In short. The rule against bias — nemo iudex in causa sua, no one may be a judge in their own cause — requires that a decision be made by someone impartial, and seen to be impartial. It binds not only the Employment Relations Authority and the Employment Court but the employer who disciplines or restructures. The law polices not just actual bias, which is rare and hard to prove, but apparent bias, because public and individual confidence in a decision depends on how it looks as much as on what was in the decision-maker's mind. The governing test comes from the Supreme Court in Saxmere.
The second limb of natural justice is the rule against bias, captured in the maxim nemo iudex in causa sua — no one should be a judge in their own cause. Its demand is simple to state and exacting in practice: the person who decides must be impartial, and must be seen to be impartial. The appearance matters because a process that looks tainted erodes confidence even where the decision-maker was, in fact, even-handed. This is the origin of the familiar principle that justice must not only be done but must be seen to be done.
In employment, the rule reaches well beyond the courtroom. It governs the Authority and the Employment Court in their own processes, but it also governs the employer who sits in judgment on an employee — in a disciplinary meeting, a restructuring decision, or an investigation. Wherever someone exercises a power to decide a matter affecting another's rights or interests, the rule against bias applies to them.
Bias, in this context, means a predisposition to decide an issue in a particular way that leaves the decision-maker's mind not properly open to persuasion. That predisposition may have many sources — a financial interest, a personal relationship, or a fixed ideological view. The law approaches it in two forms.
Actual bias is a genuine prejudice or a closed mind that in fact affects the outcome. It is rarely alleged and harder still to establish, because it requires proof of the decision-maker's actual state of mind, which is seldom available. Apparent bias, sometimes called apprehended bias, is the practical ground on which most challenges run. It asks not whether the decision-maker was actually partial, but whether the circumstances would lead a fair-minded observer to apprehend a real possibility that they might not be. The focus is on appearances, precisely because confidence in the integrity of a decision is a public good that does not survive the look of partiality. New Zealand courts have applied that standard across the employment field, from disciplinary processes in schools in Sutherland v Board of Trustees of Marlborough Girls' College [1999] NZEmpC 98 to more recent challenges such as Halse v Hamilton City Council [2023] NZEmpC 118 and Bowen v National Australia Bank Limited [2023] NZERA 735.
The working definition of bias as a predisposition that closes the mind to persuasion was put by the Employment Court in New Zealand Post Primary Teachers' Association v Board of Trustees of Kelston Boys' High School [1992] NZEmpC 2.
The definitive New Zealand test for apparent bias was settled by the Supreme Court in the Saxmere litigation: Saxmere Co Ltd v Wool Board Disestablishment Co Ltd [2009] NZSC 72, and applied on fuller facts in the Court's recall judgment, Saxmere Co Ltd v Wool Board Disestablishment Co Ltd (No 2) [2009] NZSC 122. The case itself concerned a judge's business relationship and friendship with counsel, and whether that connection created an apprehension of bias.
Whether a fair-minded lay observer, informed of the relevant facts, might reasonably apprehend that the decision-maker might not bring an impartial mind to the resolution of the question they are required to decide.
Drawn from Saxmere Co Ltd v Wool Board Disestablishment Co Ltd [2009] NZSC 72.
Applying the test is a two-stage inquiry. First, the actual circumstances said to give rise to the apprehension of bias must be identified. Second, a logical connection must be articulated between those circumstances and the feared departure from deciding the case on its merits. The standard to be met is a real and not remote possibility of bias — not a probability, but more than a fanciful or theoretical concern.
The test is applied through the eyes of a hypothetical person with defined characteristics. The fair-minded lay observer is informed — they take the trouble to learn the relevant facts and understand the context. They are intelligent and objective, neither unduly suspicious nor over-sensitive. They are taken to understand how decision-making works in its proper setting: that a judge has a duty to sit, has taken an oath of impartiality, and is expected to be independent, and that the same allowance applies, in context, to other decision-makers. Crucially, the observer is not the aggrieved party, who is often the least objective observer of all.
That objectivity matters because some forms of "leaning" are not bias at all. As the Employment Court noted in Davis v Commissioner of Police [2015] NZEmpC 38, the very business of adjudication is to be persuaded by the evidence, which in a sense produces a legitimate leaning towards the party with the better case. That is the proper work of deciding, and it is quite distinct from the impermissible partiality the rule targets.
Predetermination is closely related to bias but is a distinct concept, and it is a mistake to treat the two as interchangeable. Predetermination is the state of having already decided the outcome before the evidence and argument are heard — a mind closed in advance. A decision-maker may be predetermined without being biased toward any particular person, and may favour a party without having predetermined the result. The High Court drew exactly that distinction in Howe v Keown (High Court, 2 September 2011), a challenge to a councillor's participation in a chief-executive appointment: a strong, even firmly held, preference or point of view does not amount to predetermination, provided the decision-maker genuinely remains open to being persuaded by a better case. What is fatal is not the holding of a provisional view, but the hardening of that view into a fixed position that no longer admits genuine consideration.
