is more probable than not that rupture … will happen in respect of [his] vehicle”. But it is said that is not an answer because:

6. [136]

   Medtel Pty Ltd v Courtney (2003) 130 FCR 182; [2003] FCAFC 151 involved pacemakers which were found not to be of merchantable quality because of the increased risk of premature failure. It is said by analogy with Medtel that Volkswagen's testing suggested that the conditions required for degradation of PSAN were occurring in about 8 per cent of Volkswagen vehicles tested, which is comparable to the 5.5 per cent of pacemakers in Medtel (which had been returned to the manufacturer).

7. [137]

   The suggested analogy with Medtel is inapt.

8. [138]

   In Medtel there was evidence of an inherent risk associated with the pacemakers supplied to consumers without the particular product purchased by Mr Courtney having actually failed. The pacemakers were shown to be of greater risk of failing by reason of accelerated battery depletion than other pacemakers manufactured by the same manufacturer. The accelerated battery depletion was the consequence partial short circuits because the yellow spool solder used in the manufacturing process tended to give rise to dendritic growth: Medtel at [10]-[13]. The dendritic growth had been observed in 5.5 per cent of the pacemakers which had been returned to the manufacturer: Medtel at [14] and [30].

9. [139]

   Branson J (Jacobsen J agreeing) observed at [60] that the trial judge had found at that the use of yellow spool solder in the manufacturing process led to a “superadded risk of premature failure” by reason of partial short circuit. Her Honour rejected the notion that it would be absurd to consider a product unmerchantable where it carried “a slight but appreciable risk” of toxicity (at [72]).

10. [140]

    Her Honour emphasised that the conclusion by the trial judge that Mr Courtney’s pacemaker was unmerchantable in breach of the warranty in TPA, s 74D was not based upon its simply being a member of a class that had a statistically significant risk of failure but was, rather, based upon its exhibiting a particular physical characteristic (the yellow spool solder) that carried the risk of premature failure (at [77]). Her Honour continued at [77]:

11. [141]

    In Medtel the risk of premature failure of the battery was not a theoretical or speculative possibility. By contrast, in this case, there was no demonstrated risk of rupture (or mis-deployment) of the airbags in any of the VW Vehicles. His Honour, relevantly, found that (i) the appellant failed to establish any connection between the propensity of PSAN to degrade in the presence of moisture and temperature changes and airbag performance in Volkswagens, (ii) the appellant did not demonstrate that the asserted risk of rupture could eventuate within any meaningful timeframe, and (iii) the asserted risk was no more than a speculative possibility: at [142]-[146]. No error has been shown in these findings.

12. [142]

    It is said, unlike Protec Pacific Pty Ltd v Steuler Services GmbH & Co KG [2014] VSCA 338 at [528], that the appellant “has established on the balance of probabilities that there is a risk of rupture … in Volkswagen vehicles and that a physical feature of the vehicle (the use of PSAN as propellant) is responsible for that risk”. The submission continued:

13. [143]

    Protec involved a claim for damages for misleading or deceptive conduct based on representations concerning the suitability of a product, Bekaplast lining, a high density polyethene (HDPE) used as a long-term containment liner for concrete solvent extraction tanks at a mine site. By contrast with the circumstances of Medtel, the Victorian Court of Appeal (Tate, Santamaria and Kyrou JJA) concluded in Protec at [528], that the evidence did not establish, on the balance of probabilities, that there was a risk of premature failure nor that there was any physical feature of HPDE that was responsible for that risk. In distinguishing Medtel, the Court in Protec said at [528]:

14. [144]

    The appellant contends that a possibility of a quality that would render the good unacceptable, is sufficient to establish a contravention of s 54. But to describe the process of degradation of the PSAN propellant, if exposed to temperature changes and moisture, as one which is known to be “progressive and irreversible”, does not demonstrate that a risk of rupture is present in the airbags in the VW Vehicles or the nature of such risk. As indicated, there was no materialisation of the asserted risk of rupture within the timeframes tested by VW AG, which included vehicles up to 15 years old. That accorded with the real-world experience that there had been no such rupture in a Volkswagen vehicle anywhere in the world.

15. [145]

    Contrary to the appellant’s submission, the closed vessel tests and the tank tests did not establish the “relevant, functionally significant propensity of PSAN to degrade in Volkswagens”. His Honour found at [136] that the testing by VW AG suggests very strongly that there is no reason to predict that the airbag inflators installed in the VW Vehicles will become critical or dangerous. Further, in the summary of his conclusions at [14], his Honour found that “the appellant has not established any relevant, functionally significant, propensity of the PSAN to degrade in the airbag installed in the appellant’s vehicle so as to cause it to explode [ie rupture] …”. No error has been shown in these findings.

16. [146]

    Accepting that it follows from Medtel that it is permissible to demonstrate unacceptable quality on the basis of an inherent risk associated with a type of product, without the risk having actually materialised, it is first necessary to establish that the type of product carries the inherent risk alleged. If so, it is then necessary to consider the comparison required by s 54(2).

17. [147]

    In this case, the inherent risk alleged concerns the functional impact of the airbags on use and safety of the VW Vehicles; it is said that there is an increased risk of rupture of the airbags in the VW Vehicles. But the appellant failed to prove that the airbags in the VW Vehicles carried the inherent risk alleged. Mr Renz accepted that he could not say that degradation of PSAN would reach a point of functional significance at any identifiable point in time, let alone within the service life of any of the VW Vehicles: see [100] above. Nor does it assist the appellant’s case to describe the alleged risk as “unquantified and unquantifiable”.

18. [148]

    There was no error in the finding at [143] that the appellant failed to prove, on the balance of probabilities, that the alleged propensity or risk of PSAN to degrade in the presence of moisture and temperature changes would have materialised in any Volkswagen vehicle at any identifiable point in time.

19. [149]

    The argument that a “relevant characteristic” which makes the good of unacceptable quality can be an “unquantifiable risk of serious harm”, proceeds on the basis that:

    1. (1)

       the appellant did not show that an integrated burn rate sufficient to rupture the airbag having a particular physical consequence would be generated at an identifiable point in time;

    2. (2)

       the very fact that no identifiable point in time can be predicted for any given vehicle is part of what makes the vehicle of unacceptable quality; and

    3. (3)

       it does not matter that the appellant did not prove if and when the airbag in his vehicle would rupture or mis-deploy.

20. [150]

    As to (1) and (2) above, it is not in dispute that PSAN may absorb moisture and that this may contribute to a process of degradation which increases the burn rate for PSAN. At trial, the parties diverged as to whether the appellant had shown anything more than the theoretical potential for a form of aggressive burning of PSAN that can cause rapid inflation of an airbag in such a way as to cause a rupture. As indicated, Mr Renz accepted that he could not say that degradation of PSAN would reach a point of functional significance at any identifiable point in time, let alone within the service life of any of the VW Vehicles.

21. [151]

    Nor did the empirical testing conducted by VW AG prove the appellant’s contention that the degree of moisture absorption that affected Takata airbags in Volkswagen vehicles gave rise to the alleged propensity or risk that the airbags would rupture in a manner that causes harm to vehicle owners. As his Honour found at [142], the appellant failed to establish any connection between possible degradation in PSAN and airbag performance in Volkswagens. The contention that the “unquantifiable risk of serious harm” is not fanciful should be rejected. As his Honour found at [146], the risk that the appellant alleged was no more than a speculative possibility.

22. [152]

    As to (3) above, the appellant sought to prove that his vehicle was of unacceptable quality because it was manufactured with a particular feature (PSAN propellant in the airbag inflators) that was responsible for an increased risk of rupture compared with airbags not manufactured using PSAN. For the reasons given above, the appellant failed to do so.

23. [153]

    It is said that the unquantifiable risk makes the appellant’s vehicle of unacceptable quality because “the risk has not been excluded for the realistic life of the good”. This reverses the onus of proof. It was for the appellant to demonstrate, on the balance of probabilities, that his vehicle was of unacceptable quality. As his Honour found at [145], the appellant had not sought to demonstrate that the alleged risk could eventuate during any meaningful timeframe.

24. [154]

    It is said that contrary to his Honour’s finding at [153], there was evidence that the risk of rupture/mis-deployment was real and meaningful to a consumer. Section 54 requires an assessment of whether there is an aspect of the good that a reasonable consumer would regard as making the good not of acceptable quality. The focus on the factors in ss 54(2) and (3), which direct attention to the use and nature of the goods, emphasises that the inquiry is directed to the extent to which the goods in fact fulfil the reasonable expectations of a reasonable consumer. Relevantly, in this case, s 54 directs attention to the practical consequences for a consumer of the functional impact of the airbags on the use and safety of the vehicle.

