Duty of Care - Livent. Deloitte & Touche v. Livent Inc. (Receiver of)
Deloitte & Touche v. Livent Inc. (Receiver of) (SCC, 2017) was the seminal Livent case, which was so important to negligent misrepresentation law, and negligence law generally:
A. Duty of Care
 Traditionally, the test from Anns v. London Borough of Merton,  2 All E.R. 492 (H.L.), governed the duty analysis in decisions of this Court addressing claims for pure economic loss (Hercules; Bow Valley Husky (Bermuda) Ltd. v. Saint John Shipbuilding Ltd., 1997 CanLII 307 (SCC),  3 S.C.R. 1210; Canadian National Railway Co. v. Norsk Pacific Steamship Co., 1992 CanLII 105 (SCC),  1 S.C.R. 1021). Significantly, however, the Anns test for establishing tort liability in Canada has since been refined. In Cooper v. Hobart, 2001 SCC 79 (CanLII),  3 S.C.R. 537, this Court provided greater certainty to the law of tort by clarifying the factors which may be considered at each stage of the Anns test. While the resulting Anns/Cooper framework has yet to be applied by this Court in a case of auditor’s negligence, we adopt this statement of La Forest J. for the Court in Hercules: “. . . to create a ‘pocket’ of negligent misrepresentation cases . . . in which the existence of a duty of care is determined differently from other negligence cases would, in my view, be incorrect” (para. 21).
 We turn, therefore, to consider the test for establishing tort liability, beginning with this Court’s decision in Hercules, and the proper application of the general Anns/Cooper framework to cases of auditors’ liability.
(1) Hercules: The Anns Test
 In Hercules, this Court recognized a duty owed by an auditor in preparing a statutory audit of its corporate client. While the Court dismissed the plaintiff shareholders’ claim for lost personal investments, it consistently maintained that a claim by the corporation itself for its own losses resulting from a negligent statutory audit could have succeeded (paras. 58-59; see also paras. 1 and 60-64):
All the participants in this appeal . . . raised the issue of whether the appellants’ claims in respect of the losses they suffered in their existing shareholdings through their alleged inability to oversee management of the corporations ought to have been brought as a derivative action . . . . The duty analysis in Hercules entailed applying the then-current test for recognizing a duty of care in Canadian negligence law: the Anns test. Comprising two stages, the Anns test asked (1) whether a prima facie duty of care exists between the parties; and (2) if so, whether there are any residual policy considerations which should negate or limit the scope of the duty, the class of persons to whom it is owed or the damages to which a breach of it may give rise (Hercules, at para. 20; Kamloops (City) v. Nielsen, 1984 CanLII 21 (SCC),  2 S.C.R. 2, at pp. 10-11; Norsk, at p. 1155; Bow Valley, at para. 47).
. . . if an action is to be brought in respect of such losses, it must be brought either by the corporation itself (through management) or by way of a derivative action.
 Under the Anns test, a prima facie duty of care is recognized where a “sufficiently close relationship between the plaintiff and the defendant” exists such that “in the reasonable contemplation of the [defendant], carelessness on its part may cause damage to the [plaintiff]” (Hercules, at para. 22; Kamloops, at p. 10). In other words, where injury to the plaintiff is a reasonably foreseeable consequence of the defendant’s negligence, a duty of care would, prima facie, arise. This relationship, where present, was labelled one of “proximity” (ibid.). In Hercules, the Court provided greater particularity to the test of reasonable foreseeability which established proximity under the Anns test in the context of claims for pure economic loss arising from negligent misrepresentation or performance of a service. Specifically, it stated that proximity would inhere in a relationship where two criteria are met: (1) that the defendant should reasonably foresee that the plaintiff will rely on his or her representation; and (2) that the plaintiff’s reliance would, in the circumstances of the case, be reasonable. The Court explained that considering the plaintiff’s reliance within the test for the reasonable foreseeability of injury did not “abandon the basic tenets underlying the [Anns] formula” (para. 25). Rather, as the plaintiff’s injury in cases of pure economic loss arising from negligent misrepresentation or performance of a service stems from his or her detrimental reliance, the reasonableness of that reliance informs the determination of whether his or her injury is reasonably foreseeable (paras. 25-26). Where, therefore, the Anns test was applied to cases of negligent misrepresentation, reasonable foreseeability of injury alone, as arising from reasonable reliance, was sufficient to establish a proximate relationship supporting a prima facie duty of care (Hercules, at paras. 25 and 27; Norsk, at p. 1154; Bow Valley, at para. 61).
