Administrative Law Jurisdictional Error
Consider the following statement:
“Concepts such as procedural fairness, reasonableness, rationality, and error of law were
well understood before the modern world of law by statute came to dominate. Now, the
limits of jurisdiction are usually determined in a statutory context thus requiring the focus of
inquiry to shift from somewhat amorphous (or flexible) general law principles to principles of
The Hon Justice John Basten, ‘Judicial Review of Executive
Action: Tiers of Scrutiny or Tears of Frustration?’
(Speech delivered at the Constitutional and Administrative Law Section, NSW Bar Association, 14 May 2013 –
copy available on Moodle).
In light of Justice Basten’s remark, find and critically analyse a case which is not discussed in the
Creyke& McMillan case book (you can look here: https://jade.barnet.com.au). Do not choose
Minister for Immigration and Citizenship v Xiujuan Li (2013) 87 ALJR 618 for this purpose.
With regard to the case you have found and Justice Basten’s remark, consider the role of
jurisdictional error in Australian administrative law. Is it truly a central concept or might it be
discarded “without tears”? What might be the consequences of such a step?
CONSTITUTIONAL AND ADMINISTRATIVE LAW SECTION
OF NSW BAR ASSOCIATION
14 May 2013
JUDICIAL REVIEW OF EXECUTIVE ACTION:
TIERS OF SCRUTINY OR TEARS OF FRUSTRATION?
The Hon Justice John Basten
Judge of the NSW Court of Appeal
The original intention of this paper was to argue for the adoption of a functional
(therefore flexible) approach to judicial review by reference to a variable standard of
scrutiny. The Australian fixation with seemingly discrete grounds identified by labels,
stubbornly adhered to by courts and legislatures, has proved frustrating and
inadequate. Kirk1 showed a willingness to look past labels, but gave little guidance
as to how to fill the void. Now, having completed such an analysis, we have the
decision of the High Court in Minister for Immigration and Citizenship v Xiujuan Li2
delivered six days ago. On one view, it may mark no more than another step in a
continuous process of reformulating public law concepts in this country. On the
other hand, if one views the development of the law, like natural evolutionary
processes, as a system of punctuated equilibrium in which a period of stasis is
followed by a large step, this may mark the commencement of the next large step. I
incline to the latter view, for reasons which I will explain. However, if that view is
correct, it would be remiss of me to talk about judicial review without paying attention
to the most recent development.
Before coming to that case, it may be helpful to outline a number of broad
propositions as markers in the search for a satisfactory blueprint for judicial review,
together with some ideas as to how the search should proceed. The paper will then
consider how those views are affected by what the High Court is now saying.
First, judicial review of administrative action involves the exercise by the courts of a
supervisory role over the executive arm of government. Any expansion of the
judicial role tends to diminish the freedom of the Executive and thus affects the
relationship between those two arms of government. In constitutional terms, it
affects the separation of powers in a practical manner.
Secondly, to the extent that the courts find the mandate for a minimum standard of
judicial review in the Constitution, they impose a limit on the legislature and thus
affect the relationship between the judicial and legislative arms of government. If the
minimum content of judicial review is expanded, the power of the Parliament is to
that extent diminished. And by Parliament, one refers to both State and
1 Kirk v Industrial Court of New South Wales  HCA 1; 239 CLR 531.
2  HCA 18.
Thirdly, whatever anxiety the High Court may have about adopting a clear and
definitive statement of the grounds of review, these cases are decided on a daily
basis by busy trial judges and intermediate appellate courts, which do not have time
to consider the potential constitutional implications of their collective decisionmaking.
Fourthly, concepts such as procedural fairness, reasonableness, rationality, and
error of law were well understood before the modern world of law by statute came to
dominate. Now, the limits of jurisdiction are usually determined in a statutory context
thus requiring the focus of inquiry to shift from somewhat amorphous (or flexible)
general law principles to principles of statutory interpretation.
Fifthly, the move away from a set of labels for available grounds of review which
appears to be occurring (despite the apparent constitutionalisation of the concept of
“jurisdictional error”), is to be welcomed; courts exercising the supervisory jurisdiction
should adopt a functional and pragmatic approach, terminology to which I will return.
