Insurance Australia Limited t/as NRMA Insurance v Chowdhury

1. [1]

   The only question on this summons is one of statutory construction: whether the cost of what were loosely described as “pet care services” is outside the regime for the payment of statutory benefits under the Motor Accident Injuries Act 2017 (NSW) (the 2017 Act) — in particular, on the basis that such services are not “treatment and care” within the meaning of that Act.

2. [2]

   The plaintiff, NRMA, seeks judicial review of the determination of a non-presidential member of the Personal Injury Commission, Member Cassidy (the Member), who concluded that the cost of such services falls within the statutory benefits regime: Chowdhury v Insurance Australia Limited t/as NRMA Insurance [2025] NSWPIC 167 (‘D’). As there was no error in the Member’s conclusion, the summons must be dismissed.

3. [3]

   One matter should be noted at the outset. To the extent that the parties treated this as a test case, it concerns the care of animals kept as pets as such. It says nothing about what are termed “therapy animals”.

Background

1. [4]

   On 11 November 2024, Mr Chowdhury was injured when a vehicle came out from a side street and collided with his motorcycle. He sustained a fracture of his right ankle and a left wrist injury. He was in hospital for some five weeks, and then unable to perform certain weight-bearing activities for a period thereafter.

2. [5]

   NRMA is the third-party insurer of the vehicle Mr Chowdhury considered to be at fault for the accident. Mr Chowdhury made a claim for statutory benefits pursuant to Pt 3 of the 2017 Act on 13 November 2024. On 20 November 2024, NRMA issued a liability notice to Mr Chowdhury accepting that it was liable to pay statutory benefits. However, NRMA did not accept that some of the benefits claimed were within the statute.

3. [6]

   Mr Chowdhury had three cats. As part of his claim, Mr Chowdhury claimed for services comprising feeding and watering the cats and changing their litter. He also claimed for the costs of care for plants kept on his balcony. In a letter dated 24 December 2024, NRMA refused to pay for the pets or the plants. Mr Chowdhury sought an internal review by NRMA of that decision. On 10 January 2025, the internal review affirmed NRMA’s decision with respect to the pet care services, but allowed the plant care services.

4. [7]

   Pursuant to 7.20(1) of the 2017 Act, Mr Chowdhury then referred the dispute about whether the costs of the pet care services claimed were compensable to the Personal Injury Commission as a “medical assessment matter” within the meaning of Sch 2, s 2 of that Act. As will be seen below, the Member determined that the matter was instead a “miscellaneous claims assessment matter” within the meaning of Sch 2, s 3 and thus her power to determine it arose under s 7.42(1). The Member held a preliminary conference with the parties on 7 April 2025 and, having determined that there was sufficient information and that the issue was within a narrow scope, dispensed with the need for a formal hearing by exercising the power in s 52(3) of the Personal Injury Commission Act 2020 (NSW).

5. [8]

   It is necessary to say something of the legislative framework before addressing the Member’s decision.

Legislative framework

1. [9]

   As Mr Dominello, the Minister for Finance, Services and Property, said in the Second Reading Speech for the Motor Accident Injuries Bill 2017 (NSW) (2017 Bill) in the Legislative Assembly on 9 March 2017, the 2017 Act establishes a “hybrid scheme”: New South Wales Legislative Assembly, Parliamentary Debates (Hansard), 9 March 2017 at 1. Injured road users can claim certain statutory benefits regardless of fault. They can also claim modified common law damages in circumstances where fault can be established.

2. [10]

   Mr Chowdhury’s claim for the cost of pet care services was not for common law damages. His claim was for a statutory benefit: “treatment and care expenses” under s 3.24 of the 2017 Act. That section provides:

3. [11]

   The phrase “treatment and care” is defined in s 1.4 of the 2017 Act as follows.

4. [12]

   As NRMA submits, the use of the word “means” in this definition indicates that it is exhaustive. Accordingly, in order for the cost of the pet care services Mr Chowdhury claims to be a “treatment and care expense”, it must fall within one of the categories listed in the definition.

5. [13]

   The parties were in agreement that the only potential category which might cover Mr Chowdhury’s claim for those pet care services is “attendant care services”: Tcpt, 5 November 2025, 8.49-9.8. That term, which is central to the disposition of the summons, is defined in s 1.4 of the 2017 Act:

6. [14]

   The parties focussed particularly on the words, “assistance to people with everyday tasks” and “domestic services”.

7. [15]

   It is convenient at this point to make some general observations about the statutory language and NRMA’s argument construing it.

8. [16]

   It might be observed of the definition of “attendant care services”:

   1. (1)

      The words before the first comma contain the main operative part of the definition (in the sense of stating the meaning of the defined term).

   2. (2)

      Those words have a purposive dimension: “services that aim to provide assistance to people with everyday tasks”. The “service” may or may not consist of directly performing the identified “task”: the language of “assistance” would include but is not limited to that. Moreover, it is the “task” which is described as “everyday”, not the “service”.

   3. (3)

      The words after the first comma are capable of more than one reading. On what is the most natural reading, the words “and includes …” identify four types of services that are taken to be “attendant care services”. The words after the first comma bring within the term “attendant care services” services that do not necessarily fall within the natural and ordinary meaning of the words before the first comma (i.e., that do not necessarily, as a matter of ordinary English, involve assistance with “everyday tasks”): see McTye v Ching Yu Chang by his tutor Leo Alexander Birch [2025] NSWCA 3 at [32]-[33]. On this reading, if the “task” is clearing the gutters of a claimant’s three-storey house (where that task requires use of a harness attached to a point on the roof), that may not readily be seen as an “everyday task”. But it may well be “home maintenance”, and come within the definition on that basis.

