Federal Court of Australia

Australian Competition and Consumer Commission v Australian Institute of Professional Education Pty Ltd (in liq) (No 4) [2020] FCA 1811

ORDERS

THE COURT ORDERS THAT:

1.    The parties submit agreed or competing orders either to finalise this proceeding in accordance with this judgment, or to prepare for any further adjudication, by 27 January 2021, or such other date as may be fixed in consultation with the parties.

2.    The proceeding be listed for case management at 9.00 am on 29 January 2021, or such other date as may be fixed in consultation with the parties.

Note:    Entry of orders is dealt with in Rule 39.32 of the Federal Court Rules 2011.

REASONS FOR JUDGMENT

BROMWICH J:

1    These reasons follow the delivery of a very lengthy and detailed judgment on liability largely in favour of the applicants, Australian Competition and Consumer Commission v Australian Institute of Professional Education Pty Ltd (in liq) (No 3) [2019] FCA 1982 (liability judgment). The parties have been unable to agree how the Court should determine, for the purposes of making a compensation order in favour of the Commonwealth (the second applicant), whether the enrolment of a consumer in an online course by the respondent (AIPE) was the result of the unconscionable system of conduct or pattern of behaviour that I found had existed. This is an adjudication of that disagreement, made on the papers in response to detailed written submissions and by reference to the liability judgment. These reasons must therefore be read with that judgment. As noted below, key parts of that judgment are reproduced in a schedule to these reasons.

2    The conclusion I have reached as detailed below will be used by the Commonwealth to seek a compensation order pursuant to s 237 of the Australian Consumer Law (ACL) for the student assistance money paid by it to AIPE under the Higher Education Support Act 2003 (Cth) (HES Act) scheme known as the Vocational Education and Training Fee Higher Education Loan Program (VET FEE-HELP or VFH Scheme).

3    The applicants, the Australian Competition and Consumer Commission (ACCC) and the Commonwealth, contend that the Commonwealth should be entitled to be compensated in relation to “non-completion consumers”. They define a non-completion consumer as a consumer who did not complete a single unit of study in an AIPE online course in which they were enrolled as a student. That is, the applicants contend that AIPE should be ordered to pay compensation for the aggregate sum of VET FEE-HELP course fee payments that the Commonwealth made directly to it on behalf of each non-completion consumer. The applicants submit that this approach flows from the reasons and finding in the liability judgment that AIPE engaged in a system of conduct or pattern of behaviour that was, in all the circumstances, unconscionable within the meaning of s 21 of the ACL.

4    On that approach the compensation order would be in the order of $198 million. The final sum would then be sought to be enforced against AIPE by way of a proof of debt in its liquidation, which would affect other creditors.

5    Clause 46AA in Schedule 1A to the HES Act was inserted with effect from 1 January 2019, and provides for the re-crediting of VET FEE-HELP debts. This has taken place in relation to 12,948 of what the applicants call non-completion consumers, such that no relief in their favour is now sought under s 239 of the ACL. The Commonwealth instead seeks orders for compensation from AIPE under s 237 of the ACL for that money. AIPE does not challenge that means of redress, but does challenge the applicants’ proposed basis for determining the ambit of the class of students to which that applies and thereby the quantum of the compensation order that can be made. Clause 46AA re-crediting is useful information, and shows actual loss to the Commonwealth for which AIPE concedes a subset will be properly compensable. However, I do not consider that this assists in the task before me of determining the proportion of such money advanced to students that should be ordered to be paid by AIPE to the Commonwealth. The statutory criteria for those re-credits do not sufficiently align with the issue before me, nor sufficiently focus on the findings I have made in the liability judgment.

6    AIPE contends that the applicants’ approach in seeking compensation based upon non-completion of a single unit amounts to nothing less than using that outcome as a “proxy” for unconscionability. It contends instead that the criteria for a compensation order being made, relying on its characterisation of what was decided by the liability judgment, should be (emphasis in original):

Any student that has not completed any unit of [study] and either (a), or (b) applies to that student.

(a)    AIPE has no record of the student logging in or engaging in activity in Moodle or online for the course or unit of study in which the student was enrolled;

or

(b)    One or more of the following circumstances apply:

(i)    when they were enrolled in the Relevant Course [(a Relevant Course was defined in orders made on 1 May 2020 as a course offered by AIPE between 1 May 2013 and 1 December 2015)] they received misleading or deceptive information in relation to the operation of the VFH Scheme or the Relevant Course;

(ii)    they were offered an incentive to enrol in the Relevant Course, such as a free laptop computer, tablet, or cash;

(iii)    they did not understand that by being enrolled in the Relevant Course, they would incur a debt to the Commonwealth; and

(iv)    they could not use or access the Internet or a computer, had inadequate literacy or numeracy skills, or were otherwise unsuitable, by virtue of vulnerability or disadvantage for enrolment in the Relevant Course.

7    The difference in the approach of the parties turns on whether the additional criteria identified by AIPE should be required to be satisfied, or whether some other additional criteria should be brought to bear, consistently with the liability judgment. On AIPE’s approach, there must either be no evidence of any online activity with AIPE by a student at all, or direct evidence of such a student being actively misled, demonstrably vulnerable, or inherently unsuitable in one of four ways. Unless the first criterion can readily be satisfied, which seems unlikely for any substantial number of students as each would have had to do nothing at all as opposed to merely being wholly unsuited to the course in which they had been enrolled, the alternative second criterion is likely to be very difficult to meet, and even then only by extensive further inquiries. This is likely to be very difficult to establish given the sorts of consumers that were enrolled by AIPE as “students”.

8    If AIPE’s reasoning in support of their proposed criteria is correct, it would, in many cases including the present matter, render the proscription against a system of conduct or pattern of behaviour that was, in all the circumstances, unconscionable, toothless and without any real practical consequence. That is an outcome that I considered in evaluating the competing arguments. However, as it turned out, a conclusion in that regard was not necessary because of the nature of the case and sufficiency of the evidence that the applicants were ultimately able to adduce, a feature that may be absent in other like cases.

The key findings in the liability judgment

9    The following identification of key findings in the liability judgment is not a substitute for reading and understanding that judgment in full. Rather, it isolates for convenient reference the key findings made along the path to the ultimate conclusion that AIPE engaged in a system of conduct or pattern of behaviour that was, in all the circumstances, unconscionable. The key paragraphs for the purposes of this adjudication are reproduced in a schedule to these reasons.

Section 237 of the ACL

10    The issue of compensation orders ultimately turns on the terms of s 237(1)(a)(i) of the ACL, which provides as follows:

(1)    A court may:

(a)    on application of a person (the injured person) who has suffered, or is likely to suffer, loss or damage because of the conduct of another person that:

(i)    was engaged in a contravention of a provision of Chapter 2, 3 or 4; …

make such order or orders as the court thinks appropriate against the person who engaged in the conduct, or a person involved in that conduct.

11    There is no longer any dispute that AIPE engaged in conduct in contravention of Chapter 2 of the ACL, by breaching s 21. Nor is it disputed that the Commonwealth is an injured person and has brought an application for compensation under s 237. The live dispute concerns ascertaining the existence and the quantum of loss or damage “because of the conduct” of AIPE based on the findings made in the liability judgment.

