Norton, Louis --- "Is 'defensive practice' defensible? Chilling effects and responsible administration with regard to public authority negligence claims in New Zealand" [2018] NZPubIntLawJl 9; (2018) 5 PILJNZ 5

	Home \| Databases \| WorldLII \| Search \| Feedback Public Interest Law Journal of New Zealand

Is ^“Defensive Practice^”Defensible? Chilling Effects and Responsible Administration with Regard to Public Authority Negligence Claims in New Zealand

Negligence liability for public authorities is often hamstrung by public policy concerns negating a duty of care. A major policy concern is that monetary liability will have a “chilling effect”. Public officials are sometimes presumed to become detrimentally defensive in performing their duties for fear of incurring liability. An opposing school of thought holds that monetary liability would have the positive effect of enforcing prudence, higher standards of care and responsible administration. While the issue has been litigated extensively in comparable jurisdictions, the overall nature and scope of this “defensive practice” concern in New Zealand is less well defined. In this article, I begin by conducting a survey of New Zealand case law relying on, or dismissing, the concern and attempt to establish its current position. In the latter section, I consider what the role of the defensive practice concern should be in New Zealand. Identifying the key universal factors bearing upon the liability equation with examination of both case law and legal commentary, I attempt to set out a reasoned calculus for assessing negligence liability in a given public authority context.

* The paper on which this article was based won second prize in the 2018 LEANZ (Law and Economics Association of New Zealand) essay competition. The author wishes to thank Associate Professor Hanna Wilberg of the University of Auckland for her assistance and supervision in preparing this paper. The paper was completed in partial fulfilment of the coursework requirements for LAWHONS 742.

I Introduction

Finding that a public authority owes a duty of care in negligence exposes it to monetary liability. Consequently, public operators will become defensive in carrying on public services. This is what the landmark decision in Hill v Chief Constable of West Yorkshire leads us to believe.¹In that case, the House of Lords held that imposing liability “may lead to the exercise of a function being carried on in a detrimentally defensive frame of mind”.²The Court declined to find that a duty of care was owed by the police. For many years, the Hill principle conferred a blanket immunity on police in the United Kingdom against actions in negligence. While this trend was qualified very recently in the United Kingdom Supreme Court,³the influence of the Hill decision can be discerned across a wide range of public authority contexts. “Defensive practice” has often been applied as a pretext for dispensing with public authority liability generally.⁴Courts have left the door open to impose liability where policy considerations in favour of a duty are sufficiently strong. ⁵One such consideration is that deterrence of wrongful conduct, not defensiveness, will result from liability. This possibility was countenanced in Lord Keith’s judgment in Hill.⁶An uncertain dichotomy emerges in relation to the behavioural effect of monetary liability—with the undesirable consequences of defensive practice weighing against a finding for duty, and the positive effect of deterrence weighing in favour.

Little uniformity emerges from the relevant authorities. Within the two-stage duty of

care inquiry,⁷defensive practice concerns have often seemed to be applied arbitrarily. Much of the leading case law on defensive practice concerns and deterrence was decided in the United Kingdom. New Zealand authority is comparatively limited. This article accordingly explores both what the position on defensive practice in New Zealand currently is, and what it should be. In the first section, I canvass New Zealand case law that has either invoked the concern or dismissed it. In the latter, I examine the dichotomy between defensiveness and deterrence and attempt to set out a reasoned framework of important considerations to which the courts should have regard. The breadth of factors bearing upon the defensive practice concern is potentially limitless. The purpose of this article is to bring into focus the relevant considerations and their relative importance, yielding a calculus for assessing the applicability of the defensive practice concern. This principled application would abrogate the seemingly random, ad hoc nature of the current approach, while retaining the flexibility to deal with cases on their individual merits.

II Survey of New Zealand Case Law

The New Zealand position must be determined. I here outline a representative range of statements on the chilling effects of raising defensive practice concerns, and responsible

2 At 63. For a fuller exposition on the implications of defensive practice, see Lord Keith’s judgment in this case generally.

7 South Pacific Manufacturing Co Ltd v New Zealand Security Consultants & Investigations Ltd

administration, highlighting the positive deterrent effect of liability. I summarise the most relevant lines of reasoning emerging from the body of case law. This initial process of contextualisation highlights the volatile, and sometimes arbitrary New Zealand jurisprudential approaches to liability calculations.

A Property transactions

Gregory v Rangitikei District Council concerned the Rangitikei District Council conducting a sale of property by tender.⁸It declined to sell to Gregory, the highest tenderer, and negotiated with a different purchaser. In failing to advertise the proposal to sell the property privately, the council did not follow statutory procedures and therefore breached its duty of care to Gregory.⁹Loss was reasonably foreseeable. Defensive practice was of little concern. The duty required nothing more than compliance with statute.¹⁰McGechan J determined that the duty would improve standards of practice, and that insurance would blunt the economic burden of liability such that no detrimental influence would bear upon the authority’s behaviour.¹¹

practice/deterrence equation. It suggests that where loss-spreading measures are available, these provide a compelling reason to impose liability pursuant to the goal of responsible administration of public authorities. Gregory is also authority for the proposition that a common law duty might be warranted in the interest of deterrence if it runs tandem to a statutory duty.

B Medical practitioners

The plaintiff in S v Midcentral District Health Board (No 2) alleged negligence after she, a mental health patient, was raped by another patient in the defendant’s care.¹²William Young J held that inducement of defensive medicine was not decisive in disposing of liability.¹³Internal processes and disciplinary proceedings would be more likely to induce defensive medicine than the possibility that clinicians’ employers face a claim for exemplary damages.¹⁴The close proximity relationship justified a duty.¹⁵

This case highlights two important lines of reasoning in New Zealand jurisprudence relating to the defensiveness/deterrence calculation. First, it illustrates that judicial concerns about defensive practice are less acute when there is a strong relationship of proximity between the tortfeasor and the injured party. Secondly, the judgment contemplates the effect of liability relative to internal organisational pressures. This contributes to New Zealand commentary on the viability of accountability measures outside of negligence accountability.

