82 Nothing in the reasoning even hints at the right of indemnity descending in advance of circumstances giving rise to a fully-constituted cause of action in respect of an actual liability that is presently enforceable.
83 Findlay v Westfield Development Corporation Ltd [1972] 1 NSWLR 422 and Rheem Australia Ltd v Manufacturer's Mutual Insurance Ltd [1984] 2 NSWLR 370 demonstrate that the liability need not be one owed directly to the worker. The former case involved liability to contribute as a joint tortfeasor and the latter case related to a common law claim for loss of consortium by the injured worker's spouse. But nothing in the reasoning of those cases suggests that the employer seeking indemnity need not establish an accrued liability. Indeed, the reasoning of Glass JA (at 373) and Mahoney JA (at 376-7) in Rheem strongly infers that their Honours would confine "liability…for injury" to liability in the sense that I construe the policy.
84 SANTOW JA:
INTRODUCTION
The present case in its essentials comes to this. Did the negligent employer's statutory insurance answer at all a settled common law claim to compensate an employee who, long after that insurance and employment had ceased, was diagnosed with mesothelioma? There were three insurance policies, one for each year of employment, dating back to 1959. Each provide cover limited in amount, that limit being well below the employer's liability. This is unless subsequent legislation operated retrospectively to remove the limit and provide unlimited cover. Mesothelioma resulted from the employee's exposure to asbestos fibre at the hands of the negligent employer during that earlier period of insurance and employment. We do not know with any certainty which year during this period of three years the fatal ingestion occurred. But we do know that cumulative exposure increases the risk of contracting the disease but has no effect on its severity.
85 Then there are two further questions in the event that one or more of these policies does answer. Taking into account the Workers' Compensation Act 1926 (NSW) ("the 1926 Act"), including subsequent amendments in 1981 and 1987 removing monetary limits:
(a) does each policy now answer for an unlimited amount, so that it does not matter which of the three policies answers, or if the limit still remains because the later legislation is not retrospective, then
(b) (i) do all three policies answer so as to aggregate the monetary limit in each available to meet the appellants' claim, or
(ii) does only one policy answer, so limiting the claim to its monetary limit?
86 In answering these questions and the earlier question, I have had the advantage of reading the judgments of Spigelman CJ and Mason P in draft. I do differ on the first and third questions (whether any policy answers at all and if so, do all three answer) though not on the second (retrospective operation of subsequent legislation). Spigelman CJ and Mason P both conclude that no policy answers. This is essentially on the basis that, while injury may have occurred during the policy period, no "liability" in the sense that term is used in the policy had yet occurred. This was because, in contrast to statutory liability under the Workers Compensation legislation, which vests or accrues immediately on injury, liability at common law neither vests nor accrues until damage is suffered. That occurred only upon later onset of incapacity, when employment and insurance cover had ceased.
87 For my part, I reach a different conclusion on the meaning of that many-shaded term "liability". I do so because I consider that, read in context, the narrower meaning of "liability" (an actually accrued legal liability) is not the sense in which the policy uses that term. That narrower sense produces an unreasonable, even oppressive result which would defeat the main commercial object of the policy. That object is to ensure that employers have the financial wherewithal to meet workers' negligence claims under a comprehensive scheme for compulsory insurance also giving direct rights to workers, and itself mandated by a workers' compensation statute for the benefit of workers. By denying insurance cover to workers suffering diseases like mesothelioma, simply because injury in temporal terms occurs well before damage, would leave a large gap in that insurance safety net, contrary to the evident purpose of the legislation and the policy. Indeed while no-one could have anticipated in 1926 or in the period of insurance the potentially catastrophic effect of mesothelioma claims, there is no doubt that the intention was to include diseases of gradual onset; see s7(4) and (5) of the 1926 Act (quoted later). Such temporal separation could be even as short as a year. It could occur even between two consecutive insurance years, where there is a fresh policy for each year. That is so, though here damage happened to occur when there was no insurance policy, many years later. Yet on the narrow meaning of liability that would be fatal to insurance coverage. Thus if injury occurs in year 1 and damage in year 2 or subsequently, and even though there be a separate policy for each year, that will on that construction be fatal, as injury and liability, so defined, have not coincided but straddle two policies. I conclude that construction is not how the policy terms should be read, when their context is properly considered.