 The Anns test thereby set a low threshold at the first stage, imposing duties in relation to a nearly limitless class of persons who might rely on representations for nearly limitless purposes. Indeed, as this Court stated in Hercules, “[i]n modern commercial society, the fact that audit reports will be relied on by many different people (e.g., shareholders, creditors, potential takeover bidders, investors, etc.) for a wide variety of purposes will almost always be reasonably foreseeable to auditors themselves” (para. 32). For that reason — that is, because of the low “foreseeability” threshold for establishing a prima facie duty of care at the first stage of the Anns test — the Court looked to the second stage of the Anns test to negate or narrow the duty on the basis of the “policy consideration” of indeterminacy. It was here that the Court looked to the identity of the plaintiffs and the purpose of the audit opinion to deny liability for investment and devaluation losses of individual shareholders (paras. 27-28; see also Haig v. Bamford, 1976 CanLII 6 (SCC),  1 S.C.R. 466). Specifically, the Court found that one of the purposes of a statutory audit — that is, to “allo[w] shareholders, as a group, to supervise management and to take decisions with respect to matters concerning the proper overall administration of the corporatio[n]” (para. 56 (emphasis in original)) — would have permitted the corporate client to recover its own losses at the time of receivership had the claim been brought in the corporation’s name. As we will explain, Livent’s injury following the 1997 Audit is precisely the type of injury described in Hercules as being compensable.
(2) Cooper: Refining the Anns Test
 While this Court’s holding in Hercules remains binding authority governing an auditor’s duty of care in relation to a statutory audit, the framework by which that duty is imposed has since been refined. In the companion cases of Cooper and Edwards v. Law Society of Upper Canada, 2001 SCC 80 (CanLII),  3 S.C.R. 562, this Court revised the Anns test by distinguishing more clearly between foreseeability and proximity, and by placing greater emphasis on a more demanding first stage of the two-stage analysis (Cooper, at para. 30). While, therefore, we rely on Hercules for the general proposition that an auditor may owe its client a duty of care in relation to a particular undertaking, it is the Anns/Cooper framework to which we must have reference in identifying a principled basis for imposing liability. And, properly applied, that framework will rarely, if ever, give rise to a prima facie duty of care that could result in indeterminate liability. Accordingly, and with great respect for contrary views, there is no reason to resort to the second stage in order to negate all liability in this case.
(a) Stage One: Prima Facie Duty of Care
 In Cooper, this Court recognized that “foreseeability alone” is not enough to establish a prima facie duty of care (para. 22; see also Edwards, at para. 9). In doing so, it signalled a shift from the Anns test, which had grounded the recognition of a prima facie duty upon mere foreseeability of injury (Hercules, at paras. 25 and 27; Norsk, at p. 1154; Bow Valley, at para. 61). After Cooper, the first stage of the Anns/Cooper framework would require “something more” (Cooper, at para. 29). That “something more” is proximity (Odhavji Estate v. Woodhouse, 2003 SCC 69 (CanLII),  3 S.C.R. 263, at paras. 47-48; Childs v. Desormeaux, 2006 SCC 18 (CanLII),  1 S.C.R. 643, at para. 12; Hill v. Hamilton-Wentworth Regional Police Services Board, 2007 SCC 41 (CanLII),  3 S.C.R. 129, at para. 23; and Fullowka v. Pinkerton’s of Canada Ltd., 2010 SCC 5 (CanLII),  1 S.C.R. 132, at para. 18).
 In Cooper, the Court did not indicate whether proximity or reasonable foreseeability should be assessed first. In cases of negligent misrepresentation or performance of a service, however, proximity will be more usefully considered before foreseeability. What the defendant reasonably foresees as flowing from his or her negligence depends upon the characteristics of his or her relationship with the plaintiff, and specifically, in such cases, the purpose of the defendant’s undertaking. That said, both proximity and foreseeability of injury merit further reflection.
 Assessing proximity in the prima facie duty of care analysis entails asking whether the parties are in such a “close and direct” relationship that it would be “just and fair having regard to that relationship to impose a duty of care in law” (Cooper, at paras. 32 and 34).