   4. (4)

      On another reading, the four types of services identified after the first comma (which are described as “example[s]”) are all “attendant care services” because they fall within the words before the first comma. On that reading, the types of services identified after the first comma assist in understanding the meaning of the preceding words, and in particular the scope of “everyday tasks”. Thus, if clearing the gutters of a three-storey house is “home maintenance”, that would suggest a broad understanding of what constitutes an everyday task.

   5. (5)

      In any event, it is uncontroversial that “attendant care services” include services which fall outside the four examples after the comma. All that is required is that the service meet the description before the comma, namely, that it aims to provide assistance to people with everyday tasks.

9. [17]

   It is not necessary for the purposes of the summons to determine which of the readings in (3) and (4) is correct. The parties proceeded on the footing that, of the four examples of types of services identified in the definition of “attendant care services”, the only one that might cover the provision of the pet care services claimed is “domestic services”. There is no reason, at least in the present context, to think that domestic services would not also come within the concept of assistance with everyday tasks.

10. [18]

    “Domestic services” are not defined in the 2017 Act. However, “gratuitous domestic services” are defined in s 3.26(5) to mean “services of a domestic nature for which the person providing the service has not been paid and is not entitled to be paid”. To the extent that that definition suggests that “domestic services” are “services of a domestic nature”, it adds nothing to the ordinary meaning of the words used.

11. [19]

    The word “domestic” connects the services in some way to the home. As a matter of ordinary English meaning, domestic services might be thought to include, at least, services provided in and about the home for ordinary household purposes. Cooking, cleaning and washing are obvious examples.

12. [20]

    Defined terms are not to be construed in the abstract. The orthodox approach is to read the words of a definition into the operative provision which employs it, and then to construe the substantive enactment in its context and bearing in mind its purpose: see Workers Compensation Nominal Insurer v Sako (2025) 116 NSWLR 422; [2025] NSWCA 12 at [43]. However, because of the nested definitions here and the way in which they are expressed, including the use of examples, it is not straightforward to transpose the definitions into the operative provision, s 3.24(1). By isolating the immediately relevant language from what are otherwise lengthy definitions of both “attendant care services” and, especially, “treatment and care”, that exercise risks omitting other parts of those definitions which may have contextual significance. Bearing that in mind, and attempting to read the immediately relevant words of the definitions into the substantive enactment with maximum fidelity to the statutory language, the operative part of s 3.24(1) is relevantly to the following effect:

13. [21]

    On its face, that language — especially “everyday tasks” — states a potentially very broad entitlement.

14. [22]

    Specifically as to the application of that language in the present case, NRMA accepted that the Court could take judicial notice of the fact that having pets or companion animals is commonplace (Tcpt, 5 November 2025, 28.41); that pets are “a very valuable part of many households” (Tcpt, 37.34); and that, “[o]n the face of it” providing food and water and changing the cats’ litter “would appear to be everyday tasks” (Tcpt, 29.22). There was thus no dispute that attending to the basic needs of a domestic pet in those ways is (or at least is capable of being) an “everyday task”, as a matter of the natural and ordinary meaning of those words (Tcpt, 35.2-7). The services of feeding and watering domestic cats and changing their litter undoubtedly assist with that everyday task, and are also properly themselves described as everyday tasks. As Mr Chowdhury pointed out, s 8 of the Prevention of Cruelty to Animals Act 1979 (NSW) makes it an offence punishable by imprisonment for six months for a person in charge of an animal to fail to provide the animal with proper and sufficient food, drink or shelter which it is reasonably practicable in the circumstances for the person to provide. Whether or not it was reasonably practicable in the circumstances of his injuries for Mr Chowdhury himself to feed or water the cats or change their litter, those were undoubtedly everyday tasks.

15. [23]

    NRMA was somewhat less willing to accede to the proposition that feeding or watering the cats or changing their litter are “domestic services”. As noted above, that does not matter if the services otherwise fall within the definition of “attendant care services” because they aim to assist with an everyday task. In any event, given the widespread practice of keeping domestic pets, it is difficult to resist the conclusion that the services of feeding and watering domestic cats and changing their litter are, as a matter of ordinary English meaning, domestic services. They serve the ordinary household purpose of attending to the basic needs of a domestic pet.

16. [24]

    As will be seen, in the way in which the case was argued, the real issue is not the ordinary meaning of the words used. It is whether, despite the ordinary meaning, the legal meaning of the definition of “attendant care services” excludes the services of caring for a domestic pet. That legal meaning is said to be derived from other matters relevant to construing the statute, including the irrecoverability of the costs of such services as damages at common law; the history of the legislative schemes; and the objects of the 2017 Act and its purposes, including as disclosed in the Second Reading Speech.

17. [25]

    One final observation should be made before turning to the Member’s decision. Although the objects of the 2017 Act point in opposite directions, one aim is undoubtedly to restrain the costs of the statutory scheme. That being so, the breadth of the words “everyday tasks” and “domestic services” is some cause for caution. It is of some significance that, even if the words in the definition of “attendant care services” are given their broad ordinary meaning, various provisions of the 2017 Act operate to limit the scope of the statutory benefits.

