The practical impacts of the ADJR Act on judicial review applications (Brenda Tronson)

This is the latest post in the series on the Kerr report, cross-posted from the Australian Public Law blog.

In this post I provide an overview, from a barrister’s perspective, of the approach I am likely to take when asked for advice regarding a potential judicial review application in relation to a Commonwealth administrative decision. While the Administrative Decisions (Judicial Review) Act 1977 (Cth) (ADJR Act) had a clear impact in the early days, that has been overshadowed to some extent by the developments in judicial review under s 75(v) of the Constitution and s 39B of the Judiciary Act 1903 (Cth). However, the ADJR Act remains relevant, particularly in light of continued developments in the general judicial review jurisprudence.

The Kerr Report and its immediate impact

In 1971, when the Kerr Report was presented, judicial review of Commonwealth administrative decisions was available, but not easily accessible. The ‘normal way’ to seek judicial review was through proceedings in the High Court. While state courts could exercise jurisdiction in some judicial review matters, the High Court had exclusive jurisdiction where a writ of mandamus or prohibition, or an injunction was sought against an officer of the Commonwealth under s 75(v) of the Constitution.

There was a substantial body of law in relation to the principles of review and the remedies which could be sought. At the time of the Kerr Report, those principles were undergoing significant development across the common law world, which increased the intricacies involved in judicial review proceedings. There were also complexities arising from some of the technical requirements and questions of standing, at least in relation to some writs and remedies. Overall, the Kerr Committee concluded (at [58]):

It is generally accepted that this complex pattern of rules as to appropriate courts, principles and remedies is both unwieldly and unnecessary. The pattern is not fully understood by most lawyers; the layman tends to find the technicalities not merely incomprehensible but quite absurd. A case can be lost or won on the basis of choice of remedy and the non-lawyer can never appreciate why this should be so. …

This conclusion underlay a number of the reforms recommended in the Kerr Report, and the Ellicott Report, which followed it. Of particular relevance to this post, the reforms led ultimately to:

  • The drafting and enactment of the ADJR Act, intended to state the grounds of review and clarify standing rules, thus reducing many of the complexities of judicial review proceedings.
  • The creation of the Federal Court of Australia, a Commonwealth superior court, which has both jurisdiction under the ADJR Act and, by reason of s 39B of the Judiciary Act, functionally equivalent jurisdiction (for most purposes) as the High Court in relation to common law judicial review.

The reforms seemed to have their desired effect. This can be seen looking at the cases in 1991 – 30 years ago; 20 years after the Kerr Report; and around a decade after the Federal Court commenced to exercise jurisdiction (1977), the ADJR Act commenced (1980), and the insertion of s 39B of the Judiciary Act (1983). As Thomas Liu has explained, s 39B of the Judiciary Act is important in this context because it is the other part of the ‘dual system of federal judicial review’ described by Sir Anthony Mason.

On the basis of a review of published decisions, there appears to have been an increase in determined federal judicial review proceedings compared to the Kerr Committee’s experience, with approximately 100 first instance decisions in the Federal Court in 1991. Further, the majority of these matters were commenced pursuant to the ADJR Act, whether alone or in combination with s 39B.

In other words, 20 years after the Report and around a decade after the relevant reforms were put in place, the Federal Court and the ADJR Act were, together, providing the primary means of seeking federal judicial review. This might be said to be precisely what was intended by the Kerr Committee.

Judicial review applications today

In 2021, however, the picture is quite different.

As a matter of practice, what does an application to the Federal Court of Australia for judicial review look like 50 years after the Kerr Report?

Assuming there is a relevant decision or conduct made or done by an officer of the Commonwealth, and if it is appropriate to seek at least one of mandamus, prohibition or an injunction, my starting point will usually be to consider an application pursuant to s 39B of the Judiciary Act rather than the ADJR Act. That application will usually include grounds alleging jurisdictional error, that is, legal or procedural error which has affected the exercise of jurisdiction such that it can be argued that what has been done is not within jurisdiction or truly according to law.