12    The substantively identical predecessor to s 237, being s 87 of the former Trade Practices Act 1974 (Cth) (TPA), was considered by the High Court in I & L Securities Pty Ltd v HTW Valuers (Brisbane) Pty Ltd [2002] HCA 41; 210 CLR 109. The salient facts were concisely described by Gleeson CJ in part of [1] as follows:

The misleading or deceptive conduct involved an erroneous valuation of real estate over which a mortgage was to be given as security for a loan by the appellant. Relying upon the valuation, the appellant made the loan. The borrower defaulted, and the security, when realised, was insufficient to meet the borrower’s liability. The appellant sued for the deficiency and related losses. The appellant was found to have failed to exercise reasonable care to protect its own interests, in that it did not take proper steps to investigate the credit-worthiness of the borrower.

13    In I & L Securities, the contribution of the appellant’s carelessness in failing to check the borrower’s creditworthiness deprived it of recovering part of the loss suffered from having made the loan in reliance on the faulty valuation. That deduction was overturned by a majority of six justices of the High Court.

14    Gleeson CJ, as part of [33], commented upon s 82 in terms that were equally applicable to s 87, and thereby s 237 of the ACL:

The statutory purpose would be defeated if the remedy under s 82 were restricted to loss of which the contravening conduct was the sole cause. What is there, then, in the justice and equity of the particular case that might lead to a conclusion that the respondent should not be regarded as legally responsible for the whole of the loss, even though the contravention was a cause of the whole of the loss? Upon what principle might such responsibility be diminished? In a financing transaction, a lender takes security to protect itself against the risk of default by the borrower. One aspect of that risk is that the lender might have failed adequately to assess the borrower’s capacity to service the debt. I cannot see why, as a matter of principle, such failure by a lender should be treated, in the application of s 82, as a factor which diminishes the legal responsibility of a valuer by negativing in part the causal effect of the valuer’s misleading conduct. The statutory rule of conduct found in s 52, when applied to the relationship between a valuer and a prospective lender, gives rise to a legal responsibility in a case such as the present which extends to the whole of the loss of which the valuer’s misleading conduct is a direct cause.

15    Gaudron, Gummow and Hayne JJ said (footnotes omitted):

[52]    Section 87 may be engaged where there is a proceeding instituted under, or for an offence against, Pt VI of the Act. It may, therefore, be engaged where there is a proceeding instituted under s 82. It requires that there be a finding that a person who is a party to the proceeding has suffered or is likely to suffer loss or damage by conduct of another person that was engaged in in contravention of a provision of the specified parts of the Act. It is in those circumstances that the court may make “such order or orders as it thinks appropriate” against the contravener or a person who was involved in the contravention, whether or not the court grants an injunction under s 80, or makes an order under s 80A or s 82. The orders that may be made include, but are not limited to, the orders set out in s 87(2). A court may make an order under s 87(1) only “if the Court considers that the order or orders concerned will compensate the first-mentioned person in whole or in part for the loss or damage or will prevent or reduce the loss or damage”.

[53]    The reference to compensating “in whole or in part” for the loss or damage requires consideration of the compensatory effect that “the order or orders concerned” will have. Will the particular order that is made under s 87 (such, for example, as an order varying a contract (s 87(2)(b)) compensate for part of the loss or damage that has been sustained? Will that order prevent or reduce loss or damage that otherwise would likely be suffered? But the words “in whole or in part” do not suggest that the combination of orders that a court makes should do less than provide for full compensation for all loss and damage that is not prevented by the making of the court’s orders.

[54]    Like s 82, s 87 speaks of loss or damage suffered or likely to be suffered “by conduct of another person that was engaged in . . . in contravention of a provision” of specified parts of the Act. Section 87, like s 82, therefore requires the identification of a causal connection between loss or damage and contravention. What is the connection that must be demonstrated?

[55]    If there is a contravention of the Act and, following that contravention, a person suffers loss or damage, it may be possible to identify several features of the history of events as having contributed to the person suffering loss. To take the simple example of a person who suffers loss or damage following a person making a misleading or deceptive statement, the loss may be said to have been caused by the combined effect of the making of the statement and the reliance on it by the person who suffers loss. Sometimes it will be open to say that the person who relied on the statement was foolish to do so or, at least, did not take reasonable care to protect his or her own interests. Similarly, to take a further example, if there is a contravention of s 46 of the Act by a corporation having a substantial degree of market power deterring a person from engaging in competitive conduct in that market, it may, in some circumstances, be open to say that the person deterred could, or even should reasonably, have made some competitive response different from the response it did. In those cases it may well be that the loss or damage which has been suffered would not have been suffered but for each of the persons who suffered loss acting, or omitting to act, as they did.

[56]    There may be many acts or omissions that could be said to have contributed to the happening of an event. As has often been mentioned in learned articles on the subject of causation, the decision of a tortfeasor’s great-great grandmother to have children can be identified as one factual cause for an event which is the subject of litigation. To search for the single cause of an event is, therefore, to pursue an illusion. And, much more often than not, to speak of the “effective cause” or the “proximate cause” (or to use some similar expression) is to hide important assumptions that are made, or conclusions that are reached, about the attribution of responsibility for particular kinds of act or omission. That is why it is necessary to understand the purpose for making some inquiry about causation. Only when the purpose of the inquiry is known is it possible to identify and articulate how and why some circumstances are extracted “out of the whole complex of antecedent conditions of an event” and identified by the law as a cause of it.

[57]    In light of these considerations, it is hardly surprising that it is now well established that the question presented by s 82 of the Act is not what was the (sole) cause of the loss or damage which has allegedly been sustained. It is enough to demonstrate that contravention of a relevant provision of the Act was a cause of the loss or damage sustained.

16    McHugh J went on to draw a distinction between two causes of loss. One was where there were contributing causes of an indivisible loss, of which one was the contravening conduct. The second was where loss was divisible into parts, of which the contravening conduct was a cause of one or more parts by reference to s 82, but in terms equally applicable to s 87 and thereby s 237 of the ACL: see [69] and [89]. The contravener is liable for all of the loss in the former, but only the parts for which it had some responsibility in the latter. For the former, it did not matter that other factors played a part in bringing about the loss: [92].

17    In Awad v Twin Creeks Properties Pty Ltd [2012] NSWCA 200, Allsop P (as the Chief Justice then was), with the agreement of Macfarlan JA and Sackville AJA, observed at [43] that “[r]elief under the TPA, s 87, should be viewed not by reference to general law analogues but by reference to the rule of responsibility in the statute that is directed against misleading and deceptive conduct”, citing extensive High Court authority in support of that perhaps otherwise self-evident proposition. His Honour concluded [43] by citing, inter alia, some of the paragraphs from I & L Securities reproduced above and observing upon that basis that “[a]n approach that is limited mechanically around a but for causation enquiry will be likely not to involve a full evaluative assessment of the appropriate relief”. Allsop P also observed at [44] that if there has been a contravention of the statutory norm, many factors may influence the question of relief, such as the weight of the influence of the impugned conduct, but this is not of itself a determinative factor upon which relief turns. His Honour (at [46]) characterised an alternative basis for liability: that the conduct (there a representation) was operative, contributory and intended to be material, giving rise to relief that conformed to the rule of responsibility so as to undo the contract in contention in that case altogether. This is comfortably removed from tort causation reasoning.

The competing arguments

18    Following I & L Securities and Awad, the applicants submit that it is not necessary to ascertain with precision the loss or damage “caused” by the conduct of AIPE in the manner of a tort assessment of damages. Rather, they submit, AIPE should bear the loss by way of a compensation order in relation to a class of transactions “tainted” by its unconscionable conduct. It is enough, the applicants contend, that the contravention was a cause of the loss or damage: I & L Securities at [57]. Put another way, the system of conduct or pattern of behaviour was operative, contributory, and intended to be material: Awad at [46].

