Personal Ministerial Powers, Delegation and Soft Law: Davis v Minister for Immigration, Citizenship, Migrant Services and Multicultural Affairs; DCM20 v Secretary of Department of Home Affairs [2023] HCA 10

It is well established in Westminster systems, such as Australia, Canada and the United Kingdom, that civil servants may exercise the statutory powers of a minister: this is the Carltona principle. But the Carltona principle can be displaced by statutory language prescribing that the minister must exercise the statutory power personally. Where a ministerial statutory power is personal in nature, what role can civil servants play in its exercise? The High Court of Australia recently considered this question in Davis v Minister for Immigration, Citizenship, Migrant Services and Multicultural Affairs; DCM20 v Secretary of Department of Home Affairs, [2023] HCA 10.

In both cases, the appellant requested that the Minister use his power under s. 351 of the Migration Act to substitute a more favourable decision to the appellant than that of the Administrative Appeals Tribunal. However, the request never reached the Minister’s desk: low-level departmental officials took the view, applying the ministerial guidelines set out in a soft-law operational manual, that there was nothing exceptional or unique about the appellants’ circumstances and, thus, no basis for exercising the power under s. 351.

Section 351 is clear that the power must be exercised personally (s. 351(3)) and that where it is exercised the Minister must lay reasons before Parliament (s. 351(4)), but also that the Minister “does not have a duty to consider whether to exercise the power under subsection (1) in respect of any decision, whether he or she is requested to do so by the applicant or by any other person, or in any other circumstances” (s. 351(7)).

In their plurality judgment, Kiefel CJ, Gageler and Gleeson JJ noted that these provisions “cut both ways”: “They confine to a Minister (as distinct from the department administered by that Minister) any decision to exercise such a power in the public interest. They also confine to a Minister (as distinct from the department administered by that Minister) any decision not to exercise such a power in the public interest” (at para. 28). Accordingly, allowing civil servants to determine whether a case is exceptional or unique enough to justify the use of the power under s. 351 violated the statutory provisions: “By instructing that those cases assessed by the Department not to have unique or exceptional circumstances were to be finalised by the Department without referral, the Minister purported to entrust the dispositive evaluation of the public interest to departmental officers” (at para. 38).

The Minister could establish objective criteria to sort the wheat from the chaff (at para. 26), but, as Gordon J put it, the problem here was that the criteria were subjective and amounted to a determination “whether a departmental official or any other person might think it to be or not to be in the public interest to substitute a more favourable decision for that of the Tribunal” (at para. 99). The problem here, to put it in Edelman J’s terms, was that the system under the operational manual was “one in which no applications would be brought to the attention of the Minister unless the application met broad evaluative criteria to the satisfaction of the departmental officials” (at para. 157).

The plurality took a fairly broad view of the range of circumstances in which the Minister could limit his personal consideration of a request (based on a distinction between the procedural decision to, in effect, decide whether to decide, and the substantive decision on the merits of a request). Gordon J took a narrower view: “Prima facie, when Parliament confers a statutory power on a person, it intends that person to exercise the power. Further, subject to a contrary intention, the conferral of a statutory discretion implies a duty to consider any application that is made for the exercise of power” (at para. 92). For her, therefore, “[n]either a decision to exercise the power, nor a decision not to exercise the power, may be made by an official in the department administered by a Minister (or any other person)” (at para. 97). And Edelman J, for his part, thought the procedural/substantive distinction resembled “the bureaucratese of Sir Humphrey Appleby” (at para. 108); drawing the line between permissible assistance from civil servants and impermissible delegation of authority involves a careful, contextual exercise in all cases (at paras. 111-113). As Jagot J warned, not all “all conceptual distinctions” are “useful” (at para. 301).

Dissenting, Steward J saw the guidelines embedded in the operational manual as “a sensible administrative mechanism for the processing of requests received for the exercise of the power” in s. 351 (at para. 217). But there was no disagreement between the judges that the Minister could lawfully adopt a sorting mechanism of some sort: the problem here was that the civil servants were to apply the same criteria as the Minister, even though the power is one held personally by the Minister; they were acting “within the zone of exclusive Ministerial personal decision‑making power” (at para. 315, per Jagot J). Nothing in the judgments suggests that well-identified routine administrative matters cannot be delegated.

In any event, this case is an interesting example of a clash between first principles of administrative decision-making and the realities of mass administration. Good luck to the Minister in redrafting the operational manual.

I also commend to you Dr. Maria O’Sullivan’s post on Davis (and another important case, ENT19, which I will cover in a separate blog).

This content has been updated on July 17, 2023 at 15:27.