    1. (1)

       Section 3.24(1), which confers the statutory right to the benefit, is expressed in terms of “treatment and care for the injured person”. “For” here must mean for the benefit of the claimant. There is no reason to think that the treatment or care must be directed to the claimant’s person, or be exclusively for their benefit. “Domestic services” such as cleaning will benefit others living in the claimant’s home, as will “home maintenance”. That does not make them any less “care for the injured person”.

    2. (2)

       Section 3.24(1)(a) limits the benefit to “the reasonable cost of treatment and care”. It was submitted for Mr Chowdhury that the word “reasonable” limits not only the quantum of the expense incurred for a particular service, but also the nature of the service itself. On this view, on the assumption that the task is walking 20 dogs, the amount claimed might be a reasonable amount for performing that task. But the task of walking 20 dogs might not itself be reasonable, such that the cost of that service would also not be reasonable. It is not necessary for present purposes to determine how far this provision goes in providing a “brake” on the extent of the statutory benefit. But it is one reason for giving the words of the definition of “attendant care services” their ordinary English meaning.

    3. (3)

       Section 3.24(2) provides that benefits are not payable “to the extent that the treatment and care concerned was not reasonable and necessary in the circumstances or did not relate to the injury resulting from the motor accident concerned”. The parties were in agreement that pursuant to s 2(b) of Sch 2 to the 2017 Act, the issues under s 3.24(2) are “medical assessment matters” for the purposes of Pt 7 of the Act. The argument proceeded on the footing that the essential s 3.24(2) question is not whether the task itself is a reasonable one, but whether the claimant is medically assessed as reasonably able to perform the task without the treatment or care.

18. [26]

    In answer to the question how to construe the wide words “everyday tasks” consistently with the objects of the regime, it was submitted for Mr Chowdhury: “With sensible people interpreting them on a case by case basis”: Tcpt, 5 November 2025, 52.7. Mr Chowdhury embraced the flexibility of the statutory language “to meet individual circumstances as it should”: Tcpt, 67.16. No doubt, those involved in the practical operation of the regime should apply the statute sensibly in each particular case. Since each case will be different, the application of the provision will involve an interpretative element on each occasion. To that extent, care should be taken when using language such as “pet care services” not to lose sight of either the statutory language or the particular services for which a claim was made. “Pet care services” may be a useful term when stating the negative proposition for which NRMA contends (no benefits are payable for “pet care services”). But if NRMA’s proposition is rejected, using the converse formulation (benefits are payable for “pet care services”) is liable to distract attention from the text. The statutory question is not whether any particular service is a “pet care service”, but whether the service comes within the statutory definition of “attendant care services”. There may also be reasons of policy supporting a construction of the statutory language that allows a degree of flexibility, constrained by notions of reasonableness, to address the real needs of injured people.

19. [27]

    The question NRMA raises remains the more general one: whether the legal meaning of the words used excludes pet care services.

The PIC decision

1. [28]

   Member Cassidy identified the dispute before her as a “miscellaneous claims assessment” matter within the meaning of Sch 2, s 3(n), being an “issue of liability for a claim, or part of a claim, for statutory benefits not otherwise specified in [that] Schedule.” There is no dispute that, accordingly, the power to determine the dispute under s 7.42 of the 2017 Act was engaged.

2. [29]

   The parties proceeded on the footing that the question whether the costs of the pet care services claimed were compensable was to be determined separately from any issue about whether what was claimed in this case was reasonable and necessary in the circumstances or related to the injuries caused by the accident for the purpose of s 3.24(2). As noted above, under the 2017 Act, any dispute about the latter question was to be determined by a medical assessor in accordance with Div 7.5 and Sch 2, s 2(b) of the Act.

3. [30]

   The issue before the Member — whether the cost of the pet care services claimed was compensable as a statutory benefit under the 2017 Act — was one of statutory construction. The Member did, however, put that question in context by referring to the following evidence in relation to Mr Chowdhury’s ability to care for his cats.

   1. (1)

      A certificate of fitness completed by Dr Ewing on 13 November 2024 diagnosed Mr Chowdhury’s injuries and included an injury management plan which indicated that a six-week non-weight bearing period would be required.

   2. (2)

      An assessment by an occupational therapist, Ms O’Donoghue, that was conducted on 27 November 2024, noted that while Mr Chowdhury lived in an apartment with a flatmate, they were not close. The report further noted that

   3. (3)

      A further assessment was conducted by Ms Lum, a rehabilitation consultant and occupational therapist of AusRehab, on 14 December 2024. Ms Lum’s report, which Mr Chowdhury relied on in the PIC proceedings, noted:

   4. (4)

      Mr Chowdhury was assessed again on 16 December 2024 by Ms Khan, an exercise physiologist of AusRehab. Ms Khan’s report dated 9 January 2025 referred to Mr Chowdhury’s living arrangements and recorded “Mr Chowdhury advised he had 3 cats. He mentioned that he faced ongoing challenges in caring for his three cats such as feeding and carrying them, which he previously managed without difficulty.”

4. [31]

   Before the Member, NRMA relied upon Geaghan v D'Aubert [2002] NSWCA 260; (2002) 36 MVR 542 in support of its argument that statutory benefits were not available under the 2017 Act to recover the cost of pet care services. More will be said about Geaghan below, but it should be noted that the case held that damages at common law were not available for the value of the gratuitous care of pets. Geaghan did not expressly hold that the care of pets fell either within or outside any statutory provision.