25. [155]

    A reasonable consumer fully acquainted with the state and condition of the goods (including any hidden defects of the goods), would regard as acceptable having regard to the matters in s 54(3), that the functionality of the airbag for the use and safety of the appellant’s vehicle was the same as compared with other airbags. A merely speculative theoretical possibility of a risk of rupture of the airbag not within any meaningful timeframe, would not be regarded as unacceptable by a reasonable consumer.

26. [156]

    There was no error by his Honour in finding at [156] that the hypothetical reasonable consumer would also be acquainted with the fact that there was no reason to think that, probably, the outcomes of rupture or mis-deployment “will happen at some stage” in any Volkswagen and thus in the appellant’s vehicle. His Honour was correct to conclude that the appellant had failed to establish a breach of the consumer guarantee of acceptable quality in s 54.

27. [157]

    Grounds 1-4, 6 and 8 should be rejected.

Answers to common questions

1. [158]

   Given the conclusions on grounds 1-9, the contention in ground 10 that his Honour erred in answering questions 3, 4, 10 and 11 in the negative should be rejected.

Other grounds

1. [159]

   The above conclusions are decisive of the appeal. While not strictly necessary to consider the other grounds relating to the defence under s 271(2) and damages, this is a case which in accordance with Kuru v State of New South Wales (2008) 236 CLR 1; [2008] HCA 26 at [12], taking into account the remarks in Boensch v Pascoe (2019) 268 CLR 593; [2019] HCA 49 at [7]-[8], it is appropriate to deal with the grounds which are not decisive.

2. [160]

   Section 271(2) is a proviso in the nature of a defence, being a justification or exception. It is common ground that VW bears the onus of proving that it applies: Capic v Ford Motor Vehicle Company of Australia Pty Ltd [2021] FCA 715; (2021) 154 ACSR 235 at [741] (Perram J). The same view was taken of the proviso in the predecessor provision in TPA, s 74D(2): Effem Foods Ltd v Nicholls [2004] NSWCA 332; (2004) ATPR 42-034 at [11] (Handley JA, Tobias JA and Palmer J agreeing).

3. [161]

   The Explanatory Memorandum to the Trade Practices Amendment (Australian Consumer Law) Bill (No 2) 2010 (the 2010 Explanatory Memorandum) states at par 7.126 that s 271(2) “ensures that manufacturers are not held liable for issues with goods that are beyond their control”. The three exceptions in s 271(2) address different subject matters: sub-par (a) deals with the acts, defaults, omissions or representations by persons, other than the manufacturer, that might affect compliance with the statutory guarantee; sub-par (b) deals with events independent of human control provided they occurred after the goods left the control of the manufacturer; and sub-par (c) deals with the price charged by the supplier, that might affect compliance with the statutory guarantee.

4. [162]

   VW pleaded the exception in s 271(2)(a) in answer to the claims of the appellant and group members for any non-compliance with the consumer guarantee in s 54 (which was denied). The particulars of this defence included that Takata was responsible for the design, development and production of the Takata airbags installed in the VW Vehicles.

5. [163]

   VW relied on Mr Schade’s evidence that airbag inflators are a complex subsystem within a vehicle and require specialist expertise to design, develop, manufacture and test, and in acquiring these systems from third party suppliers, VW AG relied on their expertise and also required suppliers of airbag systems to meet quality and safety requirements for the systems specified by VW AG. Mr Schade gave evidence that:

6. [164]

   VW said that neither VW AG nor VW was responsible for the design of the airbags, and they both relied on Takata. To the extent that the use of PSAN in the Takata airbag inflators constituted a relevant defect, it was said that defect with the goods only arose because of the acts of Takata in designing and building the airbag inflators using PSAN.

7. [165]

   The appellant submitted that the exception in s 271(2)(a) did not apply because:

The primary judge’s reasons

1. [166]

   His Honour made the following findings concerning the design and manufacture of the airbags in the VW Vehicles at [35]-[38]:

2. [167]

   None of these findings are challenged on appeal.

3. [168]

   On the assumption that the appellant’s vehicle was not of acceptable quality, his Honour found at [173] that the guarantee under s 54 does not apply by reason of the exception in s 271(2)(a). It is convenient to set out in full his Honour’s reasons at [162]-[170]:

Grounds 11-14

1. [169]

   Grounds 11-14 contend:

2. [170]

   Ground 11 is a general ground. Grounds 12 and 13 challenge his Honour’s construction of s 271(2). Ground 14 incorrectly states that the answer given to question 18 was “yes” when the answer was “[g]iven the answers to Questions 3, 4 and 11, it is not necessary to answer this question”. Ground 14 depends on the outcome of the preceding grounds.

3. [171]

   It is convenient first to address the causal relationship required by the phrase “only because of” in s 271(2).

4. [172]

   The phrase “only because of” is a relational term. The approach to construing relational terms is explained by French CJ in R v Khazaal (2012) 246 CLR 601; [2012] HCA 26 at [31]. In s 272(2), the phrase refers to a causal connection between the non-compliance with the guarantee of acceptable quality and the specified matter in the three exceptions under s 271(2). The requisite nexus between the non-compliance with the consumer guarantee of acceptable quality and the specified matter in s 271(2) is that the latter must be the “only” reason for the failure of the good to comply with the consumer guarantee. Relevantly, in this case, s 271(2)(a) requires that an act, default or omission of any person, other than the manufacturer or its employee or agent, must be the “only” reason for the failure of the good to comply with the consumer guarantee.

5. [173]

   This is to be contrasted with the phrase “by reason of” in the predecessor provision in TPA, s 74D(2)(a), which is an expression that generally requires a causal connection between two matters, but it need not be direct or proximate: see, for example, in other statutory contexts, Waters v Public Transport Corporation (1991) 173 CLR 349 at 401; [1991] HCA 49 (McHugh J); Republic of Croatia v Sneddon (2010) 241 CLR 461; [2010] HCA 14 at [22] (French CJ), [53], [69] (Gummow, Hayne, Crennan, Kiefel and Bell JJ).

6. [174]

   No temporal limitation applies to the exception in s 271(2)(a) for an act, default or omission of another person, unlike TPA, s 74D(2) which explicitly imposed a temporal limitation on the specified matters in sub-pars (i) and (ii) of s 74D(2) ((i) an act or default of any person, or (ii) a cause independent of human control “occurring after the goods have left the control of the manufacturer” (emphasis added)). By contrast, this temporal limitation continues to apply to the exception in s 271(2)(b) for a cause independent of human control, and it also applies the same exception in s 259(5) for a supplier of goods.

7. [175]

   The absence of any temporal limitation in s 271(2)(a) compared to s 271(2)(b) and s 259(5), together with the drafting change to sub-par (i) of TPA, s 74D(2) strongly suggests that it is not possible to read the qualification in s 271(1)(b) and s 259(5) into the exception in s 271(2)(a). The exception in s 271(2)(a) is satisfied where the only cause of non-compliance with the consumer guarantee in s 54 is an act, default or omission of any person, other than the manufacturer or its employees or agents, occurring before or after the goods left the control of the manufacturer.

8. [176]

   The obligation under s 271 is imposed on the manufacturer of the goods. “Manufacturer” has an extended definition in s 7(1): see [42] above. As the 2010 Explanatory Memorandum states at par 2.37:

9. [177]

   It is common ground that both VW AG and VW is a manufacturer. VW AG is “a person who … produces … or assembles goods” within s 7(1)(a). And, as the importer of the VW Vehicles into Australia at a time when VW AG did not have a place of business in Australia, VW is taken to be a manufacturer within s 7(1)(e). No point was taken by VW at trial that the definition of “manufacturer” in s 7(1)(e) does not expressly state that the importer is taken to be the manufacturer for all purposes of the ACL, such as in s 271(2).

10. [178]

    In this Court, VW said that there may be a distinct question of construction of s 271(2)(a) in an action against the importer, who is taken to be a manufacturer by operation of s 7(1)(e), if the act, default or omission that made the goods non-compliant with the consumer guarantee includes that of the actual manufacturer located outside Australia. VW said that it is not necessary to determine this question as it does not arise in this case because neither VW AG nor VW was relevantly responsible for the defect in the VW Vehicles.

11. [179]

    However, in oral argument, VW implicitly conceded that in an action against the importer, who is taken to be a manufacturer by operation of s 7(1)(e), the reference to the “manufacturer” in s 271(2) included both the actual manufacturer located overseas and the importer of the goods. That should be accepted. In the circumstances of this case, the reference to the “manufacturer” in s 271(2)(a) should be read distributively to include both the actual manufacturer, VW AG, and the importer, VW. That is for several reasons.