I am not the only one taking this approach. A high-level review of decisions in judicial review proceedings in Commonwealth courts suggests that only around 10-15% of judicial review applications involve the ADJR Act. Of those where the ADJR Act is relied upon, around half are brought under both the ADJR Act and s 39B of the Judiciary Act.

Why start with s 39B?

There are three respects in which an application pursuant to s 39B of the Judiciary Act can be more flexible than an application pursuant to the ADJR Act.

First, there are fewer preconditions to the availability of the s 39B jurisdiction.

While the s 39B jurisdiction (and jurisdiction pursuant to s 75(v) of the Constitution generally) might appear to the uninitiated to be limited in terms of the available remedies, in fact, as long as at least one of the three constitutional writs is an appropriate remedy to seek, then the jurisdiction will be engaged, and the applicant will also be able to seek other remedies if necessary and appropriate.

It is not necessary to, for example, identify a decision ‘under an enactment’ (as the ADJR Act requires), which requires both the identification of a decision (whether the decision itself, or conduct related to it) and an enactment under which it is done or proposed to be done. While in many cases, identification of those preconditions is straightforward, there are some common gaps. For example, a decision not to do something might not be a decision for the purposes of the ADJR Act (because not doing something might not relevantly affect legal rights or obligations).  Steps taken under a policy which is simply part of how the relevant entity operates might also not be done ‘under an enactment’.

When such issues arise, the s 39B jurisdiction might be more clearly available and, so, the safer course for an applicant.

Second, the ADJR Act can be excluded by legislation, but the availability of the constitutional writs cannot be ousted. It is true that the jurisdiction of the Federal Court in relation to the writs can be narrowed, as has been done by s 476A of the Migration Act 1958 (Cth). But Parliament will be slow to restrict the jurisdiction of the Federal Court pursuant to s 39B, because that would likely increase the volume of applications directly to the High Court – and one of the very reasons for the Kerr Committee’s recommendation for a new court was to provide an alternative to the High Court (at [241]).

Third, for some decades and particularly since the High Court’s decision in Plaintiff S157/2002 v Commonwealth(Plaintiff S157), migration litigation has driven the development of the principles concerning jurisdictional error, in a continuation of the ‘simplifi[cation] and rationalis[ation] of the common law principles.’

Migration litigation has been a significant contributor to administrative law jurisprudence in Australia since at least the 1990s. The focus on jurisdictional error in particular came about because, between 1994 and 2001, legislative amendments progressively narrowed the scope for judicial review of migration decisions. Two developments of particular importance were:

  • The exclusion of migration decisions from the ADJR Act, and insertion of a judicial review regime specific to the Migration Act with effect from 1994.
  • The insertion of  a privative clause regime in 2001, which purported to remove almost all rights of judicial review in relation to migration decisions.

In Plaintiff S157, the High Court held that the privative clause in the Migration Act did not exclude judicial review by the High Court of a purported decision which was affected by jurisdictional error, because such a decision would not be a decision ‘under the Act’. The Court also confirmed that s 75(v) of the Constitution provides for ‘an entrenched minimum provision of judicial review’.

As explained by Liu, many of the grounds on which jurisdictional error might be alleged now reflect (or are reflected) in the grounds on which an application pursuant to the ADJR Act may be made. However, the categories of jurisdictional error are not closed.

One example of the development in the understanding of what constitutes jurisdictional error came in Minister for Immigration and Citizenship v Li (2013)(Li), in which the High Court confirmed that a decision which is unreasonable in the relevant sense will be affected by jurisdictional error. The decision also provided a basis for understanding this kind of unreasonableness (now often referred to as ‘legal unreasonableness’) as something other than ‘an exercise of power that is so unreasonable that no reasonable person could have so exercised the power’. That formulation was not only the former articulation of the common law test (commonly referred to as Wednesbury unreasonableness, after the case in which it was first stated), but is also the formulation of the test in ss 5(2)(g) and 6(2)(g) of the ADJR Act. As such, the High Court’s decision in Li provides an example of the potential for development of the understanding of jurisdictional error in circumstances where such a development might not be able to be made within the bounds of the ADJR Act, or might be slower to take hold.