In Ellis v Counties Manukau District Health Board, the question was whether the District

Health Board (DHB) owed a duty to a medically insane plaintiff to detain and prevent him

13 At [47]. Defensive medicine here may influence, for example, the decision of whether to grant a “leave of absence” in respect of a patient who is subject to an inpatient order.

from causing harm.¹⁶The plaintiff murdered his father after being released from DHB custody. Potter J stated that health professionals could become unduly defensive if “continually faced with the spectre of exposure to common law claims in negligence”.¹⁷The statutory scheme empowered health professionals with a broad discretion, and provided that detention be a last resort.¹⁸It was considered inappropriate to impose a duty of care to detain. Potter J couched part of her reasoning in the fact that other remedies were available to the plaintiff, such as lodging a complaint with the Health and Disability Commissioner, or a civil claim before the Human Rights Review Tribunal.¹⁹

Ellis underscores the close correlation between broad discretionary powers and

defensive practice concerns. Moreover, Potter J’s discussion of alternative remedies suggests that where an injured party’s claim might be vindicated through recourse to other measures than liability, this will also weigh against the deterrence argument.

C Police

The High Court in Evers v Attorney-General declined to find that a duty was owed where police were alleged to have negligently failed to stop the illegal activities of youths at a beach resort.²⁰Chambers J reviewed the English line of authority on police negligence and defensive practice.²¹Liability would not engender higher standards of care, but rather a diversion of resources from crime suppression.²²Failures to control a third party should only exceptionally attract a duty.²³The Court struck out the negligence claim.

An important feature to note is the Court’s willingness to impute to the police a privileged stature. There is substantial public interest in the robust administration of the police’s public duties, and risks attendant on excessive fettering. Evers signals that police, and perhaps other authorities providing high-utility public services, will more readily attract protection from liability. This case also provides a counterpoint to the ‘close proximity’ argument outlined above in S v Midcentral.²⁴Evers involved an alleged failure by police to control the actions of an unrelated third party. While S v Midcentral also involved a failure to control a third party, the defendant authority in that case had an established control relationship with both victim and perpetrator. Extent of control, then, is relevant in considering the weight of defensive practice concerns and the appropriateness of a duty generally.

Fyfe v Attorney General concerned an allegedly negligent arrest.²⁵McGechan J noted that it was uncontroversial that police do not owe a duty in matters of general policing and exercise of their discretion.²⁶However, he stated that “[p]olice can be liable in negligence in respect of the exercise of law enforcement functions ... liability depends upon circumstances”.²⁷This is a departure from Hill and Evers. Police negligence claims should

21 Chambers J specifically applied Hill, above n 1; Alexandrou v Oxford [1993] 4 All ER 328 (CA); and Osman v Ferguson [1992] EWCA Civ 8; [1993] 4 All ER 344 (CA).

not be arbitrarily struck out on public policy grounds.²⁸Instead, given the range of possible circumstances involving egregious police conduct, public policy demanded “robust judicial control”.²⁹

Fyfe suggests that despite the deference shown to the reasoning of the House of Lords in earlier cases,³⁰defensive practice concerns will not, in New Zealand, shield police from liability in all circumstances. Evers and Fyfe indicate that the New Zealand approach to assessing liability, both in terms of induced defensiveness and positive deterrence, turns less on the question of public authority privilege than it does other persuasive factors in the duty inquiry—like proximity. It is relevant that unlike Hill and Evers, this case did not concern a police failure to control a third party.

D Building and land consents

In Morrison v Upper Hutt City Council, an authority granted, and then revoked, a building consent because the development fell outside the requirements of the district scheme.³¹The plaintiff was granted a specified departure, but costs were higher than they would have been if the application had been granted in the first instance. There was sufficient proximity. The Court stressed that the statutory scheme gave the authority broad discretion.³²Ratepayers occasionally had to be detrimentally affected by the authority’s decisions for the scheme to be administered efficiently. ³³The question was barely justiciable and a negligence standard was inappropriate.³⁴Appeal to the Planning Tribunal was considered the appropriate recourse.³⁵The Court dispensed of liability.

Bella Vista Resort Ltd v Western Bay of Plenty District Council involved an authority

issuing consents for the construction of a lodge and conference facility.³⁶It was assumed that neighbours had consented. However, the neighbours successfully took proceedings for judicial review. The developer alleged that the council had breached its duty of care to ensure that the consent applications were correctly processed. No duty was found.

Robertson J held that the council was entitled to rely on advice that consent had been obtained. ³⁷Otherwise, the authority would be required to go behind the information placed before it, creating administrative delays.³⁸It was important that the applicants had alternative forms of redress available.³⁹

Chambers J considered that proximity requirements had been met.⁴⁰However there were justiciability issues, as the plaintiffs had asked the council to deal with their application on a non-notified basis, and they now wanted to sue the council for “having adopted the very submissions they urged on it”.⁴¹

William Young P cited the danger in dealing with public policy factors on the basis of assumptions.⁴²He ultimately held that, as in Morrison, a regulatory body carrying out a quasi-judicial function should not owe a duty.⁴³

In Monticello Holdings Ltd v Selwyn District Council, Gendall J found that a local authority was under no duty to inform a land developer that the land it had purchased was contaminated.⁴⁴A duty would fetter the authority’s ability to issue resource consents in its quasi-judicial capacity.⁴⁵He commented on the tendency for plaintiffs to seek relief from local authorities because of their perceived financial means.⁴⁶

These cases highlight several factors bearing upon the defensiveness/deterrence calculation. As in Ellis, statutory powers of discretion in these cases tended to bolster defensive practice concerns. The availability of alternative modes of redress is relevant. The cases also demonstrate that the nature of a given authority, and the extent of its powers of adjudication, are relevant to determining the strength of defensive practice and deterrence concerns. Bella Vista and Monticello particularly emphasise that quasi-judicial decision makers are typically not the subject of duties, and that in this context defensive practice concerns hold more water.

E Leaky buildings

In Attorney-General v Body Corporate 200200 (Sacramento) a body corporate sought damages in relation to leaky buildings from the Attorney-General, as statutory successor of the Building Industry Authority (BIA). ⁴⁷Proximity between the BIA and the building owners was limited.⁴⁸William Young J considered quasi-judicial authorities could not owe duties of care. Liability would induce the kind of “over-vigilance” that the applicable statutory scheme was designed to avoid.⁴⁹It would create an environment in which it would be “impracticable for building certifiers to operate”.⁵⁰

In Attorney-General v North Shore City Council (The Grange),⁵¹Arnold J considered that

the BIA performed a quasi-judicial role and the imposition of a duty would impede the flow of information to the responsible Minister. Overriding policy concerns included whether the duty would be consistent with the prescriptions of governing legislation, that quasi- legislative functions are not properly the subject of duties, that policy-oriented powers are less likely to attract duties than operational ones, and the limited extent of control exercised by the authority over the loss-causing activity.⁵²

In relation to the wider defensive practice/deterrence calculus, these authorities reaffirm the relevance of proximity, extent of control, and the particular nature of the defendant authority as discussed in the cases canvassed above.⁵³

47 Attorney-General v Body Corporate 200200 [2007] 1 NZLR 95 (CA) [Sacramento]. 48 At [61].

53 As concerns proximity, see for example S v Midcentral, above n 12; and compare Evers, above n 20. As concerns the extent of control, see Evers, above n 20. As concerns the nature of the authority, see generally Bella Vista, above n 36 and Monticello, above n 44.

