Three Types of “Material” Error?
When will a jurisdictional error be sufficiently “material” to justify the quashing of a tainted decision? This issue has prompted some discussion on the High Court of Australia, most recently in Minister for Immigration and Border Protection v SZMTA [2019] HCA 3 and especially in Hossain v Minister for Immigration and Border Protection [2018] HCA 34. This is a difficult issue (and those who do not have a predeliction for technicality and abstraction should click away now), which is obscured by the terminology of “material” error. My tentative suggestion, supported by a comparative inquiry, is that there are three different types of “material” error. Distinguishing between them would allow lawyers and judges to cut through the terminological thicket and more clearly see the underlying issues.
The terminology of “material” error tends to be used in three distinct senses. In what follows, for reasons which should become clear, I will refer generally to “materiality” rather than to “material” error. I should also say that I use the term “jurisdictional error” in a broad sense, to encompass errors of law (including failures to take account of all relevant considerations) and breaches of procedural fairness.
First, materiality can be used to denote whether an error is sufficiently serious to count as jurisdictional. The point here is that all errors are not created equal: some will have more serious consequences than others, entailing the invalidity of the decision under review; whereas some errors will be so immaterial that invalidating the decision under review would not be appropriate. As Kiefel CJ, Gageler and Keane JJ put it in their joint judgment in Hossain, “jurisdictional error is an expression not simply of the existence of an error but of the gravity of that error” (at para. 25, emphasis original):
Jurisdictional error, in the most generic sense in which it has come to be used to describe an error in a statutory decision-making process, correspondingly refers to a failure to comply with one or more statutory preconditions or conditions to an extent which results in a decision which has been made in fact lacking characteristics necessary for it to be given force and effect by the statute pursuant to which the decision-maker purported to make it.
at para. 24.
Similarly, in Cork County Council v Shackleton [2007] IEHC 241, Clarke J (as he then was), corrected a “significant error in the interpretation of a material statutory provision leading to a decision of the property arbitrator being wrong in law” (at para. 9.7, emphasis added).
Similar issues arise in respect of the distinction between “mandatory” and “directory” statutory provisions. This distinction has been deprecated by the High Court of Australia (see Project Blue Sky Inc v Australian Broadcasting Authority) (1998) 194 CLR 355) and the House of Lords (R v Soneji [2006] AC 340) but has been retained in Ireland. It is designed to serve a similar function to the distinction between jurisdictional and non-jurisdictional error, namely, assisting a court in determining whether an error is fatal to the validity of an administrative decision:
If the requirement which has not been observed may fairly be said to be an integral and indispensable part of the statutory intendment, the courts will hold it to be truly mandatory, and will not excuse a departure from it. But if, on the other hand, what is apparently a requirement is in essence merely a direction which is not of the substance of the aim and scheme of the statute, non-compliance may be excused
State (Elm Developments Ltd) v An Bord Pleanála [1981] I.L.R.M. 108 at p. 110, per Henchy J.
In determining whether to treat an error as a breach of a mandatory or merely directory provision, the materiality of the error — in terms of its consequences for the operation of the statutory scheme — will be a key consideration (see also Wei v Minister for Immigration and Border Protection (2015) 257 CLR 22 at p. 32). One way to understand the mandatory/directory distinction is to think of it in terms of the importance of the error in question: only a serious error, which undermines the statutory scheme (or, perhaps, interferes with individual rights or other legal values), will be treated as mandatory and, therefore, as having invalidated the decision; whereas a relatively immaterial error will not invalidate the decision. It is “unlikely to be an intention that the legislature is taken to have that a decision be rendered invalid by an immaterial error” (Hossain at para. 67, per Edelman J):
Ordinarily, a statute which impliedly requires that condition or another condition to be observed in the course of a decision-making process is not to be interpreted as denying legal force and effect to every decision that might be made in breach of the condition. The statute is ordinarily to be interpreted as incorporating a threshold of materiality in the event of non-compliance.
Hossain at para. 29. See also at paras. 64-56, per Edelman J.
Distinguishing jurisdictional from non-jurisdictional errors and mandatory from directory provisions is no easy task. As O’Donnell J commented in Gillen v Commissioner of An Garda Siochána [2012] IESC 3; [2012] 1 I.R. 574, “[t]he question for the court is to interpret [the legislature’s] silence on the topic” (at p. 602). In all cases, however, “[t]he question of whether a particular failure to comply with an express or implied statutory condition in purporting to make a particular decision is of a magnitude which has resulted in taking the decision outside the jurisdiction conferred by the statute cannot be answered except by reference to the construction of the statute” (Hossain, at para. 27, per Kiefel CJ, Gageler and Keane JJ; SZMTA, at para. 83, per Gageler and Gordon JJ). To put the point another way, the first type of materiality will always involve a pure question of statutory interpretation.