 Under the Anns test, proximity did not, “in and of itself, provide a principled basis on which to make a legal determination” (Hercules, at para. 23). Rather, proximity was a “label” which expressed nothing more than a “result, judgment or conclusion” (ibid.), where mere reasonable foreseeability of injury could be shown. While, under the Anns/Cooper framework, the proximity analysis has become more analytically robust, this descriptive component remains. By this, we mean that the term “proximity” is still used, in part, as a shorthand description of those categories of relationships in which proximity has already been found to exist (Cooper, at para. 23). If a relationship falls within a previously established category, or is analogous to one, then the requisite close and direct relationship is shown. So long, then, as a risk of reasonably foreseeable injury can also be shown — or has already been shown through an analogous precedent — the first stage of the Anns/Cooper framework is complete and a duty of care may be identified (ibid., at para. 36). In such circumstances, the second stage of the Anns/Cooper framework will seldom be engaged because any residual policy considerations will have already been taken into account when the proximate relationship was first identified (ibid., at para. 39; Edwards, at para. 10).
 This Court has on occasion defined previously established categories of proximity in broad terms. In Hill, for example, the Court listed “[t]he duty of care of the motorist to other users of the highway; the duty of care of the doctor to his patient; the duty of care of the solicitor to her client” (para. 25). Proximate relationships will not always, however, be identified so generally. In particular, whether proximity exists between two parties at large, or whether it inheres only for particular purposes or in relation to particular actions, will depend upon the nature of the particular relationship at issue (ibid., at para. 27; Haig, at p. 479). Indeed, and as we explain below, factors which support recognizing “novel” proximate relationships do so based upon the characteristics of the parties’ relationship and the circumstances of each particular case (Cooper, at paras. 34-35).
 It follows that, where a party seeks to base a finding of proximity upon a previously established or analogous category, a court should be attentive to the particular factors which justified recognizing that prior category in order to determine whether the relationship at issue is, in fact, truly the same as or analogous to that which was previously recognized. And, by corollary, courts should avoid identifying established categories in an overly broad manner because, again, residual policy considerations are not considered where proximity is found on the basis of an established category (Cooper, at para. 39). Analytically, this makes sense. For a court to have previously recognized a proximate relationship, second-stage residual policy considerations must already have been taken into account. When, therefore, a court relies on an established category of proximity, it follows “that there are no overriding policy considerations that would [negate] the duty of care” (ibid.). A consequence of this approach, however, is that a finding of proximity based upon a previously established or analogous category must be grounded not merely upon the identity of the parties, but upon examination of the particular relationship at issue in each case. Otherwise, courts risk recognizing prima facie duties of care without any examination of pertinent second-stage residual policy considerations.
 Where an established proximate relationship cannot be found, courts must undertake a full proximity analysis. To determine whether the “‘close and direct’ relationship which is the hallmark of the common law duty of care” exists (Saadati v. Moorhead, 2017 SCC 28 (CanLII),  1 S.C.R. 543, at para. 24, citing Cooper, at para. 32, and Donoghue v. Stevenson,  A.C. 562 (H.L.), at pp. 580-81), courts must examine all relevant “factors arising from the relationship between the plaintiff and the defendant” (Cooper, at para. 30 (emphasis in original); Edwards, at para. 9; Childs, at para. 24; Odhavji, at para. 50; Hill, at para. 24; Fullowka, at para. 26; Saadati, at para. 24). While these factors are diverse and depend on the circumstances of each case (Cooper, at para. 35), this Court has maintained that they include “expectations, representations, reliance, and the property or other interests involved” (ibid., at para. 34; Odhavji, at para. 50; Fullowka, at para. 26) as well as any statutory obligations (Cooper, at para. 38; Edwards, at paras. 9 and 13; Odhavji, at para. 56).
 In cases of pure economic loss arising from negligent misrepresentation or performance of a service, two factors are determinative in the proximity analysis: the defendant’s undertaking and the plaintiff’s reliance. Where the defendant undertakes to provide a representation or service in circumstances that invite the plaintiff’s reasonable reliance, the defendant becomes obligated to take reasonable care. And, the plaintiff has a right to rely on the defendant’s undertaking to do so (W. N. Hohfeld, “Some Fundamental Legal Conceptions as Applied in Judicial Reasoning” (1913), 23 Yale L.J. 16, at pp. 49-50). These corollary rights and obligations create a relationship of proximity (Haig, at p. 477; Caparo Industries plc. v. Dickman,  1 All E.R. 568 (H.L.), at pp. 637-38; Glanzer v. Shepard, 135 N.E. 275 (N.Y. 1922) at pp. 275-76; Ultramares Corp. v. Touche, 174 N.E. 441 (N.Y. 1931), at pp. 445-46; E. J. Weinrib, “The Disintegration of Duty” (2006), 31 Adv. Q. 212, at p. 230).
