CONNECTION AND NAVIGATION SERVICE LIMITED AND ACCIDENT COMPENSATION CORPORATION
[2024] NZHC 2737
•20 September 2024
IN THE HIGH COURT OF NEW ZEALAND WELLINGTON REGISTRY
I TE KŌTI MATUA O AOTEAROA TE WHANGANUI-A-TARA ROHE
CIV-2024-485-341 [2024] NZHC 2737
UNDER the Judicial Review Procedure Act 2016 BETWEEN
CONNECTION AND NAVIGATION SERVICE LIMITED
Applicant
AND
ACCIDENT COMPENSATION CORPORATION
Respondent
Hearing: 19&20 August 2024 Counsel:
T Mijatov, C R J O’Fee and S A Davies for Applicant C J Curran, J A Tocher and F G Wilson for Respondent
Judgment:
20 September 2024
JUDGMENT OF GRAU J
Introduction
[1] In this judicial review proceeding, the applicant, Connection and Navigation Service Limited (CNS), contends that the Accident Compensation Corporation (ACC) has unlawfully failed to follow, or take into account, its alternative dispute resolution (ADR) policy in which ACC states it prefers to use ADR to resolve disputes with claimants. CNS also contends that ACC has breached CNS’s legitimate expectation that ACC would comply with its policy, which it says is a formal statement about how ACC will behave in most cases.
[2] ACC’s position is that its ADR policy does not say what CNS asserts it to mean, and CNS has misunderstood the policy. ACC’s approach to ADR is that it supports its use in appropriate cases, but it retains a complete discretion over whether to agree to
CONNECTION AND NAVIGATION SERVICE LIMITED v ACCIDENT COMPENSATION CORPORATION [2024] NZHC 2737 [20 September 2024]
use ADR, and it is not legally required to refer claims to an external ADR provider as CNS seeks. That position is consistent with the plain wording of what is a staff policy, the statutory framework, and related documentation about the use of ADR by ACC. ACC’s position is that CNS is using public law tools, designed to hold public agencies to their well-established practices, to instead require ACC to adopt a new practice for the benefit of CNS’s business model.
Background
The parties
[3] ACC is the Crown entity founded under the Accident Compensation Act 2001 (the Act) to administer Aotearoa’s accident compensation scheme, which provides no fault personal injury cover for everyone. The Act’s broadly stated purpose in s 3 is “to provide for a fair and sustainable scheme for managing personal injury” with the overall goals of “minimising both the overall incidence of injury in the community and the impact on the community (including economic, social and personal costs)”. The purpose is to be achieved through (among other things) ensuring that when injuries occur, the focus should be on rehabilitation with the goal of achieving an appropriate quality of life through the provision of entitlements that restore a claimant’s health, independence and participation, and to ensure claimants receive fair compensation.1
[4] Another of the specified ways of achieving this purpose is in “ensuring positive claimant interactions with [ACC] by developing the Code of Claimants’ Rights” (the Code).2 The Code encourages “positive relationships between ACC and claimants” with “a partnership based on mutual trust, respect, understanding and participation”. 3 The policy at issue in this case expressly refers to the Code as governing ACC’s dealings with their clients. On the other hand, the Code itself says that any disputes about cover and entitlements, including treatment and compensation, are not covered by the Code “and continue to be addressed by the mechanisms under the Act”.4
1 Accident Compensation Act 2001, s 3(c).
2 Section 3(e); in reference to the Injury Prevention, Rehabilitation, and Compensation (Code of ACC Claimants’ Rights) Notice 2002 [The Code].
3 The Code at 1.3.
4 The Code at 1.4.
[5] ACC must comply with a range of statutory duties, including the obligations set out in s 262 of the Act, which require ACC to deliver the services required by the Act “in order to minimise the overall incidence and costs to the community of personal injury while ensuring fair rehabilitation and compensation for loss from personal injury”.5 ACC must also deliver those services in a manner that is “cost effective and promotes administrative efficiency”.6
[6] CNS provides advocacy services representing injured people in disputes with ACC. It has been in operation since March 2024. Mr Warren Forster is the director of CNS. He is a barrister and researcher, with a keen interest in the ACC system and access to justice. He says he built the business in the knowledge it was likely to be viable because it would give effect to ACC’s own policies, including its ADR policy. Presently CNS advocates for around 200 people, and on their behalf, CNS has lodged 125 review applications with ACC. Its business model relies heavily on the use of ADR. As indicated above, the crux of the complaint in this case is that Mr Forster says he expected ACC would apply its own policy in considering each of CNS’s clients’ requests for the use of ADR, but that has not been the case. ACC has declined to use ADR for all but 12 of CNS’ clients.
[7] In early 2019, Mr Forster (through a company called Simply Resolution Ltd) had successfully submitted a tender to establish and run the ADR process in the ACC dispute resolution system. This was called Talk-Meet-Resolve (TMR). TMR features in the narrative and evidence at issue in this proceeding, so it is useful to make note of its relevance at this stage.
The ACC scheme
[8] When a person is accidentally injured in Aotearoa, they (or their healthcare provider) can lodge a claim with ACC. ACC generally receives around two million claims each year. The first step is to establish whether an injury is covered by the scheme. The Act sets out what types of injuries are covered.7 Once cover is accepted ACC then determines whether the claimant is eligible for entitlements in respect of
5 Section 262(3)(a).
6 Section 262(3)(b).
7 Sections 20, 26, and 28; For example, injuries caused by a specific event are generally covered, but not those caused by a long-term gradual process (unless the injury is work-related).
their injury, for example medical treatment, rehabilitation, compensation for loss of wages/salary, or a lump sum compensation payment for permanent impairment.8
[9] A claimant who is unhappy with ACC’s decision on their claim may apply for a review of ACC’s decision. Around 9,000 reviews are filed annually. Part 5 of the Act provides the statutory review process for ACC’s decisions concerning a claimant’s injury. ACC’s decision on a claim continues to be of full effect until the review application is determined.9 The review process is intended to be quick, cost-efficient, and claimant-centric. Its key features include:
(a)Short timeframes: a date for a review hearing must be set within three months of the review application, otherwise the decision is deemed in favour of the claimant.10
(b)Minimal costs: ACC is responsible for its own costs and the claimant’s costs on review. This includes paying ADR providers. ACC may also be responsible for some representation and other costs incurred by the claimant.11
(c)Accessible appeal rights: a claimant can appeal a review decision to the District Court.12 There are no filing fees for claimants appealing to the District Court against review decisions, and ACC will ordinarily not seek costs against unsuccessful appellants.13
Alternative dispute resolution
[10] During the review process, various forms of dispute resolution can occur. One form is “internal resolution”, including so-called “early resolution”, where ACC directly engages with a claimant via a “Review Specialist” (who is an ACC staff member), and resolves the review application by agreement. Another form of resolution is by ADR, which is mainly via mediation or conciliation. The third form
8 Section 69.
9 Section 133(1).
10 Section 146.
11 Section 148.
12 Section 149. Further appeals are available, with leave, to the High Court and Court of Appeal on a question of law.
13 Section 164.
is by way of a formal review hearing, the result of which can be appealed to the District Court.