Obviously, in jurisdictions where the distinction between jurisdictional and non-jurisdictional error has been abolished (entirely in New Zealand (Peters v Davison [1999] 2 NZLR 164) and for all practical purposes in England (R (Cart) v Upper Tribunal [2012] 1 AC 663 at paras. 18, 39), this first type of materiality has much less purchase. But it retains a tenacious hold on cases involving the mandatory/directory distinction, inasmuch as the question in such cases will always be whether “on a true construction non-observance of the condition is fatal to the validity of the action” (Wade & Forsyth, Administrative Law 11th ed. (2014), at pp. 183-184). As was said in Soneji, “the emphasis ought to be on the consequences of non-compliance, and posing the question whether Parliament can fairly be taken to have intended total invalidity” (at para. 23 per Lord Steyn).
Second, materiality can be used to denote a causal link between the error complained of and the decision under review. Like negligence, reviewable error does not exist in the air: “the error has to be one which affected the actual making of the decision and affected the decision itself” (Peters v Davison [1999] 2 NZLR 164, at p. 202). What is required is a “material” error (R(Cart) v Upper Tribunal [2012] 1 AC 663, at para. 110). As Edelman J put it in Hossain “an error will not usually be material, in [the] sense of affecting the exercise of power, unless there is a possibility that it could have changed the result of the exercise of power” (at para. 72). This is distinct from the first type of materiality (SZMTA at para. 44, per Bell, Gageler and Keane JJ).
Thus, for example, when an applicant successfully demonstrates that an irrelevant factor has been taken into account by an administrative decision-maker, the applicant must also establish that the factor had a “substantial” influence on the decision (R v Rochdale Metropolitan Borough Council, ex parte Cromer Ring Mill [1982] 3 All ER 761, at p. 770, per Forbes J. See also McKernan v Employment Appeals Tribunal [2008] IEHC 40, at para. 4.6, per Feeney J). Factual errors, in jurisdictions which recognise them as grounds of review, must also be “material” (see e.g. E v Secretary of State for the Home Department [2004] QB 1044; Efe v Minister for Justice [2011] 2 IR 798).
Breaches of procedural fairness present a difficulty in this respect. In part, a counter-factual inquiry is called for (did the failure to conduct a fair hearing materially influence the outcome?), a question which is almost impossible to answer (see John v Rees [1970] Ch 345; SZMTA at paras. 47-49, per Bell, Gageler and Keane JJ). Furthermore, a procedural failing may be “so fundamental that it will be material whether or not a person is deprived of the possibility of a successful outcome” (Hossain at para. 72, per Edelman J). If one takes the view, like Lord Reed in Osborn v Parole Board [2014] 1 AC 1115, that procedural fairness is underpinned by dignitarian considerations, many procedural failings will be classified as “fundamental”. In general, however, it seems fair to say that breaches of procedural fairness are, all other things being equal, more likely to be treated as material.
Unlike the first type of materiality, which is in substance a question of law, the second type of materiality is more in the nature of a question of fact or mixed law and fact. Regardless of how it is classified or described, however, the burden of proof will be on the applicant. Demonstrating the presence of the second type of materiality is an essential component of ‘making one’s case’ in an application for judicial review:
Where materiality is in issue in an application for judicial review, and except in a case where the decision made was the only decision legally available to be made, the question of the materiality of the breach is an ordinary question of fact in respect of which the applicant bears the onus of proof. Like any ordinary question of fact, it is to be determined by inferences drawn from evidence adduced on the application.
SZMTA at para. 46, per Bell, Gageler and Keane JJ.
Third, materiality can be used to denote judicial discretion to refuse relief. In situations where there has been an error which qualifies as jurisdictional (including a breach of a statutory provision which qualifies as mandatory) and there is a causal link between the error and the outcome, the court still retains discretion to refuse a remedy. And this discretion might be exercised in situations where the error complained of was immaterial, in the sense that the decision-maker would inevitably have reached the same conclusion:
Even where the irrelevant factor played a significant or substantial part in the decision-maker’s thinking, the decision may, exceptionally, still be upheld, provided that the court is satisfied that it is clear that, even without the irrelevant factor, the decision-maker would have reached the same conclusion… There is, in theory at least, a possibility that, even if the court concludes that it ought otherwise to set aside a decision on the ground that a legally irrelevant factor was taken into account, it can nonetheless uphold the decision, if it is satisfied that it would be pointless to require the decision-maker to reconsider the question afresh, because he would reach the same answer.