12. [180]

    Ordinarily the same meaning should be given to the same words appearing in different parts of a statute unless there is reason to do otherwise: The King v Jacobs Group (Australia) Pty Ltd (formerly known as Sinclair Knight Merz) [2023] HCA 23 at [25], referring to Registrar of Titles (WA) v Franzon (1975) 132 CLR 611 at 618 (Mason J); [1975] HCA 41. Neither party suggested that there was a reason to do otherwise in s 271(1) and (2) with respect to the word “manufacturer”.

13. [181]

    The word manufacturer in s 271(2) is to be construed in light of the context, the general purpose and policy of s 271 which is to give consumers rights and protections that previously (prior to the TPA) were not available against a manufacturer. The extended definition of manufacturer in s 7(1) is flexible enough to refer in s 271(1) and (2) to both the actual manufacturer located outside Australia and the importer of the goods, where the circumstances mentioned in s 7(1)(e) are satisfied.

14. [182]

    One should strive to read provisions of a statute in a harmonious way: Project Blue Sky Inc v Australian Broadcasting Authority (1998) 194 CLR 355; [1998] HCA 28 at [69]-[71]. A narrow meaning of the word “manufacturer” in s 271(2) which excludes the actual manufacturer located outside Australia, when the importer satisfies the extended definition of manufacturer, would not be consistent with the purpose of s 271(1) which is to make “the” manufacturer liable to compensate the consumer for the failure to comply with the consumer guarantee of acceptable quality in s 54.

Application of the “other person” exception in s 271(2)(a)

1. [183]

   The appellant relies upon two reasons which are said to demonstrate error in his Honour’s finding that the assumed defect only arose because of the act of Takata in deciding to use PSAN as the propellant: at [164].

2. [184]

   Ground 12, as explained in written submissions, contends that the effect of his Honour’s construction of s 271(2)(a) is to transform manufacturer liability for breach of consumer guarantees from strict liability into fault-based liability, if and to the extent that subcontractors are engaged. It is said that this would allow an easy means for manufacturers to circumvent the consumer guarantees.

3. [185]

   This ground misstates the exception in s 271(2)(a). The exception directs attention to the fact or facts which made the goods of unacceptable quality under s 54. Having identified that fact or facts, the question is whether the goods were not of acceptable quality only because of, relevantly, an act, default or omission of any person, other than the manufacturer or an employee or agent of the manufacturer.

4. [186]

   The exception in s 271(2)(a) draws the line at the conduct in respect of which the manufacturer is strictly liable as its own conduct and that of its employees or agents. It was not suggested that Takata was an agent of VW AG. It was common ground at trial that Takata was a third party supplier to VW AG. The words in s 271(2)(a) are clear and must be given effect according to their terms without the unwarranted gloss suggested by the appellant. If the sole cause of what made the goods non-compliant with the consumer guarantee in s 54 is an act, default or omission of a third party supplier to the manufacturer, there is no reason to deny the operation of the exception to a manufacturer. Ground 12 should be rejected.

5. [187]

   Ground 13 contends that non-compliance with the guarantee could be “only because of” the act, default or omission of another if the manufacturer had no control over the good between the relevant act, default or omission and the supply. The matter was put a little differently in writing. It is said that the exception in s 271(2)(a) only applies when the breach of the consumer guarantee is triggered solely by matters in respect of which the manufacturer had no “involvement or responsibility” in what made the goods non-compliant with the consumer guarantee.

6. [188]

   The appellant said that the “manufacturer” had at least the following “involvement” in the non-compliance with the consumer guarantee in respect of the Takata airbags: (i) it arranged for Takata to manufacture the airbags and permitted Takata to select the propellant (and presumably accepted proof from Takata that the airbag module was designed so as to ensure consistent functional safety over the service life of the vehicle); (ii) it installed the airbags so manufactured into the vehicles; and (iii) it released each vehicle for supply.

7. [189]

   The “release” argument was put in these terms in the appellant’s submissions in reply:

8. [190]

   The argument is based on the acts of VW AG in “choosing” Takata to manufacture the airbags, in “installing” the airbags as manufactured by Takata, and in “releasing” the VW vehicles in that state. But the argument confuses the steps taken by VW AG in obtaining and installing components manufactured by third party suppliers and releasing the goods in that state for supply to consumers, which are a necessary precondition of liability for any failure to comply with the consumer guarantee of acceptable quality, with the question of whether the vehicle was non-compliant with the consumer guarantee only because an act, default or omission of any person, other than the manufacturer. A similar distinction was drawn by Gummow J in Elna Australia Pty Ltd v International Computers (Aust) Pty Ltd (No 2) (1987) 16 FCR 410 at 419 between necessary preconditions and an effective cause in the context of claims for “loss or damage” under TPA, s 82(1).

9. [191]

   There was no error by his Honour in rejecting the “choice” or “selection” of third party supplier argument at [170]. The circumstance that VW AG “chose” Takata as its airbag supplier was not what made the VW Vehicles of unacceptable quality, assuming it constitutes a relevant defect in the Takata airbags to have used PSAN as the propellant.

10. [192]

    Nor was there an error by his Honour in rejecting the installation argument at [165]. The circumstance that VW AG installed the Takata airbags is not what made the VW Vehicles of unacceptable quality, on the present assumption. As his Honour noted at [165], there was no suggestion that the manner of airbag installations was causative of the appellant’s vehicle not being of acceptable quality. The related argument that VW AG “chose” to have the particular component (the airbag module) installed in the vehicle, is another way of putting the installation argument.

11. [193]

    Similarly, there was no error by his Honour in rejecting the “supply” argument. The circumstance that VW AG released the goods to VW for supply to consumers is not what made the goods non-compliant with the consumer guarantee, on the present assumption. As his Honour observed at [167], the premise of the exception in s 271(2)(a) is that the manufacturer has supplied goods to a consumer.

12. [194]

    The argument that VW AG “permitted” Takata to select the propellant was not put in those terms at trial. But in any event, this ignores the unchallenged findings at [38] that the use of PSAN as the propellant was not specified by VW AG, and the design choice to use PSAN was made by Takata.

13. [195]

    It is said that VW AG had “involvement or responsibility” in what made the goods non-compliant with the consumer guarantee because it “presumably” accepted proof from Takata that the airbag module was designed so as to ensure consistent functional safety over the service life of the vehicle, as required by the “lifetime requirements” specification. This was a reference to the “lifetime requirements” specification in relation to the airbag module contained in the Book of Requirements which VW AG gave Takata. But it was not put to Mr Schade in cross-examination that VW AG omitted to take any relevant step in this regard. Nor was it suggested in submissions that there was any default or omission by VW AG in accepting proof from Takata that the airbag was designed to meet the “lifetime requirements” specification.

14. [196]

    No error has been demonstrated in his Honour’s contingent finding that the exception in s 271(2)(a) applied. Grounds 11 and 13 should be rejected. Ground 14 challenging the answer to question 18 should also be rejected.

15. [197]

    Common questions 17, 19 and 20A articulated the damages claimed on behalf of group members under two heads of damage:

    1. (1)

       Questions 17 and 19 asked, “can the Group Members have suffered any loss or damage, in the form of a reduction in the ‘true value’ [of the vehicle] as at the date of purchase, attributable to the [vehicle] not being of” “merchantable quality” (Q17) or “acceptable quality” (Q19), in circumstances in which the airbag has been or can be replaced at no cost to the Group Member?

    2. (2)

       Question 20A asked, “can the Group Members have suffered any loss or damage, in the form of disappointment/ distress and/ or anxiety”, in circumstances in which the airbag has been or can be replaced at no cost to the Group Member?

16. [198]

    In final written submissions at trial, three potential approaches to the appellant’s claim for damages under s 272(1) were identified:

    1. (1)

       reduction in value: award the appellant the difference between the price paid and the “true value” at the time (of supply);

    2. (2)

       “left in the hand”: award the appellant the difference between the price paid and the current value of the vehicle; and

    3. (3)

       “do the best”: the Court should “do the best” to arrive at a damages figure by an award of either:

17. [199]

    It is not necessary to refer to his Honour’s reasons for rejecting the appellant’s claim under s 272(1)(b) for damages for disappointment, distress or anxiety, as there is no challenge to this finding.

The primary judge’s reasons

1. [200]

   His Honour rejected the appellant’s claim under s 272(1)(a) on two bases.

2. [201]

   First, the appellant was only entitled to damages for loss actually suffered because the question posed by s 272(1)(a) is what damage the appellant has now suffered “for” any reduction in value below the purchase price “resulting from” any failure by VW to comply with the s 54 guarantee. Reference was made to Kizbeau Pty Ltd v WG & B Pty Ltd (1995) 184 CLR 281 at 291; [1995] HCA 4 at [190], [193].