88 Indeed such a result also involves a further anomaly. It appears accepted in the reasoning of Spigelman CJ and implicitly of Mason P that such a gap would only occur in common law negligence claims, but not in statutory compensation claims. The statutory policy would evidently cover the latter. This is because the authorities establish that an employer's liability under the Act accrues when injury occurs, not when consequential incapacity arises. Accordingly, while compensation in a practical sense could hardly be claimed before any damage is suffered (how could it be quantified?) a declaratory judgment could perhaps be sought though in NSW that practice (based on Privy Council authority in Ogden Industries v Lucas (1969) 118 CLR 32) had long fallen into disuse; see Mills "New South Wales Workers' Compensation (Butterworths, 1979) at 192-3. This is because the statutory policy has nonetheless been held, correctly in my view, to respond at the time of injury, even if incapacity arises later for what is, in reality also a liability yet to manifest itself in any damage. See for example Australian Iron & Steel Ltd v Coal Mines Insurance Pty Ltd (1952) 52 SR(NSW) 47 at 50-51, 52, 55-6; Geraldton Building Co Pty Ltd v May (1977) 136 CLR 379 at 384, 404; Manufacturers Mutual Insurance Ltd v National Employers Mutual General Insurance Association Ltd (1991) 6 ANZ Ins Cas 61-038 at 76,964 (hereafter MMI v NEM); National General Insurance Co Ltd v South British Insurance Co Ltd (1982) 149 CLR 327 at 334-335; State Mines Control Authority v Government Insurance Office (NSW) (1964) 65 SR(NSW) 258 at 264-265, 268; Dillingham Engineering Pty Ltd v National Employers' Mutual General Insurance Association Ltd [1971] 1 NSWLR 578 at 586-587.
89 These questions must ultimately turn on the proper interpretation of these policies according to their contractual terms as embodied in statute, namely the Workers Compensation Act 1926 (NSW) and its regulations. That requires first establishing when "injury" occurred. The policy then requires that in addition there be "liability" on the part of the employer in the dual sense of liability to pay compensation and liability independently of the Act. The central issue is whether the latter requires an actual legal liability by way of completed cause of action against the employer, or is satisfied by something less, namely a liability accruing due, but yet to manifest itself in onset of incapacity. Indeed in one sense, damage in the sense of injury has occurred, by reason of the ineluctable progress towards mesothelioma.
90 Answering these questions is made more difficult by the fact that mesothelioma is a disease whose aetiology of cause, behaviour and prognosis is still not completely understood, though our knowledge has expanded over time. The question of its causation is sometimes expressed in terms of whether particular exposure, or its prolongation, has brought about a material increase in risk of injury. But in Australia more is required. The risk must come home so as to implicate the defendant, by showing on the balance of probabilities that the defendant at the least "materially contributed" to the injury suffered; Mason P in Bendix Mintex Pty Limited v Barnes (1997) 42 NSWLR 307 at 312-320. While we do not know precisely why it is that one employee suffers mesothelioma but not another under similar conditions, nor when precisely that employee suffered a fatal ingestion of fibre, the law nonetheless accommodates a degree of imprecision in these matters. It does so by what Mason P calls a "robust and pragmatic approach" to proof of causation.
"It can be demonstrated that the common law is not unsympathetic to the plight of plaintiffs who are faced with multiple defendants yet uncertain as to which of them was legally responsible, where it appears that not all of them were. First, proof that any defendant caused or contributed to injury or damage by negligent breach of a duty of care will suffice to impose liability on that defendant. In other words, the law readily embraces the notion that several persons may bear legal responsibility for the one injury. Secondly, the trier of fact is entitled (indeed encouraged) to take a "robust and pragmatic approach" to proof of causation. The inability to call lay or expert evidence that shows the precise way in which something has happened is not fatal: see Adelaide Stevedoring Co Ltd v Forst (1940) 64 CLR 538 at 563-564, 569; Tubemakers of Australia Ltd v Fernandez . Thirdly, slight evidence may suffice to persuade the trier of fact that liability has been sheeted home to a defendant where the facts are particularly within the knowledge of that party: Parker v Paton (1941) 41 SR (NSW) 237; 58 WN (NSW) 189; Hampton Court Ltd v Crooks (1957) 97 CLR 367 at 371-372. Fourthly, a plaintiff who has sued multiple defendants one or more of whom may be liable is entitled, upon showing prima facie that at least one defendant may be responsible, to insist that the court hear the whole of the evidence before entertaining submissions by any other defendant that no case has been established against that defendant: Broken Hill Proprietary Co Pty Ltd v Waugh (1988) 14 NSWLR 360 at 372. This principle may be vital to a plaintiff who needs to seek an extension of time in which to bring proceedings against a number of defendants: see Briggs v James Hardie & Co Pty Ltd (1989) 16 NSWLR 549. Fifthly, there is a special "rule" relating to a pre-existing medical condition, although it probably reflects little more than a commonsense or lay approach. It is that:
'... where a plaintiff has ... made out a prima facie case that incapacity has resulted from the defendant's negligence, the onus of adducing evidence that his incapacity is wholly or partly the result of some pre-existing condition or that incapacity, either total or partial, would, in any event, have resulted from a pre-existing condition, rests upon the defendant.' ( Purkess v Crittenden (1965) 114 CLR 164 at 168.)