R (FDA) v Work and Pensions Secretary [2013] 1 WLR 444, at paras. 68-69, per Lord Neuberger MR
This type of materiality is conceptually distinct from the first and second types. As Nettle J explained in Hossain, an error of law may be jurisdictional and material but the issue of the consequences that flow from this combination of factors is a separate matter. Consider, for instance, a situation:
…where a decision maker is required to make a decision by reference to a single specified criterion and, in error, addresses himself or herself to the wrong criterion. In such a case, the decision maker’s error will be a jurisdictional error – a failure to exercise the jurisdiction of deciding the question according to the applicable criterion – regardless of whether one can say that, if properly directed and having determined the application by reference to the correct criterion, the decision maker would have been bound to make the same decision.
Hossain at para. 40.
This is part of the ordinary remedial discretion exercised by a reviewing judge (see e.g. R v Commonwealth Court of Conciliation and Arbitration; Ex parte Ozone Theatres (Aust) Ltd (1949) 78 CLR 389, at p. 400; Harelkin v University of Regina [1979] 2 SCR 561, at p. 576; Bulk Gas Users Group v Attorney-General [1983] NZLR 129 at p. 136; Stefan v Minister for Justice Equality and Law Reform [2001] 4 IR 203, at p. 217; R (New London College) v Secretary of State for the Home Department [2013] 1 WLR 2358, at paras. 45-46). I do not think it is particularly helpful to characterise this exercise as backward-looking or forward-looking (Hossain at para. 43, per Nettle J and at para. 74, per Edelman J. See e.g. R (Nadarajah) v Home Secretary [2002] EWHC 2595 (Admin), at para. 30, refusing relief because the illegality at issue had subsequently been legitimised, such that the same result would have been reached on remittal and also GXL Royalties v Minister of Energy [2010] NZCA 185). The key point is that the discretion will have to exercised on the facts of each individual case.
The point of setting out this typology is to illuminate the tendency, in some recent Australian cases, to blur these distinct types of materiality.
For example, in Minister for Immigration and Border Protection v WZARH[2015] HCA 40, Gageler and Gordon JJ rolled types one and two together: “the implied condition [to give a reasonable opportunity to be heard] which governs the exercise of the Minister’s statutory powers of consideration is material, so as to justify the grant of declaratory relief by a court of competent jurisdiction, if it operates to deprive the offshore entry person of ‘the possibility of a successful outcome'” (at para. 56. See also para. 60, not clearly distinguishing between types two and three).
The joint judgment in Hossain came perilously close to rolling types two and three together: “the threshold of materiality would not ordinarily be met in the event of a failure to comply with a condition if complying with the condition could have made no difference to the decision that was made in the circumstances in which that decision was made” (at para. 30). This prompted Nettle J to observe (correctly in my view): that “there may be a number of circumstances in which an error is jurisdictional despite not depriving a party of the possibility of a successful outcome” (at para. 40).
Most significantly, Nettle and Gordon JJ seemed to conflate types two and three in SZMTA. While they were quite right to observe that deciding whether or not to exercise discretion to refuse relief “entails a different and separate exercise from the identification of jurisdictional error” (at para. 85), which “must remain distinct” (at para. 88), their insistence that materiality is always “a question of statutory construction” (at para. 89) would eliminate any role for the second type of materiality (causal connection): “The only place for that kind of analysis (about the materiality of the error to the applicant) is in the exercise of the court’s discretion whether to grant relief after jurisdictional error is made out” (at para. 90).
Nettle and Gordon JJ had two particular concerns. First, they doubted the applicability of the second type of materiality to Australia, on the basis that it is a creature known only to the English law of judicial review which, having more or less abolished the distinction between jurisdictional error, is now very different from Australian administrative law (at para. 91). As I have shown, however, the second type of materiality can be perceived in Ireland as well and other Commonwealth systems. Second, they considered that it would be inappropriate to place the burden of proving materiality on the applicant (at paras. 93-95). I accept that where the third type of materiality is concerned, the burden should be on the decision-maker “to establish that the relief would be futile in the applicant’s circumstances” (at para. 93). But as I have observed above, demonstrating a causal link between error and result is part of the case an applicant for judicial review must make out, as courts across the common law world have recognised. This is often difficult, especially in the case of abuse of discretion (see e.g. CREEDNZ Inc v Governor-General [1981] 1 NZLR 172; R (Bancoult) v Foreign and Commonwealth Secretary (No. 3) [2018] 1 WLR 973) but is nonetheless a well-settled feature of the common law of judicial review of administrative action.
I hope this tentative typology is helpful. All comments very welcome.
This content has been updated on June 10, 2019 at 10:28.