F Child protection

Attorney-General v Prince and Gardner involved an alleged failure to protect a child by omitting to investigate a complaint made under the Children and Young Persons Act 1974 .⁵⁴Proximity was established.⁵⁵Richardson P, for the majority, considered a common law duty to investigate would enhance the obligations placed on the department in the relevant legislation.⁵⁶Henry J dissented, holding that no duty was owed on the grounds that “potential tortfeasors may act defensively, to the possible ultimate detriment of the young person”. ⁵⁷He emphasised the discretion exercised by social workers and the problematic width of the duty.⁵⁸

Prince demonstrates a tension that is not easily resolved. Strong rationales underlying

a push for positive deterrence (such as pre-existing statutory duties⁵⁹and close proximity relationships ⁶⁰), and those tending to add weight to defensive practice concerns (the unimpeded exercise of discretionary powers)⁶¹are both represented. That the court was divided reinforces the apparent uncertainty in the defensiveness/deterrence balancing act, and the weight afforded to these factors in judicial contemplation of public authority liability.

In B v Attorney General the Privy Council considered whether a duty was owed to a

father suspected (later cleared) of sexually abusing his daughters.⁶²Further, it had to be determined whether a duty was owed to the children for trauma from being removed from their father’s custody.⁶³Lord Nicholls held that, in accordance with Prince,⁶⁴a common law duty to the children should be recognised.⁶⁵No duty was owed to the father. Given that the interests of an alleged perpetrator and a victim are incompatible, it would be impermissible to find a duty owed to the former at the same time as one owed to the latter.⁶⁶The interest of the child must prevail. This case therefore raised a strong conflict of duties formulation of defensive practice.

Conflict of duties is elaborated on at length in the latter section of this article. For immediate purposes, it is helpful to think of conflicting duties as a counter-point to the pre-existing duty argument raised in Gregory and Prince. The pre-existing duty argument involves a common law duty reinforcing an existing, complementary duty imposed by statute in the interest of more responsible administration. Conversely, the conflict of duties situation weighs strongly against the imposition of a common law duty. Where a proposed duty would conflict with the existing public duty of a given authority, this will necessarily impede that authority’s agency—de facto defensive practice resulting.

60 Weighing in favour of deterrence in cases such as S v Midcentral, above n 12.

G Parolee supervision

Hobson v Attorney-General raised the question of whether the Department of Corrections was liable in negligence for failing to supervise parolee William Bell, who had carried out a robbery and triple homicide at a Returned and Services Association.⁶⁷William Young P held that liability would induce a higher likelihood of offending. Over-cautiousness on the part of probation officers would impede the rehabilitation of parolees.⁶⁸The proposed duty would conflict with the statutory purposes of the parole scheme to reintegrate offenders.⁶⁹The accident compensation scheme excluded the injured party, Ms Couch, from seeking compensatory damages. ⁷⁰Appropriate redress was not a counterweight to policy considerations that might be compromised by a duty of care.

Chambers J considered risk minimisation efforts could cripple efficiency across a range of responsibilities involving detention, supervision and monitoring. ⁷¹He expressed scepticism that a duty would promote higher standards of administration. ⁷²He cited proximity and floodgates concerns, and the discretion exercised by the authority as pointing against liability.⁷³The lack of direct control exercised by the department over the offender was important.⁷⁴

Hammond J, dissenting, questioned whether the “tort dress had been overshrunk” in New Zealand. Ms Couch primarily sought public accountability and Hammond J considered that this could only be appropriately vindicated by imposition of a duty.⁷⁵He held that the negligence claim should not be struck out and, referring to the work of Basil S Markesinis and others, ⁷⁶urged that judges not be too quick to invoke defensive practice at the expense of tort law’s wider objectives of deterrence.⁷⁷

identified in the above cases. They also show how these factors bear upon the defensive practice question. William Young P and Chambers J canvass proximity and extent of control issues, the importance of the authority’s discretion, and the existence of a possible conflict between the proposed duty in negligence and the statutory framework underlying the parole scheme. They are persuasive in arguing that given New Zealand’s accident compensation scheme, matters of appropriate compensation should not weigh heavily against defensive practice concerns where the claim concerns personal injury. Hobson suggests that the presence of some or all of these factors should overwhelm the type of accountability/deterrence argument put forth by Hammond J.

The Supreme Court was more equivocal. Hobson came before the Court on appeal.⁷⁸Elias CJ held that dispensing of liability on the basis of presupposition was erroneous.⁷⁹She was also critical of the determination that a common law duty would be inconsistent

76 See Basil S Markesinis and others Tortious Liability of Statutory Bodies: A Comparative and Economic Analysis of Five English Cases (Hart Publishing, Oregon, 1999) at 109.

with the statutory framework—this approach would immunise the Probation Service from actions in negligence, no matter how great the want of care.⁸⁰A common law duty would instead “march hand in hand” with its statutory responsibilities.⁸¹Elias CJ considered it premature to strike out the claim. It was possible that there was a sufficient proximate relationship to found a duty given the high amount of control the Probation Service exercised over Bell.⁸²

Tipping J, for the majority, agreed that it was arguable that a duty was owed. However, in his estimation the case, as pleaded, did not disclose sufficient proximity.⁸³Tipping J touched on the policy arm of the duty inquiry only lightly. He declined to make any determinations as to the import of any factors existing on both sides of the policy equation.⁸⁴

These two decisions highlight judicial divergence on the question of pre-existing duties in the public authority context. William Young P’s judgment in Hobson held that a common law duty would conflict with the existing parole scheme—a conflicting duties formulation of defensive practice.⁸⁵Elias CJ in Couch v Attorney-General asserted in the alternative that the duty contended for would complement the Probation Service’s existing charges. That fact that these two judges came to different conclusions as to the claim’s viability demonstrates that any proposed duty should be scrutinised: assessing whether its effect in practice would be to reinforce prescribed obligations, or, alternatively, to fetter a contrasting set of obligations.