Labels: A brief historical excursus
Over the last 35 years the discourse of judicial review has been dominated, some
would say blighted, by the influence of the Administrative Decisions (Judicial Review)
Act 1977 (Cth) (“the ADJR Act”). It applies only to decisions under Commonwealth
enactments and then only to decisions of an administrative character. Despite its
initial educative value, the list of grounds in s 5 has beguiled both pleaders and
judges. It even proved irresistible to the Parliament, which, in enacting and then
amending a limited form of statutory judicial review of migration decisions, conferred
on the Federal Court power to review on the basis of some ADJR-style grounds, but
not on others.3 Thus, whilst permitting review for failure to comply with statutory
procedures, it sought to deny the availability of review for breach of general law rules
of natural justice. Similarly, whilst purporting to permit review on the basis that the
decision-maker “did not have jurisdiction to make the decision”, or that the decision
“was not authorised by this Act”, it purported to withdraw the power to review on the
basis that the exercise of power was “so unreasonable that no reasonable person
could have so exercised the power”.
These distinctions were writ in water. The availability of judicial review cannot
depend on semantic labels of imprecise scope. Reliance on such an approach will
produce arbitrary and capricious results. Judges who have a ‘feel’ for judicial review
may well produce consistent outcomes, not by application of labels, but by applying
an intuitive understanding of principles derived from experience. The conceptual
incoherence of the traditional labels is readily established. That is revealed by two
3 See Migration Act 1958 (Cth), s 476, as introduced by the Migration Reform
Act 1992 (Cth), up to the time of its repeal by the Migration Legislation
Amendment (Judicial Review) Act 2001 (Cth).
First, the exercise under the Migration Act failed in its attempt to corral the grounds
and hence diminish the availability of judicial review.4
The second example is more complex but revealing. We are familiar with the
propositions from Azzopardi v Tasman UEB5, a case involving an appeal limited to
error in point of law. Glass JA stated that perversity of reasoning in determining
facts is not a form of legal error. There were statements to similar effect in the
judgment of Mason CJ in Bond,6 referring to Menzies J in Ex parte White7, that “want
of logic is not synonymous with” error of law. But how are these statements to be
reconciled with the willingness of Latham CJ in Hetton Bellbird Collieries8 to set aside
a decision where it depended on an opinion found to be “capricious” or “arbitrary”?
The answer must lie elsewhere than in the labels. Most powers nowadays find their
origins in statutes. Arguably the limits of a statutory power must be found in the
constituting statute. But if, as we are consistently exhorted by the High Court, we
pay close attention to the words of the statute, we will usually find no reference to the
phraseology of judicial review case law. Accordingly, if we need to resolve fine
questions about the boundaries of legality in a contested case, it is usually not
possible to find any express reference in the power-conferring statute.
But if that source is unavailable, the only available alternative is the general law, the
language of which is the source of the uncertainty.
An alternative explanation
The answer is to be found in a combination of three sources. The first is the source
we initially rejected but must revisit: namely the statute. An important constraint on
the exercise of any power is its purpose. No power may be exercised for a purpose
foreign to that for which it was conferred. That purpose can only be derived from the
terms of the statute, in accordance with settled principles of statutory interpretation.
That in turn has a direct link to the established categories of judicial review, not only
by reference to improper purpose, but also to the requirement to take into account
relevant (that is mandatory) considerations and ignore irrelevant (that is prohibited)
considerations. Identification of such factors will depend intimately upon the relevant
purpose. Sometimes these factors are expressly stated (though not usually
exhaustively) but otherwise they are to be implied from the statutory context, in the
same way that the language and context will elucidate the purpose.
4 Minister for Immigration and Multicultural Affairs v Yusuf  HCA 30; 206
5 Azzopardi v Tasman UEB Industries Ltd (1985) 4 NSWLR 139.
6 Australian Broadcasting Tribunal v Bond  HCA 33; 170 CLR 321 at
7 R v District Court; Ex parte White  HCA 69; 116 CLR 644.
8 R v Connell; Ex parte Hetton Bellbird Collieries Ltd  HCA 42; 69 CLR
The second source of assistance is found in the judgments of Brennan J. It is
impossible to avoid the conclusion that relevant limitations (including the principle
just stated in respect of extraneous, foreign or improper purpose) must be located in
the general law. However, a third source of assistance is necessary to explain the
scope and operation of the general law. It is the functional, institutional or structural
(the label is unimportant) underpinning of the constitutional framework described as
“the rule of law”, as it operates in the Australian political system. It is that which is
captured in the canonical statement of Brennan J in Quin’s case9.