19    The applicants submit that the payment by the Commonwealth of VET FEE-HELP assistance to the non-completion consumers reflected the objects in s 2-1(d) of the HES Act, relevantly being to support students undertaking vocational education and training of the type provided by AIPE. Those payments to those consumers can be seen to be money lost in each case in which the payment made has not achieved any kind of educational training, in the sense that the consumer has not even completed a single unit of study. But how is AIPE said to be responsible for this outcome?

20    On the applicant’s argument, the class of non-completion consumers sufficiently connects the unconscionable conduct found to have taken place to the loss that has been suffered by the Commonwealth. On this argument, the conduct was a cause of the loss, because the Court’s finding in relation to that conduct was that it constituted a system or pattern that contravened s 21 of the ACL and that AIPE’s unconscionable businesses practices, on the balance of probabilities, likely affected all, or the significant majority, of consumers who were subject to them.

21    Implicit in this argument is that the loss was not just an unhappy event occurring in relation to a smattering of students, or even a substantial number of students, unrelated or insufficiently related to an unconscionable enrolment system, but rather flowed from and is explained by AIPE’s conduct as determined adversely to it in and by the liability judgment. Put another way, the argument is that the loss was relevantly “because of the conduct” of AIPE, even if that conduct might fall short of what would be required by the rigours of tort causation.

22    The applicants argue that this approach is also consistent with the focus of s 21, and thus the focus of the Court’s liability determination, being on AIPE’s conduct, not on the focus of the individual characteristics of the individual consumers. The applicants rely upon the observation at [59] of the liability judgment that requiring “granular evidence of particular transactions or individuals would defeat the strengths of s 21(1) via s 21(4)(b) of the ACL in not requiring proof of individual disadvantage, or requiring pervasive evidence for all points in time”. To allow a more general approach at the liability stage, only to demand more at the loss or damage stage, is said to defeat the whole purpose of a broad provision such as s 21.

23    The applicants submit that the criterion of non-completion of a unit of study is consistent with the “broad remedial object” of ss 21 and 237 of the ACL and “best captures the proper ambit for redress having regard to the nature of AIPE’s contravening conduct and, in particular, the design of AIPE’s enrolment system to limit consumers’ knowledge and understanding of the census date (and thereby opportunities for their enrolment to be cancelled)”. They point to an enrolment system that was declared to be directed to persons who were vulnerable. They submit that this is consistent with the approach taken by Gleeson J in Australian Competition and Consumer Commission v Cornerstone Investment Aust Pty Ltd (in liq) (No 5) [2019] FCA 1544 (referred to in the liability judgment paragraphs reproduced in the schedule to these reasons as Empower, being the college run by the respondent), especially at [99] in the same kind of case.

24    The applicants submit that the approach urged by AIPE would require an enquiry to be made into the particular circumstances of enrolment of each of the 12,948 non-completion consumers. This would first require identifying those in AIPE’s proposed category (a), and then contacting the balance and having those who can be contacted identify whether they were subject to specific and granular aspects of AIPE’s unconscionable conduct as listed in (b)(i) to (iv). This is charactered as not capturing the extent of AIPE’s unconscionable conduct as found by in the liability judgment, and as being uncertain and inequitable and resulting in under-recovery by the Commonwealth. The applicants submit that such a granular approach was only justified for consumers who completed one or more units of an AIPE online course, defined as unit completion consumers.

25    There is no dispute between the parties as to what needs to happen for the Commonwealth to recover any loss or damages in relation to the unit completion consumers. An agreed process is being undertaken to determine whether any were subject to misleading or deceptive enrolment information, offered an incentive to enrol, did not understand that they were incurring a debt to the Commonwealth, or if the course in which they enrolled was not suitable. These reasons are therefore not concerned with those unit completion consumers, being some subset of 828 identified consumers.

26    In support of the approach the applicants advocate for in relation to non-completion consumers, they assert that those consumers obtained no benefit at all from enrolment at AIPE. As such, they assert it is open to the Court to infer that the non-completion consumers were subject to the contravening conduct found to have taken place, in that they were unsuitable for enrolment, were enrolled without understanding that they would incur lifetime debts to the Commonwealth, or did not understand the importance of the census date.

27    The applicants contend that the following conclusions can be drawn from the liability judgment (omitting footnotes):

(a)    AIPE engaged in a system of conduct or pattern of behaviour that was, in all the circumstances, unconscionable within the meaning of s 21 of the ACL.

(b)    An overwhelming majority of consumers enrolled by AIPE as VET FEE-HELP funded students failed to complete their courses and were not suitable to be enrolled and so were not genuine or bona fide students.

(c)    AIPE was fully aware that most of the consumers who were enrolled as online students were not partaking of the course in which they were enrolled, and that this was largely the situation from the moment that enrolment took place.

(d)    The class of consumer targeted by AIPE to be enrolled was characterised by a generalised aspect of vulnerability with an inherently higher risk of being unsuitable to be enrolled as a student, in the sense of being less able to detect defects in the false information they were given, and to resist becoming enrolled in a course despite being unsuitable.

(e)    AIPE’s business model focused on maximising revenue through government VET FEE-HELP loans, with little or no ambition to provide actual training to more than a very small proportion of consumers enrolled as students. A very high proportion of consumers enrolled never actively participating in study (at least 70%) was a profitable and desirable outcome that was effectively encouraged, or at least not meaningfully discouraged.

(f)    Results from the McGrathNicol survey found that for a significant number of the 6,066 course enrolments that AIPE reported for 2015, there was a very low level of engagement with … AIPE’s online learning system. The McGrathNicol survey found that [15%] of students had not logged into the online learning system, and [76%] of students had logged in on four different days or less. Only [10%] of students had logged in on more than 10 different days. This contrasts with students who completed courses, who were found to have had an average of 58 days active, indicating those students who had logged in on less than 10 days may not have significantly advanced toward course completion. This was overwhelmingly, if not entirely, the product of AIPE’s enrolment system that enrolled unsuitable consumers.

(g)    The vast majority of students enrolled were incapable of participating in the courses in which they were enrolled.

(h)    Surveys of students following enrolment were characterised by low contact rates and response rates.

28    The above is a reasonably comprehensive, but not exhaustive, summary of the burden of the liability judgment, and is reflected in the paragraphs of that judgment reproduced in the schedule to these reasons. However, it evades to an extent the causation question posed by s 237 of the ACL.

29    The applicants further contend that the Commonwealth’s loss or damage is the total sum of VET FEE-HELP assistance paid to AIPE in discharge of the liability to pay VET tuition fees of non-completion consumers in respect of whom the amounts are no longer recoverable as a result of AIPE’s contravening conduct. They submit that this loss was suffered as a result of the widespread system of unconscionable conduct found by the Court. They therefore submit that the order they seek, and the basis for it, is “appropriate” as required by s 237(1).

30    AIPE’s contrary position, aided by the form of order sought as reproduced at [6] above, is expressed with some conciseness and can be restated with the same efficiency. AIPE characterises the applicants’ argument as being that “non-completion” be the sole criterion for determining the class of students in respect of whom the Commonwealth is to be compensated on the basis that it is a “proxy” for unconscionability. It submits that this proposal fails to grapple with the wording of s 237 and in particular the role that the phrase “because of the conduct” must play.

31    AIPE contends that for non-completion to be an appropriate proxy, there would need to be a finding that every non-completion student’s enrolment, and thereby loss to the Commonwealth, was caused by its contravening conduct, which is inconsistent with the applicants’ position in relation to the unit completion students. That approach, AIPE says, cannot be accepted as there is no evidence or finding that supports non-completion alone as a suitable proxy for unconscionability. AIPE contends that its alternative criteria rely on the findings of contravening conduct in the liability judgment, rather than any concept of a “proxy”.