[11] It is important to note here that, generally speaking, ADR is a term that can refer to any method of resolving a dispute that does not involve going to court or a statutory adjudication. In the context of the ACC scheme and this proceeding, ADR more specifically refers to the assistance of an independent party (such as TMR, an external provider of ADR services), to conduct a conciliation or mediation.
[12] The only reference to ADR in the Act itself is in a section that deals with regulation-making powers. Section 328A(1)(a) provides that regulations may be made setting out a framework or rules (or both) governing the conduct of ADR. Despite that this section was added to the Act in 2005,14 no regulations have yet been made.
[13] Section 328A(2) provides that in the absence of regulations in force under subs (1)(a):
… the use of alternative dispute resolution and the manner in which it is to be conducted is a matter for agreement between [ACC] and the claimant.
ACC’s ADR policy
[14] ACC has a policy about the use of ADR in its dispute resolution process (the Policy). Given its brevity, and its centrality to the proceeding, it is set out in full below:
Summary Objective
Our alternative dispute resolution (ADR) options include mediation,
facilitation, and conciliation. ACC usually uses conciliation or evaluative mediation as the type of ADR. ACC can refer disputes or difficult situations with clients to Talk-Meet-Resolve for ADR. A client can also ask for ADR. We prefer to use ADR to resolve disputes if there was more than one possible outcome.
Occasionally when the dispute is related to an issue at review and an Independent Reviewer provider has already been engaged, Fairway Resolutions Ltd or Independent Complaint and Review Authority (ICRA) will be engaged to hold conciliation to assist ACC in finding a resolution or way forward.
…
14 By the Injury Prevention, Rehabilitation, and Compensation Amendment Act (No 2) 2005 .
Policy
1.0 Rules
a The Code of ACC Claimants’ Rights governs our dealings with clients.
Clients applying for review can be invited to attend some form of ADR. ADR is also available for any customer outside of the review process.
ACC uses conciliation or evaluative mediation, but other forms of mediation are available when required.
2.0Exceptions
a We may not offer ADR when we receive a review application or reapplication, if:
•the Remote Claims Unit manages the claim.
•the review application or reapplication is from a risky client or a vexatious or continual review applicant.
If you think there is a case for ADR, you must discuss it with a team manager, a resolution specialist or a business unit manager.
[15] The Policy is an internal one. It is intended to be used by ACC staff to guide them on the use of ADR. ACC says it was not drafted with the public in mind, nor was it intended to be used or relied on by the general public. It is not sent to claimants. The evidence of ACC’s Principal Advisor of Resolution Services, Ms Greer Dines, is that, with the exception of Mr Forster, she cannot recall a claimant or representative ever relying on the ADR Policy.
[16] The genesis of the ADR Policy is disputed. The evidence for CNS says it came in at a time of a change in ACC’s approach to ADR in 2019. Mr Forster says he understood it was to “give effect to client choice to use ADR”. ACC says its ADR Policy was already in existence before 2019 and there was no “sea-change” in 2019. ACC’s evidence is that the Policy was added to ACC’s internal staff system, ProMapp, in May 2018, but it did not go “live” for ACC staff to see until February 2019. ACC has been unable to ascertain when the Policy was first created, but the substantive wording of it has remained unchanged from 2018 to 2024, contrary to CNS’s allegation that the Policy implemented a change in ACC’s approach. CNS says that this timeline is consistent with its position that there had been a change in attitude, given the Policy had not been available to staff until 2019.
What does the Policy mean?
Positions of the parties
[17] The parties do not dispute the lawfulness of the Policy. Nor do they dispute the principles that apply to the interpretation of a policy. But they do have different interpretations of the Policy. They also have different interpretations of what the other’s case is.
[18] CNS’s position (as best as I can understand it) is that the ADR Policy generally requires ACC to agree to use an external ADR provider whenever a claimant requests it, subject to the limited specified exceptions provided for in the Policy (which involve either concerns about safety,15 or where the request is from a risky, vexatious or continual claimant).
[19] CNS says ACC has misunderstood its case, which is not that the claimant has sole discretion over whether to use ADR. Instead, CNS says the practical effect of the Policy is that ADR will tend to be appropriate where the claimant also elects to use ADR because ACC has already indicated its preference for ADR in the Policy. When ACC has agreed by its Policy that it will prefer to use ADR, the claimant’s election to use ADR relies on the claimant first being adequately informed about ADR. Thus, in practical terms, the first thing a Review Specialist needs to do when faced with an application for review is to tell the claimant about ADR and that ACC prefers to use it. If the claimant elects to use ADR, then it will normally be the appropriate avenue. At the hearing, Mr Mijatov conceded that CNS’s position, if accepted, would require Review Specialists and claimants to engage an external provider and book in a time for ADR the moment a claimant chooses ADR (which could be at the time of the first contact between the parties), even if this choice came before the Review Specialist attempted to resolve the dispute with the claimant directly. I note at this point that there is an obvious logical difficulty in CNS’s case, when it appears to suggest that ACC does retain a discretion in each case and can choose not to engage in ADR, but at the same time also suggests ACC has already exercised its discretion via the Policy such that ACC effectively cannot choose not to engage in ADR if the claimant requests it.
15 It was explained at the hearing that the remote claims unit was set up after the murder of an ACC case manager to deal with clients who pose a risk to the safety of ACC staff.