5. [32]

   In deciding that pet care services were compensable by way of statutory benefit under the 2017 Act, the Member drew assistance from the following obiter observations of Basten JA in Teuma & Anor v C P & P K Judd Pty Ltd [2007] NSWCA 166 at [98] in relation to the decision in Geaghan:

6. [33]

   The Member found at D[58]-[61]:

Consideration

1. [34]

   A substantial part of NRMA’s argument on the summons sought to draw support from the operation of the two preceding statutory motor accident regimes and from cases decided in the context of those regimes. The argument was subject to a degree of skilful refinement in the course of the hearing: see, e.g., Tcpt, 5 November 2025, 5.41-6.14; 7.1-7; 12.22-26; 25.8-26.9; 41.12-17. The propositions on which NRMA ultimately relied as to the construction of s 3.24(1) of the 2017 Act reduce to some combination of the following:

   1. (1)

      Damages are not recoverable at common law for the value of gratuitous pet care services.

   2. (2)

      Language in predecessor statutes that regulated common law claims for damages for the value of gratuitous care in general was “designed to cover Griffiths v Kerkemeyer care and only Griffiths v Kerkemeyer care” (referring to Griffiths v Kerkemeyer (1977) 139 CLR 161; [1977] HCA 45) (Tcpt, 5 November 2025, 25.36), and should therefore be understood as not extending to pet care.

   3. (3)

      The 2017 Act was enacted against that background, and uses similar or identical language to that of the predecessor statutes.

   4. (4)

      The objects and the secondary materials for the 2017 Act are inconsistent with any intention to expand the scope of compensation beyond what was available at common law, and express a purpose of reducing the cost of the scheme.

   5. (5)

      Parliament should accordingly be taken to have intended that pet care services stand outside the definition of “attendant care services” in the 2017 Act.

2. [35]

   Given the prominence of the earlier statutory motor accident regimes in NRMA’s argument, it is first necessary to explain why those regimes and the cases decided under them, most notably Geaghan, cast little light on the present question, which is the construction of s 3.24(1). Indeed, to the extent that the earlier regimes shed light at all, they do so in a way that assists Mr Chowdhury.

3. [36]

   Section 72 of the Motor Accidents Act 1988 (NSW) (the 1988 Act) provided (as at the date of the accident the subject of the decision in Geaghan):

4. [37]

   The purpose of s 72 was thus not to confer a right. To the contrary, the provision operated to limit Griffiths v Kerkemeyer claims for damages for the value of gratuitous care. The words in subs (1A), “services of a domestic nature or services relating to nursing and attendance”, were part of the mechanism which curtailed the common law right. The specific purpose of those words, which were not defined in the 1988 Act, was to capture Griffiths v Kerkemeyer claims in order to subject them to the requirements of the section. In that context, there is no sensible reason to conclude that the Parliament intended those words to have a legal meaning in the 1988 Act that was narrower than their natural and ordinary meaning. In particular, if damages might be recoverable at common law for the value of the gratuitous care of pets, it is difficult to see any reason why Parliament would have left such damages unaffected by the restrictions in s 72, while subjecting claims for the value of all other gratuitous services of a domestic nature to the provision.

5. [38]

   The evident purpose of the words, “services of a domestic nature or services relating to nursing and attendance”, was thus to cast the net widely. Section 72 was intended to have an ambulatory operation, capturing whatever claims for damages for gratuitous care the common law allowed. NRMA rightly accepted that the pre-2017 provisions had an ambulatory operation (specifically in the context of the corresponding term “attendant care services” in s 128 of the Motor Accidents Compensation Act 1999 (NSW) (the 1999 Act), discussed below: Tcpt, 5 November 2025, 33.17-21).

6. [39]

   NRMA nevertheless maintained that “the common law is the guide to how that definition is to be construed”: Tcpt, 33.27-8. That submission should not be accepted. Once it is acknowledged that the 1988 and 1999 provisions were intended to capture any and all gratuitous services for which damages were available at common law, whatever those services might be, it is difficult to see how the meaning of the statutory provisions (their connotation) could be limited by the matters upon which the provisions happened to operate from time to time (their denotation, i.e., the gratuitous services for which damages had been held to be available at common law from time to time). That being so, the proposition that the language of s 72 should be understood as not extending to pet care services must be rejected.

7. [40]

   It should be noted that the 1988 Act did not use the defined term “attendant care services”. Section 72 was thus not expressed in terms of “everyday tasks”.

8. [41]

   However, the 1999 Act did use that language. Section 3 defined “attendant care services” as follows:

9. [42]

   That definition was identical to the definition of the same term in s 1.4 of the 2017 Act (and there was again no definition of “domestic services” in the 1999 Act). However, the defined term, “attendant care services”, played a very different role in the 1999 Act from its role in the 2017 Act.

10. [43]

    Section 128 of the 1999 Act originally provided:

11. [44]

    That section performed a similar function to s 72 of the 1988 Act. It significantly restricted the recovery of common law damages for the value of gratuitous care. The defined term, “attendant care services”, again served the purpose of capturing Griffiths v Kerkemeyer claims in order to subject them to the requirements of s 128. Again, there is no reason to think that Parliament intended to give that ambulatory definition a narrow operation.

12. [45]

    That being so, the proposition for which NRMA contends — that the defined term, “attendant care services”, in the 1999 Act should be understood as not extending to pet care services — must be rejected.

13. [46]

    (For completeness, it should be noted that s 128(3) was later amended in a way not presently relevant. Section 128 was then reenacted as s 141B in a form which for present purposes was relevantly similar.)