32    AIPE accepts that an inference may be drawn that consumers enrolled in an online course who did not log in or participate in that course were enrolled by reason of its contravening conduct. It submits that those consumers are therefore an appropriate class in respect of whom compensation to the Commonwealth should be paid. AIPE submits that no finding was made in the liability judgment that the quality of teaching to students who participated in the courses was not sufficient or the courses were inadequate, with the proceedings being concerned with the enrolment of consumers who were not bona fide or genuine and who did not participate in AIPE’s courses. AIPE submits that the position is different for consumers who, after enrolling did take steps to participate, because that tends to show that the consumer intended to enrol in the course of study. In addition to suffering from a lack of logic, for the reasons detailed below, those propositions entail an unwarranted gloss on the conclusions that I reached. This is not least because participation was expressly found to vary along a spectrum, rather than to be categorically present or absent: see [636]–[637] of the liability judgment, reproduced below.

33    Be that as it may, those arguments are said by AIPE to be consistent with the approach of Murphy J in Director of Consumer Affairs Victoria v Domain Register Pty Ltd (No 2) [2018] FCA 2008 at [32(a)] (emphasis added by AIPE):

In my view a subset of the 21,089 persons who acquired a “.com” domain name in the relevant period are likely to have suffered loss or damage by reason of Domain’s misleading conduct, and the orders operate to undo, as best as possible, the damage done by Domain’s contraventions. I consider the orders are appropriately tailored to the circumstances of the case, including because:

(a)    section 240(1) provides that the Court may have regard to the conduct of the non-party consumer since the contravention occurred. In circumstances where there is no evidence that the renewal notices are themselves misleading, renewal of a “.com” domain registration by a person tends to show that they intended to acquire it in the first place. In my view it is only appropriate to make a non-party consumer redress order in relation to those persons who acquired a “.com” domain name and did not subsequently renew it. This means that only 9,851 of the 21,089 persons who acquired a “.com” domain name will be eligible to claim a refund;

…

34    AIPE therefore submits that, applying this reasoning, it is open to find that consumers who participated in the courses intended to enrol in them and it is not possible to draw the inference that they were unconscionably captured by a system that enrolled unsuitable consumers. They argue that for consumers who did participate in the course in which they were enrolled, loss to the Commonwealth must be assessed by reference to other contravening conduct of AIPE directed towards that consumer. For this reason, it is submitted that the same process should apply to unit completion students as to non-completion consumers who participated by as little as logging on, treating both as participating consumers. They submit that participating consumers compensation should be limited to those who met one or more of each of the four criteria proposed which were elements of the overall unconscionable system of conduct or pattern of behaviour found against it, reproduced at [6] above.

35    The applicants object to these proposed criteria on the bases that they raise practical difficulties and that it is inequitable for AIPE to retain VET FEE-HELP payments in respect of consumers who have not completed a single unit of study. AIPE characterises the former objection as inappropriate and the latter as irrelevant. Rather, the necessary causal connection, according to AIPE, must be found before the power under s 237 is available.

Consideration

36    The starting point is an overall characterisation of the VFH Scheme and AIPE’s enrolment system:

(1)    The Commonwealth paid AIPE VET FEE-HELP assistance on behalf of “students” enrolled, with the objective of enabling those students to undertake vocational education and training.

(2)    The evident purpose of the VFH Scheme is that such students would gain skills and employment and, in most cases, when their income rose above the statutory threshold, would be able to repay the loan (being the VET FEE-HELP assistance plus 20%).

(3)    A necessary part of the VET provider’s obligation was to enrol suitable students, both for the sake of those students, and for the sake of the viability of the VFH Scheme overall.

(4)    AIPE subverted the scheme by its unconscionable conduct in relation to the enrolment of, relevantly, some of the non-completion consumers, but not all such students. A further limitation is required to ensure the necessary causal connection required by s 237.

(5)    Payments made on behalf of those students were lost once they could not even complete a single unit of study, if, on the balance of probabilities, that occurred due to an unconscionable enrolment system causing consumers to be enrolled who should never have been enrolled in the first place and therefore never had a VET FEE-HELP payment made, or resultant debt incurred. If so, a relevant and significant contributor sufficient for liability would be AIPE’s unconscionable conduct.

37    Following the reasoning in both I & L Securities and Awad, an approach that is limited mechanically around a “but for” causation enquiry will be likely not to involve a full evaluative assessment of the appropriate relief. The salient question is whether AIPE’s unconscionable system of conduct or pattern of behaviour was in some way sufficiently causally related to all of the loss sustained by the Commonwealth as the applicants contend.

38    AIPE contends that inferential weight can be placed on the absence of logging in or participation, but denies that any similar inference can be drawn from the fact of failure, en masse, to complete a single unit of study by reason of failing to log in, or participate in any way. The basis for that difference in reasoning is, curiously, itself inferential. I am asked to conclude that a consumer, enrolled as a student, taking steps to participate (it seems no matter how perfunctorily or ineffectively), supports an inference that they intended to enrol in the course of study. It is not apparent to me why both the first circumstance of taking no step at all enables the necessary inference to be drawn, and the second circumstance of the presence of any such participation, no matter how perfunctory, denies the same inference being drawn. With respect, that seems to me to be an arbitrary exercise in line-drawing which I am not compelled to engage in.

39    That said, I consider that there is some merit in aspects of AIPE’s argument. A level of participation which meets the threshold of actively participating in study seems to me to make unsafe the inference that a consumer should never have been enrolled in the first place, and that such an enrolment took place because of AIPE’s unconscionable system of conduct or pattern of behaviour. Put the other way around, I consider that the necessary inference can only be safely drawn when there is something more than the bare fact of non-completion of a unit of study.

40    Relevant to that approach, the liability judgment included the following:

[636]    In mid-February 2015, after AIPE stopped using GLS [(Global Learning Support, a Philippines-based student support company)], Ms Casale [(AIPE’s Online Student Services Team Leader from 3 November 2014 to 6 March 2016)] got her team to go through all consumers recorded in CRM and to classify them according to three different criteria (using the word “consumers” in place of “students”):

[637]    This is clear evidence that no more than 10% of consumers enrolled as students with AIPE were within the range of being plainly suitable in the sense of even attempting to participate. At least 70%, and probably more out of the remaining 20%, it may readily be inferred, were not suitable to be enrolled, and/or were not genuine students because, for example, they had been told that the course was “free” and there was no need to participate, but that signing up was necessary in order to be given (and later, “loaned”) a “free” laptop. This evidence, in the context of all the evidence, gives a sound foundation for concluding that something in excess of 70% of VET FEE-HELP payments should not have been made in the relevant period.

41    I further held:

[694]    The applicants make the following 10 submissions, which I accept, about the proper use that can be made of the evidence from Ms Benton [(AIPE’s Student Services Manager from January 2014)], Ms Casale and Ms Qudsia:

First, up to the beginning of 2015 (more than half way through the Relevant Period), AIPE outsourced its student induction process to a call centre in the Philippines which sometimes did not call students before they incurred debts and often did not clearly explain to students that they would incur debts. Mr Luhr [(AIPE’s Business Operations Manager)], at least, was aware of these problems before Ms Casale was recruited in November 2014. The issues were also discussed with Mr Khanche although the exact timing is unclear.

Second, AIPE had no process in place until early 2015 for checking whether students had adequate literacy, language and numeracy skills before enrolling them. Staff were concerned that many students seemed to have inadequate skills and made these concerns known to senior managers.