It is commonly said that the principle of separation of powers operates at the federal,
but not at the state level. However, that is an oversimplification. Not only is the
separation of powers blurred at points in the federal system, but it is not entirely
absent from state jurisdiction. No one doubts the importance of the independence of
the State judiciary, nor was it novel when constitutionally protected characteristics of
State Supreme Courts were discovered in Chapter III. Quin was a case in State
jurisdiction: the limitation on the role of the court exercising supervisory jurisdiction
over administrative action involved an understanding of the respective spheres of
executive and judicial power.10
The ‘no evidence’ ground also fits within this analysis. The idea that a decision
cannot be arbitrary or capricious rests on two assumptions. One is that the decisionmaker
must be affirmatively satisfied as to a particular state of affairs in order to
exercise the power in a particular way. The second is that satisfaction as to a
particular matter must be based on information or material which could be logically
probative of such a state of satisfaction. Thus, if a visa is to be granted only to a
person with a well-founded fear of persecution in his or her country of nationality,
considerations which have no logical connection with that test must be disregarded,
as must be factual material not relevant to that issue.
The most coherent way to describe the principles we have been discussing is to
identify a purported decision which fails to comply with them as a constructive failure
to exercise the power involved. The underlying principles are those of rationality
(without which the rule of law is gutted of content) and the separation of powers
(which imposes an important limit on judicial power, in circumstances where no right
of appeal exists).
There remain three questions which are not resolved by this analysis. The first is
whether the rationality principle permits (or requires) that a decision be set aside only
if the result is “unreasonable” or also if the process is flawed. That is, whether the
concepts of irrationality and illogicality apply to an objective analysis by the reviewing
court of the information available to the decision-maker and the result, or whether it
may or should involve an analysis of the decision-maker’s process of reasoning.
The second question, which also arises from the foregoing analysis, is to ask
whether, if the review process in fact extends to an analysis of the decision-maker’s
9 Attorney-General (NSW) v Quin  HCA 21; 170 CLR 1, at 35-36.
10 See also Brennan J in Kable v Director of Public Prosecutions (NSW) 
HCA 24; 189 CLR 51.
process of reasoning, that imposes an obligation to provide reasons for a decision
and, where such an obligation exists, set a standard as to the ‘adequacy’ of the
reasons to be provided.
The third question concerns the nature and content of procedural fairness in this
(a) The rationality principle
Starting with the first question, the language of “unreasonableness” has a varied and
volatile ancestry. To describe a decision as unreasonable, even manifestly
unreasonable, may be, as Gleeson CJ and McHugh J have noted, to do no more
than express vigorous disagreement.11 If there is a valid distinction between merit
review and patrolling the boundaries of legality, scrutinising for unreasonableness
risks blurring it into inutility. To expand the label to “conduct so unreasonable that no
reasonable person could so conclude” does little more than emphasise that the
concept is inherently vague and to demand a strong rather than a weak state of
satisfaction to permit intervention. Similarly, to distinguish between an evaluative
factual judgment and an exercise of power (true Wednesbury unreasonableness) is
to differentiate two areas of operation, without refining the test.12
At times, high level unreasonableness appears to be equated with irrationality: see
Crennan and Bell JJ in SZMDS.13 However, despite areas of overlap, the concepts
are distinct. The point of distinction may be illustrated by adapting a biblical story.14
Suppose the Pharaoh believed an augury that Jews in Egypt would rebel and that
the male leader of the rebellion had just been born. The plan to kill all newborn
males in the Jewish population may have been a manifestly unreasonable response,
but it was not irrational. The process of reasoning, from perceived problem to
proposed solution, was rational; the proposed solution would only be reasonable,
however, if deemed to be an appropriate response to the problem; if
disproportionate, the solution might be described as “unreasonable”.
Unreasonableness draws upon an assumed, but usually unarticulated set of
standards, believed to be commonly held within a particular community. No doubt
rationality is itself a culturally determined process; we might treat those who believed
the Mayans had correctly predicted the end of the world as irrational, but accepting
the perceived problem, the solutions some chose were not irrational.
In criminal appeals we may ask whether a jury verdict was unreasonable or
unsupportable on the evidence. Given the high state of satisfaction to be reached by
the jury (beyond reasonable doubt) and the anxiety of the administration of criminal
justice to avoid a miscarriage, this exercise is understandable. However, we do not
11 Minister for Immigration v Eshetu  HCA 21; 197 CLR 611 at .
12 Cf Gummow J in Eshetu at ; in other areas of discourse the functional
distinction is disregarded – Singer v Berghouse  HCA 40; 181 CLR 201
13 Minister for Immigration and Citizenship v SZMDS  HCA 16; 240 CLR
14 Exodus, Ch 1.
describe the verdict as a “purported verdict”, nor as a nullity, nor as invalid: it is a
decision which is effective until set aside on appeal. In Australia we have not
adopted proportionality as a test of legality of administrative decisions,15 as opposed
to delegated legislation.16 To be consistent, we should also be sceptical of
“unreasonableness”, and for the same reason – that is because each requires an
assessment by the reviewing court of the material available to the decision-maker,
by a flexible, if not undefinable, standard of community values.