14. [47]

    Geaghan and cases following it featured prominently in NRMA’s submissions as to the proper construction of the 2017 Act. However, none of those cases was decided under the 2017 Act. The following should be noted.

    1. (1)

       Geaghan was decided in 2002, in the context of the 1988 Act, which, as noted above, did not use the defined term “attendant care services”. Geaghan itself accordingly did not construe the words “everyday tasks”.

    2. (2)

       Nor did Teuma, which followed Geaghan. Teuma was also decided in the context of the 1988 Act as a notional determination for the purposes of s 151Z of the Workers Compensation Act 1987 (NSW).

    3. (3)

       Woods v Collins [2019] SASCFC 146; (2019) 90 MVR 439 was decided in the context of s 141B of the NSW 1999 Act, which did use the defined term “attendant care services” and thus incorporated the language of “everyday tasks”. But the Full Court in Woods did not construe that language. Instead, Kourakis CJ, with whom the other members of the Court agreed, simply stated at [31], as to time spent caring for two dogs: “The Judge correctly excluded the time in accordance with the decision in Geaghan v D’Aubert.”

    4. (4)

       Schofield v Hopman [2017] QSC 297 at [195] simply applied Geaghan in giving a negative answer to the question whether “the Griffiths principle” should be “extend[ed] to the care provided to pets”.

    5. (5)

       Makaroff v Nepean Blue Mountains Local Health District [2019] NSWSC 715 was decided under the Civil Liability Act 2001 (NSW). The plaintiff claimed 11 hours per week for horse maintenance and feeding, said to be part of her business: at [547]. For present purposes, the decision does no more than acknowledge at [553] that, if the horses were to be characterised as pets or a hobby, “the plaintiff is not entitled at law to recover with respect to their future care: see Geaghan at [61]-[66].”

    6. (6)

       NRMA also relied to some extent on CSR Ltd v Eddy (2005) 226 CLR 1; [2005] HCA 64, especially at [61] (as to “Where is the line to be drawn?”). Gleeson CJ, Gummow and Heydon JJ, distinguishing Griffiths v Kerkemeyer, held that a plaintiff who had been rendered unable to provide gratuitous services for another person could not recover as damages the commercial value of those services. That is far from this case.

15. [48]

    None of those cases construed the statutory language with which the present summons is concerned — still less in the context of the 2017 Act itself. NRMA nevertheless submitted that Geaghan and the cases following it were dispositive of that question. To oversimplify this strand of NRMA’s argument, it ran as follows: in the context of the 1988 Act Geaghan decided that pet care was not recoverable; the 1988, 1999 and 2017 Acts are in pari materia (that is, they “deal with the same subject matter along the same lines”: SZTAL v Minister for Immigration and Border Protection (2017) 262 CLR 362; [2017] HCA 34 at [24] per Kiefel CJ, Nettle and Gordon JJ); it follows that pet care is not recoverable as a statutory benefit pursuant to s 3.24(1) of the 2017 Act. The argument should not be accepted.

16. [49]

    The respondent in Geaghan was injured in a motor vehicle accident on 10 March 1998: [5]. The applicable motor accident legislation was the 1988 Act. At the relevant time (see at [13]), the form of s 72(2) of the 1988 Act was:

17. [50]

    As noted above, Geaghan did not address the meaning of the term “attendant care services”: that language was simply not part of the 1988 Act.

18. [51]

    Nor did Geaghan decide that pet care services were not “services of a domestic nature” within the meaning of s 72(1A). Geaghan did not expressly address that question at all. Indeed, as will be seen, to the extent that Geaghan implicitly addressed the question, it suggests that pet care services were “services of a domestic nature”.

19. [52]

    Instead, insofar as Geaghan expressly said anything about the availability of compensation for pet care, that occurred at an anterior stage in the analysis, before any question of interpretation of the words “services of a domestic nature”. As seen above, s 72 operated to exclude from an award of damages compensation for the value of certain services provided by certain persons, except in accordance with the section. But before the statute could operate to restrict or exclude compensation for the services, there first had to be a common law right to damages for the value of the services. Geaghan determined that there was no such common law right: Griffiths v Kerkemeyer damages for gratuitous pet care were not recoverable at common law.

20. [53]

    The central reasoning was at [62] and [66], where Stein JA said as follows:

21. [54]

    As to [62]:

    1. (1)

       The words “[i]t might be that no relevant ‘need’ of a plaintiff to the provision of such a service … has been established” were not referring to any aspect of s 72. That language was directed to the common law: Griffiths v Kerkemeyer damages are recoverable only in respect of “needs”: see CSR Ltd v Eddy at [20] per Gleeson CJ, Gummow and Heydon JJ.

    2. (2)

       The words “it may be that if a plaintiff was unable to look after a pet, or lost a hobby, that loss more appropriately sounded in general damages” were again referring to a right to recover at common law. (This reasoning appears to assume the existence of such a right. But the right to recover general damages was restricted under the 1988 Act. The right to bring a claim in damages at common law is also significantly restricted under the 2017 Act.)

    3. (3)

       The words “[i]t may also be that such a loss may be seen as too remote a damage to be laid at a defendant tortfeasor’s door” were again concerned with common law principles, not the statute.