Third, even after AIPE introduced a LLN test it permitted its largest agent (Acquire Learning) to run its own test, the contents and application of which were not visible to AIPE’s staff.

Fourth, there were significant deficiencies in the way AIPE’s LLN test was administered, including (in an unknown but significant number of cases) the test being completed while the recruiter was with the student.

Fifth, it was only in early 2015 that AIPE introduced a process for verifying that a student actually wanted to do the course in which he or she was enrolled before the census date arrived (the withdrawal policy designed by Ms Casale).

Sixth, AIPE’s brokers had a strong interest in all of the students they recruited remaining enrolled until they passed their census dates (at which point the student would incur a VET FEE-HELP debt and the agent would become entitled to a commission). AIPE, through its CEO Mr Khanche, was unwilling to upset agents and appears to have shared their enthusiasm for having as many students as possible pass their census dates. So, for example, the policy referred to in the previous paragraph was reversed by Mr Khanche when it led to a significant fall in the number of students passing their census dates. Mr Khanche also rejected the proposal that students should only be allowed to attempt the LLN test once.

Seventh, simple measures such as attempting to contact students prior to census date (and reversing their enrolment if they could not be contacted) led to significant falls in enrolments. That must have made it obvious to AIPE’s managers (if it was not obvious already) that agents were recruiting significant numbers of students who were not genuine or suitable.

Eighth, senior managers at AIPE were well aware that only a small proportion of online students were engaging to any extent with their courses. It was only in about October 2015, when AIPE’s registration was in jeopardy, that Mr Khanche expressed concern about the low completion rates of online students.

Ninth, when AIPE’s student numbers increased very significantly, the number of people it employed to teach students and assess their work did not. Even with thousands of students on its books AIPE employed only a handful of trainers, and they spent part of their time making orientation calls. AIPE was not equipped to deliver training to the thousands of students in respect of whom it was collecting VET FEE-HELP payments.

Tenth, AIPE received a large number of complaints and requests for refunds. Common themes of these complaints included a lack of explanation about the financial consequences of enrolment and the promise of “free” laptops by agents. Senior staff were regularly informed about the number and nature of complaints. In many cases AIPE acceded to requests for cancellation of enrolment; but it did not reform its processes in any significant way, or do anything to assert tighter control over agents. Mr Khanche worked on the principle that “we don’t need to worry about the ones who don’t complain”.

[695]    The applicants submit that these points, derived from the employee evidence, when taken together, strongly point to a business model focused on maximising revenue through government VET FEE-HELP loans, with little or no ambition to provide actual training to more than a very small proportion of consumers enrolled as students. This submission mirrors the point made earlier about the windfall profit opportunity to be had by maximising VET FEE-HELP revenue and minimising marginal costs by the delivery of services. Whether that was the intention, or an outcome that AIPE was simply willing to facilitate, does not greatly matter, given that an intention to have unconscionable conduct, whether by a system or otherwise, is not required to be established. I comfortably infer that this affected the way in which recruitment was able to, and was allowed to, take place and that this reality informed the approach taken by the most senior officers at AIPE, including in particular its CEO, Mr Khanche and its Business Operations Manager, Mr Luhr. A very high proportion of consumers enrolled never actively participating in study – at least 70% – was a profitable and desirable outcome that was effectively encouraged, or at least not meaningfully discouraged, to take place. Changes designed to reduce the prospect of this occurring were not permitted to continue if they had this effect.

42    Thus the liability judgment contains factual findings and thereby a sound path for drawing the necessary causal conclusion, namely that never actively participating in study, essentially after the census date as I have found, was a by-product of the deliberate design of the enrolment system to maximise revenue, while at the same time minimising expenditure.

43    This approach has the virtue of requiring more than the applicants contend is necessary, obviating the risk of an insufficient causal connection or an over-estimate of loss. However, it does not take it to the extreme in the other direction posited by AIPE that any degree of participation at all is reflective of a decision to enrol that was untainted in a causal way by the unconscionable system of conduct or pattern of behaviour found to have been in place. While the finding in [637] of the liability judgment is that at least 70% of VET FEE-HELP payments should not have been made in the relevant period, it is safest to limit the finding to that 70% cap. With a significant error margin in favour of AIPE, this ensures that the realistic possibility that, in relation to some consumers, not completing a unit of study was due to factors unrelated to AIPE’s unconscionable enrolment system, is excluded from the loss calculation. This also eliminates any realistic risk of it being able to be said that the Commonwealth is able to choose what it is to be repaid by such a consumer, and thereby make AIPE liable to repay, through the mechanism of re-crediting VET FEE-HELP debts. I will therefore reduce the amount of compensation to be paid in this area of dispute to 70% of the money paid in respect of students in the relevant period who did not complete a unit of study. That will produce a compensation figure in the order of $139 million, although the final precise figure should be able to be agreed upon between the parties.

Associate:

Dated:    17 December 2020

SCHEDULE

Key passages from Australian Competition and Consumer Commission v Australian Institute of Professional Education Pty Ltd (in liq) (No 3) [2019] FCA 1982

…

[6]    This proceeding is not about the quality of teaching provided by AIPE to consumers who were bona fide or genuine students and participated as such in the courses it provided, but about the enrolment of consumers whom the applicants contend did not meet that description. The label “student” can only safely be applied to those consumers who enrolled for AIPE’s online VET FEE-HELP funded courses and were suitable to be so enrolled. In these reasons, I therefore refer in a neutral way to “consumers enrolled as students”.

[7]    The applicants’ case turns on:

(1)    the way in which AIPE marketed to and enrolled consumers as students in its online VET courses, including through the use of recruiter organisations (agents) paid by substantial commissions, with employees in the field (recruiters); and

(2)    how it dealt with consumers who had been enrolled as students, especially in relation to a census date after which the debt to the Commonwealth was incurred.

…

[12]    The outcome that an overwhelming majority of consumers enrolled by AIPE as VET FEE-HELP funded students were not suitable to be enrolled in the first place and so were not genuine or bona fide students, while formally disputed, was a conclusion that is difficult to avoid. It is a conclusion that I ultimately reach on the balance of probabilities based on, inter alia, the disputed enrolment data relied upon by the applicants and the vast number, in excess of 11,000, who incurred VET FEE-HELP debts who failed to complete their courses, with many never being active students. The live question is whether this outcome was the product of unconscionable conduct, or only, for example, a consequence that flowed naturally from an education loan funding scheme that made this likely to occur no matter what, and therefore for which AIPE could not and did not bear any legal responsibility.

…

[585]    On all of the evidence it is inherently unlikely that the systematic omission of information about the meaning and importance of the census date, until after the ASQA [(Australian Skills Qualification Authority)] audit in April 2015, was accidental. To the contrary, it can readily be inferred that this was a design feature of AIPE’s enrolment system, which had the predicable effect of minimising the rate of withdrawals prior to the census date, and thus maximising VET FEE-HELP revenue. It must have been readily apparent to AIPE that providing such information, and doing so in a clear and forthright way, would necessarily increase the chance that a consumer who had been enrolled as a student would:

(1)    become overtly aware of the incurring of a VET FEE-HELP debt;

(2)    become aware that any representation that the course was “free” or that other benefits such as a laptop were “free”, was at best misleading, because a debt would be incurred;

(3)    realise that active steps would have to be taken to avoid that debt being incurred; and

(4)    take steps to withdraw from a course before the census date, or be better placed to resist becoming enrolled in the first place,

which would necessarily affect the stream of VET FEE-HELP money being provided by the government.