3. [202]

   Applying this approach, his Honour concluded at [195] that the appellant has not suffered any financial loss, giving five reasons: (i) the airbag has since been replaced at no cost; (ii) the appellant has had full use of the vehicle since he acquired it; (iii) there has never been any performance issue with the airbag associated with the alleged defect; (iv) the appellant has not been involved in any accident so as to trigger deployment of the airbag; and (iv) the appellant continues to own a vehicle that is as valuable as an ordinary car of that age and type: at [196]-[201]. His Honour found that “[t]he effect of such replacement [of the airbag] is that such defect, as may hitherto have affected the vehicle’s value, has been removed”: at [205].

4. [203]

   Second, the appellant had not proved damage in any of the three ways in which the appellant put his claim in final submissions: (1) reduction in value, (2) “left in the hand”, or (3) the court should “do the best” to arrive at a damages figure: at [184]-[187].

5. [204]

   As to (1), the appellant’s case was that a discount of about 30 per cent to the purchase price should be applied, had the true facts been known. This discount was based on Prof Michelle Baddeley’s survey-type evidence that the market price of affected new Volkswagen vehicles would have been about 30 per cent lower than the actual price, had the risks associated with the Takata airbags been disclosed to the market.

6. [205]

   His Honour gave detailed reasons for finding at [223] that no weight could be given to the discrete choice experiments conducted by Prof Baddeley, including the fundamental flaws in her work (at [230]), and the methodological problems which caused the surveys to be “confusing” (at [249]-[251]), none of which reasons were challenged on appeal. His Honour concluded at [252] that the experiments were “deeply flawed”, of no value to the task at hand, and cast no light on the “true value” of the appellant’s vehicle at the time he purchased it.

7. [206]

   As to (2), his Honour rejected the “left in the hand” claim because: (i) it would put the appellant in the same financial position as he was when his vehicle was a brand new car, (ii) the argument was first raised in closing submissions, was not pleaded, was not opened on and was not the subject of any evidence, and (iii) the framework of s 272 is not apt to accommodate damages calculated this way: at [259]-[262].

8. [207]

   As to (3), his Honour rejected each of the suggested “do my best” approaches because they did not take account of the fact that such “defect” as may have been present in the appellant’s vehicle had been remedied, and the approaches assumed that the appellant is now in a position of having suffered financial loss when he had not: at [265]-[267].

Grounds 15-20

1. [208]

   The appellant did not press ground 15 which contends that his Honour erred in his approach to the award of damages under s 272(1).

2. [209]

   Grounds 16-20 contend:

3. [210]

   These grounds fall into two categories. Some grounds challenge the rejection of the appellant’s individual claim. In this category, grounds 16, 17 and 18(a) and (b) challenge the rejection of the reduction in value claim, alternatively, ground 18(c) challenges the rejection of the “left in the hand” claim, ground 18(d) was not the subject of specific submissions, and grounds 18(e), (f) and (g) advance a new claim for consequential loss under s 272(1)(b) for excess financing costs and excess taxes.

4. [211]

   Other grounds are relied upon in support of both the appellant’s individual claim and the claims of the group members. Ground 19 raises an issue of construction as to whether subsequent events may be taken into account in assessing any reduction in value under s 272(1)(a). Grounds 18(e), (f) and (g) relating to the claim for consequential loss under s 272(1)(b) are also relied upon on behalf of group members. Ground 20 challenges the answers to questions 17, 19 and 20A, and depends on the outcome of the preceding grounds.

5. [212]

   It is convenient to address these grounds in the following order: (1) the proper approach to the assessment of reduction in value damages under s 272(1)(a), (2) the appellant’s individual claim, and (3) the new damages claim under s 272(1)(b).

The proper approach to the assessment of reduction in value damages

1. [213]

   Section 271(1) creates a statutory cause of action. It entitles affected persons to damages from the manufacturer for non-compliance with the consumer guarantee of acceptable quality in s 54. Section 272(1) states the type of damages a consumer is entitled to claim under s 271, being any reduction in value of the goods “resulting from” the failure to comply with the statutory guarantee (s 272(1)(a)) and any other reasonably foreseeable loss or damage, including consequential loss (s 272(1)(b)).

2. [214]

   The word “damages” in ss 271 and 272 is used in the sense of statutory compensation. Two issues of construction arise in relation to s 272(1)(a):

   1. (1)

      the point in time for assessing damages for any reduction in the value of the goods; and

   2. (2)

      whether events or information obtained after the time of supply can be taken into account.

Time for assessment of any reduction in value

1. [215]

   It was and is common ground between the parties that any reduction in value of the goods is to be assessed by reference to the time of acquisition of the goods, that is, the time of supply. That is consistent with the statutory language in s 272(1)(a), given that each of the integers by which reduction in the value of the goods is assessed is a price referable to the time of supply. It also accords with views expressed in:

2. [216]

   After this Court reserved judgment on the appeal, the Full Federal Court expressed a qualification to the time for assessment of reduction in value damages in Toyota Motor Corporation Australia Ltd v Williams [2023] FCAFC 50; (2023) 408 ALR 582. Although Moshinsky, Colvin and Stewart JJ accepted that “generally” or in “most cases” it will be appropriate to assess damages for any reduction in the value of the goods by reference to the time of supply (at [98], [100]), their Honours said that, depending on the circumstances, the assessment of whether the claimant has suffered loss or damage “may require” “a departure from the time of supply or an adjustment to avoid over-compensation” (at [99]).

3. [217]

   Two main reasons were given why this may be necessary. One is that the word “damages” in s 272 imports a requirement that the applicant for relief must have suffered loss or damage, and this requirement is not to be displaced by the “explicit or implicit reference to the time of supply” in s 272(2)(a)(i) and (ii) (at [99]). The other is that, like the cases involving general law damages, the overarching consideration to award appropriate compensation may require flexibility in applying the general rule that damages are assessed as at the time of supply (at [102]-[103]).

4. [218]

   It is not necessary to consider the qualification stated in Toyota v Williams that a “departure” from assessment of damages under s 272(1)(a), at the time of supply, is appropriate in some cases. Although the decisions in the Toyota cases were drawn to this Court’s attention by the appellant (with the consent of VW), neither party sought leave to withdraw their submissions, which were common ground, that the assessment of damages for any reduction in the value of the goods is by reference to the time of supply.

Whether subsequent events can be taken into account

1. [219]

   Assuming the use of PSAN in the airbag inflator was a latent defect, the parties diverged as to whether his Honour erred in having regard to the subsequent replacement of the airbag, at no cost, when assessing whether the appellant had suffered damage under s 272(1)(a). The essential difference between the parties is whether in assessing statutory damages of the kind referred to in s 272(1)(a) of the ACL, it is an error to have regard to common law principles for the assessment of damages.

2. [220]

   The appellant accepted that to take account of the replacement of the airbag at no cost would be fatal to his case. In resisting that approach, which the primary judge adopted, the appellant said that the language of s 272(1)(a) does not permit an assessment of any reduction in value which takes account of the goods having been being “fixed”.

3. [221]

   It is said that (i) general law notions of compensation should not be imported into s 272 by the use of the word “damages” in that provision; (ii) disregarding subsequent events coheres with s 271(6); (iii) any reduction in value “can happen theoretically”; and (iv) the replacement of the airbag was simply a benefit which the appellant received in mitigation of loss and should be brought to account.

4. [222]

   No separate or different argument was advanced by the appellant in relation to the words “loss or damage” and “compensation” in s 74D(1) of the TPA.

5. [223]

   VW submits that the analysis of the primary judge was plainly correct: any notional reduction in value based on abstract propositions could not be shown to have caused the appellant any loss in circumstances where the appellant had experienced no detrimental impact of any supposed defect and the defect had been wholly rectified, at no cost, before any risk of manifestation had arisen.

6. [224]

   It is said, by analogy with the High Court’s approach in Kizbeau to assessing damages under s 82 of the TPA, that the primary judge was right to take into account the subsequent replacement of the airbag at no cost in analysing whether the appellant was entitled to damages under s 272(1)(a).

7. [225]

   It is well-established that “[t]he primary task is to construe the relevant statutory provisions, rather than to draw an analogy with claims under the general law”: Jonval Builders Pty Ltd v Commissioner for Fair Trading (2020) 104 NSWLR 1; [2020] NSWCA 233 at [35] (Leeming JA, Bathurst CJ and Meagher JA agreeing), referring to the joint judgment in Murphy v Overton Investments Pty Ltd (2004) 216 CLR 388; [2004] HCA 3 at [44]. The same point was made in I & L Securities Pty Ltd v HTW Valuers (Brisbane) Pty Ltd (2002) 210 CLR 109; [2002] HCA 41 at [84] (McHugh J) and Marks v GIO Australia Holdings Ltd (1998) 196 CLR 494; [1998] HCA 69 at [32], [38] (McHugh, Hayne and Callinan JJ).