Having considered the availability of judicial review pursuant to s 39B, I will then consider whether an application should be made pursuant to the ADJR Act, whether in addition to the s 39B application or instead of it.

Factors which will influence my approach in this respect include:

  • Whether the ADJR Act permits remedies more suited to my client’s grievance, for reasons such as those described by Liu.
  • Whether the decision-maker is a Commonwealth authority for the purposes of the ADJR Act, but might not be an officer of the Commonwealth for the purposes of s 39B. This might apply to some bodies corporate, which are generally not officers of the Commonwealth for the reasons outlined by Katzmann J of the Federal Court at [41]-[43] of Australasian College of Cosmetic Surgery Limited v Australian Medical Council Limited (2015).
  • Whether other differences in jurisprudence mean that an application pursuant to the ADJR Act is appropriate, although the convergence of the principles means this is likely to have a very limited effect.

Overall, I will also have regard to the best way to persuade the Court. Because of way the jurisprudence has developed in this area, the language and arguments in relation to s 39B jurisdiction (or to s 75(v) jurisdiction generally) are familiar to lawyers and to the Court. Finding a strong argument based on something familiar (with any extension being incremental and, hopefully, analogous to what has already been done) is much more likely to be in a client’s interests than launching into the unknown.

Will the trend towards greater use of the s 39B jurisdiction continue?

The development of the jurisdictional error jurisprudence can result in a narrowing of its availability, as well as an expansion. A recent example of this is found in the High Court’s decisions concerning materiality in Hossain v Minister for Immigration and Border Protection(2018), Minister for Immigration and Border Protection v SZMTA(2019)and MZAPC v Minister for Immigration and Border Protection(2021).

The majority view arising from these decisions is clear: there is no jurisdictional error unless the error is material, in the sense that it must have deprived the applicant of a ‘realistic possibility that the decision in fact made could have been different’ without the error, and the applicant for review bears the onus of proof in this regard.

Given the reasoning as to the requirement for materiality before jurisdictional error is established, there is a live question as to whether an applicant must demonstrate materiality in order to establish one of the grounds in the ADJR Act. As such, we might see an increase in the number of applications pursuant to the ADJR Act (either solely, or in addition to s 39B) while the jurisprudence in this respect is developed.

This particular question might make little difference in the long run. In relation to at least some ADJR Act grounds of review, it is likely that materiality will be required. For instance in the Federal Court decision in Mohamed v Secretary, Department of Education, Skills & Employment (No 2)(2020), O’Bryan J considered materiality in determining whether the grounds in s 5(1)(e) and (f) were made out. This is another example of the convergence between the two streams of judicial review jurisprudence.

In any event, orders under the ADJR Act are discretionary.  The materiality of any breach or error is likely to affect the exercise of discretion at this stage, even if the courts ultimately decide that materiality is not directly relevant to ADJR Act grounds. In those circumstances, there might be little difference in outcome, although the point of the reasoning process at which materiality becomes relevant might make a difference in relation to the onus of proof.

For these reasons, I do not predict that the materiality jurisprudence will cause a divergence of the streams, or a sustained increase in ADJR Act applications. But it does provide an example of how jurisprudential developments, combined with the availability of statutory judicial review, might continue to influence changes in practical and procedural choices.

Acknowledgment and thanks for research assistance: Elizabeth Lambert, Barrister, Eight Wentworth Chambers, Sydney.

Brenda Tronson is a barrister at Level 22 Chambers in Sydney.

This content has been updated on October 25, 2021 at 01:55.