That line — provisional view permissible, closed mind not — is what the Authority enforced in Ryan & Paekau v Waipa Youth and Whanau Trust [2007] NZERA 147, where the decision-makers were found to have closed their minds and were no longer giving genuine consideration to the issues before them. Genuine consideration of the employee's response is, in any event, a statutory requirement under section 103A of the Employment Relations Act 2000, so a predetermined decision fails the test of justification directly.
Predetermination most often surfaces not in the final decision but in the investigation that feeds it. Where investigators go looking for a complaint, draw on old or unproven matters, or frame their report in prejudicial and loaded language, the danger is that the apparent neutrality of the fact-finding is lost and the ultimate decision-maker is steered toward a conclusion before the employee has answered. The Employment Court and the Authority have scrutinised investigative conduct of this kind in the COVID-19 vaccination dismissals, including Pilgrim v Attorney-General [2022] NZEmpC 138, and in disciplinary investigations such as AJY v Chief Executive of the Department of Corrections [2022] NZERA 169. The lesson is consistent: an investigation is not a neutral preliminary that can be conducted carelessly, because a tainted investigation can predetermine the decision that rests on it.
The most common structural source of apparent bias in the workplace is the fusion of incompatible roles — the same person who receives the complaint also investigates it, builds the case, and then sits in judgment. The difficulty is human and obvious: a person who has gathered the evidence and, in effect, prosecuted the case will struggle to switch roles and weigh that very case impartially. Even where they manage it, the structure itself undermines the appearance of fairness.
The Employment Court was blunt about this in Russell v Wanganui City College [1999] NZEmpC 65: it remains objectionable for the person who has carried out the inquiries and built the case against an individual to also sit in judgment on them. The fact that a rule or policy may permit the same person to receive the initial complaint does not save the position if that person's later conduct shows them to have become partisan or over-zealous in pursuing the case.
The risk is structural, and so is the answer. Good practice — reflected in many well-drafted internal policies — separates the functions so that no single person carries the matter from accusation to verdict. The person who receives the complaint may make a preliminary inquiry. For serious allegations, an independent investigator, external or an internal manager with no connection to the matter, should be appointed to produce a factual report. A separate and impartial decision-maker then considers that report alongside the employee's response before deciding. Conducting the hearing before the actual decision-maker, rather than relying on second-hand reports, reinforces the separation.
The value of recognising the problem early is illustrated by JSC v Landcorp Farming Limited t/a Pāmu [2024] NZERA 162, where a manager recognised that his close working relationship with the employee under investigation made it hard for him to investigate and remain independent, and the correct course — appointing an external investigator — was taken.
The rule against bias asks two things of every decision-maker, the employer included: be impartial, and be seen to be impartial. The law rarely needs to prove an actually prejudiced mind, because the Saxmere standard catches the decision that a fair-minded, informed observer would reasonably see as carrying a real possibility of partiality. Around that core sit two recurring failures — the closed, predetermined mind, and the fusion of complainant, investigator, and judge in a single person. Both are avoidable, and the means of avoiding them is the same: keep the mind genuinely open, and keep the roles genuinely apart. A decision-maker who does neither will find that even a well-founded outcome cannot survive the appearance of the way it was reached.
Legislation cited is linked to the current consolidated text on the New Zealand Parliamentary Counsel Office website at legislation.govt.nz. Case authorities are linked to the Working for Workers case law and reference repository at caselawandreference.workingforworkers.nz, except Saxmere Co Ltd v Wool Board Disestablishment Co Ltd [2009] NZSC 122, linked to the judgment published by the Courts of New Zealand, and Howe v Keown (High Court, 2 September 2011), which is not held in the repository; its citation should be confirmed against the official report before it is relied upon in any filed document.
The right to an impartial decision-maker is not confined to the workplace. Whenever a person or body with power over you decides a matter affecting your job, your livelihood, or your rights — an employer, the Accident Compensation Corporation (ACC), an incorporated society, a tribunal, or any public body exercising a statutory power — they must bring an open and unbiased mind to it, and must not be judge in their own cause. When the same person accuses, investigates, and decides, or when the outcome was fixed before you were heard, that is the failure of natural justice this article describes, and it can be challenged.
Working for Workers stands up for people facing unfair and unreasonable process, across four areas of practice:
Fairness in Law · Accident Compensation Corporation (ACC) Law · Employment Law · Incorporated Societies Evaluation
Whatever the decision-maker, if you are up against a biased or predetermined process, get in touch and we will help you work out where you stand.