[20] ACC says CNS has mischaracterised ACC’s position. CNS is saying ACC is required to agree to use an external ADR provider whenever a claimant requests it (subject to limited specified exceptions). But ACC does not operate such a policy and is not required to. Rather, ACC retains complete discretion over whether to offer or agree to use an external ADR provider to resolve a dispute. But this does not mean ACC has sole discretion either—a claimant also has complete discretion to decline ADR if it is offered or suggested by ACC. ACC’s position is that its understanding of its Policy is consistent with its plain wording, the statutory framework, and related documentation about the use of ADR by ACC.
Legal principles
[21] Expressions of policy can give rise to enforceable obligations and their interpretation can constitute a question of law subject to judicial oversight and intervention.16 The starting point is that the Policy must be read in light of the governing legislative provision. The relevant interpretative approach to be taken is set out by the Court of Appeal in Patel v Chief Executive of the Department of Labour (Patel):17
A policy document, such as the one in issue, is not to be construed with the strictness which might be regarded as appropriate to the interpretation of a statute or statutory instrument … It must be construed sensibly according to the purpose of the policy and the natural meaning of the language in the context in which it is employed …
[22] The Supreme Court has said recently that “isolating and de-contexualising individual provisions in a manner that does not fairly reflect the broad intent of the drafters must be avoided”.18
The plain or natural meaning of “prefer”
[23] The ADR Policy is in two sections. The first is a “Summary” describing the “Objective” of the Policy and listing the “Owner” and “Expert” associated with it. The second is entitled “Policy”, which contains “Rules” and “Exceptions” to those rules. ACC’s stated preference for ADR that is at the heart of this case is in the “Objective” section of the Policy.
16 Pora v Attorney-General [2017] NZHC 2081, [2017] 3 NZLR 683 at [90] and [105].
17 Patel v Chief Executive of the Department of Labour [1997] NZAR 264 (CA) at 271.
18 Royal Forest and Bird Protection society of New Zealand Inc v New Zealand Transport Agency
[2024] NZSC 26 at [79].
[24] ACC points to the meaning of “prefer” as being “to like better or value more highly”.19 CNS says to “prefer” something means to place it before or ahead of an alternative, noting it comes from the Latin “to place before, especially in terms of esteem”.20
[25] CNS argues that preference language means, given a choice between ADR and formal review, ACC will elect ADR ahead of formal review. In contrast, ACC says the Policy indicates an organisational attitude of support for ADR, but that attitude does not govern the discretion contained in the “Rules” section. It is not a commitment to act in a certain way.
[26] In my view, there is in reality little difference between the parties as what the word “prefer” actually means. What is clear, however, is that the objective of the Policy is not expressed in mandatory language, as would be expected in an objective, as opposed to a rule. As I will go on to discuss below, when it is placed in the context of the Policy as a whole, ACC’s statement that it “prefers to use ADR to resolve disputes if there was more than one possible outcome” is a statement of a goal or aspiration. It is not a requirement for a particular action (that is, to engage in ADR). It is a statement of what ACC’s attitude to ADR is.
Context of the use of “prefer” in the Policy
[27] Looking at the “Objective” section in its entirety, it starts with an explanation of what ACC’s ADR options include. It then states that ACC “can refer disputes or difficult situations with clients to [TMR, who, as above, had won the tender to establish and run ADR services] for ADR”, and that a client “can ask for mediation”.21 Then the paragraph concludes with the stated preference to use ADR to resolve disputes, “if there was more than one possible outcome”.
[28] Despite its argument that ACC is effectively bound by its stated preference to use ADR, CNS says the Policy means ACC does not have to go to ADR for every dispute (for example, if a claim was patently hopeless). CNS says the exceptions
19 Citing Collins English Dictionary (13th ed, HarperCollins, Glasgow, 2018) at 1573; Merriam- Webster Dictionary “Prefer” (10 August 2024) Merriam Webster < and Oxford English Dictionary “Prefer” (2007) Oxford University Press < Citing Vicky Aldus and Mary O’Neill (eds) Chambers Concise Dictionary (Chambers Harrap, United Kingdom, 2009) at 959.
21 Emphasis added.
contained in the Policy (and the Policy as a whole) reinforce ACC’s preference for ADR in cases involving genuine and arguable disputes. ACC says the Policy reflects the reality that ADR is typically less appropriate when there is no genuine flexibility available to ACC in response to the review application, for example, if ACC’s decision had to be overturned or where the Act directed the outcome (such as where cover is expressly excluded under the Act).
[29] As above, the stated preference for ADR is in the “Objective” section of the Policy. It is not in the operative section of the Policy which contains the “rules”. I agree with CNS that it is a statement of ACC’s preference to use ADR, which must be a preference to use ADR over the formal review process. However, if ACC intended that ADR was to be the default position, with narrow exceptions, I would have expected that expressed preference to have been clearly set out in the rules and phrased in mandatory terms. For example, the Policy could have been phrased as an instruction to staff that they “must” offer ADR to claimants as soon as they become engaged with a claimant for review, and immediately proceed down that track—which is what Mr Mijatov at the hearing ultimately accepted was CNS’s position. Instead, all of the wording in the “Objective” section is highly discretionary.
[30] Turning to the “Policy” section and the rules therein, there is a mixture of prescriptive and discretionary wording. The first part states that the Code of ACC Claimants’ rights “governs” dealings with clients. That leaves no room for discretion although, as I indicated above, the Code itself excludes disputes about cover and entitlements. Thus, this part of the rules relates to the process, rather than the substance of the interaction with the claimant about a review. Then there is a repeat of the discretionary language relating directly to the actual use of ADR, which sets out that a claimant applying for review “can be invited” to attend ADR (as well as it being available for any claimant outside of the review process). Again, that language does not lend itself to an interpretation that ACC has pre-committed to ADR as the starting point in a review process. In my view, it is more naturally read as an instruction to ACC staff to consider the use of ADR if they think it appropriate. The final part of the rule describes again the types of ADR ACC generally uses, but states that other forms are available “when required”.
[31] Finally, there are the “Exceptions”. They are not hard and fast exceptions either. Rather ACC “may not” (rather than “must not”) offer ADR in two defined
circumstances that relate to claimants who generally either present a safety risk to staff or are vexatious. The Review Specialist may still offer it, though. But if they do think there is a case for ADR (despite the attendant risk involved), they are instructed to discuss it with a manager. Those exceptions are consistent with the aspirational objective of the Policy and with its discretionary nature.