91 Establishing causation in Australia for diseases of gradual onset, though aided by this approach, nonetheless requires more of the claimant than merely to establish that "a particular matter [exposure to asbestos] cannot be excluded as a cause"; per Beazley JA in Bendix Mintex Pty Limited v Barnes (supra) at 339.
92 That warning is apposite when it comes to considering the extent of the evidentiary gap allowed to be overcome for the employee who suffered mesothelioma in the United Kingdom in Fairchild v Glenhaven Funeral Services Ltd [2002] 3 WLR 89. The House of Lords had to deal, not with several successive insurance policies and insurance periods, but with several successive careless employers and employment periods. Under the current state of medical knowledge, onset of the disease could not be attributed exclusively to one of several successive employers, all equally careless. The evidence, given the limitations of medical knowledge, did not enable the Court to establish which employer was responsible for the employee's fatal ingestion of asbestos fibre leading to his mesothelioma, though it did establish that any one of these employers must have been responsible. Fairchild held that in those special circumstances the claimant need not prove on the balance of probabilities that a particular employer's tortious conduct, being one of several employers who carelessly exposed the employee to asbestos fibre, caused or materially contributed to the claimant's injury. It suffices if the claimant can prove on the balance of probabilities that the defendant chosen to be sued, materially contributed to the risk of the injury (mesothelioma) that the claimant suffered (as distinct from materially contributing to the injury itself). Such an elision (from causing injury to merely causing its risk) is not permitted under Australian authority to overcome difficulty in establishing causation.
93 The present case, as I have said, does not involve multiple employers or different insurers over an extended period. Rather it involves the one insurer's successive insurance policies, one for each year over three years, each with its own monetary limit on cover. If under the statutory wording, the policies can answer at all, the further questions posed by their financial limits are fundamentally temporal problems of causation. They concern when the relevant event or events have occurred which render the policy liable to answer. In each or any of the three years, did the events in question cause both injury and employer liability, within the meaning of those terms in the policy? Is there any basis for apportionment between the three policies, given that there was only one insurer throughout? Problems of apportionment were left unsolved by Fairchild as Professor Stapleton explains, in her article "Lords a'leaping evidentiary gaps" in (2002) 10 Torts Law Journal 1 at 24-30. If there were no retrospective removal of these monetary limits, the question turns on whether the first year's policy for 1959 did meet the total claim, if it meets any claim at all; or whether the three policies combine in aggregate to meet the total claim in the amount of $180,000. If the former, then only $60,000 is available under that policy's monetary limit. If the latter, then $180,000 is available, by aggregating the monetary limit of $60,000 per policy (for 1959, 1960 and 1961 respectively).
94 In determining whether the policy answers at all, under its statutory terms both "injury" and "liability" of the employer must occur and coincide with the term of the one policy. The term "injury", like "liability", has its own range of possible meanings, to which the effect of cumulative exposure is relevant. Does "injury" in the sense used in the policy, occur when the fibre is first ingested or only after a sufficient cumulative exposure over time, or when the disease first manifests itself? What can we conclude as a matter of probability, though our knowledge is incomplete? When, following injury to the worker, is the employer "liable to pay" in respect of the employer's "liability" to the worker independently of the Act? Has that occurred during the currency of the policy? The answers to these questions bear on which policy answers, if any policy does, or whether all do. While we know that the minimum exposure to asbestos dust capable of triggering its occurrence can be the ingestion of one fibre, the expert evidence of Professor Henderson has been accepted in other cases, including an important case in the Dust Diseases Tribunal which went on appeal to this Court. According to his evidence in that case, which was accepted, prolonged exposure to asbestos fibre does accentuate the risk of its occurrence ("the cumulative effect" theory); E M Baldwin & Son Pty Limited v Plane & Another (1998) 17 NSWCCR 434 at [95-6] per Fitzgerald A-JA. No argument was put to the contrary of that proposition in the present case. It was accepted in argument there was a practical need to draw upon expert evidence accepted in other cases in the Dust Diseases Tribunal (or on appeal therefrom) in this specialised area of its jurisdiction; compare s25(3) of the Dust Diseases Act 1989 (NSW). During oral submissions (T, 26) the parties agreed that insofar as the court needed to have before it evidence of the cause of mesothelioma in relation to the inhalation of asbestos dust and fibre, it could assume that the evidence of Professor Henderson as set out by Fitzgerald AJA in Baldwin & Son Pty Limited v Plane (supra) at [89] to [96] was accepted by the Tribunal and should be accepted by this Court, including in particular the cumulative effect theory. I shall return to this aspect later.