8. [226]

   Nevertheless, the assessment of damages for the relevant statutory purpose may be informed by a consideration of common law principles for the assessment of damages, particularly where those principles relate to a claim with a close analogy to the claim being considered. In such circumstances, general law principles may provide “useful guidance” or be an “appropriate guide”: Henville v Walker (2001) 206 CLR 459; [2001] HCA 52 at [18] (Gleeson CJ), [130] (McHugh J). As Gleeson CJ explained at [18], in the context of claims for damage under s 82 of the TPA, general law analogies may provide useful guidance:

9. [227]

   The statutory cause of action in ACL, s 271 and its predecessor TPA, s 74D(1), does not operate on any contract. However, as Perram J said in Capic at [891], by reference to the observations of Edelman J in Moore v Scenic Tours Pty Ltd (2020) 268 CLR 326; [2020] HCA 17 at [63]-[67] in relation to the similar provisions of the ACL (s 267(3) and (4)), s 272(1)(a) and (b) resemble the two components of compensatory damages available at common law for breach of contract: (1) compensation directly for the performance interest, and (2) compensation for consequential losses. Perram J observed in Capic at [891] that this might suggest that the point of an award of damages is to put the claimant “in the position she would have been in if the [thing acquired] had been of acceptable quality”. The appellant did not argue to the contrary.

10. [228]

    The task of the Court under statutory provisions such as 272(1) “is to select a measure of damages which conforms to the remedial purpose of the statute and to the justice and equity of the case”: Henville v Walker at [18] (Gleeson CJ). The 2010 Explanatory Memorandum at par 7.120 notes that the primary contact point for consumers when they buy goods is the supplier (against whom remedies are provided in Div 1 of Pt 5-4 of the ACL). The purpose of the statute in Div 2 of Pt 5-4 of the ACL, so far as presently relevant, is to provide consumers with a remedy against the manufacturer of goods when goods fail to meet the standard required by the guarantee of acceptable quality.

11. [229]

    Section 272(1)(a) entitles an affected person to recover compensation directly against the manufacturer for the performance interest that the goods meet the standard required by the guarantee of acceptable quality. By analogy with contract damages, the aim of this component of damages is to provide the consumer with the difference between the value of what was promised under the consumer guarantee of acceptable quality and the value of what was received: Moore v Scenic Tours at [64] (Edelman J).

12. [230]

    The measure of damages under s 272(1)(a) is the difference between the true value of the good as at the time of supply (resulting from the failure to comply with consumer guarantee) and the price paid for it, or the average retail price, whichever is lower. Given this comparison between the price paid and the true value, as distinct from market value, subsequent events are potentially relevant because they may illuminate or indicate or reflect the true value at the time of supply.

13. [231]

    The consideration of subsequent events in the assessment of damages is not unusual or novel. In HTW Valuers (Central Qld) Pty Ltd v Astonland Pty Ltd (2004) 217 CLR 640; [2004] HCA 54, the joint judgment observed at [39] that Kizbeau at 291-296 (Brennan, Deane, Dawson, Gaudron and McHugh JJ) pointed out that, in many fields of law, assessments of compensation or value at one date are commonly made taking into account all matters known by the later date when the Court’s assessment is being carried out. That included, as Kizbeau noted at 291, that in an action in deceit, “subsequent events that arise from the nature or use of the thing itself” may be looked at “insofar as they illuminate the value of the thing as at [the date of acquisition]”. Kizbeau endorsed that approach in relation to s 82 of the TPA when the Court is assessing damages by comparing the price and the real value of the asset at the date of the acquisition.

14. [232]

    There is no reason why the same approach should not be taken in relation to ACL, s 272(1). However, not all subsequent events will be relevant. Subsequent events as a consequence of inherent factors present in the thing itself at the acquisition date, which “illuminate” or indicate or reflect the true value as at the acquisition date, must be taken into account. Subsequent events as a consequence of independent, extrinsic, supervening or accidental factors must not be taken into account: HTW Valuers v Astonland at [40]. Whether a subsequent event truly illuminates or indicates or reflects the measure of loss earlier suffered depends on all the circumstances: HTW Valuers v Astonland at [40], citing Potts v Miller (1940) 64 CLR 282 at 298 (Dixon J); [1940] HCA 43 and Gould v Vaggelas (1985) 157 CLR 215 at 220 (Gibbs CJ); [1985] HCA 68.

15. [233]

    None of the arguments advanced by the appellant for the contrary view is compelling.

16. [234]

    The word “damages” in s 272(1) makes clear that the provision is concerned with compensation for loss or damage: Capic at [714]; Toyota v Williams at [99]. There is no inconsistency between assessing damages for any reduction in value, at the time of supply, and taking into account subsequent events which illuminate or indicate or reflect the value of the good at the acquisition date.

17. [235]

    Contrary to the appellant’s submission, s 271(6) is not determinative that subsequent events must be disregarded in an action under s 271(1). These provisions deal with different subject matters: (i) s 271(1) is concerned with the manufacturer’s failure to comply with the consumer guarantee of acceptable quality in s 54, and (ii) s 271(6) is concerned with the manufacturer’s refusal or failure, within a reasonable time, to repair or replace the goods if the consumer has required the manufacturer to do so under an express warranty. That the subject matter of s 271(6) concerns an express warranty by a manufacturer to repair or replace goods, says nothing of the circumstance that in an action under s 271(1) to recover damages of the kind referred to in s 272(1)(a), the replacement of the goods at no cost may illuminate or indicate or reflect the true value of the goods at the date of purchase, in assessing damages under 272(1)(a).

18. [236]

    The appellant frankly acknowledged that the consequences of disregarding subsequent events under s 272(1)(a) are a “little bit” odd. Thus, it is accepted that to disregard such events would have the consequence that an affected person would be entitled to damages for any failure to comply with the consumer guarantee of acceptable quality, even if their vehicle was sold before the latent defect was discovered. Further, given the broad definition of an “affected person” in s 2(1) includes a person who acquires goods from a consumer (other than for the purpose of re-supply), that entitlement to damages would extend to a similarly unaware purchaser in the entire chain of persons who purchased the vehicles second-hand before the defect was identified. Such consequences would be inconsistent with the overarching consideration that the amount of compensation for any reduction in value be appropriate: Toyota v Williams at [99].

19. [237]

    Further, to disregard subsequent events which illuminate or indicate or reflect the true value of the good at the time of supply would be inconsistent with the “universal rule” that a plaintiff cannot recover more than he or she has lost: Baxter v Obacelo (2001) 205 CLR 635; [2001] HCA 66 at [57] (Gummow and Hayne JJ).

20. [238]

    As Kirk JA (Ward P agreeing) said of the circumstance in Scenic Tours Pty Ltd v Moore [2023] NSWCA 74 at [43] that some claimants against a supplier of services had been awarded compensation under ACL, s 267(3) and (4) for a necessary cost of the contract being fulfilled (being expenditure on airfares to travel to the cruise destination), whilst also obtaining compensation to put them in a position as if the consumer guarantees had been fulfilled:

21. [239]

    Similarly, in Vautin at [294], Derrington J said that double recovery is not permissible in a claim against a supplier of goods for compensation under ACL, s 259(4). There is no reason to consider that s 272, like similar remedial provisions within Pt 5-4 of Ch 5 of the ACL, such as s 267 and s 259, was intended to be inconsistent with that “strong principle” referred to in Scenic Tours v Moore at [43].

22. [240]

    Given the resemblance between the two components of damages referred to in s 272(1)(a) and (b) and compensatory damages available at common law for breach of contract, it is appropriate to draw upon general law notions of compensation, by way of analogy, when assessing statutory damages under s 272(1): HTW Valuers v Astonland at [39]. If the replacement of the airbag in the appellant’s vehicle is as a consequence of intrinsic or inherent factors in the nature of the vehicle, it is an event which must be taken into account in assessing under s 272(1)(a) any reduction in value damages at the time of supply.

23. [241]

    It is not in dispute, as VW submits, that the appellant acquired a mechanically complex product – a vehicle – that requires maintenance in order to remain safe for use, and frequently has parts repaired and even replaced during its lifetime. Given that a vehicle should remain safe for use, it is inherent in the nature of a vehicle that upon recognising a latent defect, the defective part will be repaired or replaced as necessary. Assuming the airbag in the appellant’s vehicle was a latent defect, that is what occurred in this case.

