The Makasa proscription and the error in the 2017 decision
23 It follows from the foregoing that the assessment of materiality in this case is affected by the ambit of the Makasa finality proscription, and the High Court's identification in Makasa of what is needed to enliven the re-exercise of the power by way of "subsequent events or further information" so as to provide a "different factual basis" for the formation of the suspicion that BRO18 did not pass the character test. A proper understanding of Makasa therefore frames the counterfactual exercise required to determine whether there was a realistic possibility of a different outcome had the error in the 2017 decision not been made.
24 The High Court explained further what was meant by "subsequent events or further information" amounting to a "different factual basis" for exercising the power in s 501(2) at [48] (footnote embedded):
Turning then to the scheme of the [Migration] Act and the AAT Act, it is important at the outset to recognise that nothing in the legislative scheme indicates an intention to displace the application of s 33(1) of the [Acts Interpretation Act 1901 (Cth)] to the power conferred by s 501(2) of the Act to the extent that subsequent events or further information not previously before the Minister or a delegate provide a different factual basis upon which to form a reasonable suspicion that a visa holder does not pass the character test. A new sentence of imprisonment amounting by operation of s 501(7)(c), or contributing by operation of s 501(7)(d), to the coming into existence of a new substantial criminal record within the meaning of s 501(6)(a) is an example. A new conviction providing a reasonable basis for making a revised assessment of the visa holder's "character" - his or her "enduring moral qualities" (Minister for Immigration and Multicultural Affairs v Gunner (1998) 84 FCR 400 at 408; Minister for Immigration and Multicultural Affairs v Jia Legeng (2001) 205 CLR 507 at 529 [65]) - under s 501(6)(c) is another example. Section 501(6)(c) was not argued to be engaged in this appeal.
25 A number of points emerge from Makasa at [48]:
(a) The "subsequent events or further information" relate to something that was not previously before the Minister (or delegate) so as to provide a "different factual basis upon which to form a reasonable suspicion that a visa holder does not pass the character test". It follows that determining whether there has been a re-exercise of the power in s 501(2) in conformity with Makasa may in some cases require a comparison to be made between the original foundation for the formation of the suspicion and the different foundation for the formation of the suspicion later relied upon. In this case, the re-exercise foundation was confined to a subset of the original foundation, so clearly did not conform with the Makasa requirement.
(b) The re-exercise of the power in s 501(2) may be one that entails making a "revised assessment" of the suspicion reasonably able to be formed about the visa holder's "character" under s 501(6)(c) upon a differently constituted foundation by reason of additional information, rather than a fresh assessment upon a wholly new foundation, excising the past.
(c) That revised assessment is of the visa holder's "enduring moral qualities", a phrase that apparently has its origins in the judgment of Lee J in Irving v Minister for Immigration, Local Government and Ethnic Affairs (1996) 68 FCR 422 (Davies, Lee and RD Nicholson JJ) at 431-2, drawing a distinction between this and good standing, fame or repute. The phrase "enduring moral qualities" as used in Irving was applied in Minister for Immigration and Ethnic Affairs v Baker (1997) 73 FCR 187 at 197 (Burchett, Branson and Tamberlin JJ), Minister for Immigration and Multicultural Affairs v Gunner (1998) 84 FCR 400 at 408 (Heerey, Lindgren and Emmett JJ), and Minister for Immigration and Multicultural Affairs v Jia Legeng [2001] HCA 17; 205 CLR 507 at [65] (Gleeson CJ and Gummow J, Hayne J agreeing). Gleeson CJ and Gummow in Jia at [65] said, citing Irving and Baker, that "enduring moral qualities" was concerned with "disposition rather than general reputation", and that "past conviction of serious crime did not necessarily mean, without examination of any other matters, that a person was of bad character at the time of decision-making". Their Honours referred to the observation by Latham CJ that a person "may be guilty of grave wrongdoing and may subsequently become a person of good character": In re Davis (1947) 75 CLR 409 at 416. This reasoning suggests that "enduring moral qualities" is pointing to a holistic assessment of a person, so far as the criteria in s 501(6) permit this, rather than a process that entails permanently disregarding indicia of character that has been previously considered.