[32] Read as a whole, in my view, the interpretation CNS urges on the Court is a strained one, taking a single sentence from the non-operative “Objective” section of the Policy and then using the exceptions to seek an interpretation that is, in essence, a pre-agreement by ACC that it will always go straight to ADR if the claimant wishes to, whether or not ACC considers the case is suitable for that pathway.
[33] I also take into account that the ADR Policy is an internal policy directed at ACC review staff, not an outward manifestation of a set position. That status also suggests to me that the preference for ADR contained in the Policy is aspirational only. And it is consistent with the natural meaning of “prefer”. A preference is a course that a person would rather take, not a course that they will necessarily take, or must take. As a very simple example, it is likely that many (or possibly most) people would prefer to stay in bed rather than get up early to go to work. But most people nevertheless get up early (or earlier than they would prefer to) and go to work. The natural meaning of “prefer” also means that the Policy is not pointless, as CNS asserts it would be on ACC’s interpretation, when its purpose appears to be to provide guidance (and encouragement) to staff on the use of ADR.
[34] The ultimate question is whether ACC has, through this Policy, effectively self- limited its otherwise unconstrained statutory discretion whether to choose to agree to ADR in each individual case. As I will come to below, the statutory context of the ADR Policy instead suggests CNS’s interpretation would amount to ACC fettering the discretion it has been accorded by Parliament.
Statutory context
[35] Turning to the wider statutory context, in my view, ACC’s interpretation is consistent with the only statutory reference to ADR in s 328A of the Act, which requires that the use of ADR, and the manner in which it is to be conducted, is a matter for agreement. There is no statutory limit to what is self-evidently a wide discretion.
The plain words of this provision suggest that both the claimant and ACC must consciously agree to follow an ADR process having had regard to the circumstances of the particular claim sought to be reviewed.
[36] CNS’s interpretation is not consistent with the plain words of the provision. Although CNS was at pains to suggest ACC still had to agree to ADR, the reality is that CNS’s position requires ACC to start at a point where it has already agreed to use ADR in all but a limited number of cases.
[37] To interpret the Policy in the manner CNS contends would amount to a fettering of ACC’s wide statutory discretion. The point is made well by the Court of Appeal in Criminal Bar Association of New Zealand Inc v Attorney-General:22
[118] … When a statute confers a discretion on a particular person, it cannot be altered by means other than statutory amendment. The adoption of policy guidance might be administratively convenient for a decision maker, and can advance rule of law values such as consistency and certainty in decision making. However, a policy which guides the exercise of a discretion will inevitably fetter that discretion to some extent. If that policy guidance crosses the line between legally acceptable limits on the exercise of discretion and those which are not legally acceptable, it “fetters” the discretion and is unlawful.
[119] … reliance on policy is not unlawful, but blind following of policy is. Each case must therefore be decided on its own merits, and it should be open to the person exercising the discretion to find that the policy does not apply in a particular case. If a policy is so phrased to admit of no exceptions, it is unlawful. Any policy must be based on factors and purposes relevant to the power, and must not be unreasonable. Acting pursuant to an overriding policy can also be viewed as a failure to consider relevant factors, for example the merits of the particular case.
[38] It would not be realistic, or appropriate, for ACC to agree to ADR before knowing much about the claim or the claimant, as it leaves no room for judgment informed by the circumstances of the particular review. Public law does not permit such abdications of discretion: a public body cannot unlawfully fetter itself by adopting and applying rigid policies that do not reflect the width of the discretion intended by Parliament.23
22 Criminal Bar Association of New Zealand Inc v Attorney-General [2013] NZCA 176, [2013] NZAR 1409 (footnotes omitted, emphasis added).
23 Christiansen v Director-General of Health [2020] NZHC 887, [2020] 2 NZLR 566 at [47].
Turning tides?
[39] As indicated above, part of CNS’s argument is that in 2021 there was a change in ACC’s approach to ADR, and the Policy was simply a statement of its new position. At this point, it is useful to go back in time a little and set out some further background that CNS relies upon.
[40] In 2015, advocacy group Acclaim Otago (led by Mr Forster) released a report about ACC dispute resolution processes in which it identified barriers to access to justice. The Ministry of Business, Innovation, and Employment then commissioned an independent review of Acclaim’s report. The 2016 review, by Mirian Dean KC, found some of Acclaim’s concerns were valid and recommended improvements to existing practice (the Dean Report).24 In 2016, Cabinet accepted all the recommendations in the Dean Report.
[41] In 2017, ACC established the Resolution Services Team as part of a restructure to the ACC review process to develop more consistent decisions and management of reviews, and to improve timeliness of resolution. Review Specialists were provided with additional scope to evaluate and revisit ACC decisions.
[42] As noted above, it is also relevant that in 2019 TMR was successful in its tender to provide ACC with ADR services.
[43] CNS then relies on the following evidence which it says supports its argument that ACC had a shift in attitude in 2019 regarding the use of ADR:
(a)During the tender process for TMR, Mr Forster’s evidence is that ACC met with him and explained that, although they were unsure about the volumes of cases that would go through to ADR, giving people the choice to use it would result in an uplift in the number of cases. He deposed that ACC confirmed with him that claimants would be given the choice of using ADR, but no choice of provider, in that all ADR cases would go to TMR (save for limited exceptions such as conflicts of interest).
24 Miriam Dean Independent Review of the Acclaim Otago (Inc) July 2015 Report into Accident Compensation Dispute Resolution Processes (Ministry of Business, Innovation and Employment, May 2016).
(b)ACC and TMR met in June 2019 to agree on how conversations between ACC staff and claimants would be scripted to give people the choice to use the ADR service.
(c)Mr Pieter Burger, a Senior Review Specialist at ACC at the relevant time, gave evidence that, up to and around 2019, the choice of whether to use ADR was the decision of individual review specialists on behalf of ACC. Mr Burger says he was part of the team that was consulted about changing the ADR Policy in 2019; the change being that ADR became the preferred method of dispute resolution. He deposed that, from 2019, Review Specialists had written delegated authority to use ADR and had a wide range of delegation to settle matters.