[Note that [585(2)] was an additional, and not an indispensable, part of the reasoning in the above paragraph.]

…

[592]    During her time at AIPE, Ms Qudsia [VET coordinator at AIPE] was aware from the CRM records that the numbers of consumers enrolled as students in AIPE’s online courses had rapidly increased during 2014 to several thousand. As enrolment numbers increased, she proposed to her managers at AIPE that they should employ more trainers. However other than the staff already referred to, AIPE did not employ any additional trainers in her team. This was a telling sign that there was no expectation that increased enrolments would result in a proportionate increase in participating students.

…

[595]    Ms Qudsia estimates that about 20-30% of the consumers enrolled as students that she spoke to in her time in the online business at AIPE communicated to her that they were unfamiliar with computers or the software on the laptops provided to them. It is worth noting that this was necessarily 20-30% of enrolled students who were participating, or at least attempting to participate in courses, and are therefore most likely reflective of the more capable consumers.

…

[598]    The unchallenged evidence of Ms Qudsia establishes that there was a steep increase in online student numbers 2014. That is, her evidence is not confined to the period that she was in that role. The six month period that she was there is most revealing. Most importantly, it supports the inference that AIPE were not expecting that a significant enough proportion of consumers who were enrolled would in fact become active students in need of student support. That is, the expectation was that, despite thousands of consumers being enrolled, only a handful of online trainer and support workers would be required. Such an expectation is consistent with a system in which a high proportion of consumers would be enrolled, would incur a VET FEE-HELP debt, but would never in fact be active students for whom support would be required.

…

[631]    About three to four weeks after the new policy was implemented, Mr Khanche [AIPE’s CEO] came to her desk and they had a conversation to the following effect:

Mr Khanche:    Why do we have so many withdrawals?

Ms Casale:    We need to make sure the students are suitable and want to do the course so the new policy is to contact them and make sure they want to do the course and are engaging before census.

Mr Khanche:    Communicate to your team that everyone passes the census date. We can’t afford this many withdrawals.

Ms Casale:    But this will increase the number of appeals.

Mr Khanche:    I don’t care.

[632]    As with conversations between Ms Benton [(AIPE’s Student Services Manager from January 2014)] and Mr Khanche, this conversation makes it abundantly clear that the profitability of AIPE’s business model, as reflected in its enrolment system, significantly depended on consumers who were not suitable, or not genuine, being enrolled as students and remaining enrolled until the census date so that a VET FEE-HELP payment would be made and the corresponding debt incurred. This conclusion is reinforced by further evidence below about Mr Khanche’s attitude towards the adverse effect on consumers of AIPE’s enrolment practices.

…

[635]    Thus Mr Khanche was clearly not concerned with the underling integrity of AIPE’s enrolment system, and how it was implemented. He was expressly on notice that consumers were not even starting the courses in which they had been enrolled as students. He regarded that the risk of ASQA investigating non-compliance with its standards [was] a risk that he was willing to take. There had to be a very powerful countervailing incentive for this to be a risk that he was willing to take. The only incentive that is apparent on all of the evidence is the VET FEE-HELP revenue from loans advanced to such students. The raw profit from such revenue not accompanied by the cost of providing services to non-participating consumers enrolled as students had to have been obvious to Mr Khanche and thus to AIPE.

[636]    In mid-February 2015, after AIPE stopped using GLS [(Global Learning Support, a Philippines-based student support company)], Ms Casale [(AIPE’s Online Student Services Team Leader from 3 November 2014 to 6 March 2016)] got her team to go through all consumers recorded in CRM and to classify them according to three different criteria (using the word “consumers” in place of “students”):

[637]    This is clear evidence that no more than 10% of consumers enrolled as students with AIPE were within the range of being plainly suitable in the sense of even attempting to participate. At least 70%, and probably more out of the remaining 20%, it may readily be inferred, were not suitable to be enrolled, and/or were not genuine students because, for example, they had been told that the course was “free” and there was no need to participate, but that signing up was necessary in order to be given (and later, “loaned”) a “free” laptop. This evidence, in the context of all the evidence, gives a sound foundation for concluding that something in excess of 70% of VET FEE-HELP payments should not have been made in the relevant period.

…

[644]    As discussed above, VET FEE-HELP supported course providers should not have been enrolling consumers with language, literacy or numeracy skills that were below that required to complete the course, even if this was an obvious implicit requirement, rather than an express requirement, of the scheme. But to adhere to this was necessarily to limit the pool of consumers enrolled in the first place. As this and Ms Benton’s evidence demonstrates, not only did AIPE’s enrolment system not have any language, literacy and numeracy barrier to enrolment until early 2015, when such a barrier was introduced it was done in a way that was perfunctory and illusory, and only implemented in the shadow of closer investigation of enrolment problems by ASQA, culminating in audits. AIPE, via Mr Luhr [(AIPE’s Business Operations Manager)] and Mr Khanche in particular, was not concerned about consumers who had a real risk of not being suitable to be enrolled as students.

…

[653]    Once again, Ms Casale’s attempts to do something about a clear and obvious problem with consumers being enrolled as students with AIPE despite not having the necessary language, literacy and numeracy skills were disregarded. This again is a telling indicator that revenue, and profit, were the dominant considerations. Enrolling unsuitable students was both an accepted feature of the enrolment system, and was not regarded as a problem by those in charge at AIPE.

…

[665]    I attach little or no weight to such self-serving policy statements in the absence of evidence that this was what actually occurred, which was sparse and indicated that this only seemed to take place in extreme cases. There is ample evidence to indicate that AIPE was replete with policy statements which were not matched by evidence to demonstrate that this went beyond being part of a paper system that did not reflect the day-to-day reality of maximising enrolments and retaining them until the census date was the priority, ahead of consumer suitability.

…

[669]    Ms Casale remembers that in May 2015, her team had a meeting which Ms Abraham [Ms Benton’s replacement as Student Services Manager from June 2015 to January 2016)] and Mr Khanche attended as well. She was working from home at the time and had dialled in. She remembers Mr Khanche said to those at the meeting words to the effect:

This is a business. You are not being paid to make moral judgments. If a student wants to take on three debts for three courses then that’s the student’s choice.

[670]    This was a clear direction by the CEO to senior AIPE staff as to how they were to manage the enrolment process. The situation that Mr Khanche referred to was revealing evidence as to how far he, and thus AIPE, was prepared to permit a situation to develop as part of the enrolment process, especially given that this indicated the latitude he considered appropriate to be given to agents and their recruiters acting on behalf of AIPE. Indeed, as Ms Benton’s evidence (at [558]) shows, this was not even a hypothetical scenario. Agents clearly had a very wide authority to exercise in pursuing the maximum number of enrolments possible, treating the evidence in favour of AIPE as establishing that the line was drawn only at outright lies and overtly false information.

[671]    The view expressed by Mr Khanche that concern over a consumer enrolling in three courses and taking on three VET FEE-HELP debts was, pejoratively, an inappropriate “moral judgment”, is a very strong indication of AIPE’s flagrant disregard for the vulnerabilities of the disadvantaged target consumers. The scenario strongly points to a consumer having no idea what they are getting themselves into, as was the case with the man with physical and mental disabilities whom Ms Benton recalled coming to AIPE’s office with a support worker (at [558]). It would not simply reflect the consumer’s “choice”, as Mr Khanche disingenuously suggests. If this scenario was not a source of concern, then it is fair to infer that the unconscionable enrolment of inappropriate consumers in general was not going to be a source of concern for AIPE. This and many other facts in this case contrast starkly with the circumstances in Kobelt, where the final resolution in favour of the respondent in that case was quite finely balanced. Like in Empower (at [750]-[751]), it shows a “callous indifference” to the consumer protection considerations that are necessary when the system is directed to people who are vulnerable to being misled or deceived.