26 In the second example given in Makasa at [48], a new conviction provides a reasonable basis for making a revised assessment of a visa holder's character within the terms of s 501(6)(c) upon a different factual basis, so as to reconsider what the person's "enduring moral qualities" are. This does not provide a clean slate excluding the past. It requires there to be additional information beyond that which has previously been considered so that the exercise of the power is enlivened once again upon a different foundation. As in Makasa, there was no suggestion in this case of reliance upon s 501(6)(c) in relation to the character test, with the reliance being confined to s 501(6)(a).
27 The same could be said of any event that is relevant to any of the character test grounds in s 501(6). This allows any sentence previously considered in the exercise of the power in s 501(2) to be reconsidered in the context of a new event, or becoming aware of further information, including about past events, relevant to character. A further prison sentence following a new conviction is the most obvious means of there being a different factual basis, not least because of the aggregating operation of s 501(7)(d): Makasa at [48]. The previously considered sentence and a new sentence might, for example, combine to pass the 12-month threshold for a substantial criminal record, which the sentence previously considered alone did not do. But additional information that had existed at the time of the previous exercise of the power, but not before the Minister (or delegate), including casting new light on those previously considered events, might also permit a revised assessment.
28 The soundness of the above reasoning can be tested by considering the alternative. If, in the course of re-exercising the power under s 501(2), the Minister could not consider a prison sentence that had already been considered as part of the prior exercise of the power, along with a new event or new information having a bearing on character so as to create a different factual basis, that would require the second assessment to be conducted upon a fictional or blinkered basis that no prior event relevant to character had taken place at all. It is difficult to envisage the High Court intending such an outcome, especially given the reference in [48] to a revised assessment, not to a wholly new assessment, and to the creation of a new substantial criminal record, either by a new sentence on its own meeting the threshold for a substantial criminal record, or by a new sentence contributing to the creation of such a record with a prior sentence or sentences by the operation of the aggregation provision in s 501(7)(d). This is the better way to read the reference to "contributing" in [48] of Makasa, rather than it referring to a substantial criminal record founded only on new sentences not previously considered.
29 Moreover, to ignore a relevant past event in the context of new information would not be an assessment of enduring moral qualities, but rather only of transient moral qualities in an isolated point in time, even though s 501(6)(c) was not relied upon in this case. This is contrary to the reasoning in Jia at [65], referred to at [25(c)] above.
30 Applying the above reasoning as to what Makasa permits to the present situation, there would have been no error in the Minister basing the 2017 finding of a suspicion that BRO18 did not pass the character test in part upon the previously considered 12-month prison sentence imposed by the Perth District Court on 8 December 2008, provided it was accompanied by consideration of additional information relevant to the character test to constitute, in combination, a different factual basis for the re-exercise of the power in s 501(2). The post-2009 sentences, or any of them, meet that description. The combination of the previously considered sentence and the new sentence(s) would be a different factual basis for the exercise of the power in s 501(2).
31 Viewed in this way, the error in the making of the 2017 decision lies not in having regard to the 2008 sentence in the formation of the suspicion per se, but in having regard only to that sentence, being part of what had been considered in making the 2009 decision, being all three 2008 sentences, such that there was no different factual basis in the sense of new or additional information deployed at the suspicion formation stage as required by Makasa for the re-exercise of the power in s 501(2). The fact that the post-2009 sentences were considered, albeit only at the discretionary second stage, is highly relevant to the assessment of materiality, not least because it has a bearing on the predictive aspect of the counterfactual analysis required. It is inherently more likely that the post-2009 sentences would have been considered, on a counterfactual analysis, at the suspicion formation stage, given that they were considered at the discretion stage, than if they had been wholly disregarded or not known.