The concept of proportionality has recently been described as imported into the
currently contentious ‘second limb’ of Lange.17 That has much to recommend it: but
judging the constitutional validity of legislation which impinges on the implied
freedom of political speech is a far cry from judicial review of administrative action.
(b) Reasons and reasoning
This brings me to the second question, namely review of the process of reasoning.
Once it is accepted that review for irrationality applies to the reasoning process, one
must infer a legal obligation for a decision-maker to act rationally. But it is not
possible to assess such a process accurately without a record of the decisionmaker’s
reasons. It would seem to follow that effective judicial review requires a
duty on all decision-makers to give reasons. However, the High Court has not
accepted that conclusion.18 That judgment constituted an exercise, not of logic, but
of pragmatism. The giving of reasons, at least for decisions which determine
disputes, is seen as an essential characteristic of judicial power as exercised by
judges, (though not by juries). A line is drawn between judicial and administrative
decision-making, although it is somewhat porous.19 But the rationale for that
distinction is not explained in terms of principle, but of history, tradition and,
ultimately, pragmatism. The Court declined in Osmond to impose what might have
been a significant economic and resource intensive obligation on all decision-makers
within executive governments across the country. If that were to be done, the Court
held, it was to be done by legislatures. Section 13 of the ADJR Act was not to be
applied to the State executives by osmosis (or trickle-down, depending on one’s
Absent reasons, the scope for judicial review is restricted, but not destroyed. In clear
cases, a reviewing court is able to decide that an impugned decision could not have
been made on the available information if the law had been correctly understood and
15 Aronson & Groves “Judicial Review of Administrative Action” (5th ed, 2013) at
373-377; Spigelman CJ in Bruce v Cole (1998) 45 NSWLR 163 at ; Dawson J
in Cunliffe v Commonwealth (1994) 182 CLR 272 and James v Military
Rehabilitation and Compensation Commission  FCAFC 95; 186 FCR
134 at - (Keane CJ, Middleton and Gordon JJ) .
16 South Australia v Tanner  HCA 3; 166 CLR 161at 168.
17 Wotton v Queensland  HCA 2; 86 ALJR 246 at  (Kiefel J).
18 Public Service Board (NSW) v Osmond  HCA 7; 159 CLR 656.
19 Campbelltown City Council v Vegan  NSWCA 284; 67 NSWLR 372 at
rationally applied.20 However, when reasons are available a new problem arises: if
the declared reasoning is flawed, but the same result could have been reached by a
logical and rational process, is the decision to be set aside? The answer given by
the courts is that it must be, unless it was the only one reasonably open.
Further, what if the reasons are inadequate to allow the reviewing court to determine
whether the law has been correctly understood and logically applied? Assuming the
decision was open or available, should the court conclude that error has not been
demonstrated? There may be two answers to this question, depending upon the
relief sought. If the relief sought is mandamus, the court might well conclude that, a
legal duty not having been fully exercised, the matter should be sent back to the
decision-maker for completion of the task. On the other hand, if the relief sought is a
quashing of the decision, the court may have to consider, according to Project Blue
Sky21 principles, whether the statutory obligation to give reasons was accompanied
by a sanction rendering the decision invalid in the absence of “adequate” reasons.
(c) Procedural fairness
The third question involved procedural fairness. The duty to accord procedural
fairness is undoubtedly a central element of the rule of law. It is concerned, not so
much with the process of decision-making, as with the institutional process. Affected
parties are entitled to an assessment of their claims or defences by an impartial
tribunal. They must have an opportunity to present their respective cases and an
opportunity to rebut matters material to the outcome and adverse to their interests.
In Dranichnikov,22 in a passage affirmed in SZJSS23, Gummow and Callinan JJ said:
“To fail to respond to a substantial, clearly articulated argument relying
upon established facts was at least to fail to accord Mr Dranichnikov
To call that a constructive failure to exercise jurisdiction is entirely appropriate, in
accordance with the preceding analysis; to call it a failure to accord procedural
fairness is to intrude procedural fairness into the decisional process where it has no
justifiable role and can blur functional roles of the repository of the power and of the
It is necessary now to add a footnote: there has been a major omission from the
preceding discussion – no mention has been made of “jurisdictional error”. That is
not only deliberate: it is part of my thesis. The phrase “jurisdictional error” has been
20 Avon Downs Pty Ltd v Commissioner of Taxation (Cth)  HCA 26; 78
21 Project Blue Sky v Australian Broadcasting Authority  HCA 28; 194
22 Dranichnikov v Minister for Immigration and Multicultural Affairs  HCA
26; 197 ALR 389.