22. [55]

    As to [66], the first sentence repeated the reference to those three possible reasons, each of which concerned matters going to the availability of a right to recover common law damages for the value of pet care services, rather than going to any aspect of the 1988 Act. In what amounted to an attack on the conclusion in Geaghan as to the unavailability of common law damages for gratuitous pet care, it was submitted on behalf of Mr Chowdhury that [66] did not identify which of those three reasons, or what combination of them, led to the conclusion stated in that paragraph. It was also submitted that in this paragraph, as elsewhere in the judgment, Stein JA appeared to conflate keeping a pet with a hobby (although it was acknowledged that at other points, such as at [58], [60] and [61], his Honour expressed himself in terms of “pets or a hobby”).

23. [56]

    For present purposes, the points made in those submissions make no difference to the conclusion to be drawn from Geaghan when it is read as a whole, namely, that it held that the value of pet care services provided gratuitously was not recoverable as damages in an action at common law in accordance with Griffiths v Kerkemeyer. That was in substance the conclusion reached by Ipp JA, with whom Hodgson JA agreed, in Teuma at [48] as to the effect of Geaghan: that it “held that Griffiths v Kerkemeyer … services did not include the provision of care to an injured person’s pets or a hobby”. Basten JA’s view as to the ratio of Geaghan, expressed in obiter at [98] of Teuma, was set out above.

24. [57]

    One further observation should be made about Geaghan. As noted above, the purpose served by the words “services of a domestic nature or services relating to nursing and attendance” in s 72 of the 1988 Act was to subject Griffiths v Kerkemeyer claims to the restrictions for which s 72 provided. There is no reason to think that those words were intended to be given a narrow meaning which would exclude claims for pet care services. That those words should be given a broad construction is consistent with the way in which the cross-appeal proceeded in Geaghan. The primary judge in that case had allowed five hours per week of gratuitous (non-pet-related) domestic services in the past and for five years into the future: [56]. The appellant successfully argued on the main appeal that s 72(2) of the 1988 Act prohibited the award of damages for gratuitous domestic services as the services had been, or were to be, provided for fewer than six hours per week.

25. [58]

    On the cross-appeal, the cross-appellant argued that the primary judge should have awarded one to two hours per week for the care by her husband of three dogs and one horse: at [53], [55]. With the benefit of those one to two hours per week when coupled with the five hours per week allowed by the primary judge, the cross-appellant would meet the six-hour requirement in s 72(2)(a). That is to say, the cross-appeal proceeded on the implicit premise that (assuming that the value of care for the dogs and horse was recoverable in damages at common law under Griffiths v Kerkemeyer), such care would come within “services of a domestic nature” for the purposes of s 72. No part of the reasoning of Stein JA rejected that premise. To the contrary, the reasoning at [56]-[67] implicitly accepted the premise, but (as discussed above) decided the case at an anterior stage in the analysis, namely, that damages for the value of gratuitous pet care were not recoverable at common law.

26. [59]

    Geaghan thus at least implicitly supports the claimant, not NRMA.

27. [60]

    That being so, there appear to be two main possibilities as to the significance of the 1988 Act and Geaghan for the question presently before the Court.

    1. (1)

       The 1988 Act and the implicit premise of the cross-appeal in Geaghan assist Mr Chowdhury. The words, “services of a domestic nature”, had a broad meaning in the 1988 Act which would have extended to pet care (if damages had been available at common law for gratuitous pet care). In the 2017 Act, the similar words “domestic services”, dealing with a somewhat similar subject matter, in the definition of “attendant care services” should be given the same broad meaning as “services of a domestic nature” had in s 72 of the 1988 Act.

    2. (2)

       Alternatively, no conclusion should be drawn about the 2017 Act from the words “services of a domestic nature” in the 1988 Act, or from Geaghan. The purpose served by the words “services of a domestic nature” in s 72 of the 1988 Act (casting the net widely so as to restrict recovery of Griffiths v Kerkemeyer damages for the value of gratuitous care) is so radically different from the purpose of the words “domestic services” in the 2017 Act (providing part of the foundation of the s 3.24(1) right to receive statutory benefits) that whatever meaning was given to those words in s 72 of the 1988 Act does not assist in interpreting the 2017 Act.

28. [61]

    There is a good deal of force in the second proposition above, particularly given the importance of purpose in statutory construction: see Interpretation Act 1987 (NSW), s 33. On that view, the divergence in the purposes of s 72 of the 1988 Act and s 3.24(1) of the 2017 Act would render the in pari materia principle of statutory construction inapplicable. The two Acts might fairly be said to “deal with the same subject matter”, but they do not do so “along the same lines”. The distinct purpose of s 3.24(1) might still result in a similarly broad construction, but that would not be because of any similarity with the language or subject matter of s 72.

29. [62]

    However, the force of the argument that the in pari materia principle must yield in light of the divergence in purposes is somewhat diminished when one comes to the corresponding provisions in the 1999 Act. As has been seen above, s 128 (later s 141B) of the 1999 Act performed a similar function to s 72 of the 1988 Act: it restricted the right to recover common law damages for the value of gratuitous care. But in doing so, the 1999 Act used more detailed language (the term “attendant care services” and its definition) than had the 1988 Act. The ordinary English meaning of the language used in that definition was broad, and given the purpose of s 128, it was again clearly intended to operate broadly so as to capture common law claims for gratuitous services.

30. [63]

    The broad language of the 1999 Act was then reproduced exactly in the 2017 Act. The adoption of that identical detailed language is at least one reason to give the language of the 2017 Act a similarly broad interpretation. In other words, it supports rather than undermines a construction of the definition of “attendant care services” based on the broad natural and ordinary meaning of the statutory language.

31. [64]

    Against that background, it is convenient to return to the propositions in NRMA’s argument identified above.