H Ministerial error

Takaro Properties Ltd v Rowling concerned the Minister of Finance’s alleged negligence in refusing to grant consent for a scheme that would raise Takaro’s capital.⁸⁶Takaro then went into receivership. Lord Keith reiterated that it was hoped that liability would induce higher standards of care, but that in some situations harmful consequences would arise that is, chilling effects).⁸⁷As the Minister’s decision was capable of being described as having a policy rather than operational character, the alleged negligence skirted suitability for judicial resolution.⁸⁸The Court considered that errors of law by a Minister or public authority could only rarely be described as negligent, and that a Minister would not usually be under a duty to seek legal advice.⁸⁹Considered wholly, it was dubious that liability in negligence was appropriate.⁹⁰

Takaro is relevant in illustrating the correlation between justiciability issues and defensive practice concerns. Much as in Morrison, Bella Vista and The Grange,⁹¹Takaro

affirms that where the exercise of powers is policy-oriented rather than operational, or where negligence liability may otherwise be an inappropriate measure, judicial fears of defensive practice will tend to be more acute.

91 Morrison, above n 31 at 338; Bella Vista, above n 36 at [90]; and The Grange, above 51 at [32].

The authorities disclose little consistency on the position on defensive practice in New Zealand. The policy concern has been applied or dispensed with on a case by case basis— often seemingly at whim. Particular judges are also internally inconsistent, without much indication of which factors decisively weighed for or against the concern in different contexts. ⁹²There is little to guide us in determining whether defensive practice is an appropriate consideration in a particular situation, and to what extent it should insulate public authorities from liability.

The most concrete observation is that defensive practice/deterrence concerns are often parasitic on the strength of other factors in the duty inquiry. Where there is a strong proximity relationship or other policy factors pointing toward a duty, the deterrent effect of liability is often stressed.⁹³Equally, where proximity is weak, or other policy concerns militate against the duty, this has tended to strengthen defensive practice considerations. ⁹⁴The overriding impression is that defensive practice/deterrence arguments are accessory concerns that are not usually determinative of liability per se. Possible exceptions are cases involving a strong conflict of duties, or where the discretion of the authority is not to be interfered with.⁹⁵

If it is true that defensive practice is a secondary or even tertiary consideration, this would go some way in explaining the inconsistent state of New Zealand jurisprudence. Peripheral to commonly cited policy factors like floodgates arguments, proximity, and justiciability issues, defensive practice has proved to be the perpetual bridesmaid in the greater liability equation. Uncertainty, to some extent, is part and parcel of tort law. The scattershot nature of the approach to defensive practice in New Zealand, however, must be considered unsatisfactory. The irresistible conclusion to be drawn from the New Zealand authorities is that there is scope to proceed on a more reasoned basis than has so far been apparent.

III The Defensiveness/Deterrence Dichotomy

I here consider what the New Zealand position should be by outlining the most important factors to which the court should have regard in assessing liability. This exposition allows us to both critique and explain much of the existing body of New Zealand case law. It also shines a light on important considerations that New Zealand courts have not tended to expressly countenance.

A The empirical dimension

A common reservation about the validity of the defensive practice concern is the apparent lack of empirical evidence supporting it.⁹⁶Fears of over-cautiousness on the part of the public authority often arise from supposition. The countervailing argument—that

92 See, for example, William Young J’s vacillations on defensive practice between S v Midcentral, above n 12; Bella Vista, above n 36; Sacramento, above n 47; and Hobson, above n 67.

96 See, for example, Claire McIvor “Getting Defensive about Police Negligence: The Hill Principle, the Human Rights Act 1998 and the House of Lords” (2010) 69 CLJ 133 at 135.

improved performance and responsible administration will be the outcome of liability— must be subject to the same uncertainty.⁹⁷

The Law Commission of England and Wales has examined the available evidence on the empirical issue.⁹⁸It concluded that defensiveness in response to liability was a fear more imagined than real. ⁹⁹Public bodies consider liability in decision making, but in practice do not significantly alter behaviour. ¹⁰⁰The Commission recognised that deterrence of negligent behaviour could result in improved public services.¹⁰¹However, it considered that any general inference of the behavioural influence of liability was difficult to draw.¹⁰²The Commission qualified these conclusions by conceding that any observable change in public authority behaviour would likely reflect defensive practice rather than effective administration.¹⁰³

Jef De Mot and Michael Faure adopt an economic framework to analyse chilling effects.¹⁰⁴They argue that such effects will occur when a potential injurer takes risk-averse measures beyond the socially and economically optimal level. ¹⁰⁵Socially desirable activities may cease where the benefit accruing to the injurer is overwhelmed by the costs of precaution and liability. In the case of a public authority, this fear is much more acute. Unlike private tortfeasors, who would pay all the costs associated with precaution themselves, public authorities usually balance the (externalised) costs of inaction and intervention. ¹⁰⁶The authority accordingly becomes inclined to go to the extreme in preventing risks, as it does not bear the precaution cost itself.

The Law Commission characterises this situation as related to the question of appropriate incentives. Effective administration should result from incentivising authorities to the degree that they invest sufficiently in harm reduction. Defensive administration would result from over investment to this end.¹⁰⁷A related difficulty is that it is normally a high threshold to establish breach on the part of a public authority. This means that the typical authority would not tend to feel obligated to invest heavily in harm prevention where it can instead protect itself through compliance to a reasonable standard of care.¹⁰⁸

The defensive practice concern in New Zealand is amorphous. This exacerbates the possibility for chilling effects to occur. Where the standards to which a public authority must comply are established and definitive, the authority only needs to ensure that its behaviour meets that standard. Where the law is ambiguous, chilling behaviour is more likely.¹⁰⁹A lack of information about the limits of liability or the standard expected makes ascertaining the appropriate level of investment in harm-reduction difficult—leading

98 Law Commission Administrative Redress: Public Bodies and the Citizen (Law Commission, Consultation Paper No 187, June 2008).

104 Jef De Mot and Michael Faure “Public authority liability and the chilling effect” (2014) 22 Tort L Rev 120.

authorities to maximise investment to the extent that “de facto defensive administration is the result”.¹¹⁰

Uncertain legal standards and externalisation of costs related to risk-aversion therefore create an environment in which strong chilling effects may come to bear.¹¹¹As public bodies are not subject to competitive market forces, any induced defensiveness affecting the quality of service provision would not be challenged by individuals transferring their custom elsewhere. Public authorities, then, lack compelling disincentives to invest heavily in risk-aversion measures.¹¹²De Mot and Faure conclude that “chilling effects are not a complete illusion”.¹¹³

Norman G Poythress and Stanley L Brodsky conducted a study on the deterrent function of tort in the context of mental health practitioners.¹¹⁴They concluded that fear of litigation reduces quality of care generally, increases risk of institutionalisation and reduces staff-to-patient ratios.¹¹⁵The pathogenic effect of litigation-induced fear led to indiscriminate practices by clinical staff such as refusing to discharge patients where discharge was reasonable and warranted.¹¹⁶The authors stress that different contexts will produce different cost-benefit profiles, and as such these findings might not be universally applicable. ¹¹⁷From these studies, however, it appears that bare assumptions that defensive practice is merely illusory are not well-founded.