24. [242]

    Contrary to the appellant’s submission, this is not a question of quantifying a mitigating benefit the appellant has received. Whether there is “any” reduction in the value of the appellant’s vehicle, at the time of supply, directs attention to the nature of the failure to comply with the consumer guarantee. Here, the asserted problem with the appellant’s vehicle that is said to have reduced its value, at the time of supply, has been fully rectified, at no cost, in a manner that a reasonable consumer would have always expected given the nature of goods of this kind, if the vehicle contained a latent defect. On the present assumption, that was an event that arose from the nature of the vehicle.

25. [243]

    It is said that the fatal flaw in VW’s submission is “assuming away breach to get to a conclusion of no damage”. This misunderstands the submission. The premise of the submission is that the appellant’s vehicle is not of acceptable quality at the time of supply because the use of PSAN as a propellant is a latent defect. The submission points to the rectification of the assumed latent defect in a manner which a reasonable consumer would always have expected, given the nature of the goods, as a subsequent event which illuminates or reflects or indicates the value of the appellant’s vehicle at the time of supply.

26. [244]

    Nor is it an answer for the appellant to say, as he does, that any reduction in value “can happen theoretically”. Aside from the obvious difficulty in assessing damages based on a “theoretical” reduction in value, to award damages on that basis would be unprincipled. It would place the appellant in a superior position to that which he would have been in had the guarantee of acceptable quality been fulfilled by VW, assuming a failure to comply with the consumer guarantee in s 54. The appellant would have received a vehicle which he has used without incident and, which he accepts is, by reason of replacement of the airbag, as valuable today as it would have been had it been supplied originally with non-defective airbags. The appellant cannot recover more than his actual loss such that “double recovery” is impermissible. Nothing in s 272(1) suggests otherwise.

27. [245]

    There was no error by his Honour in taking into account the replacement of the airbag, at no cost, when assessing damage for any reduction in value of the appellant’s vehicle as referred to in s 272(1)(a). Ground 19 should be rejected.

The appellant’s damages claim in this Court

1. [246]

   In writing, the appellant advanced the three claims for damages under s 272(1) (see (1), (2) and (3)(a) below). In oral argument, a further claim was advanced (see (3)(b) below):

   1. (1)

      reduction in value: a claim under s 272(1)(a) for the difference between the purchase price (about $40,000) and the “true” value of the vehicle at the time of supply;

   2. (2)

      “left in the hand”: a claim based on what was “left in the hand”, which is said to be a proxy for damages under s 272(1)(b), being a loss of $15,000 calculated as the difference between the purchase price (about $40,000) and the current value of the appellant’s vehicle at trial which is said to be $25,000; and/or

   3. (3)

      consequential loss: a claim under s 272(1)(b) for (a) “excess” financing costs (interest) and “excess” taxes (GST and stamp duty) on the difference between the purchase price and the “true” value of the vehicle at the time of supply; and (b) interest on the cost of repair.

2. [247]

   In oral argument in reply, senior counsel for the appellant conceded that “a lot of the methodology that we float in our written submissions falls foul of the statutory definition, and I accept that”. Given this concession, some of the appellant’s claims can be dealt with relatively briefly.

Reduction in value

1. [248]

   In writing, the appellant contended that the true value of his vehicle, at the time of supply, was either: (i) nil, (ii) a value as assessed by reference to Prof Baddeley’s discrete choice experiments, or (iii) a value assessed by the Court as a matter of impression.

2. [249]

   It is said, by application of market theory, that the appellant’s vehicle had no value at the time of supply, and therefore he should recover his full purchase price. As the appellant acknowledged, this submission was not put at trial. Indeed, in final submissions at trial, senior counsel for the appellant expressly disavowed this argument stating, “I’m not putting a submission, as tempting as it is, that the car was worth zero”.

3. [250]

   The short answer to the “nil” value argument is that it was not pressed in this Court in oral submissions. Senior counsel for the appellant frankly acknowledged that the argument was “too artificial” and said nothing in support of it. Grounds 17 and 18(a) should be taken not to have been pressed.

4. [251]

   Had it been necessary to determine VW’s objection to this claim as a new point on appeal, the objection should be upheld. This claim could possibly have been met by calling evidence at the hearing: Suttor v Gundowda Pty Ltd (1950) 81 CLR 418 at 438; [1950] HCA 35 and Coulton v Holcombe (1986) 162 CLR 1 at 7-9; [1986] HCA 33.

5. [252]

   The critical issues at trial were whether it was possible to reason from the discrete choice experiments of Prof Baddeley and, if so, what conclusions might properly be drawn as to any reduction in value of the appellant’s vehicle, at the time of supply, assuming a failure to comply with the guarantee of acceptable quality. It is said that his Honour erred in placing no weight on Prof Baddeley’s evidence (ground 16), and that he should have placed “some” weight on that evidence (ground 18(b)).

6. [253]

   The appellant accepts that there were imperfections in the discrete choice experiments of Prof Baddeley, which could have been done better. However, it is said that it is obvious (by recourse to common sense or basic economic theory) that the appellant’s vehicle was worth less at the time of supply than he paid for it.

7. [254]

   No attempt was made in oral argument to sustain the broad challenge in ground 16 to the finding that no weight should be given to the evidence of Prof Baddeley. Nor did the appellant challenge the findings that there were fundamental flaws in Prof Baddeley’s work (at [230]), that she had “allowed herself to become something of an advocate for” the appellant and group members (at [227]), and that there were methodological problems with the surveys (at [249]-[251]).

8. [255]

   To the extent that the appellant urged the Court to give “a tiny bit of weight” to what Prof Baddeley said in relation to the discrete choice experiments, the appellant did not identify any error in his Honour’s findings. Grounds 16 and 18(b) should be rejected.

9. [256]

   Next, it is said that the Court must “do the best” it can, even where precise evidence of value is not available, and the exercise is inherently impressionistic, referring to statements in Capic at [883]-[884]. This submission has two limbs.

10. [257]

    It is said that his Honour ought to have accepted the results of the discrete choice experiments as a starting point and then, if appropriate, adjust for unreliability. Alternatively, it is said that even if his Honour did not err in rejecting the results of the discrete choice experiments, he should nevertheless have made an assessment, “as a matter of impression”, referring to Capic at [882]-[890].

11. [258]

    It is not in doubt that “mere difficulty in estimating damages does not relieve a court from the responsibility of estimating them” and “[w]here precise evidence is not available, the Court must do the best it can”: Commonwealth v Amann Aviation Pty Ltd (1991) 174 CLR 64 at 83 (Mason CJ and Dawson J); [1991] HCA 54.

12. [259]

    In Placer (Granny Smith) Pty Ltd v Thiess Contractors Pty Ltd [2003] HCA 10; (2003) 196 ALR 257 at [37]-[38], Hayne J (Gleeson CJ, McHugh and Kirby JJ agreeing) distinguished between a case where the plaintiff cannot adduce precise evidence of what has been lost and a case where, although apparently able to do so, the plaintiff has not adduced such evidence. His Honour said of the former kind of case at [38]:

13. [260]

    This case falls into the latter category. The appellant sought to prove the “true” value of his vehicle, at the time of supply, and of all vehicles affected by the same alleged defect, by reference to the evidence of Prof Baddeley which was wholly rejected.

14. [261]

    This is not a case where damage has been proved but the evidence does not enable precise quantification of it and the Court is required to “do its best”: Paino v Paino [2008] NSWCA 276 at [76] (Hodgson and McColl JJA, Young CJ in Eq agreeing at [113]). The appellant failed to prove his loss and its quantum by adducing precise or reasonably precise evidence of any reduction in value of his vehicle at the acquisition date. That circumstance does not permit the primary judge to make an assessment by placing “some weight” on evidence that was “unreliable”, or make a jury assessment as to any reduction in value at the time of supply: Catalano v Managing Australia Destinations Pty Ltd [2014] FCAFC 55; (2014) 314 ALR 62 at [31].

15. [262]

    Inconsistently with the submission that this Honour should have chosen a figure “as a matter of impression”, the appellant said that if VW had not fixed the airbag, the “obvious” measure of any reduction in value under s 272(1)(a) is how much it would cost to fix the airbag in the appellant’s vehicle. The appellant acknowledged that this was not the way the case was run at trial, however, such an approach to assessing any reduction in value was open. In Vautin at [300]-[302], Derrington J treated the repair cost of the good (a vessel) as the prima facie amount of the reduction in value.

16. [263]

    Of course, had the appellant claimed loss for himself and on behalf of group members based on the cost of repair, the claim would have faced substantial difficulties, given that the airbags have been replaced, at no cost to the appellant and group members, and the appellant and each group member had used their vehicle since purchase without incident. But that does not undermine the conclusion that the appellant failed to adduce precise, or reasonably precise evidence of loss, being the cost of repair, when he was able to do so. It simply highlights that VW would have had a ready answer to such a claim.