32. [65]

    (1) Damages are not recoverable at common law for the value of gratuitous pet care services.

33. [66]

    On the authority of Geaghan, it should be accepted that damages are not recoverable at common law for the value of gratuitous pet care services. But that is not necessarily the correct prism through which to view the issue of common law damages. The scheme for payment of statutory benefits for treatment and care under the 2017 Act is a scheme for reimbursement of expenses incurred. The corresponding type of claim at common law would be one for damages to recover costs actually incurred for pet care services provided commercially. The parties were unable to identify any case addressing such a claim. But as a matter of principle, one type of loss that a plaintiff who has suffered negligently caused personal injury is generally able to recover is actual financial loss, in the sense of amounts actually paid or liabilities actually incurred. I did not understand NRMA to dispute that, subject to questions of causation and remoteness, such a claim might be maintainable at common law. It was not suggested by NRMA that Geaghan barred such a claim. Nor was it suggested that the provisions of the 1988 Act (s 72) or the 1999 Act (s 128, later s 141B) which regulated Griffiths v Kerkemeyer claims operated upon a claim to recover costs actually incurred for commercial pet care.

34. [67]

    It follows that the unavailability of a claim for gratuitous pet care at common law is not a strong reason for excluding claims for pet care from the regime for statutory benefits in the 2017 Act. In particular, it is not at all clear that if the 2017 Act permits recovery of expenses actually incurred for commercially provided pet care, that would be a new type of compensation which was not recoverable at common law.

35. [68]

    (2) Language in predecessor statutes that regulated common law claims for damages for the value of gratuitous care in general was designed to cover Griffiths v Kerkemeyer care only and should be understood as not extending to pet care.

36. [69]

    The substance of this issue has been addressed above. The ambulatory language in the 1988 and 1999 Acts should be understood as having a broad meaning which caught whatever was recoverable by way of Griffiths v Kerkemeyer damages. The relevant provisions in those Acts should be understood on the basis that if the cost of gratuitous pet care was recoverable at common law, it would have been caught.

37. [70]

    (3) The 2017 Act was enacted against that background, and uses similar or identical language to that of the predecessor statutes.

38. [71]

    In light of (2), the fact that the 2017 Act uses identical language to the 1999 Act is, if anything, unhelpful to NRMA’s argument that the words in the definition of “attendant care services” should not be given their natural and ordinary meaning.

39. [72]

    (4) The objects and the secondary materials for the 2017 Act are inconsistent with any intention to expand the scope of compensation beyond what was available at common law, and express a purpose of reducing the cost of the scheme.

40. [73]

    NRMA relies on various statements in the Second Reading Speech for the 2017 Bill which referred to an expectation that there would be a reduction in green slip premiums because of the new scheme. (It was not disputed that the Second Reading Speech was available for that purpose.) The Minister said that “[t]he introduction of statutory and no-fault benefits under part 3 will reduce legal costs and the adversarial nature of the scheme because injured people will no longer have to lodge a common law damages claim to get compensation for their injuries.” But the fact that legal costs may be reduced does not mean that the overall cost of the scheme would necessarily be reduced.

41. [74]

    Indeed, as was submitted for Mr Chowdhury, several aspects of the regime for statutory benefits (not limited to treatment and care) in the 2017 Act expanded the scope of compensation, compared to the position both at common law and under the 1988 and 1999 Acts. Putting to one side the regime for modified common law damages under the 2017 Act, and at the risk of some oversimplification and omission of certain qualifications, the regime for statutory benefits has the following significant features.

    1. (1)

       Section 3.1(2)(a) provides that statutory benefits are payable, subject to Pt 3, “whether or not the motor accident was caused by the fault of the owner or driver of a motor vehicle in the use or operation of the vehicle”. That was a radical development. In practice, it means that benefits are available for up to one year regardless of fault, as explained below.

    2. (2)

       The statutory benefits include weekly payments on account of lost earnings: ss 3.6-3.10. Such benefits continue on a no-fault basis for one year (in the Act’s original form, for 6 months). After that period, the weekly benefits cease if “the motor accident was caused wholly or mostly by the fault of the person”, or if “the person’s only injuries resulting from the motor accident were threshold injuries”: s 3.11(1). By s 3.11(2), “A motor accident was caused mostly by the fault of a person if the contributory negligence of the person in relation to the motor accident … was greater than 61%”. By section 1.6, “threshold injuries” are soft tissue injuries and psychological or psychiatric injuries that are not recognised psychiatric illnesses.

    3. (3)

       The effect of those provisions is that, provided that the injury is more than a “threshold injury”, such as Mr Chowdhury’s broken bones, even a claimant who is 50% contributorily negligent will receive weekly payments for 12 months. After that period, weekly payments may continue (subject to reduction for contributory negligence pursuant to s 3.38) for up to 5 years, depending upon the degree of permanent impairment and whether the person’s injury is the subject of a pending claim for damages: s 3.12.