Empirical evidence receives limited acknowledgment in New Zealand case law. Most judgments avoid referencing such evidence entirely, and baldly assert that either defensiveness or deterrence is likely to result from liability. Greater attention to the considerations detailed above is one way the courts can introduce greater coherence to the law as it relates to behavioural influence on public authorities. For example, if we re-evaluate Hobson ¹¹⁸in light of the Poythress and Brodsky study, the arguments of William Young P and Chambers J hold more water than was later credited by the Supreme Court. One of Elias CJ’s main bones of contention in Couch was that the defensive practice argument was predicated on hypothetical facts.¹¹⁹Parolee supervision can be viewed as a similar context to mental health practitioners, who also exercise a similar supervisory function in the public interest. We may reason by analogy that similar quality-of-care impacts and service dilution to that described in the study would have materialised had liability eventually been imposed in the Hobson and Couch litigation. The Supreme Court’s determination that policy considerations disclosed no justification for summary dismissal, then, may have been at least partially faulty. Attentiveness to the body of empirical evidence should mitigate the possibility of flawed decisions being arrived at on the basis of supposition.

114 Norman G Poythress and Stanley L Brodsky “In the Wake of a Negligent Release Law Suit: An Investigation of Professional Consequences and Institutional Impact on a State Psychiatric Hospital” (1992) 16 Law and Human Behavior 155.

B Quantum of damages

Substantial damages attract defensive behaviour. Plaintiffs have less incentive to pursue small claims. Larger claims induce higher risk aversion measures, and small claims are not likely to strain the budgets of large public departments. ¹²⁰Low payouts attract little publicity, so the adverse reputational effects of claims entering the public domain are diminished.¹²¹While it is reasonable to assume that this holds true for positive effects of deterrence, De Mot and Faure argue that liability’s sanction is comparatively weak.¹²²They note that only a fraction of victims recover damages from the tort system, and that due to internal budgetary reasons, liability costs are not always proportionately charged to the responsible department.¹²³

Quantum of damages is not expressly related to the question of deterrence and defensiveness in the New Zealand case law. Greater clarity might be achieved by the courts explicitly contemplating what amount of damages would produce the best set of incentives and sanctions on public authorities. It is important consider whether, as De Mot and Faure argue, substantial payouts might be against the greater public interest in tending to engender defensive behaviour more so than deterrence of wrongful conduct.

C Insurance

Frequently, damages awards are paid by the defendant authority’s insurance company. This blunts the deterrent effect of liability.¹²⁴Defensiveness will not always be abrogated in the same way. De Mot and Faure state that emotional costs can be substantial in certain circumstances—for example, mental health clinicians are generally more willing to shoulder the emotional cost of adverse litigation attendant on defensively declining to release a potentially dangerous patient than they would the emotional cost of releasing a patient who then harms third parties.¹²⁵While insurance may mitigate defensive practice in a financial sense, it cannot indemnify emotional, reputational and other nonfinancial costs which may also bring to bear significant chilling effects.¹²⁶

Gregory is authority for the proposition that where emotional costs are not likely to

flow from litigation, insurance should tend to overwhelm the chilling effect argument.¹²⁷McGechan J considered that insurance would defray the financial burden such that defensive practice was not a concern. His judgment is not without flaws. He opined that liability should promote improved performance and more effective administration. If insurance is an equalising agent, as McGechan J holds, it follows that the behavioural effect of liability should be neutral.

This contradiction typifies the confusion in New Zealand law in terms of defensive practice and deterrence concerns. Gregory suggests that liability will not attract defensive behaviour where insurance is available, but it will engender increased deterrence. This is a non sequitur, and undesirable in terms of clarity in the law. It only affirms the impression, previously noted, that liability’s behavioural effect will tend to be framed as either

defensiveness or deterrence in light of other prevailing considerations. In future, New Zealand courts should carefully contemplate the behavioural influence that loss-spreading measures are likely to have, and allow defensive practice concerns their proper place in that contemplation.

D The defendant authority

The Law Commission report states that a public authority’s level of legal knowledge, conscientiousness and competence influences its ability to respond to liability appropriately.¹²⁸A competent authority should be able to order policies and systems so that responsible administration results. This would be organisation-specific. It would be extremely difficult to definitively circumscribe the distribution of competence within and between different public authority contexts.¹²⁹

Dissemination of relevant information within an authority’s organisational structure is important. Schwartz stresses that litigation will not deter misconduct if the authority does not have robust information practices.¹³⁰If public officers are not informed of potential liability attaching to an act or omission, they are unlikely to modify their conduct. Accordingly, she argues that the provision of information relating to litigation across an authority’s organisational structure should engender positive changes in the deterrent effect of actions for damages.¹³¹Chilling effects may arise from lack of certainty around reasonable standards. Therefore, effective informational structures would help to provide this certainty. This would reduce the potential for defensive practice. John Hartshorne, Nicholas Smith and Rosemarie Everton also argue that the attitude of the authority will impact significantly on whether deterrence or defensiveness results from liability. ¹³²Employee training, professional development and reinforcement of correct procedure should instil a respectful appreciation of liability, rather than excessive apprehension.

New Zealand courts have tended not to expressly consider questions of public authority knowledge, competence and information structures in liability assessments. They have, however, consistently placed importance on the nature of the authority’s discretion, and the statutory framework in which it operates. A common theme in the case law is that authorities with strong discretionary powers, or which are of a quasi-judicial nature are not normally liable in negligence. Ellis, Fyfe, Morrison, Bella Vista, Monticello, Sacramento and The Grange are all authorities for the proposition that where a given authority has a strong discretion empowered by statute, or is of a quasi-judicial character, defensive practice concerns carry much more weight.¹³³Conversely, where a common law duty is thought appropriate to buttress statutory duties, deterrence has been stressed.¹³⁴

130 Joanna C Schwartz “Myths and Mechanics of Deterrence: The Role of Lawsuits in Law Enforcement Decisionmaking” (2010) 57 UCLA L Rev 1023 at 1085.