32 Even if that is not correct, and Makasa does require that the 2008 convictions be wholly disregarded in the formation of the suspicion, as already noted, the post-2009 sentences in aggregate resulted in the imposition of 42 months of imprisonment, greatly exceeding the 12-month threshold in s 501(7) for a substantial criminal record for the purposes of s 501(6)(a). Thus, even if, on the counterfactual analysis, reliance on the 2008 convictions is required to be put entirely to one side, that could not have made any difference because the same outcome would inevitably have resulted by reason of the already known and already evaluated post-2009 sentences, leading to the formation of the same suspicion that BRO18 did not pass the character test. The same exercise of the discretion to cancel his visa would inevitably have followed, such that there is no realistic possibility that the overall result would have been different.
33 BRO18's argument in defence of the primary judge's decision cannot be accepted, primarily because of its dependence on disregarding altogether the 2008 sentence that the Minister relied upon for suspicion formation, without allowing the post-2009 sentences (or any one of them) to be part of what was considered at the suspicion formation stage. This is a formulation of materiality reasoning that allows only for subtraction, and not reasonable addition. The inquiry is not so narrow. The counterfactual is to be determined by "reasonable conjecture within the parameters set by the historical facts that have been determined on the balance of probabilities": MZAPC at [38]. That analysis is not limited to pencilling out the error and considering what the result would be. Materiality can be absent, for example, where a decision-maker failed to take into account a mandatory consideration if that consideration could not have materially affected the decision that was made: see Hossain v Minister for Immigration and Border Protection [2018] HCA 34; 264 CLR 123 at [30]. BRO18's objection to such counterfactual analysis on the basis that it is speculative must also be rejected. This argument overlooks the fact that a counterfactual analysis necessarily asks the court to speculate on what could have occurred had the error not taken place, but that such an inquiry is anchored in the determined facts.
34 It follows that the Minister's submissions should be accepted and those for BRO18 rejected. The Minister's submission as to errors made by the primary judge at J[68] and J[73] concerning the application of counterfactual reasoning to ascertain materiality should also be accepted. On the primary basis that Makasa proscribes consideration of any of the 2008 sentences in the formation of the required suspicion in the re-exercise of the power in s 501(2) only if nothing further was also considered that amounted to a different factual basis for that to take place, it is clear that the post-2009 sentences inevitably would have been considered at that stage, and accordingly the error in confining consideration to the 2008 conviction was not material.
35 Even on the alternative basis that any consideration of any of the 2008 convictions in the suspicion formation stage of the re-exercise of the power in s 501(2) was forever forbidden, the same position is reached. In that event, while BRO18's position that any consideration of any of the 2008 convictions in the suspicion formation stage of a re-exercise of the power is proscribed, it remains clear that the post-2009 sentences inevitably would have been considered at that stage, and accordingly the error in confining consideration to the 2008 conviction was still not material.
36 Put in positive terms, there is no compelling reason to doubt that had the Minister been aware of the Makasa prohibition on having regard to any of the 2008 sentences on the formation of the suspicion, or on having regard only to the 2008 sentences for that purpose, by reason of them having been deployed for that purpose as part of the 2009 decision, he would have considered the post-2009 sentences in whole or in part in the formation of the suspicion. That would have led inexorably to the conclusion that BRO18 could not possibly pass the character test given his post-2009 aggregate sentences totalled 42 months, being over three times the 12-month threshold. It follows that formation of the suspicion that BRO18 did not pass the character test based on the post-2009 sentences was not just reasonable, but unavoidable and inevitable. BRO18 could not possibly have satisfied the Minister to the contrary. The discretion undoubtedly would have been enlivened. In contrast to the kind of error made by the decision-maker in Nguyen v Minister for Home Affairs [2019] FCAFC 128; 270 FCR 555 as described at [51], there is not a "clear causal link" between the kind of error that occurred and the outcome of the decision. If the error, understood in either way, had not taken place, it could not realistically have produced a different result as to the formation of the suspicion that BRO18 did not pass the character test.
37 When it then came to the second stage and the exercise of the discretion, the result would clearly enough inevitably have been the same: cancellation of BRO18's visa. That is especially so given the chain of events leading to the 2017 cancellation decision, including notices of intention to consider cancellation which referred to the post-2009 sentences, which were not responded to.