23 Minister for Immigration and Citizenship v SZJSS  HCA 48; 243 CLR
described as a conclusory term, but none the worse for that.24 For some time I have
accepted that view, but I no longer do. Casting a cloak over the body of principles
governing judicial review is to obscure the functional analysis which must be
understood and applied to ensure judicial review stays within its proper constitutional
boundaries. Jurisdictional error is a label now used in the courts daily by those who
do not realise that there can usually be judicial review for non-jurisdictional errors of
law.25 Jurisdictional error has been cast adrift from its original function in
administrative law and is now a freestanding constitutional concept encapsulating a
limit on the legislative powers of the States. Kirk’s case, which caused this loss of
direction, went further; it dismissed the semantic labels by which the heads of
jurisdictional error had been identified in Craig,26 as non-exhaustive examples. That
step may be applauded, but no alternative analysis was provided. I have expressed
concern as to that omission in the past.27 The universal response, however, is a
demand for an articulated alternative. This discussion is offered as a possible next
step in the exercise.
Against this background, it is necessary to consider how the High Court addressed
these topics in Minister for Immigration and Citizenship v Xiujuan Li.28 This was not a
promising case, on one view, for a restatement of fundamental principle. The
applicant first applied for a visa on 10 February 2007. Her entitlement turned on an
assessment of her skills as a cook. That assessment was to be made by a specified
authority and depended upon her work experience in Australia. Her first application
contained false information, although it seems not to have been determined whether
the responsibility lay with the applicant or her migration agent. When challenged,
she conceded the falsity of the information and then sought a further assessment.
That assessment was unfavourable, but she claimed that it was based upon an error
and sought time to seek a review of the assessment. The Migration Review
Tribunal, considering her visa application, three years after it had been made,
decided that she had had enough time to demonstrate compliance with the criteria.
In January 2010, it rejected her application. It gave some brief reasons for
proceeding to determine the application, noting that she had “been provided with
enough opportunities to present her case”, apparently considering that it was entitled
to rely upon the assessment which had been provided, albeit the unfavourable
assessment had been challenged.29
24 M Aronson, “Jurisdictional error without the tears”, in Groves and Lee,
Australian Administrative Law – Fundamentals, principles and doctrines
(Camb UP, 2007) Ch 21 and Aronson and Groves, n 15 above, at [1.140] and
25 Absent an effective privative clause.
26 Craig v South Australia  HCA 58; 184 CLR 163.
27 “The supervisory jurisdiction of the Supreme Courts”, (2011) 85 ALJ 273.
28  HCA 18.
29 At .
The High Court unanimously held that that decision was so unreasonable as to
demonstrate invalidity. Presumably the only decision reasonably available to the
Tribunal was to grant the adjournment.
On the facts, some might think the outcome surprising, but it is sufficient for present
purposes to focus on the reasoning of the Court, which delivered three separate
judgments. Hayne, Kiefel and Bell JJ wrote together, whilst French CJ and Gageler
J each wrote separately.
Although this was a straightforward case involving the exercise of a discretionary
power, which invited consideration according to the well-established principles of
Wednesbury unreasonableness, no member of the Court was satisfied to adopt that
approach. In the result, the reasoning extends to 125 paragraphs and 50 pages.
In the search for an authoritative statement of principle, it is necessary to focus on
the joint reasons. There are four aspects of the joint reasons which invite
consideration. First, the joint reasons drew upon the background and language
found in Wade & Forsyth, Administrative Law.30 This course required caution. Not
only is English public law significantly affected, in sometimes unarticulated ways, by
the Human Rights Act, but more fundamentally in general law terms, by the
abandonment of a discrete concept of ‘jurisdictional error’, usually traced to
Anisminic.31 In reliance on Wade & Forsyth, the joint reasons adopted the language
of “the rule of reason” and the “legal standard of reasonableness”, which is largely
the language of Lord Greene in Wednesbury Corporation. However they continued
“The legal standard of unreasonableness should not be considered as
limited to what is in effect an irrational, if not bizarre, decision – which
is to say one that is so unreasonable that no reasonable person could
have arrived at it – nor should Lord Greene MR be taken to have
limited unreasonableness in this way in his judgment in Wednesbury.”
The reasons then refer to a passage in Wednesbury Corporation which is identified
by Wade & Forsyth as demonstrating that “[t]he rule of reason has thus become a
generalised rubric covering not only sheer absurdity or caprice, but merging into
illegitimate motives and purposes, a wide category of errors commonly described as
‘irrelevant considerations’, and mistakes and misunderstandings which can be
classified as self-misdirection, or addressing oneself to the wrong question”.32 The
joint reasons appear to accept this explanation at  in the following terms:
“The more specific errors in decision-making, to which the courts often
refer, may also be seen as encompassed by unreasonableness.”