132 John Hartshorne, Nicholas Smith and Rosemarie Everton “‘Caparo Under Fire’: A Study into the Effects upon the Fire Service of Liability in Negligence” (2000) 63 MLR 502 at 520.

133 Ellis, above n 16, at [175]; Fyfe, above n 25, at [58]; Morrison, above n 31, at 338; Bella Vista, above n 36, at [75]; Monticello, above n 44, at [88]; Sacramento, above n 47, at [93]; and The Grange, above n 51, at [53].

A related argument is that immunity is justified where the risk of defensive practice by a given authority is thought to have significant negative implications for the public interest.¹³⁵Police therefore have been the most obvious beneficiary of immunity from actions in negligence. This reasoning has been evident in New Zealand.¹³⁶It is difficult to delimit which authorities should attract a privileged immunity. Justifications for these immunities are ringing hollow—widely criticised in legal commentaries and departed from in other jurisdictions.¹³⁷While the New Zealand courts have not followed the House of Lords’ example such that defensive practice operates as a blanket immunity, there nevertheless remains room for a more reasoned approach in assessing liability. By integrating considerations of legal competence and information into existing inquiries into statutory frameworks and discretion, the New Zealand courts will be better placed to make principled assessments about which authorities should be insulated on the basis of defensive practice.

E Quality of conduct and proximity

The kind of wrong committed and harm suffered are relevant considerations in determining whether liability will promote defensiveness or deterrence. The Law Commission notes that “[i]llegality can take a number of forms ... it should not be assumed that they are all equally responsive to a specific mode of deterrence.”¹³⁸Liability could incentivise higher standards of care in cases where the authority has been inattentive.¹³⁹The same consequence is unlikely where the alleged negligence stems from a lack of resources.¹⁴⁰Where the harmed party has suffered a personal injury, this may give rise to a stronger call for deterrence. In New Zealand, this tends not to be a relevant consideration because of the no-fault accident compensation scheme. ¹⁴¹Liability would only result where the authority’s conduct was so outrageous that exemplary damages would be warranted. ¹⁴²This was evident in Hobson, where William Young P considered that defensive practice concerns were strong, and the accident compensation scheme meant that concerns of appropriate redress were not a counterweight.¹⁵⁴

The nature of the claim in relation to the impugned conduct affects the relative strength of defensive practice concerns. Certain types of conduct are scarcely justiciable. Takaro and The Grange suggest that where the exercise of powers can be described as having a policy rather than operational character, a duty should not properly be found.¹⁴³Morrison held that a negligence standard was inappropriate in a case of alleged error in interpretation.¹⁴⁴These cases suggest that where questions of justiciability are at issue, defensive practice concerns appear to carry more weight.

135	Hill, above n 1, at 63.
136	Evers, above n 20, at [34].
137	See, for example, Claire McIvor, above n 96, at 133.
138	Law Commission, above n 98, at [B.9].
139	At [B.10].
140	At [B.10].
141	Accident Compensation Act 2001, s 317.
142	Section 319.
154	Hobson, above n 67, at [127].
143	Takaro, above n 86, at 709; and The Grange, above n 51, at [32].
144	Morrison, above n 31, at 338.

Proximity is important. Where the authority is “causally peripheral”, the defensive practice concern offers a compelling reason to deny a finding of duty. ¹⁴⁵This most obviously comes to bear in cases involving an authority’s failure to control a harm-causing third party. This reasoning is evident in the body of New Zealand case law, for example Evers and Hobson.¹⁴⁶In the former case, it was found that a police failure to control did not warrant a duty. The finding of a duty in the case of a positive negligent arrest by police in Fyfe signals that the public authority context is not determinative. This proposition is supported by the international authorities. The United Kingdom Supreme Court, in Robinson v Chief Constable of West Yorkshire Police, ¹⁴⁷belatedly disturbed the longstanding blanket immunity of police conferred by Hill. This case also dealt with a negligent arrest and found police could properly be liable in negligence for positive negligent acts.¹⁴⁸The vital component of Hill warranting insulation from liability was that the police there had failed to protect a victim (an omission).¹⁴⁹

In weak-proximity situations—such as where a failure to control or protect has arisen from an omission—joint and several liability operates inequitably. The relative causal responsibility of the authority and the primary tortfeasor is lopsided.¹⁵⁰Jane Stapleton stresses that this problem is particularly acute where the peripheral party is a public authority. Deep-pocketed authorities are sometimes improperly made to pay damages out of budgets that should be applied to public duties.¹⁵¹Public authorities are tasked with upholding the public interest and should therefore not be exposed to liability out of proportion with their actual responsibility.¹⁵²Gendall J, in Monticello, held that this was an important consideration that tended to add weight to concerns of defensive practice.¹⁵³

The proximity question is a two-way street. Hanna Wilberg has identified that where a close proximate relationship exists, courts have tended to favour deterrence. ¹⁵⁴For example, where there is a discernible assumption of responsibility on the part of the authority, the defensive practice concern has carried little weight.¹⁵⁵Similar reasoning is evident in the New Zealand context—S v Midcentral and Prince both involved close proximity relationships, and in both cases deterrence and high standards of administration were stressed in a finding for duty. ¹⁵⁶Wilberg suggests that defensive practice should not be regarded as a de facto immunity from liability, but a qualified protection that is subject to several factors—with proximity being a key feature in the overriding deterrence/defensiveness calculation.¹⁵⁷

145 Jane Stapleton “Duty of care: peripheral parties and alternative opportunities for deterrence” (1995) 111 LQR 301 at 310.

154 Hanna Wilberg “Defensive practice or conflict of duties? Policy concerns in public authority negligence claims” (2010) 126 LQR 420 at 435.

F Conflict of duties

The formulation of the defensive practice concern affects its strength. Defensive practice can manifest as a conflict of duties.¹⁵⁸As stated earlier, where an alleged duty of care is incompatible with the strong existing public duty of a given authority, this provides sufficient reason to decline to impose that alleged duty. Wilberg explains that a conflicting duties situation necessarily entails a form of defensive practice. Where a statutory scheme treats one competing interest as paramount, any duty that would conflict with that interest directly interferes with the public authority’s ability to perform its primary function. ¹⁵⁹Wilberg argues that defensive practice is more readily defensible in terms of “authority and ... inherent persuasiveness” where it is formulated as a strong conflict of duties argument.¹⁶⁰

Strong conflict of duties situations often arise as an incompatibility between the interests of plaintiff suspects and victims, the latter of which certain authorities have public duties to protect.¹⁶¹Wilberg says that a conflicting duties argument may also arise even where the plaintiff is a victim. This may be so if the relevant public duty is not limited to victim protection, and cannot require protection of all potential victims at all times (that is, an individual duty conflicting with a broader duty owed to the public generally).¹⁶²Perhaps broadly similar reasoning can be attributed to William Young P in Hobson. ¹⁶³He considered that proposed duties to supervise or warn would be inconsistent with the public interest in parolee reintegration, and the corresponding duties imposed by the governing legislation. As Wilberg argues, however, this formulation of the doctrine is less persuasive. ¹⁶⁴Indeed, in Couch Elias CJ attacked this approach as conferring an unjustifiable blanket immunity that may be based on erroneous assumptions.¹⁶⁵It seems fair to conclude that, to date, the weaker conflicting duties argument has not been effective in negating liability in the New Zealand context.