17. [264]

    Nor is there is an analogy between this case and Capic where Perram J said at [883]-[884] that it was “obvious” that Ms Capic’s vehicle was worth less at the time of its acquisition than she had paid for it, and that “[t]he process of assessing what the value of a chattel was 9 years in the past when the defects it suffered from are risks not actualities is inherently impressionistic”. Capic is distinguishable on the basis that there was expert evidence in that case, which Perram J accepted, that the claimant’s vehicle was worth less at the acquisition date than she paid for it (at [883]).

18. [265]

    The statement in Capic that this was “obvious” reflected the particular findings in that case that the transmission in the claimant’s vehicle posed certain risks related to components within it, which were found to be “real” risks (at [887]) based on proven component failures (at [884]) and a “high” risk of problems with normal operating characteristics of the vehicle (at [888]) in relation to gear rattling and a slight shudder at low speeds. By contrast, in this case, the appellant framed the asserted risk of rupture as “unquantifiable”, there was no proven rupture in the real-world experience, and there was no expert evidence that the appellant’s vehicle was worth less at the acquisition date than he paid for it.

19. [266]

    Ground 18(b) should be rejected.

“Left in the hand” approach

1. [267]

   The appellant accepted that his Honour’s conclusion at [254]-[255] that the “left in the hand” measure of loss did not fall within s 272(1)(a) “appears to be correct”. That concession was properly made. The enquiry as to the true value at the date of supply is conceptually distinct from enquiring into the current value of the relevant good, which is the wrong question for the purpose of s 272(1)(a): Capic at [880].

2. [268]

   In writing, the appellant said that the “left in hand” approach is a proxy for assessing the loss or damage suffered by the affected person because of the failure to comply with the guarantee in s 54 and therefore falls within s 272(1)(b). By analogy with the “no transaction” cases, such as HTW Valuers v Astonland, it is said that the appellant seeks to be put in the position he would have been in had he not spent his money on acquiring a Volkswagen vehicle by receiving the purchase price that he would not have paid and bringing to account the value of his vehicle that he still holds.

3. [269]

   The appellant did not mention the “left in the hand” approach in oral argument in chief, nor in reply after VW stated in oral argument that it assumed that this claim was not pressed. This claim should be taken not to have been pressed.

4. [270]

   If the claim had been pressed, it should be rejected for two reasons.

5. [271]

   The appellant’s contention that VW was not prejudiced by the late notice of this unpleaded claim cannot be accepted. When the claim was first raised in closing submissions, the appellant conceded that there was “no direct evidence of sales in 2021”, and pointed to the highest prices from the eleven sales recorded in the Pickles sales data covering 2015-2018 for second-hand Volkswagen Passats manufactured on the same date as the appellant’s vehicle at trial. This is the precise prejudice referred to by his Honour at [260]-[261].

6. [272]

   Nor was the belated “offer” by the appellant on the last day of trial (1 June 2021) that VW appoint an expert who would inspect the vehicle and prepare a report on its market value an adequate solution to the prejudice occasioned to VW. There is no reason not to accept senior counsel’s statement at trial, which was repeated in this Court, that had the appellant claimed this loss at some point before closing submissions, VW would have put on evidence going directly to the question of the vehicle’s current value.

7. [273]

   The appellant should not be permitted to advance a new point on appeal which could possibly have been met by calling evidence at the hearing: Suttor v Gundowda Pty Ltd at 438; Coulton v Holcombe at 7-9.

8. [274]

   The “left in the hand” claim assumes a “no transaction” case, which is inconsistent with the rights given in s 272(1) which are premised on the transaction having taken effect: Capic at [891]. So much was correctly accepted by the appellant in oral argument.

9. [275]

   Further, the premise of this claim under s 272(1)(b) is that the appellant acquired a vehicle that was less valuable than the purchase price. But this is inconsistent with s 272(3), which indicates that any claim for loss or damage alleged to have been “suffered through a reduction in the value of the goods” must be made under s 272(1)(a), not s 272(1)(b). As noted, the appellant accepts that the “left in the hand” claim does not come within s 272(1)(a).

10. [276]

    Ground 18(c) should be rejected.

Consequential loss: interest, GST and stamp duty

1. [277]

   The claim for “excess” financing costs (interest) and “excess” taxes (GST and stamp duty) is put as a claim for consequential loss under s 272(1)(b). The appellant acknowledged that no claim was made at trial for GST and stamp duty but said that a claim for interest was made.

2. [278]

   It is said that the appellant should be permitted to advance the claims for interest and GST on appeal, as these claims “depend on resolution of a question of principle, not on any factual matter not debated at trial”. The appellant pointed to documentary evidence that he financed the purchase of his vehicle, paying $6,133 in interest over five years and GST of $3,617.

3. [279]

   It is also said that whilst the appellant did not press his claim for stamp duty on appeal, given the absence of evidence that he paid stamp duty on the purchase, the group members might be able to claim stamp duty after this Court “resolve[s] the question of principle” concerning whether those amounts “are recoverable”.

4. [280]

   The claim for interest was put differently at trial. As his Honour observed at [265(b)], the appellant contended in final submissions that one “potential approach” to the “do my best” assessment of damages was that the court should award “interest” at an appropriate rate on the difference between the price paid and the true value “over the period” during which the vehicle was defective, being the period 2013 to 2019, before the airbag in the appellant’s vehicle was replaced.

5. [281]

   The stated premise of this approach was that replacement of the airbag in the appellant’s vehicle in 2018 reduces, but does not fully offset, that part of the appellant’s loss that comprises the difference between the purchase price and true value, and some allowance should be made for the fact that for most of the period of the appellant’s ownership, he has not had the car he paid for. That claim was not pressed on appeal. In this Court, the “interest” claim was put in two ways; neither was run at trial, nor pleaded, nor specified in the common questions.

6. [282]

   “Excess” financing costs: It is said that the appellant and group members are entitled to damages referable to the interest on the difference between the purchase price and the “true” value of the vehicle, at the time of supply. This assumed that the appellant had established by evidence that his vehicle’s value was less than what he paid for it, at the time of supply, by reason of the defective airbags, and that he sustained detriment arising from that state of affairs. That is not so, for the reasons given above in rejecting the reduction in value claim.

7. [283]

   Interest on the “cost of repair”: It is said that fixing the airbag is a proxy for paying reduction in value damages, and the appellant and group members are entitled to damages under s 272(1)(b) for consequential loss for being “out of my money” for the period the airbag was not fixed. The premise of this claim is unsound. The appellant was never “out of my money” for the cost of repair, nor were any of the group members. There can be no consequential loss for interest on the cost of repair of the airbag, given the replacement of the airbags was at no cost to the appellant and group members.

8. [284]

   It is said that the appellant’s failure to contend at trial that the available damage under s 272(1)(b) included “excess” GST and stamp duty paid in respect of the purchase is not fatal for three reasons: (i) a distinction must be drawn between the appellant’s individual claim and claims on behalf of group members, (ii) the new claims sought to be raised on appeal involve questions of law, and (iii) the common questions can now be amended to reflect the legal issues actually common to the appellant and group members.

9. [285]

   As to the first matter, the distinction between a plaintiff’s role as a party advancing an individual claim and the representative role of a plaintiff in representative proceedings under Pt 10 of the Civil Procedure Act is not contentious: see, for example, BHP Group Ltd v Impiombato (2021) 286 FCR 625; [2021] FCAFC 93 at [17]-[18]. As observed in Searle v Commonwealth of Australia (2019) 100 NSWLR 55; [2019] NSWCA 127 at [236] (Bell P, Bathurst CJ and Basten JA agreeing):

10. [286]

    However, this distinction is no explanation for the appellant’s failure to make a claim for himself and on behalf of group members for GST and stamp duty.

11. [287]

    As to the second matter, even if it be assumed that the new claims only involved a question of law, “[a] party does not have a right to insist that a new point be decided on appeal simply because all of the facts having been established beyond controversy or the point is one of construction or of law”: Multicon Engineering Pty Ltd v Federal Airports Corporation (1997) 47 NSWLR 631 at 645 (Mason P, Gleeson CJ and Priestley JA agreeing). It always remains a question of whether it is “expedient and in the interests of justice” to entertain the point: Multicon at 645 (Mason P), quoting Water Board v Moustakas (1988) 180 CLR 491 at 497; [1988] HCA 12.

12. [288]

    When considering that question, it has been said that “the finality of litigation and the importance of parties being bound to the cases they make at trial should never be overlooked”: Branir Pty Ltd v Owston Nominees (No 2) Pty Ltd (2001) 117 FCR 424; [2001] FCA 1833 at [38] (Allsop J, Drummond and Mansfield JJ agreeing), referring among others to Metwally v University of Wollongong [1985] HCA 28; (1985) 60 ALR 68 at 71. See also: TAL Life Ltd v Shuetrim; Metlife Insurance Ltd v Shuetrim (2016) 91 NSWLR 439; [2016] NSWCA 68 at [166] (Leeming JA, Beazley P and Emmett AJA agreeing).