95 This present case also poses the question whether the negligent employer's "liability", as that term is used in this policy, is limited to the actual crystallised liability when the disease manifested itself, or does it extend back to the time when the asbestos fibres were first ingested, as would lead inexorably to the onset of mesothelioma, assuming the worker did not die from other causes? The time of onset of symptoms is conventionally said to be when the employee's cause of action is complete, as only then is damage said to occur. The respondent therefore contends that that is when "liability" first occurs, within the meaning of the policy, as it is only then that the employer can invoke the policy to seek indemnity for that liability. The respondent then relies on this event being outside the policy period.
96 On the other hand the appellants contend that the trial judge was in error in accepting that narrower meaning of "liability". The appellants contend that "liability" for the purposes of the policy arises when it is inchoate or merely potential. Alternatively, I consider whether it arises
(i) when the liability, though yet to crystallise, is essentially unavoidable following sufficient ingestion, or
(ii) when the employer is exposed to, or at risk of, liability to make (future) payment to the worker.
97 One might compare the first kind of liability to the liability to tax which is yet to accrue, but inevitably will fall upon the tax-payer at the close of the year of income, when the tax can then be calculated and an assessment issue; compare Newton v Federal Commissioner of Taxation [1958] 98 CLR 2 at 7 (Privy Council). Such a liability is not a contingent liability in the sense of it depending on any uncertain contingency. Here too we know, though in retrospect, that there was no further uncertain contingency required. Onset of disease was, we now know, unavoidable after the fatal ingestion. Though it took many years to come home, it worked its inexorable way to that outcome.
98 That intermediate sense of "liability" picking up claims accruing due is in no way strained. It aptly fits the context of a compulsory insurance scheme comprehending, for the benefit of worker, diseases of gradual onset. If such a gap exists in the insurance coverage, this would be as oppressive as a new authority succeeding to the assets and liabilities of the old, but not those liabilities by way of claims accruing but not yet accrued; see Crimmins v Stevedoring Industry Finance Committee (1999) 200 CLR 1, discussed later. It is wrong to treat this case as the paradigm, for in many situations the period between injury and damage may be much shorter. Moreover if there be a second policy only when damage manifests, it will not answer either, as injury pre-dates. Thus in Kelly v Norwich Union Fire Insurance Ltd (1990) 1 WLR 139 a policy was held not to answer which was in force when damage to a building occurred as the relevant event bringing that about (incursion of water) pre-dated the policy. Importantly the Court of Appeal emphasised that the crucial event triggering policy cover was the event of incursion of water, not the later damage. That is an approach which applies by analogy here, where the event of injury from exposure to asbestos fibre is the essential trigger, not the advent of damage from later established legal liability, as distinct from an earlier accruing liability. That of course needs to be tested by reference to the precise policy wording of the policy here under consideration and I deal with that later in this judgment. Essentially I conclude that the policy is triggered by injury and an accrued or accruing liability, though the latter be yet to crystallise in the damage that constitutes a completed cause of action. This has the beneficial result that statutory compensation and common law liability independently of the Act are not artificially differentiated, based on statutory compensation supposedly not requiring damage, in terms of qualifying for insurance cover.
99 The starting point to elaborate on the answer to these questions must be the contractual terms of the statutory policy in its statutory context. Account must be taken of subsequent amendments or re-enactments, but only to the extent, if at all, that they have retrospective effect. The terms so construed must then be applied to the present circumstances. While legal principles and authority have determined in other cases when "liability" and "injury" occur, those terms ultimately depend for their interpretation on the policy terms themselves. I turn, however, first to the salient facts, which are not in dispute.