 Rights, like duties, are, however, not limitless. Any reliance on the part of the plaintiff which falls outside of the scope of the defendant’s undertaking of responsibility — that is, of the purpose for which the representation was made or the service was undertaken — necessarily falls outside the scope of the proximate relationship and, therefore, of the defendant’s duty of care (Weinrib; A. Beever, Rediscovering the Law of Negligence (2007), at pp. 293-94). This principle, also referred to as the “end and aim” rule, properly limits liability on the basis that the defendant cannot be liable for a risk of injury against which he did not undertake to protect (Glanzer, at pp. 275 and 277; Ultramares, at pp. 445-46; Haig, at p. 482). By assessing all relevant factors arising from the relationship between the parties, the proximity analysis not only determines the existence of a relationship of proximity, but also delineates the scope of the rights and duties which flow from that relationship. In short, it furnishes not only a “principled basis upon which to draw the line between those to whom the duty is owed and those to whom it is not” (Fullowka, at para. 70), but also a principled delineation of the scope of such duty, based upon the purpose for which the defendant undertakes responsibility. As we will explain, these principled limits are essential to determining the type of injury that was a reasonably foreseeable consequence of the defendant’s negligence.
(ii) Reasonable Foreseeability
 Assessing reasonable foreseeability in the prima facie duty of care analysis entails asking whether an injury to the plaintiff was a reasonably foreseeable consequence of the defendant’s negligence (Cooper, at para. 30).
 Broadly speaking, reasonable foreseeability concerns the likelihood of injury arising from the defendant’s negligence (Donoghue, at p. 580). This inquiry is not amenable to, and does not require, actuarial precision. The jurisprudence gives content, however, to the foreseeability inquiry, providing courts with guidance. In the abstract, a defendant’s negligent misrepresentation or performance of a service could potentially give rise to innumerable injuries tangentially cascading from the originally contemplated service. This was so in Hercules, where the Court recognized that an auditor’s statement could be relied upon by a potentially limitless number of individuals (e.g., shareholders or takeover bidders), for a potentially limitless array of purposes (e.g., investments or takeover bids), any of which could result in various foreseeable injuries.
 As we have already observed, however, reasonable foreseeability of injury is no longer the sole consideration at the first stage of the Anns/Cooper framework. Since Cooper, both reasonable foreseeability and proximity — the latter expressed in Cooper as a distinct and more demanding hurdle than reasonable foreseeability — must be proven in order to establish a prima facie duty of care. And, in cases of negligent misrepresentation or performance of a service, the proximate relationship — grounded in the defendant’s undertaking and the plaintiff’s reliance — informs the foreseeability inquiry. Meaning, the purpose underlying that undertaking and that corresponding reliance limits the type of injury which could be reasonably foreseen to result from the defendant’s negligence.
 As a matter of first principles, it must be borne in mind that an injury to the plaintiff in this sort of case flows from the fact that he or she detrimentally relied on the defendant’s undertaking, whether it take the form of a representation or the performance of a service. It follows that an injury to the plaintiff will be reasonably foreseeable if (1) the defendant should have reasonably foreseen that the plaintiff would rely on his or her representation; and (2) such reliance would, in the particular circumstances of the case, be reasonable (Hercules, at para. 27). Both the reasonableness and the reasonable foreseeability of the plaintiff’s reliance will be determined by the relationship of proximity between the parties; a plaintiff has a right to rely on a defendant to act with reasonable care for the particular purpose of the defendant’s undertaking, and his or her reliance on the defendant for that purpose is therefore both reasonable and reasonably foreseeable. But a plaintiff has no right to rely on a defendant for any other purpose, because such reliance would fall outside the scope of the defendant’s undertaking. As such, any consequent injury could not have been reasonably foreseeable.