[672]    In the context of evidence of similar comments made by Mr Khanche, it is clear that he, and thus AIPE, saw nothing wrong with the provision of information that was incomplete or even misleading, deceptive or contextually false, provided it was not overtly false.

…

[676]    Ms Casale knew that it was difficult to get more “students” into the “green” category of students actively progressing in a course because she was unable to contact many consumers who had enrolled in courses. If she was able to make contact, on most occasions they were not interested in doing the course.

…

[682]    The unchallenged aspects of the evidence of Ms Casale, along with that of Ms Benton and Ms Qudsia, gives a strong and sound foundation for concluding that the reality of AIPE’s enrolment system bore little resemblance to the benign situation painted by the policy documents alone. While I have perused all of the policy documents in evidence, and considered more closely those to which my attention was directed, the best that could be said of most of them is that they were aspirational rather than any accurate reflection of the enrolment system they purported to convey. At best they reflect what Ms Benton, Ms Casale and Ms Qudsia endeavoured to achieve, being a weak breeze against the strong headwinds of maximising revenue and profit.

…

[686]    The main conclusion to be drawn from this evidence is that, despite the efforts of the employee witnesses and some of their staff to improve AIPE enrolment practices to avoid unsuitable consumers being enrolled as students, the highest level of AIPE was concerned with high levels of enrolment rather than even modest levels of successful study. The overall conclusion is that Mr Khanche and thus AIPE were fully aware that most of the consumers who were enrolled as online students were not partaking of the course in which they were enrolled, and that this was largely the situation from the moment that enrolment took place.

[687]    The proper conclusion to draw on the evidence is that those at the very top of the AIPE hierarchy were only willing to have the appearance of taking steps to confine the imposition of VET FEE-HELP debts to genuine students, but did not in fact want or allow that to happen if it had any significant impact on enrolments. Mr Khanche took active steps to reverse changes that reduced enrolments of that kind, having been told expressly that this would have the effect of causing unsuitable consumers to be enrolled as students. The overall effect and quality of the evidence satisfies me at, and indeed beyond, the balance of probabilities, that this was not merely an unfortunate outcome of a poor quality enrolment system, or even mere inadvertence, but an outcome that [was] known and was advantageous to AIPE.

[688]    It was an active part of AIPE’s enrolment system that a very high proportion of consumers who were enrolled would pass the census date and incur a VET FEE-HELP debt in circumstances where it was not just predictable that they would never need to be given any student support, but also an outcome that was highly profitable. This outcome was an accepted and even anticipated part of AIPE’s business model.

[689]    It is important to note that there was no evidence to indicate that a significant number of consumers enrolled as students ever commenced to participate in the courses in which they had enrolled: Ms Casale’s evidence about what an interrogation of AIPE’s student records revealed about participation rates points quite to the contrary. The survey evidence considered below helps to verify the conclusions otherwise reached, rather than being an independent or indispensable source or foundation for such conclusions.

[690]    AIPE’s acceptance of the enrolment of unsuitable students is also reflected in staffing practices that are difficult to reconcile with any other understanding. There was no substantial increase in staff numbers, and certainly nothing like the three- to five-fold increase in enrolment numbers, and the overall numbers remained quite paltry. The staffing numbers evidence enables it to be safely concluded that it is more likely than not that AIPE understood that the great bulk of consumers enrolled as students who were withdrawn after census were unsuitable to be enrolled, or were not genuine students (such as by reason of being told enrolment was a means of getting a “free” laptop). Most of those consumers should never have been the subject of a VET FEE-HELP payment or debt.

[691]    That business model, given its predictable effect on vulnerable prospective consumers, was unconscionable. It was incumbent on AIPE to conduct any recruitment from this disadvantaged pool of consumers in a way that did not take advantage of their vulnerability. That was the legal and practical framework in which AIPE was operating.

[692]    While aspects of disadvantage pleaded by the applicants may be seen to turn on a specific species of vulnerability, as the case unfolded the more generalised aspect of that vulnerability came to the fore. That is, the vulnerability of a general class of consumer with an inherently higher risk of being unsuitable to be enrolled as a student, rather than having features specific to one consumer but absent from another. The outcomes and events pointing to that general vulnerability, and thus particular and special disadvantage, was not left to speculation, or generalisation, but rather was repeatedly raised by each of the three employee witnesses, sought to be rectified and thwarted or otherwise undermined by the most senior decision-makers at AIPE, most notably, Mr Khanche and Mr Luhr.

[693]    As already made clear, AIPE’s approach of attacking the reliance by the applicants on only select agents to examine how AIPE’s enrolment system worked was misconceived. The employee evidence shows that there was an inherent character of the system that applied regardless of the agent or recruiter involved. That is, … AIPE was on notice that a high proportion of its consumers enrolled as students were not genuine or [bona fide] and that this was often the result of nefarious practices by agents, including through the use of false and misleading information. Yet any check or balance introduced to rectify this situation was resisted or quickly erased. It was powerful evidence of what AIPE’s enrolment system knowingly allowed to take place.

[694]    The applicants make the following 10 submissions, which I accept, about the proper use that can be made of the evidence from Ms Benton, Ms Casale and Ms Qudsia:

First, up to the beginning of 2015 (more than half way through the Relevant Period), AIPE outsourced its student induction process to a call centre in the Philippines which sometimes did not call students before they incurred debts and often did not clearly explain to students that they would incur debts. Mr Luhr, at least, was aware of these problems before Ms Casale was recruited in November 2014. The issues were also discussed with Mr Khanche although the exact timing is unclear.

Second, AIPE had no process in place until early 2015 for checking whether students had adequate literacy, language and numeracy skills before enrolling them. Staff were concerned that many students seemed to have inadequate skills and made these concerns known to senior managers.

Third, even after AIPE introduced a LLN [(language, literacy and numeracy)] test it permitted its largest agent (Acquire Learning) to run its own test, the contents and application of which were not visible to AIPE’s staff.

Fourth, there were significant deficiencies in the way AIPE’s LLN test was administered, including (in an unknown but significant number of cases) the test being completed while the recruiter was with the student.

Fifth, it was only in early 2015 that AIPE introduced a process for verifying that a student actually wanted to do the course in which he or she was enrolled before the census date arrived (the withdrawal policy designed by Ms Casale).

Sixth, AIPE’s brokers had a strong interest in all of the students they recruited remaining enrolled until they passed their census dates (at which point the student would incur a VET FEE-HELP debt and the agent would become entitled to a commission). AIPE, through its CEO Mr Khanche, was unwilling to upset agents and appears to have shared their enthusiasm for having as many students as possible pass their census dates. So, for example, the policy referred to in the previous paragraph was reversed by Mr Khanche when it led to a significant fall in the number of students passing their census dates. Mr Khanche also rejected the proposal that students should only be allowed to attempt the LLN test once.

Seventh, simple measures such as attempting to contact students prior to census date (and reversing their enrolment if they could not be contacted) led to significant falls in enrolments. That must have made it obvious to AIPE’s managers (if it was not obvious already) that agents were recruiting significant numbers of students who were not genuine or suitable.

Eighth, senior managers at AIPE were well aware that only a small proportion of online students were engaging to any extent with their courses. It was only in about October 2015, when AIPE’s registration was in jeopardy, that Mr Khanche expressed concern about the low completion rates of online students.