A footnote to that passage refers to s 5(2) of the ADJR Act, as an example of the
tabulation of “more specific errors”. Whether or not that is helpful, may be doubted.
30 (10th ed, 2009).
31 Anisminic Ltd v Foreign Compensation Commission  2 AC 147.
32 Wade & Forsyth, pp 303-304.
It appears to abandon the distinction between jurisdictional error and other errors of
Secondly, the joint reasons referred to Minister for Aboriginal Affairs v Peko-
Wallsend33 stating that Mason J “considered that the preferred ground for setting
aside an administrative decision which has failed to give adequate weight to a
relevant factor of great importance, or has given excessive weight to an irrelevant
factor of no importance, is that the decision is ‘manifestly unreasonable’.”
It should perhaps be assumed that there is a typographical error in this passage: to
the administrative lawyer familiar with Peko-Wallsend, an “irrelevant” consideration is
a prohibited consideration, namely the taking into account of an extraneous factor.
Mason J actually referred to giving “excessive weight to a relevant factor of no great
The joint reasons then continued:
“Whether a decision-maker be regarded, by reference to the scope
and purpose of the statute, as having committed a particular error in
reasoning, given disproportionate weight to some factor or reasoned
illogically or irrationally, the final conclusion will in each case be that
the decision-maker has been unreasonable in a legal sense.”
This passage adds “error in reasoning” and “reasoned illogically or irrationally” as if
these were self-evidently grounds of invalidity, and regardless of the statutory
context. If they are to become such, it is surely necessary at least to acknowledge
the fact that there is, as earlier noted, authority in the High Court which is squarely to
Thirdly, the joint reasons noted Mason J’s remark that “guidance may be found in the
close analogy between judicial review of administrative action and appellate review
of a judicial discretion”, referring to House v The King.34 The principle derived from
that authority was in fact limited to the proposition that “unreasonableness” may be
an inference “drawn from the facts and the matters falling for consideration”. What
the High Court actually said in House v The King35 was as follows:
“It may not appear how the primary judge has reached the result
embodied in his order, but, if upon the facts it is unreasonable or
plainly unjust, the appellate court may infer that in some way there has
been a failure properly to exercise the discretion which the law
reposes in the court of first instance. In such a case, although the
nature of the error may not be discoverable, the exercise of the
discretion is reviewed on the ground that a substantial wrong has in
33  HCA 40; 162 CLR 24 at 41.
34 Li at ; Peko-Wallsend at 42.
35  HCA 40; 55 CLR 499 at 505.
This principle is well-understood in the area of judicial review of administrative
action, especially in cases where no reasons are available.36 This is not so much a
principle of judicial review, or the specification of a ground of review, but rather
identification of a process by which error may be inferred from a patent mismatch
between the material available to the decision-maker and the result.
It is worth reminding oneself in this context that analogies are fraught with risk, if
contextual difference is not borne in mind. Whatever the similarities of language
used by the courts in respect of appellate jurisdiction and judicial review of
administrative action, the functions in each case are separate. House, and similar
cases dealing with civil jurisdiction, are not applied in relation to appeals limited to
questions of law, but operate where the appellate court is conducting an appeal by
way of rehearing on the facts. Secondly, although the language of discretion is
commonly applied to a binary decision (such as granting or refusing an
adjournment), where based on considerations requiring evaluative judgment, it is
being applied to decisions which could legitimately fall anywhere within a range.
These exercises are different. House was dealing with a sentencing appeal and
therefore fell into the latter category.
Fourthly, there is a passage in the middle of the discussion referred to above in the
joint reasons where there is apparent endorsement, albeit obiter, of “a proportionality
analysis by reference to the scope of the power”.37 Giving “excessive weight – more
than was reasonably necessary” to a particular consideration, was described as “an
obviously disproportionate response … by which a conclusion of unreasonableness
may be reached”.
Because the reasoning was undeveloped it might be unnecessary to say more about
it, except that it will undoubtedly be relied upon by counsel in future cases.
Again, there was no reference to the doubts cast on the use of proportionality as a
basis of review in earlier authority. Nor was there acknowledgement of two
important considerations in this area: first, proportionality review requires an
assessment of the merits and thus, at least in Australian jurisprudence, strays
beyond the proper scope of judicial review, as a matter of principle. Secondly, it
involves heightened scrutiny of administrative decisions. The reason why these
dicta cannot be entirely disregarded is that one senses a lowering of the barrier to
judicial review of administrative decisions, except in the reasons of Gageler J.