13. [289]

    In this case, it would not be “expedient” or “in the interests of justice” to permit the appellant to advance for himself and on behalf of group members new claims for GST and stamp duty which were not run at trial, nor pleaded, nor specified in the common questions. This is for several reasons.

14. [290]

    One is that there is no explanation for why these new claims were not advanced at trial, including for the delay in raising the new claims: Aon Risk Services Australia Ltd v Australian National University (2000) 239 CLR 175; [2009] HCA 27 at [98], [102]-[103] (Gummow, Hayne, Crennan, Kiefel and Bell JJ). The appellant has not identified any sufficient reason why he and group members should not be bound by the forensic choices made at trial, including as representative plaintiff as reflected in the common questions: Water Board v Moustakas at 497.

15. [291]

    Another is that there is no reasoning by his Honour on the complex question of whether and how damages for GST and stamp duty payments fit within s 272(1) of the ACL. The complexity is apparent from Capic at [891]-[898], [909]-[916], where Perram J dealt with a similar claim for consequential losses.

16. [292]

    Another reason relates to the third matter relied upon by the appellant. The premise of the submission that the common questions can now be amended is unsound. This ignores the effect of s 179 of the Civil Procedure Act which is addressed in the next section below.

17. [293]

    Further, even if the appellant was permitted to advance his new claim for GST on appeal (noting that he did not pursue his individual claim for stamp duty), for the reasons already given at [282] above, the premise of the appellant’s claim for GST has not been established.

Amendment to the common questions

1. [294]

   The appellant said that VW’s criticism – that the common questions as framed by the parties and adopted by the primary judge do not raise the questions of GST and stamp duty – treats the procedural device of common questions as the master rather than the servant. This submission overlooks that Pt 10 of the Civil Procedure Act creates its own kind of “statutory estoppel” in s 179, as the similar provision in Part 4A of the Supreme Court Act 1986 (Vic), s 33ZB, was described in Timbercorp Finance Pty Ltd (in liquidation) v Collins (2016) 259 CLR 212; [2016] HCA 44 at [52] (French CJ, Kiefel, Keane and Nettle JJ).

2. [295]

   Section 179 requires that a judgment given in representative proceedings must describe or otherwise identify the group members affected by it and provides that the judgment “binds all such persons” (subject to s 162 which is not presently relevant, concerning any person who has opted out of the proceedings). It is established that group members in a representative proceeding are only bound by the determination of the claims giving rise to the common questions, but not with respect to the plaintiff’s individual claims: Timbercorp at [53].

3. [296]

   The way the statutory scheme works to bind group members (as non-parties) to an order made by the Court in proceedings under Pt 10 of the Civil Procedure Act is by operation of s 179: Dyczynski v Gibson (2020) 280 FCR 583; [2020] FCAFC 120 at [249] (Murphy and Colvin JJ) and [391] (Lee J); Ethicon Sàrl v Gill (2021) 288 FCR 338; [2021] FCAFC 29 at [393] (Jagot, Murphy and Lee JJ). This is one of the reasons why it is critical for the orderly conduct of a representative proceeding that prior to an initial trial there is specificity in what common questions are being determined at trial: Gill v Ethicon Sàrl (No 3) [2019] FCA 587; (2019) 369 ALR 175 at [9] (Lee J). This occurred in the present case. The interlocutory decision and orders of Sackar J made on 21 June 2019, as varied on 22 May 2020, identified the common questions for determination at trial.

4. [297]

   It is not in dispute, as VW submits, that questions 17, 19 and 20A articulated in detail the damages claimed on group members’ behalf, and were carefully framed to facilitate a determination, on a common basis, of the heads of loss that were or were not claimable by the group members. The framing of the common questions in this case is to be distinguished from Scenic Tours, where Appendix C records that common question 12, relevantly, posed the broader question: “What heads of damage are recoverable for a claim for compensation under s 267 of the ACL?”.

5. [298]

   Nor is it in dispute, as VW submits, that the effect of s 179 of the Civil Procedure Act is that the group members (as non-parties) are relevantly bound by the statutory estoppel on the issues raised by questions 17, 19 and 20A: Timbercorp at [52]; Dyczynski at [249], [391]; Ethicon Sàrl v Gill at [393].

6. [299]

   The appellant’s sole response to the statutory estoppel on the issues raised by questions 17, 19 and 20A, is to suggest that the common questions as framed can now be amended “to reflect the resolution of the legal issues actually common to the appellant and group members”. Searle does not assist the appellant.

7. [300]

   Searle relevantly involved an unsuccessful application by the Commonwealth for leave to appeal against the discretionary decision of the trial judge declining to answer common questions 22 and 23 in the negative or at all: at [238]. That was in circumstances where, although no formal order had been made by the trial judge amending the terms of the common questions to reflect the way in which the representative party had ran the case at trial, the trial judge had found Mr Searle’s abandonment of negligent misrepresentation and deceit claims had the effect of withdrawing questions 22 and 23 from determination in the hearing but did not effect an abandonment of those claims on behalf of all group members: at [233].

8. [301]

   The power to amend the terms of the common questions was not in issue in Searle, given the concession on appeal by the Commonwealth. The Court (Bell P, Bathurst CJ and Basten JA agreeing) held at [237] that it was not only within power for the trial judge to decline to answer questions 22 and 23, but entirely appropriate. As Bell P said at [225]:

9. [302]

   Nor is the trial judge’s power to amend the common questions in issue in this case. As said in Ethicon Sàrl v Grill at [56] (Jagot, Murphy and Lee JJ) in relation to Pt IVA of the Federal Court of Australia Act 1976 (Cth) which is, relevantly, in similar terms to Pt 10 of the Civil Procedure Act:

10. [303]

    By contrast, the task of an appellate court by way of rehearing pursuant to s 75A of the Supreme Court Act 1970 (NSW) is the correction of error: Branir at [22], [30]. In this case, (i) there is no challenge to the interlocutory decision and orders of Sackar J made on 21 June 2019, as varied on 22 May 2022, identifying the common questions for determination at trial, (ii) the appellant did not make any application during the trial or following delivery of the principal judgment for leave to amend the amended statement of claim and the common questions to include the new claims now sought to be raised regarding GST and stamp duty, and (iii) there is no ground of appeal contending that the primary judge erred in not amending the common questions.

11. [304]

    Nor did the appellant apply before the primary judge to set aside the final orders made on 16 September 2021 on the ground that the Court below “has apparently proceeded according to some misapprehension of the facts or the relevant law and that this misapprehension cannot be attributed solely to the neglect or default of the party seeking the rehearing”: Autodesk Inc v Dyason (No 2) (1993) 176 CLR 300 at 303 (Mason CJ); [1993] HCA 6.

12. [305]

    Whilst this Court on an appeal by way of rehearing has the same powers as the primary judge, including concerning amendment: Supreme Court Act, s 75A(6), there is no occasion to consider amending the common questions following the trial and dismissal of the proceedings. The appellant did not make any application for leave to amend the notice of appeal to seek such relief in respect of a matter which the appellant for himself and as the representative party had not run at trial, nor pleaded.

13. [306]

    Aside from the absence of a formal amendment application, no circumstances were identified by the appellant that would justify this Court reopening the damages issues based on the common questions following the trial and judgment below. That would constitute a marked departure from the way the parties conducted the trial with respect to the common questions. This is not a case where the appellant for himself and on behalf of group members was not afforded an opportunity of being heard on the common questions which proved to be decisive in the determination of the proceedings before the primary judge: Autodesk Inc v Dyason (No 2) at 308 (Brennan J).

14. [307]

    There was no error by his Honour in the answers given to questions 17, 19 and 20A. Grounds 18(e), (f) and (g) and 20 should be rejected.

Ground 21: costs

1. [308]

   The appellant did not advance any written or oral submissions in support of ground 21 which challenged the indemnity costs order made on 16 September 2021 relating to one aspect of the damages claim. This ground can be taken not to have been pressed.

Conclusion and Orders

1. [309]

   The appeal has failed. There is no reason why costs should not follow the event: Uniform Civil Procedure Rules 2005 (NSW), r 42.1.

2. [310]

   I propose the following orders:

   1. (1)

      Appeal dismissed.

   2. (2)

      Appellant to pay the respondent’s costs.

3. [311]

   LEEMING JA: I agree with Gleeson JA.

4. [312]

   WHITE JA: I agree with Gleeson JA.