 We add this. Under the Anns test, the Court recognized that auditors may owe a prima facie duty of care to an innumerable number of parties on the basis of reasonable foreseeability alone (Hercules, at para. 32). We acknowledge that the Anns/Cooper framework, when applied to cases of negligent misrepresentation, will give rise to a far narrower scope of reasonably foreseeable injuries and, therefore, a narrower range of prima facie duties of care. This is no indictment of the Anns/Cooper analysis. Rather, it was the very purpose and effect of this Court’s instruction in Cooper that “something more” than mere foreseeability is required at the first stage of the Anns/Cooper framework. By requiring examination of the relationship between the parties as we have just discussed, Cooper gave Canadian courts a more complete array of legal tools to determine whether it is “just and fair” to impose a prima facie duty of care.
(b) Stage Two: Residual Policy Considerations
 Where a prima facie duty of care is recognized on the basis of proximity and reasonable foreseeability, the analysis advances to stage two of the Anns/Cooper framework. Here, the question is whether there are “residual policy considerations” outside the relationship of the parties that may negate the imposition of a duty of care (Cooper, at para. 30; Edwards, at para. 10; Odhavji, at para. 51).
 By “residual”, we mean that such considerations “are not concerned with the relationship between the parties [already considered at stage one], but with the effect of recognizing a duty of care on other legal obligations, the legal system and society more generally” (Cooper, at para. 37; see also Edwards, at para. 10). To the extent, therefore, that stage one of the prima facie duty of care is said to engage “policy” considerations arising from the relationship between the parties — i.e., the recognition that it is sound “policy” to only hold defendants liable for negligence when they are in a proximate relationship with the plaintiff and when the injury suffered was reasonably foreseeable (see Cooper, at para. 25) — such “policy” considerations are not revisited at stage two (ibid., at para. 28). Indeed such reconsideration would be both redundant and analytically confusing (ibid., at para. 29).
 Cooper, and in particular, its strict delineation between “factors arising from the relationship [between the parties]” (para. 30 (emphasis in original)) and factors that “are not concerned with the relationship between the parties” (para. 37) has impacted the stage at which certain factors are considered within the Anns/Cooper framework. For example, principles that were traditionally considered at the second stage of the Anns test in cases of negligent misrepresentation, such as (1) whether the defendant knew the identity of the plaintiff or the class of plaintiffs who would rely on its representation; and (2) whether the reliance losses claimed by the plaintiff stem from the particular transaction in respect of which the statement at issue was made (Hercules, at paras. 27 and 40; Bow Valley, at paras. 55-56), are no longer considered at the second stage. This is because, as we have explained, these factors arise from the relationship between the parties and are, therefore, properly accounted for under the first stage proximity and reasonable foreseeability analysis.
 What, then, remains to be considered at the second stage of the Anns/Cooper framework? In Cooper, this Court identified factors which are external to the relationship between the parties, including (1) whether the law already provides a remedy; (2) whether recognition of the duty of care creates “the spectre of unlimited liability to an unlimited class”; and (3) whether there are “other reasons of broad policy that suggest that the duty of care should not be recognized” (para. 37). In this way, the residual policy inquiry is a normative inquiry. It asks whether it would be better, for reasons relating to legal or doctrinal order, or reasons arising from other societal concerns, not to recognize a duty of care in a given case.
 The place within the Anns/Cooper framework of this policy inquiry is significant. It follows the proximity and foreseeability inquiries. The policy inquiry assesses whether, despite the proximate relationship between the parties, and despite the reasonably foreseeable quality of the plaintiff’s injury, the defendant should nonetheless be insulated from liability (Cooper, at para. 30; Odhavji, at para. 51). That it would limit liability in the face of findings of both proximity and reasonable foreseeability makes plain how narrowly it should be relied upon (Cooper, at para. 30, citing Yuen Kun Yeu v. Attorney-General of Hong Kong,  1 A.C. 175 (P.C.); Edgeworth Construction Ltd. v. N.D. Lea & Associates Ltd., 1993 CanLII 67 (SCC),  3 S.C.R. 206, at p. 218). Only in rare cases — such as those concerning decisions of governmental policy (Cooper, at paras. 38 and 53) or quasi-judicial bodies (ibid., at para. 52; Edwards, at para. 19) — should liability be denied when a defendant’s negligence causes reasonably foreseeable injury to a plaintiff with whom he or she shares a close and direct relationship. In light of the above, the stage at which certain factors are considered in the Anns/Cooper framework is material.