    4. (4)

       With respect to treatment and care, as has been seen, the statutory benefits are principally addressed in s 3.24. Again, the benefits continue on a no-fault basis for up to one year. After that the benefits cease if the person is wholly or mostly (i.e., 61%) at fault or suffers from no more than a threshold injury: s 3.28. Otherwise, subject to various restrictions, the benefits continue after the first year. This regime may have significant advantages from the claimant’s perspective when compared to a lump sum award of common law damages. As was submitted for Mr Chowdhury:

42. [75]

    NRMA’s response was to acknowledge that “the Act expanded the class of people who could receive compensation because it did extend benefits to those at fault”, but to argue that “the kind of compensation that would be available to the extended class was not intended to be extended under the new regime”: Tcpt, 5 November 2025, 84.18-23. The premise – that damages had not previously been available for commercially provided pet care – is contestable. But even assuming that premise, there is, however, no express provision of the Act, nor even an express statement in the Second Reading Speech, which supports the submission. In light of the acknowledged statutory expansion of compensation in one dimension (as to the class of persons who could benefit), it becomes difficult to identify an affirmative statutory intention not to expand compensation in a different dimension (the type of expense that would be recoverable). To the contrary, as the Minister said in the Second Reading Speech, “The people injured on our roads will benefit from broader coverage and greater benefits.”

43. [76]

    As to the objects of the statutory motor accident regimes, it has often been said that they pull in opposite directions. The 2017 Act is no different. For example, the object in s 1.3(2)(b) (“to provide early and ongoing financial support for persons injured in motor accidents”) is in some tension with the object in par (d) (“to keep premiums for third-party policies affordable … by limiting benefits payable for soft tissue injuries and psychological or psychiatric injuries that are not recognised psychiatric illnesses”). Approaching the question of statutory construction by reference to considerations at that level of generality is unproductive. More useful is a consideration of what the operative provisions actually do. As has been seen above, the operative provisions of the 2017 Act in significant respects expanded the availability of compensation, particularly in relation to treatment and care. Far from being inconsistent with any intention to expand the scope of compensation, the operative provisions of the 2017 Act did just that.

44. [77]

    At that point it becomes difficult to maintain that the statutory text should be given a legal meaning different from its ordinary English meaning, by reference to the prior state of the law.

45. [78]

    NRMA did, however, point to the object identified in s 1.3(3)(c)(iii). Subparagraphs (i) and (ii) first refer to the premium pool from which claims are paid, and state that the setting of appropriate premiums requires a large measure of stability and predictability regarding the future cost of claims. Subparagraph (iii) goes on to provide: “It must be acknowledged in the application and administration of this Act— … (c) … (iii) that stability and predictability require consistent and stable application of the law”. To the extent that this is put forward as suggesting that the 2017 Act should be construed as having the same substantive effect as the 1999 Act, it is difficult to reconcile with the many departures from the 1999 Act made in 2017. The same is true of the relationship between the 2017 Act and the common law (see below). Section 1.3(3)(c) is better understood as being directed to stability in the interpretation of the 2017 Act itself.

46. [79]

    (5) Parliament should accordingly be taken to have intended that pet care services stood outside the definition of “attendant care services” in the 2017 Act

47. [80]

    This conclusion was said to flow from the earlier steps in the argument. For the reasons given above, those earlier steps should in large measure be rejected.

48. [81]

    But even if the earlier steps in the argument were accepted, the conclusion would not necessarily follow. It is true that matters of legislative history may have a significant bearing on the interpretation of legislation: ENT19 v Minister for Home Affairs (2023) 278 CLR 75; [2023] HCA 18 at [86] per Gordon, Edelman, Steward and Gleeson JJ; see also at [63]. But in this case, as the Minister made clear in the Second Reading Speech, the purpose of the 2017 Bill was to start with a clean slate. The 1999 Act was being sent “straight to the wrecking yard”, to be replaced by “a brand-new bill”. It is true that the new bill recycled some of the language from the earlier regimes. But that does not mean that the 2017 Act should be construed in a way that limits the new system of statutory benefits to compensation for losses that would have been available under the statutory regime that had just been scrapped, or that would be available in a claim for damages at common law.

49. [82]

    As to the latter point, NRMA submitted that “when the 2017 Act was introduced, it was designed to be harmonious and consonant with the common law position and … certainly not to provide extended benefits”: Tcpt, 5 November 2025, 27.9-13. NRMA relied on the statement in Balog v Independent Commission Against Corruption (1990) 169 CLR 625; [1990] HCA 28 at 635-636 that “where two alternative constructions of legislation are open, that which is consonant with the common law is to be preferred”. Whatever force the Balog principle of statutory construction may have in other circumstances, it does not assist here. For present purposes, given that NRMA’s argument was based on Geaghan, “the common law” must mean the law as to common law damages for gratuitous care. That is the very thing the 2017 Act set out to abrogate, replacing it with a new scheme of statutory benefits providing compensation for the costs of treatment and care.

Conclusion and orders

1. [83]

   No sufficient reason has been shown for giving the definition of “attendant care services”, once read into the substantive enactment in s 3.24(1), a legal meaning different from its ordinary English meaning. The observations of Basten JA at [98] of Teuma, albeit in the very different context of the availability of a claim for gratuitous care at common law, are apposite. The evident purpose of s 3.24(1) is, as the definition of “attendant care services” makes clear, to secure assistance to injured persons with everyday tasks, including in and around the home. Providing food and water for Mr Chowdhury’s cats and changing their litter were such tasks. Services that aimed to assist him as an injured person with those tasks fell comfortably within the definition of “attendant care services”. That construction advances the purpose of s 3.24(1). NRMA has not established error in the Member’s construction.

2. [84]

   If, as a matter of policy, it is considered desirable to exclude the reasonable cost of caring for pets from the scheme of statutory benefits, it will be a relatively straightforward thing to do so. The definition of “treatment and care” provides that that term “does not include any treatment, care, support or services of a kind declared by the regulations to be excluded from this definition”.

3. [85]

   The order of the Court is: the summons is dismissed with costs.