Conflict of duties is invoked frequently in the context of child protection. Courts have tended to find that no duty is owed to a parent suspected of abuse, as that duty would clash with the public duty of social workers to uphold the interests of possible victim children. This principle has been seen in operation in the United Kingdom, Australia and Canada.¹⁶⁶In New Zealand, B v Attorney-General is an authority for the strong conflicting duties formulation of defensive practice acting to negate liability.¹⁶⁷

There are other public authority contexts in which a conflict of duties situation could operate to dispose of liability. Wilberg argues that cases like Elguzouli-Daf v Commissioner

158 Cherie Booth and Daniel Squires The Negligence Liability of Public Authorities (Oxford University Press, New York, 2006); and Wilberg, above n 154.

of Police of the Metropolis and Brooks v Commissioner of Police for the Metropolis,¹⁶⁸which were dismissed on the basis of defensive practice generally, could be better justified on the basis of a strong conflict of duties.¹⁶⁹These cases concerned the functions of the Crown Prosecution Service and the police force respectively, and applied the core principle of Hill in a straightforward manner. However, given that prosecutors and police are under a public duty to protect victims, and that these cases involved proposed duties to potential suspects, the principle of conflicting duties should come into play.¹⁷⁰

The conflicting duties argument has been countenanced by Australian courts. ¹⁷¹Equally, however, the conflicting duties doctrine in the police context has been dismissed in Canada.¹⁷²The majority in Hill v Hamilton-Wentworth Regional Police Services Board held that a duty of care to suspects would not place police officers under incompatible obligations.¹⁷³There was a greater interest in the avoidance of wrongful convictions, and the suspect (as a member of the public) shared in the greater public interest of diligent investigation.¹⁷⁴McLachlin CJ considered that a duty would have positive implications in terms of responsible administration, in the same breath dismissing the possibility that chilling effects could arise.¹⁷⁵Chilling effects were speculative and tort liability would have no adverse effect on police capacity to investigate.¹⁷⁶Her approach to the conflict question is similar to the dismissive attitude adopted by Lord Bingham in his dissenting judgment in JD v East Berkshire Community Health NHS Trust.¹⁷⁷

With respect to the Chief Justice, it is difficult to escape the sense that she has fallen into the same kind of loose reasoning that she criticises. It is unclear why she would readily dismiss concerns of defensive practice as conjectural, while at the same time asserting that increased prudence, higher standards and more effective administration would result from liability. Such arguments are equally subject to the empirical question. It is possible that McLachlin CJ’s personal convictions regarding corrective justice coloured her reasoning, and that her judgment was massaged to achieve a desired outcome. Charron J’s minority judgment represents a more sensible application of principle. She rightly identifies that a suspected criminal’s interest in being left unmolested by police will always be in conflict with the wider public interest in crime investigation.¹⁷⁸She concludes, therefore, that a private duty of care should not be recognised to a suspect under investigation.¹⁷⁹

Insofar as the responsible administration argument is relevant in finding a duty, a strong conflict of duties argument would usually negate it. While a general defensive practice concern is subject to the ambiguity of a deterrence/defensiveness calculation, the strong conflict of duties situation poses an unambiguously negative threat. ¹⁸⁰Wilberg further argues that a strong conflict of duties effectively negates the empirical question in

168	Elguzouli-Daf, above n 4; and Brooks v Commissioner of Police for the Metropolis [2005] UKHL
	24[2005] UKHL 24; , [2005] 1 WLR 1495.
169	Wilberg, above n 154, at 429 and 445.
170	At 429.
171	Tame v New South Wales [2002] HCA 35, (2002) 211 CLR 317 at [231].
172	Hill v Hamilton-Wentworth Regional Police Services Board [2007] SCC 41, [2007] 3 SCR 129.
173	At [40] per McLachlin CJ.
174	At [41].
175	At [43].
176	At [57].
177	JD v East Berkshire, above n 182.
178	Hill, above n 172, at [140].
179	At [153].
180	Wilberg, above n 154, at 438.

that where there is any possibility of bringing about such a strongly negative consequence, there is a large incentive not to take the risk.¹⁸¹McLachlin CJ’s points in favour of liability where police are subject to conflicting duties therefore erode.

It is important that the legislative intent of Parliament is not undercut by the courts, and this rationale provides additional support to a conflict of duties argument where the overriding public duty is one conferred by statute.¹⁸²Accordingly, the strong conflict of duties situation provides a relatively unimpeachable argument against finding a duty of care. Where it can be properly identified, this should in most cases be sufficient to override public interest in deterrence. Furthermore, the courts should be cautious of invoking the relatively weaker general formulation in situations where examination would reveal the conflicting duties argument to be appropriate. Presumably, there will be a range of circumstances that would constitute a conflict of duties warranting protection from liability. In New Zealand, increased judicial scrutiny and recognition of the different forms of defensive practice should provide clarity in explaining why certain public authority contexts should attract protection more than others.¹⁸³

G Alternatives

It is worth considering to what extent certain objectives of tort law, such as deterrence and responsible administration, may be achieved by alternative means. Public authorities may be deterred from improper conduct through internal procedures, negative media exposure, and judicial review. De Mot and Faure similarly argue that tangential consequences of liability can strongly incentivise either deterrence or defensive behaviour.¹⁸⁴The reputational loss following from an adverse verdict can have chilling effects where the media frames the authority’s conduct in an unflattering light. ¹⁸⁵The extent of this will again depend on the particular nature of the defendant authority. For example, Simon Halliday, Jonathan Ilan and Colin Scott state that sectors like social work, education and policing would tend to attract greater political costs than would an authority responsible for upkeep of roads.¹⁸⁶