 In this case, the Chief Justice finds that, if it were necessary to proceed to the second stage of the Anns/Cooper framework, she would insulate Deloitte from liability based on the residual policy consideration of indeterminacy (para. 166). We concede that indeterminate liability may, in some cases, be a legitimate residual policy consideration (Cooper, at paras. 37 and 54; Hercules, at para. 31). In our view, however, rarely, if ever, should a concern for indeterminate liability persist after a properly applied proximity and foreseeability analysis (Saadati, at para. 34; Fullowka, at para. 70). Robust application of stage one of the Anns/Cooper framework should almost always obviate concerns for indeterminate liability. This follows from an appreciation of what indeterminate liability, as a concept, actually means.
 Indeterminate liability is liability of a specific character, not of a specific amount. In particular, indeterminate liability should not be confused with significant liability (Gross v. Great-West Life Assurance Co., 2002 ABCA 37 (CanLII), 299 A.R. 142, at para. 38). Certain activities — like flying commercial aircraft, manufacturing pharmaceutical drugs, or auditing a large corporation — may well give rise to significant liability. But such liability arises from the nature of the defendant’s undertakings and of the severe but reasonably foreseeable scale of injury that can result where such undertakings are negligently performed. This explains the significant compensation which these high risk undertakings typically attract. It also explains why contractual disclaimers limiting liability may often be warranted (Edgeworth, at p. 220). In contrast, the liability arising from these “high risk” undertakings may only be characterized as “indeterminate” if the scope of such liability is impossible to ascertain (Black’s Law Dictionary (10th ed. 2014), sub verbo “indeterminate”). In other words, liability is truly “indeterminate” if “the accepted sources of law and the accepted methods of working with those sources such as deduction and analogy — are insufficient to resolve the question” (M. V. Tushnet, “Defending the Indeterminacy Thesis”, in B. Bix, ed., Analyzing Law: New Essays in Legal Theory (1998), 223, at pp. 224-25). More specifically, there are three pertinent aspects to so-called “indeterminacy” in these cases: (1) value indeterminacy (“liability in an indeterminate amount”); (2) temporal indeterminacy (“liability . . . for an indeterminate time”); and (3) claimant indeterminacy (“liability . . . to an indeterminate class”): Hercules, at para. 31, citing Ultramares, at p. 444. Naturally, when a claim has value, temporal, and claimant indeterminacy, our legal tools are insufficient to resolve the quantum of infinite damages that will flow from such a claim.
 All this said, it would be very difficult for liability of an indeterminate character, so understood, to survive a robust analysis of proximity and foreseeability at the first stage of the Anns/Cooper framework. In cases of negligent misrepresentation or performance of a service, the requisite proximity analysis will address claimant indeterminacy because the class of claimants is determinate, including only those in respect of whom the defendant undertook to act. Likewise, foreseeability, which is constrained by the purpose of the undertaking in question, should address concerns about value indeterminacy, because the value of damages is limited — that is, determined — by the reasonably foreseeable quality of the injury (Hercules, at para. 32). Finally, proximity and foreseeability should both address temporal indeterminacy since the longer the period of time over which injury is said to have occurred, the less likely the defendant undertook to protect against it and the less foreseeable the injury, taken as a whole. Hence Cardozo C.J.’s statement in the oft-cited Ultramares decision that a duty which gives rise to indeterminacy “enkindle[s] doubt whether a flaw may not exist in the implication of a duty that exposes to these consequences” (p. 444; see also Weinrib, at p. 231; Beever, at p. 275). In other words, a finding of indeterminate liability at the damages stage strongly suggests that a legal error occurred at the duty stage, since a finding of a prima facie duty of indeterminate scope underlies the resulting indeterminate liability.
 We one final point. Indeterminate liability is a residual policy consideration, nothing more. The presence of indeterminacy need not be dispositive of liability in all cases. To approach the analysis otherwise would transform indeterminate liability from a policy consideration into a policy veto. While indeterminacy may militate against liability, other policy considerations — such as the immense profit margins that “high risk” actors often benefit from, or the extent to which “high risk” actors voluntarily assume the risk of indeterminate liability — may ultimately justify maintaining that liability, despite its indeterminacy (Beever, at p. 293). Even, therefore, in the rare case where indeterminate liability survives the proximity and foreseeability inquiries, it is not automatic that such indeterminacy will necessarily govern (Fullowka, at para. 70). Indeed, any so-called “indeterminate liability” which survives stage one of the Anns/Cooper framework presumably arises from the risk against which the defendant voluntarily undertook to protect the plaintiff and, therefore, may justly and fairly result in liability.