(d)In 2020, TMR had found that people were not being given the choice to use ADR. ACC proposed to survey its staff to understand the issues involved in the process of referral to TMR. Forty Review Specialists participated. ACC’s summary of the results included that there was a “significant unconscious bias around which customers and issues will be suitable for ADR” and there were “some gaps in understanding and appreciating how effective ADR through TMR can have a positive impact on the customer journey”.
(e)In 2021, Ms Dines and Mr Forster discussed “where to next with ADR”. Ms Dines acknowledged that there was “limited information available in ACC resources for customers to understand that ADR is an option for resolving their disputes”. Ms Dines then set out a number of steps ACC intended to take to remedy this, for example by providing further web resources and changing the automatic email response sent to those making a review application.
[44] ACC’s position is that CNS has mischaracterised this evidence which does not indicate that ACC’s approach to ADR had changed.
[45] Ms Dines deposed that the “consultation” team Mr Burger says he was part of was actually a “test group”, formed and consulted on operational changes concerning ADR in 2019. Ms Dines had also been part of this group and did not recall discussions
about changing the substantive nature of the Policy. The relevant operational changes included matters such as timeframes for reviews and the use of case conferences. But there was no change to ACC’s approach to, and discretionary use of, ADR as a matter of principle.
[46] Ms Dines also says that the 2020 survey was for the purpose of determining perceptions of TMR amongst Review Specialists. It was not a survey about ADR more broadly, nor was it undertaken to understand why customers were not being given their entitlement to choose ADR, as Mr Forster alleges. And, in the email exchange between Mr Forster and Ms Dines, ACC is still clear that the intention was to increase the uptake of ADR with the “right customers and the right types of review”, supporting ACC’s position that ADR would be used only in appropriate cases.
[47] My view is that the evidence shows ACC has been willing to engage with perceived issues regarding the use of ADR and to work towards finding solutions to them. I would agree with CNS that an inference can be drawn that ACC has, at least since 2019, indicted an intention to promote ADR use. However, as is apparent, I do not think this is determinative of the issue in CNS’ favour. Put simply, the fact that ACC has indicated an intention to promote the use of ADR with claimants (ostensibly by increasing their knowledge of its availability) does not outweigh the other contextual and interpretative factors pointing to a discretionary and non-mandatory interpretation of the Policy.
Other documents
[48] I agree with the submission for CNS that an objective interpretation of the ADR Policy does not hinge on other ACC documents, especially where these other documents post-date the Policy. But I consider it is also the case that that the broader ADR material further supports ACC’s interpretation of the Policy as preserving ACC’s discretion to refuse to engage in ADR and maintaining the use of ADR as a matter for open discussion. For example:
(a)The Supplier Operational Guidelines refer to ADR being “agreed” between the parties. ACC’s responsibilities include “Coming to agreement with the customer to participate in ADR together”. While the Guidelines also state that claimants are given “the option for [ADR]
as a method of dispute resolution”, I do not agree this means, as CNS contends, it is the customer who ordinarily gets to choose whether to use ADR. Instead, the Guidelines reflect that ADR is a live issue or open question requiring prospective agreement between ACC and the claimant in each case.
(b)The “Conduct Initial Customer Contact” internal process document states that a Review Specialist should contact the claimant to explain “the focus for ACC on seeking resolution” and how the outcome of that “conversation” should “help the Review Specialist determine whether conciliation or a resolution offer may be appropriate”.
(c)The short-form ADR guide provided to Review Specialists in 2020 contains various references to the discretionary use of ADR, including how the guide can be used “to help you decide whether ADR would help to resolve … a dispute” and that “you and your customer both agree ADR would be beneficial”.
(d)The Resolution Services Review Guide 2022, an internal induction guide for Review Specialists, describes “resolution pathways”, stating there are different ways of obtaining a resolution short of formal review hearing, and that “you may try one or a combination of these options”. It describes ADR as a “tool that we can use” to assist resolution without the need for a formal review hearing.
[49] Similarly, communications to claimants also align with ACC’s position. For example:
(a)ACC’s automated email response to a review application refers to different resolution options “based on what works best for you and your circumstances including … Alternative Disputes Resolution”.
(b)Fact Sheet ACC255 states that the claimant could “talk [with ACC] … about whether ADR might be right for you”.
[50] I note this second fact sheet document, called “Kōrero mai – Get in touch with us” is a fact sheet that is said to accompany all ACC decisions. Accordingly, it is provided before any ACC claimant applies for a review. CNS accepts that it demonstrates ACC has a discretion whether or not to use ADR, but CNS says the key to the case is that this discretion is not untrammelled, and it is guided by a formal Policy that says ACC prefers ADR. In contrast, ACC says the document makes it clear that a person can request ADR, but they do not get a right of election. Instead, the suitability assessment is informed by discussion and is an open matter for consideration, assessment, discussion, and ultimately agreement by both parties. In my view the “Kōrero mai” document reflects ACC’s position that ADR is a matter for agreement between ACC and the claimant, rather than ADR being a mandated starting point.
[51] As well, there is a fact sheet that accompanies ACC’s acknowledgement of receipt of a review application which is sent to the claimant. This notice called “Reviewing an ACC decision – What you need to know” (ACC8041) states that ACC will do its best to reach a resolution at every stage of the review process and “If we can’t resolve things together, we can involve an independent conciliator or reviewer”. It goes on to state under the heading “Alternative Dispute Resolution” that sometimes using an independent conciliator can assist resolution and “If we both agree to attend [ADR], we can ask an independent conciliator to arrange a meeting at a time that suits everyone”, at which the conciliator can help to find a solution. Again, this is the language of mutual discussion and agreement, rather than an ACC-adopted starting point of the use of ADR.
[52] There is also information about ADR in the ACC review process on the websites of third parties—the Citizens Advice Bureau, Community Law, TMR, the Independent Complaint and Review Authority (ICRA), and FairWay Resolutions Ltd. The first two are organisations that provide advice to the public generally and the latter three are independent providers of review and/or mediation services for ACC. All state that mediation is voluntary if both claimant and ACC are willing.