An authority usually has relationships with the public, other authorities, and regulators.¹⁸⁷Liability will influence the relative esteem the authority is held in by these parties. The strength of the influence will depend on several factors bearing upon these relationships—such as whether the authority’s conduct is politically damaging or expedient, whether claims are responded to quickly and satisfactorily, whether claims are settled out of court or litigated in the public domain, and whether the performance indicators in relation to liability set by regulators are respected by the authority.¹⁸⁸

As reputational pressures can arguably chill the public authority’s behaviour, accountability pressures beyond liability can also promote responsible administration.¹⁸⁹

186 Simon Halliday, Jonathan Ilan and Colin Scott “The Public Management of Liability Risks” (2011) 31 OJLS 527 at 548.

Halliday, Ilan and Scott argue this point in terms of “bottom-up pressures” acting upon the authority, such as complaints processes, and “top-down pressures” such as performance audits and reviews.¹⁹⁰They argue that accountability regimes outside the usual resort to tort measures can in this way “weaken the grip of liability risks on public service delivery”.¹⁹¹Laura CH Hoyano, conversely, casts internal inquiries in an uncertain light in the wake of Brooks.¹⁹²She argues that the inquiry into the murder of Stephen Lawrence proved that post-mortem examinations may be subject to deficiencies, and that the disciplinary processes undertaken by authorities may lack independence. ¹⁹³That her argument is situated specifically in the context of police is an important qualification. These concerns may not necessarily come to bear as strongly in other public authority contexts. This should be a determination for the courts, assuming they are in possession of adequate information about the relevant authority’s internal systems and procedures. New Zealand case law is divided on the suitability of alternative means of accountability and deterrence. In Prince, Henry J noted that accountability could not be satisfactorily achieved through other means like making a complaint to the Ombudsman, ministerial and parliamentary oversight, or judicial review.¹⁹⁴Moreover, in S v Midcentral, William Young J considered that disciplinary proceedings for psychiatric staff who “negligently” released dangerous patients would be far more likely to promote unnecessarily defensive medicine than the possibility that their employers face a claim for exemplary damages.¹⁹⁵However, Ellis, Bella Vista, and Morrison made clear that measures like complaints processes and causes of action against other parties could be persuasive

It is not clear what particular features of the cases made alternatives to tortious liability more compelling in some contexts than others. This is typical of confused and internally inconsistent jurisprudence on defensive practice and deterrence concerns as they relate to public authorities. New Zealand courts can shore up these deficiencies in future cases by expressly contemplating the appropriateness of monetary liability with respect to likely reputational costs of litigation, availability of valid alternatives, and the scope of political accountability in a given public authority context. These factors all bear directly on the strength of the defensive practice concern and are underrepresented in the body of New Zealand law.

IV Conclusion

Are defensive practice arguments defensible? How significant is the public interest in deterring an authority’s wrongful conduct? What should the approach to negligence liability be in New Zealand in light of the competing considerations of chilling effects and improved administration? These are hefty questions, to which there are no clear-cut answers. Each consideration in the overall calculation seems to have a countervailing

193 Laura CH Hoyano “Policing Flawed Police Investigations: Unravelling the Blanket” (1999) 62 MLR 912 at 934.

196 Ellis, above n 16, at [171]; Bella Vista, above n 36, at [60]; and Morrison, above n 31, at 338.

argument, making any definitive approach to policy tenuous. Any arguments about the appropriate ambit of the defensive practice concern must be subject to the qualification that a single, unified formula is likely impossible.¹⁹⁷

My aim has not been to propose an absolute solution. Rather, I have sought to show the most universal variables bearing upon the strength and validity of the concern in different public authority contexts. This forms a useful heuristic for calculating whether liability is appropriate in a given situation. I have used these variables to both explain and criticise the existing body of New Zealand law dealing with defensive practice and deterrence concerns. I have also shown how some of these important factors are comparatively underdeveloped and unrepresented in existing jurisprudence. I have suggested that deliberate and express contemplation of these neglected factors should allow for a more reasoned approach to weighing defensive practice against deterrence in the wider liability equation.

The charge may be levelled that a broad, multifactorial framework could have the effect of making the liability equation even more fragmented and needlessly complex. I answer that some kind of taxonomy must surely be better than the confused assortment of authorities that currently passes for jurisprudence on defensive practice and deterrence. Furthermore, this proposed system is to be regarded as a gestalt rather than a definitive checklist. The overall calculation will be highly context-sensitive. As Halliday, Ilan and Scott observe, a number of factors must be considered when “thinking, context by context, about the promise or threat of tort law”.¹⁹⁸It is impossible to quantify the relative weight of each of these factors in the duty equation. This would be a determination for the courts in light of the particular circumstances.

It is clear that there is still much to be established about the way in which public authorities modify behaviour in response to competing incentives and disincentives (financial, reputational, political or otherwise). Bare assertions about the influence of legal challenge and monetary penalties on decision making must therefore be made cautiously.¹⁹⁹On balance, however, it is fair to state that reliance on defensive practice as a wholesale exclusionary rule for liability is untenable.²⁰⁰This does little justice to the nuances involved in the wider duty inquiry. Equally, broad brush statements asserting the desirability of deterrence and responsible administration are often plagued by a seemingly wilful blindness toward the legitimate rationales underlying the defensive practice concern.

The body of authority considering defensiveness and deterrence in the policy arm of the duty question has left the law in a rather confused state. This must be the kind of unsatisfactory state of affairs that Burrough J contemplated when he characterised public policy as “a very unruly horse ... once you get astride it you never know where it will carry you”.²⁰¹However, attention to the factors identified in this article should ameliorate this concern. As De Mot and Faure conclude, due contemplation of the relevant factors will help to “[delineate] the types of cases in which a restrictive approach is justifiable ... [this]

could significantly limit the scope for the chilling effect argument to be incorrectly decided”.²⁰²Ultimately, while the duty calculation will inevitably be subject to a complex range of difficulties and considerations, this is no reason to shy away from attempting to tame the policy horse. The continued search for clarity in tort law is one from which we should not be deterred.

Public Interest Law Journal of New Zealand

Norton, Louis --- "Is 'defensive practice' defensible? Chilling effects and responsible administration with regard to public authority negligence claims in New Zealand" [2018] NZPubIntLawJl 9; (2018) 5 PILJNZ 5

I Introduction

II Survey of New Zealand Case Law

A Property transactions

B Medical practitioners

C Police

D Building and land consents

E Leaky buildings

F Child protection

G Parolee supervision

H Ministerial error

III The Defensiveness/Deterrence Dichotomy

A The empirical dimension

B Quantum of damages

C Insurance

D The defendant authority

E Quality of conduct and proximity

F Conflict of duties

G Alternatives

IV Conclusion