Ninth, when AIPE’s student numbers increased very significantly, the number of people it employed to teach students and assess their work did not. Even with thousands of students on its books AIPE employed only a handful of trainers, and they spent part of their time making orientation calls. AIPE was not equipped to deliver training to the thousands of students in respect of whom it was collecting VET FEE-HELP payments.

Tenth, AIPE received a large number of complaints and requests for refunds. Common themes of these complaints included a lack of explanation about the financial consequences of enrolment and the promise of “free” laptops by agents. Senior staff were regularly informed about the number and nature of complaints. In many cases AIPE acceded to requests for cancellation of enrolment; but it did not reform its processes in any significant way, or do anything to assert tighter control over agents. Mr Khanche worked on the principle that “we don’t need to worry about the ones who don’t complain”.

[695]    The applicants submit that these points, derived from the employee evidence, when taken together, strongly point to a business model focused on maximising revenue through government VET FEE-HELP loans, with little or no ambition to provide actual training to more than a very small proportion of consumers enrolled as students. This submission mirrors the point made earlier about the windfall profit opportunity to be had by maximising VET FEE-HELP revenue and minimising marginal costs by the delivery of services. Whether that was the intention, or an outcome that AIPE was simply willing to facilitate, does not greatly matter, given that an intention to have unconscionable conduct, whether by a system or otherwise, is not required to be established. I comfortably infer that this affected the way in which recruitment was able to, and was allowed to, take place and that this reality informed the approach taken by the most senior officers at AIPE, including in particular its CEO, Mr Khanche and its Business Operations Manager, Mr Luhr. A very high proportion of consumers enrolled never actively participating in study – at least 70% – was a profitable and desirable outcome that was effectively encouraged, or at least not meaningfully discouraged, to take place. Changes designed to reduce the prospect of this occurring were not permitted to continue if they had this effect.

…

[706]    The evidence considered below establishes to my satisfaction that the dominant reason for AIPE commencing training was a reaction to complaints and eventually increased scrutiny of VET providers, rather than any real attempt to diminish enrolment revenue and profits from unsuitable consumers. If, contrary to that conclusion, it was meant to be a genuine response to the sorts of issues that must have been well-known by then, it was at best half-hearted. It was a mindset that was inherently unlikely to address squarely the conflict of interest on the part of both agents/recruiters and by AIPE between:

(1)     maximising revenue by aggressive marketing strategies, including

approaching consumers unsolicited, and hard sell-approaches and incentives coupled with inaccurate, incomplete and even contextually false, deceptive or misleading information (putting to one side overtly false information); and

(2)     ensuring that only genuine and suitable consumers with some reasonable

possibility of being able to study AIPE online courses successfully were enrolled.

…

[760]    The evidence of Ms Benton and Ms Casale, as summarised and analysed, makes it clear that Mr Khanche knew full well about the conduct of AIPE’s agents and their recruiters, both by being told about the problems directly, and because of complaints that were drawn to his attention. This made it abundantly clear to him that AIPE was enrolling a high proportion of consumers who were simply incapable of undertaking the courses. Thus AIPE, especially through its CEO, knew full well that its enrolment system did not screen students for suitability, or ensure that they were aware they had been enrolled in the first place or had a VET FEE-HELP debt. Rather, it is clear, via Mr Khanche’s reaction, that the focus was to keep consumers enrolled past their census date, and that this reality was reflected in very low participation rates from the outset.

…

[762]    McGrathNicol randomly selected the student records for 100 consumers enrolled as students out of 6,057 reported by AIPE for 2015, and tested the enrolment documentation held for those students. This related to 6,066 courses due to a small number of multiple enrolments. With that testing, consideration of the student records for all 6,057 consumers was able to be examined to produce the following “engagement testing results” at [1.4.2] of the report which “identified on how many days a student logged into the LMS [(learning management system)], and how many different parts of the online course a student had interacted with”, producing the following results:

McGrathNicol found that for a significant number of the 6,066 course enrolments that AIPE reported for 2015, there was a very low level of engagement with the LMS. For example, our testing found that [15%] of students had not logged into the online LMS, and [76%] of students had logged in on four different days or less. Only [10%] of students had logged in on more than 10 different days. This contrasts with students who completed courses, who we found had an average of 58 days active, indicating those students who had logged in on less than 10 days may not have significantly advanced toward course completion.

[763]    McGrathNicol then analysed and sought to match student activity logs with HEIMS [(Higher Education Information Management System)] data so as to calculate the dollar value of units engaged with by consumers enrolled as students in 2015 alone, with the report concluding on this topic:

Based on the methodology as outlined in Section 6.5.5, from a total of $114,123,050 VET FEE-HELP loans in HEIMS, according to the Moodle activity logs there were:

•    course enrolments with no evidence of activity or engagement in a unit in Moodle with a value of $49.2 million;

•    units in HEIMS with no evidence of engagement in Moodle with a value of $33.9 million; and

•    a total value of loans with no evidence of engagement in Moodle of $83.1 million.

[764]    I conclude that this was overwhelmingly, if not entirely, the product of AIPE’s enrolment system that enrolled unsuitable consumers, with most of this amount being a windfall profit after the deduction of actual operating expenses. This supports the conclusion independently reached that a very high proportion of VET FEE-HELP loans in 2015 (going a month beyond the relevant period) are attributable to AIPE’s unconscionable enrolment system. I can see no reason for inferring that it would have been any better in the preceding two years of the relevant period, especially considering AIPE’s submissions that its system improved towards the end of the relevant period. Even with all the evidence that AIPE says was missing, no precise figure attributable to this conduct would necessarily have been possible. Mass consumer cases of this kind cannot descend to that degree of granular detail while remaining viable as proceedings.

…

[772] Before concluding, it is necessary, for completeness, to address some of the arguments that AIPE made regarding the case against it (some of which have been mentioned above). AIPE argued that it was substantially the same as the case in Unique, referring to certain gaps identified by the Full Court in that case (at [254] to [256]) and relying upon them to identify what were said to be similar gaps, asserted to be fatal in the case brought by the applicants. This approach did not engage with much of the substance or detail of the applicants’ case, ignoring the evidence from its former employees and ignoring the enrolment data. AIPE made the following submissions (reproduced almost in full in italics), pointing to the asserted gaps and other deficiencies in the evidence adduced and relied upon by the applicants, with my summary comments as to each:

    …

(6)     There being no evidence as to the vulnerability of the consumers enrolled as students, with it being asserted that there was instead a “silent assumption” by the applicants that consumers from disadvantaged background were vulnerable, relying on what was said in Unique at [235] as to not making stereotype assumptions of that kind. This overlooks what the Full Court said in Unique at [135] to [136]. The applicants’ [unconscionability] case relied upon general attributes of the disadvantaged consumers who were targeted to be enrolled, in the sense of being less able to detect defects in the information they were given, and to resist becoming enrolled in a course despite being unsuitable. Any member of society who was told, for example, that a course was “free” when it was not free of a debt burden, or that a laptop was for use in such a course (or otherwise) was “free” when in fact it too was provided in the context of and upon the condition that a debt burden was assumed, could well be encouraged to enrol in such a course. The burden of the applicants’ case is that persons from disadvantaged backgrounds, who were targeted in accordance with the liberalised VET FEE-HELP scheme, were more likely to be misled by such conduct and therefore to become enrolled, even if they were also not suitable to be enrolled in the first place. This was not an improper or silent assumption, let alone one relying upon stereotypes, but rather a normal understanding of differential capacity, and therefore vulnerability, to be misled by conduct of the kind that the applicants rely upon.

…