[53] The short point is that if ACC had adopted an ADR policy that means what CNS says it means, that would be expected to be made clear (or at least hinted at) in claimant-facing documents and in information by third parties aimed at ACC
“consumers”. But it is not. Instead, ADR is consistently pitched as being an option that can be agreed to by both parties.
[54] CNS further points to correspondence between ACC and Mr Forster as strongly indicating that client choice is the touchstone for whether ADR will be used:
(a)Ms Dines had emailed Mr Forster in August 2021 stating that ACC “now want to increase ADR offering” which could be achieved by “Giving customers choice in how they resolve disputes with ACC”.
(b)Ms McKenzie, the manager of Resolution Services at ACC, sent an email to Mr Forster in August 2021 saying, “Resolution services are really committed to ensuring that customers have choice in how they resolve disputes with ACC and that the great outcomes from Talk-Meet- Resolve continue”.
[55] Ms Dines has explained that her use of “choice” referred to providing more information to claimants about ADR so they could make their views known, instead of relying on Review Specialists to educate them. Ms McKenzie’s email to Mr Forster is clearly about the same meeting that preceded Ms Dines’ email, in which there was a discussion about increasing the number of referrals to TMR, but ACC did not agree with a proposed script to actively promote TMR. In my view, this correspondence does not have the meaning CNS seeks to attach to it. When read in context, the correspondence suggests it is only about increasing claimant awareness of ADR as an option to resolve disputes, not more.
Parliamentary materials
[56] CNS also relies on ACC responses to Select Committee questions in March 2022 as supporting its interpretation of the Policy. The Select Committee summarised ACC’s response to a question as to why more reviews were not being referred to ADR:
ACC said that alternative disputes resolution is not always the most suitable approach. Its process for a claim review is first to attempt to resolve it directly with the client; second to offer the client the choice between alternative disputes resolution or a formal review hearing; and finally to have a formal review hearing.
[57] ACC objects to the use of this material, submitting it infringes s 11 of the Parliamentary Privilege Act 2014. I do not need to come to a concluded view on this point, given my findings above. But it would appear that it does, based on the terms of ss 11(d) and (e). The authority CNS relies on was concerned with interpreting legislation, in particular whether an enactment was inconsistent with a right affirmed and protected by the New Zealand Bill of Rights Act 1990.25 I do not understand it to stand as authority for the proposition that parliamentary proceedings may be used to interpret an internal policy document of a Crown entity. In contrast, in Gama Foundation v Controller and Auditor-General of New Zealand,26 the Court found that reporting of the Auditor-General was covered by parliamentary privilege when that reporting had been actively considered and addressed by a Select Committee.27 In this case, CNS is similarly seeking to use evidence presented before, and engaged with by, a Select Committee to support its position in this proceeding.
[58] In any case, if I am wrong about that, my view is that this is another example of CNS seeking to use the word “choice” to mean more than it does, when it is read in context. Although ACC’s answers to the Select Committee include that it will offer the client the choice between ADR or a formal review hearing, I do not consider the answers support the proposition that the client is the party who will ordinarily determine whether a review proceeds by ADR or another process.
Summary on the meaning of the Policy
[59] I do not accept CNS’s interpretation of the ADR Policy, based on my examination of the words of the Policy, and having regard to the relevant statutory and documentary context. There are no countervailing policy reasons that I can apprehend to suggest CNS’s interpretation should be preferred. This conclusion has a significant impact on the judicial review grounds that CNS advances. When I have found the Policy does not mean what CNS says it means, the grounds fall away, and the parties effectively acknowledged this in the hearing. That being said, I will provide some comments on the grounds of review, given the novel nature of some arguments raised.
25 Attorney-General v Taylor [2017] NZCA 215, [2017] 3 NZLR 24.
26 The Gama Foundation v Controller and Auditor-General of New Zealand [2024] NZHC 2735.
27 At [123] and [132]. See also Chamberlain v Attorney-General [2017] NZHC 1821, [2017] NZAR 1271 at [76].
Ground 1: failure to comply with the Policy
[60] CNS argues that individuals have a right to have decision-makers make determinations in accordance with that decision-maker’s policy, and, where a decision-maker fails to do so, they will have acted unlawfully. This doctrine is grounded in concepts of fairness and good public administration.28 In the case of Mandalia v Secretary of State for the Home Department (Mandalia), the Supreme Court of the United Kingdom has recognised it as related to, but distinct from, legitimate expectation.29 CNS submits this case has been endorsed (in passing) by the Supreme Court in New Zealand, as well as in the High Court. 30
[61] The Mandalia case itself related to the declining of an application for an extension of a visa.31 The visa applicant did not provide the required documentation. But the UK border agency had a process instruction, requiring it to ask an applicant for the relevant information if the application was liable to be refused solely for its absence, which it did not do.32 The applicant had no knowledge of the instruction and could not therefore rely on a breach of a legitimate expectation. But the border agency was required to comply with its own policy, and its failure to do so, for no good reason, was unlawful.33
[62] CNS says this doctrine is applicable to the present case, with the result that ACC’s failure to comply with its Policy is unlawful.
[63] This ground fails from the outset when I have found that CNS’s interpretation of the Policy is misconceived (or not available). If, as I have found, ACC’s policy does not require it to agree to ADR (or to start from a point of preference for ADR with the result that it will ordinarily go directly to a referral to ADR), then it cannot be acting unlawfully if it fails to do so. Put another way, ACC cannot be acting unlawfully by failing to act in accordance with CNS’s interpretation of ACC’s policy.
28 R (Nadarajah) v Secretary of State for the Home Department [2005] EWCA Civ 1363 at [68].
29 Mandalia v Secretary of State for the Home Department [2015] UKSC 59, [2015] 1 WLR 4546 [Mandalia] at [31].
30 By Elias CJ in Ngāti Whātua Ōrākei Trust v Attorney-General [2018] NZSC 84, [2019] 1 NZLR 116 at [75], fn 62; and Gwyn J in Te Pou Matakana Ltd v Attorney-General [2021] NZHC 2942, [2022] 2 NZLR 148 at [98], fn 26.