What is intriguing in Li’s case is that her challenge to the adverse assessment by the
authority was ultimately successful: had it not been, there would have been no
purpose in the judicial review of the adjournment decision, because without a
favourable assessment she could not have ultimately succeeded in her visa
application before the Tribunal. Of course, to rely upon a later event to challenge the
refusal of the adjournment would not be permissible reasoning. On the other hand,
the later event raises a question as to whether the real issue was the failure of the
Tribunal to undertake some assessment of the strength of her claim which was, not
merely that she was challenging the adverse assessment, but that she had good
36 See Avon Downs at 360.
37 At -.
prospects of success. (That challenge was made to a different body.) One might
infer from the brief reasons given by the Tribunal (in effect, that “enough was
enough”) that the Tribunal not only did not undertake that task, but thought it
unnecessary or inappropriate. That might require the reviewing court to ask two
further questions, namely: first, whether if the Tribunal did adopt that approach, it
erred in law and, secondly, assuming it was in error, whether that was the kind of
error which constituted a failure of the Tribunal to exercise its statutory jurisdiction.
If, as the Court reminded us, the consequence of a failure involves an exercise in
statutory interpretation, in accordance with Project Blue Sky, then might not the
legislative provisions vesting procedural control in the Tribunal, taken with the
privative clause in s 474 of the Migration Act, provide a strong statutory indication
that such procedural errors were not intended to carry with them the consequences
The judgments of French CJ and Gageler J might be thought to accord more closely
with conventional authority in this country, with respect to both the statutory
provisions and judicial review generally. However, each adopted a course which
sought to establish a grand unifying theory based on what was described by French
CJ as “the framework of rationality”. French CJ also flirted with concepts of
“proportionality”,38 whilst eschewing the need to consider either the gap, or the
bridge, between that concept and rationality.
Gageler J, whilst referring at one point to the rather more flexible approach adopted
in Canada in Dunsmuir v New Brunswick39 was emphatic in adhering to the limited
degree of scrutiny undertaken under Wednesbury unreasonableness.40 Referring to
the family law cases of Lovell v Lovell41 and Norbis v Norbis,42 Gageler J noted that
“failure to give adequate weight to relevant considerations really amounts to a failure
to exercise the jurisdiction actually entrusted to the Court”. He noted, by reference to
Norbis, that it was harder to satisfy that standard “particularly when the
administrative discretion is wide in its scope or is affected by policies of which the
Court has no experience”. He was disinclined to think that those difficulties affected
the decision of the Tribunal to refuse an adjournment.43
Gageler J (and I paraphrase), held that no great deference (my word) need be
accorded to the Migration Review Tribunal reviewing an adjournment application,
because that is close to the exercise of a judicial function. I am troubled by that
approach: in some cases it may work, but in others, by abstracting a highly particular
aspect from a broader context, there is a risk that vibrant colour will be reduced to a
Why, then, is Xinjuan Li an important decision? First, because all members of the
Court sought to locate broad underpinning principles of judicial review, which are
38 At .
39  1 SCR 190.
40 At .
41  HCA 52; 81 CLR 513.
42  HCA 17; 161 CLR 513.
43 At .
often somewhat glibly, described simply as inherent in the rule of law. That exercise
is to be welcomed. Secondly, Gageler J drew attention to an insightful discussion by
Sir Philip Sales, a member of the English High Court.44 Sales J adopted an approach
to the development of the law in this area which is both flexible and incrementalist.
He is wary about proportionality. He stated:45
“The adoption of proportionality as a general standard of review would
represent a significant change in the law involving a shift in the
balance of power between the judiciary and the political branches of
That is because, as he recognised, proportionality contemplates “a more intensive
form of judicial review than the rationality/Wednesbury reasonableness standard”.46
On the other hand, Sales J emphasised a degree of flexibility implicit in the common
law standards and statute law stating:47
“However, the qualifications or conditions which are identified as being
presumptively implied into statutory powers may themselves be opentextured
(such as the duty to act fairly or in accordance with natural
justice or, indeed, the duty to act rationally) rather than narrowly
formulated, so allowing the courts a margin of evaluation in working
out how they should be applied on particular facts or in the face of
changing social expectations and standards.”
This element of flexibility was also accepted by Gageler J in Xinjuan Li, referring to
the comment of the US Supreme Court in Universal Camera Corp v National Labor
Relations Board48 that “[a] formula for judicial review of administrative action may
afford grounds for certitude but cannot assure certainty of application”. The formula
adopted should not, the Supreme Court suggested, be used “as an instrument of
Curiously, Universal Camera Corp appears to have been cited in Australian cases on
only two previous occasions. The earlier, a judgment of Davies J in the Federal
Court, considered whether the “no evidence” ground of judicial review should extend,
in accordance with US authority including Universal Camera, to sufficient evidence
as a reasonable mind might require as adequate support for a conclusion.49 Davies J
accepted that “the Wednesbury principle applies if a decision is so unsupported by
the facts that no reasonable decision-maker would have made it”, but nevertheless,
44 “Rationality, proportionality and the development of the law” (2013) 129 LQR
45 At 225.
46 At 226.
47 At 228.
48 340 US 474 at 488-489 (1951).