31 Mandalia, above n 29.
32 At [22]–[28].
33 At [29].
[64] It is also the case that Mandalia has not been endorsed in this jurisdiction. At best, it has been referred to in passing. It is but the last of a number of cases cited in the “see also” part of a footnote in a minority decision of Elias CJ (writing for herself only) in a case that was about the lawfulness of a policy itself (which is not at issue in the present case). Nor did the High Court endorse it in Te Pou Matakana Ltd v Attorney-General, beyond noting a submission by the applicant in that case. The High Court did not “agree” with that submission, as CNS suggests. Instead, the Court decided the case on the well-established principles of legitimate expectation.
[65] Nor can Mandalia be said to have been endorsed in the United Kingdom either. Rather, subsequent cases have been hesitant to adopt its principles to policies generally, beyond rigid prescriptive (and secret) policies in the context of decisions with human rights implications such as detention or deportation. For example, the Court of Appeal of England and Wales has held that Mandalia is an “unusual case”, explaining that Government policies can take many forms and “do not create legal rights as such”.34
[66] The present case is not such a case involving rigid or secret policies. It involves procedure. Even if I had accepted CNS’s interpretation of the ADR Policy, a failure to use ADR as the first port of call does not spell the end of a claimant’s review of an ACC decision, let alone have the implications that were apparent in the United Kingdom cases that have applied Mandalia. In addition, recognising a new stand- alone ground of judicial review for the failure to follow the type of policy at issue in this case would appear to me to be a step too far. It would have the effect of cutting across established legal principles applying to the judicial review grounds of mandatory relevant consideration and breach of legitimate expectation in New Zealand. Breach of legitimate expectation requires showing that reliance on a policy is reasonable, and a policy will only be a mandatory relevant consideration where its position as such is expressly or impliedly indicated in the relevant statutory scheme. It would run counter to those established grounds to allow the principles in Mandalia to apply and circumvent these requirements.
[67] I note here the evidence about the number of reviews resolved by the various options for resolution. Based on the figures for the period of July 2023 to May 2024:
34 See R (Good Law Project Ltd) v Prime Minister [2022] EWCA Civ 1580, [2023] 1 WLR 785 at [56]–[58]; citing R (A) v Secretary of State for the Home Department [2021] WLR 3931 at [3].
(a)10,727 review applications were filed with ACC and there were 9,409 outcomes (about 87 per cent).
(b)Around 73 per cent of the outcomes were resolved by agreement (6,868 in total), including by ACC overturning the decision, a settlement being reached, or withdrawal of the review:
(i)3,939 were resolved at early resolution, before the appointment of an external reviewer or ADR provider. Early resolution accounts for 42 per cent of the total review outcomes. The average timeframe for resolution of these cases was 35 days.
(ii)2,929, or 31 per cent, were resolved once an external provider had been engaged, including cases resolved at ADR as well as by internal resolution that happened after a provider was engaged. The average timeframe for resolution of these cases was 140 days.
(c)This leaves 2,541 cases where the outcome was determined at a final hearing; around 27 per cent of the total review outcomes. The average timeframe for determination of these cases was 216 days. ACC is typically successful in 70–75 per cent of review hearings.
[68] Across the wider period of July 2018 to May 2024, ACC’s average figures are that:
(a)it receives 9,035 review applications annually and delivers 8,956 outcomes;
(b)65 per cent of outcomes are resolved by agreement with a customer, whether by early resolution (37 per cent) or after an external reviewer or ADR provider had been engaged (28 per cent, using ADR or internal resolution), with 35 per cent of cases being determined at a review hearing; and
(c)nine per cent of total reviews go to ADR; this equates to 14 per cent of remaining reviews once the cases resolved by early resolution are excluded from the calculation.
[69] CNS’s evidence is that, since it began its operation, around 200 people have become clients and CNS has lodged 125 review application with ACC on their behalf. Of those only 12 have been referred to ADR. ACC’s evidence is that it has in fact agreed to ADR in around 20 per cent of those applications and has already resolved around 50 per cent of CNS’s applications.
[70] There was other evidence from CNS alleging a sudden increase in ACC’s willingness to use ADR since CNS filed its proceedings. ACC says, given that the proceedings were filed after only nine weeks of CNS being in operation (and CNS says it filed around 50 new review applications a week), the rate of resolution of review applications from CNS clients is consistent with the usual time ACC takes to carry out its established dispute resolution processes.
[71] I cannot resolve that issue and do not need to. What is tolerably clear from the numbers, however, is that ACC appears to use a range of dispute resolution processes to reach outcomes with the vast majority of claimants. Early resolution accounts for a significant proportion of resolutions. The number that go on to a formal review process is also significant, at around 27 per cent. Around one-third of reviews are resolved either through ADR or by agreement after an ADR provider has been engaged. It may well be the case that ACC could make greater use of ADR. Given its stated preference for ADR in its staff Policy, it might be expected that raising staff awareness and training staff as to the Policy’s use would be expected, and would increase ADR use. But that is a staff training issue; it is operational—it does not mean ACC is using (or failing to use) its Policy in a manner that is unlawful. But the premise of CNS’s argument, that ADR is more efficient than a review hearing so should be the preferred method of dispute resolution, is based on an incorrect assumption when the majority of review applications are being dealt with other than by a review hearing.
Ground 2: failure to consider the Policy as a mandatory relevant consideration
[72] CNS argues that, because a relevant policy is a mandatory relevant consideration a statutory decision-maker must take into account, Review Specialists
must take the ADR Policy into account every time they make a decision about whether to refer a case to ADR. When there is no evidence that Review Specialists do so, it renders their decisions unlawful. CNS says the Court cannot be satisfied all Review Specialists are aware of the Policy, and that even if they are aware of it, they are told to consider ACC’s preference for ADR when making decisions on whether to use it. The Court also cannot be satisfied as to whether Review Specialists do consider ACC’s preference in fact. Instead, Review Specialists are making decisions about resolution pathways for a variety of reasons, including reasons that are entirely unrelated to the ADR Policy, such as workload, personal views and understanding of the ADR process, assumptions about the appropriateness of disputes for ADR, and whether the original decision is robust.
[73] This ground must also fail when it is premised on a version of a Policy that ACC has not adopted. But even if CNS’s interpretation was preferred the Policy is not, in my view, a mandatory relevant consideration.
[74] Whether a policy will form a mandatory relevant consideration is clearly expressed in the case of CREEDNZ Inc v Governor-General:35
What has to be emphasised is that it is only when the statute expressly or impliedly identifies considerations required to be taken into account by the authority as a matter of legal obligation that the Court holds a decision invalid on the ground now invoked. It is not enough that a consideration is one that may properly be taken into account, nor even that it is one which many people, including the Court itself, would have taken into account if they had to make the decision. …
Questions of degree can arise here and it would be dangerous to dogmatise. But it is safe to say that the more general and the more obviously important the consideration, the readier the Court must be to hold that Parliament must have meant it to be taken into account. …
[75] The Act, neither expressly, nor by implication, requires the ADR Policy (whether on ACC’s interpretation of it or CNS’s) to be taken into account. Because no regulations have been made, the only thing the Act says about ADR is that its use is a matter for agreement between ACC and the claimant. No parliamentary intention to require ACC to turn its mind to its own internal policy can therefore be discerned. Instead, s 328A anticipates regulations could be made to govern the conduct of ADR.
35 CREEDNZ Inc v Governor-General [1981] 1 NZLR 172 (CA) at 183.
Regulations would create mandatory considerations, but an internal staff Policy does not.
[76] The Act can be contrasted with legislation which expressly requires policy to be a consideration. For example, in the Resource Management Act 1991 (RMA), subpt 1 of pt 5 expressly allows for the creation of “national policy statements”. Regional councils are empowered to prepare objectives and regional policies in s 30, and some sections of the RMA require compliance with objectives and policies (for example, s 77M). If the Policy in this case was intended to have such a binding effect on the processes of ACC, this could have been made clear in similarly unequivocal terms as in the RMA.
[77] Nor is this a situation where, for example, legal principles of clear contextual relevance (such as those in respect of tikanga Māori or international law obligations) might be an obvious matter a decision-maker must take into account despite not being expressly statutorily identified.36
[78] Finally, I note the evidence on this point, or lack thereof. As ACC correctly points out, it is for CNS, who has pleaded this ground, to prove a systemic failure to consider ADR, not for ACC to prove there is no systemic failure. CNS’s evidence does not establish a failure to consider ADR. At most, it relates to attendance at ADR. ACC’s evidence is that Review Specialists are trained to identify appropriate resolution opportunities and determine whether ADR may be appropriate, and this would ordinarily occur after first considering and/or pursuing a resolution directly with the claimant. That would appear to be a reasonable position for ACC to take when it is a steward of public funds, and it is statutorily required to deliver its services in a cost-effective manner (and one that promotes administrative efficiency).37
Ground 3: legitimate expectation
[79] The requirements to establish a legitimate expectation are well established in New Zealand and are not in dispute. They are set out by the Court of Appeal in Comptroller of Customs v Terminals (NZ) Ltd:38
36 See, for example, Ngāti Whātua Ōrākei Trust v Attorney-General (No 4) [2022] NZHC 843, [2022] 3 NZLR 601 at [597]; and Thomson v Minister for Climate Change Issues [2017] NZHC 733, [2018] 2 NZLR 160 at [157].
37 Accident Compensation Act, s 262(3).
38 Comptroller of Customs v Terminals (NZ) Ltd [2012] NZCA 598, [2014] 2 NZLR 137.
[125] Where legitimate expectation is raised, the inquiry generally has three steps. The first is to establish the nature of thecommitment madeby the public authority whether by a promise or settled practice or policy. This is a question of fact to be determined by reference to all the surrounding circumstances. A promise or practice that is ambiguous in nature is unlikely to be treated as giving rise to a legitimate expectation in administrative law terms.
[126] The second is to determine whether the plaintiff’s reliance on the promise or practice in question is legitimate. This involves an inquiry as to whether any such reliance was reasonable in the context in which it was given.
[127] The third, and often most difficult part of the inquiry, is to decide what remedy, if any, should be provided if a legitimate expectation is established.
[80] For an expectation to arise from practice or policy, “its existence and content must … be established to the level of a commitment or undertaking”, meaning it “must be both unambiguous, and settled in the sense that it is regular and well established”.39 This is a high threshold. Reliance on a promise or settled practice must be “reasonable in the context in which it was given”.40 A legitimate expectation is to be distinguished from “a mere hope that a course of action will be pursued”.41
[81] CNS’s case on this ground of review is that ACC has made a commitment to its preference to use ADR by adopting its Policy, which would affect all ACC claimants, including CNS’s clients. CNS has relied on the Policy; Mr Forster established the business in reliance on it. That reliance was legitimate and reasonable, given it is reasonable to expect ACC to comply with its own formal policy that it has chosen to adopt, and the Policy is the culmination of a change in direction by ACC towards an increased willingness to use ADR. It is therefore reasonable for a person in Mr Forster’s position to consider the Policy as a statement of intention and rely on ACC to follow it.
[82] Because I have found the Policy does not have the meaning CNS seeks to ascribe to it, this ground of review cannot succeed either.
[83] There has been no clear or unambiguous undertaking or settled practice by ACC that it will always, or almost always, or even generally agree to ADR when a client asks for it. As I have found, this staff-focused Policy only sets out ACC’s
39 Green v Racing Integrity Unit Ltd [2014] NZCA 133, [2014] NZAR 623 at [14].
40 Comptroller of Customs v Terminals (NZ) Ltd, above n 38, at [126].
41 Green v Racing Integrity Unit Ltd, above n 39, at [15].
attitude or aspiration towards the use of ADR to resolve review applications. It is not a promise to use ADR.
[84] Nor could it be considered reasonable for CNS to rely on its interpretation of the Policy in the context of this case. Virtually always agreeing to the use of ADR has never been ACC’s practice and all of the publicly available material of other informed stakeholders does not support this perception. On Mr Forster’s own evidence, it would appear he himself did not consider ACC was following what he saw as a change in approach to the use of ADR. That motivated his involvement with ACC in respect of the survey in 2020 to understand ACC staff perceptions of TMR and in developing a new ADR strategy in 2021 to address staff not using conciliation.
[85]This ground also fails.
Result
[86]None of the grounds of review are made out. Thus, no question of relief arises.
[87] If the parties are not able to agree on costs the applicants are to file a memorandum within 10 working days of this decision, with the respondent having 10 working days to respond.
Grau J
Solicitors:
Duncan Cotterill, Wellington for Applicant Russell McVeagh, Wellington for Respondent
0
12
0