49 CA Ford Pty Ltd (T/as CA Ford Castors) v Comptroller-General of Customs
 FCA 560; 34 ALD 123.
noting the importance of not usurping the function of fact-finding reposed in the
administrative body, declined to adopt a “sufficient evidence” standard.50
The second and much more recent case is a judgment of Robertson J in the Federal
Court, relying on Universal Camera for the proposition that jurisdictional error was
not to be rejected on the basis that the alleged error did not fall within “fixed
categories or formulas”, which were to be seen as servants rather than masters.51
The call for flexibility is revealed as a need for a functional approach. To an extent,
this may be recognised in the now frequent admonitions to pay “close attention” to
the provisions of the governing statute. That, however, is only part of the exercise.
It is also necessary to have regard to the nature of the interest which is in contention,
the nature, including possible specialist expertise, of the tribunal or officer in whom
decision-making power has been reposed and practical considerations as to the
consequences of any tightening or relaxation in judicial control. If courts focus too
narrowly on individual cases, they will miss the ramifications for the administration of
government generally, which have both practical and constitutional elements.52
Sales J further noted that, “domestic judicial review took off in the 1960s”, referring to
the English trilogy of Ridge v Baldwin, Padfield and Anisminic. He continued:53
“In each case the House of Lords took itself to be engaged in the
interpretation of the statute in question, but the way in which it gave
effect to it was based on values more favourable to individual interests
and contrary to the claims of executive bodies than might have been
Similar developments occurred later in Australia but are undeniable. What is lacking
in this country is the thorough assessment of the legitimacy of the process and an
assessment of whether an appropriate balance has yet been achieved.54 Close
attention to the terms of particular statutory provisions will not provide guidance as to
broader “rule of law” values, which transparency of approach requires. These values
can only be properly articulated by the courts: they set the rules. Xinjuan Li appears
to demonstrate a growing consciousness of the need to engage with standards and
intensity of review. The development is to be welcomed.
50 The American principle derives from the express language of the
Administrative Procedures Act 1947, which finds no reflection in the ADJR
Act or other similar Australian statutes.
51 Minister for Immigration and Citizenship v SZRKT  FCA 317 at .
Mining Universal Camera is becoming positively fashionable: Allianz Australia
Insurance Ltd v Shamoun  NSWSC 579 at  (McCallum J)
52 85 ALJ 273 at 293-294, referring to the work of JL Mashaw.
53 At 229.
54 Achieving a balance is not, of course, the end point in a process, but merely a
Let me finish with a morsel for further reflection – harking back to CA Ford, an antidumping
case. It reminds one that perhaps the most important national interests
arise in the field of economic regulation. Yet administrative law is being developed
by reference to a discrete area of asylum-seeker related decisions and migration
decisions more generally. The interests at stake are often important issues of
individual human rights arising under the Refugees Convention. They may well
provide a litmus test for the health of our civil society, but other areas of executive
action may call for a differential application of principle.
Some years ago, though less frequently now, State judges used to ask whether it
was necessary to read “all these migration cases”. A standard reply was “Yes, they
are where you find modern statements of administrative law”. But that may not be
the full answer. That thought raised a novel question (for me): why are
administrative law processes developing in some areas, but not others? One
answer may lie in the other half of the 1970 administrative law reforms: namely the
development of administrative appeal tribunals, with further appeals limited to
questions of law. The burgeoning of modern administrative law cases may be
occurring where there are no statutory rights of appeal, and especially where there
are strong privative clauses. As both these features have merely forced review
applications into a different (and now constitutionally protected) procedural channel
one may wonder why governments do not abandon such techniques and confer a
general single right of merit review of administrative decisions, combined with
statutory rights of appeal limited to questions of law.
Whether that course is ultimately embraced or not, the central point is that, in a
functional approach, the interests at stake should influence the intensity of scrutiny of
administrative decision-making. If “reasonableness” is to become the underpinning
principle of judicial review, greater attention will need to be paid to the intensity of
scrutiny accorded different kinds of decision. The result may be a binary choice
between validity or invalidity, but the standard applied can e flexible. And courts
must be prepared to state whether a particular aspect of executive power calls for
heightened scrutiny, or something less, and why.