AustLII Home | Databases | WorldLII | Search | Feedback

Precedent (Australian Lawyers Alliance)

You are here:  AustLII >> Databases >> Precedent (Australian Lawyers Alliance) >> 2020 >> [2020] PrecedentAULA 14

Database Search | Name Search | Recent Articles | Noteup | LawCite | Author Info | Download | Help

Bant, Elise --- "A road map to decision causation: In misleading conduct and failure to disclose cases" [2020] PrecedentAULA 14; (2020) 157 Precedent 4


A ROAD MAP TO DECISION CAUSATION

IN MISLEADING CONDUCT AND FAILURE TO DISCLOSE CASES

By Professor Elise Bant

Statutory concepts of causation have been front and centre in a series of recent Federal Court and Supreme Court decisions addressing cases of misleading conduct.[1] The statutes all provide compensatory remedies for loss or damage suffered ‘because of’, ‘by’ or ‘as a result of’ misleading conduct.[2] A particularly difficult question is how a failure to disclose some fact or matter can cause a person to suffer loss or damage. How can a person rely on something that was not said?[3]

This article explains some simple truths about decision causation, the idea of ‘reliance’, and omissions. With these essential concepts in place, it is possible to identify and navigate the very difficult questions facing practitioners advising clients and the courts deciding cases.

SOME BASIC CAUSAL CONCEPTS

Factual versus legal causation

The concept of ‘commonsense causation’[4] is now understood as both unhelpfully vague[5] and collapsing two quite different enquiries.[6] The first is the question of ‘factual causation’: whether the putative cause bears some explanatory relation to the existence of some outcome that in fact occurred. This is what Professor Jane Stapleton calls a question of ‘historical involvement’.[7] Did the bullet fired from the defendant’s gun play some part in bringing about the plaintiff’s gunshot injury? Did the fire lit by the defendant play some part in bringing about the destruction of the barn? Did the defendant’s misleading conduct play some part in bringing about the plaintiff’s decision to purchase the defendant’s wares?

The second is a normative question of legal responsibility (sometimes called ‘legal causation’): should the defendant be held liable for the outcome that in fact occurred? This is a question of the defendant’s scope of liability. Private law practitioners are most familiar with this second enquiry through tortious concepts such as remoteness, mitigation and apportionment on the basis of fault. The idea is that although a person might have been factually implicated in the historical genesis of some harm (so factual causation is satisfied), the extent of their responsibility for the harm that they have caused is a separate question. Was the defendant legally responsible for the plaintiff’s injury when he did factually cause the plaintiff injury by shooting the gun, but the incident occurred when the plaintiff carelessly handed the gun to the defendant?

This two-step enquiry is also found in cases that are not based on defendant wrongdoing. Suppose a plaintiff seeks restitution of the value of a mistaken payment.[8] The mistake might have been ‘spontaneous’ on the part of the plaintiff (she simply misunderstood some fact or matter), or it might have been induced by a third party or by the defendant. Whatever the case, the first question is whether, as a matter of fact, the plaintiff made the payment ‘by’ (as a result of, because of) a mistake, and the second is the defendant’s scope of liability to make restitution of the value of the payment. Here, we find less familiar but still operative scope of liability-style considerations: for example, whether the plaintiff took an undue risk in making the payment.[9]

In this article, ‘causation’ means factual causation: the ‘historical involvement’ of some alleged cause in the result that in fact occurred. The separate normative question of the defendant’s responsibility for that result is called the ‘scope of liability’ enquiry.

Decision causation and reliance

We can further loosely divide the cases into, on the one hand, cases involving sequences of ‘involuntary’ events and, on the other, cases of ‘decision causation’. Most of us are familiar with involuntary sequence causation from tort and criminal law: the impact of a bullet fired from a gun, a fire in a barn, and so on. Decision causation by contrast refers to the impact of some fact or matter on the process of decision-making. Where a plaintiff says that she has ‘relied’ on the defendant’s misleading conduct, for example, she is generally making a claim of decision causation.

However, reliance is a tricky label because it may imply some additional, normative aspect – that the plaintiff trusted the defendant, for example, or that the defendant was under some moral or legal duty to the plaintiff. For this reason, decision causation more precisely describes the factual causal enquiry.

Also, plaintiff reliance is only one way of indicating the presence of causation or contribution in cases involving misleading conduct.[10] For example, a defendant might tell the plaintiff’s customers that the plaintiff’s shop has closed down in order to lure the customers to his own business.[11] Here it is the customers, not the plaintiff, who have relied on the misleading conduct. However, absent any other scope of liability reasons, it remains clear that the defendant’s misleading conduct has caused or contributed to the plaintiff’s lost profits.

Similarly, a plaintiff may purchase goods or shares at an inflated price because the market has been misled. Here, causation does not involve reliance by the plaintiff, except to the extent that she may (actively or tacitly) assume that the market is efficient and thus that the price reflects the true value of the asset. Rather, it is the third-party market, comprising hundreds or thousands of individuals, which has ‘relied’ on the misleading conduct. For example, misleading statements contained within a prospectus or other disclosure document issued by the defendant may have caused the market price of shares to be artificially inflated. The plaintiff suffers loss when, having purchased shares at that inflated price, the true position is revealed, the market adjusts and the share prices correspondingly fall.[12] As Beach J has recently held in PT Patrol,[13] there is nothing in the statutory language of causation to suggest that such indirect or market-based causation is excluded.[14]

‘But for’ causation and the ‘a factor’ test of contribution

The last basic point is to distinguish two tests of causation or contribution. The familiar ‘but for’ test uses a counterfactual world where the alleged cause (a bullet fired from the defendant’s gun, a fire lit by the defendant, the defendant’s misleading conduct) did not exist. To construct this counterfactual world properly, we need a fairly clear idea of all the main, relevant factors that together produce the result that occurred. For example, where a barn is burned down, relevant factors would include things like temperature, humidity, wind, and the presence of fuel and other sources of fire on the day, together with the defendant’s act of lighting a fire. The aim of this enquiry is to determine whether it is more likely than not that the alleged cause (the defendant’s act of lighting the fire) was necessary for the particular outcome that is the subject of the claim (the destruction of the barn). If, in the hypothetical world, the absence of the fact or matter would have led to a different outcome (the barn could be unscathed, or less damaged), then causation is established. Importantly, it is only possible and legitimate to use this test where the abstracted, metaphysical world accurately reflects the conditions found in our real world. This is often not the case where decision causation is involved.

By contrast, the ‘a factor’ test asks whether, as a matter of real-life history, a certain fact or matter played some role (was ‘a’ factor) in the process that led to the result that in fact occurred. On this approach, the alleged cause does not need to be a ‘but for’ or necessary cause of the result that in fact occurred; it is enough if it formed one of the historical components that, together with all other factors actually present on the day, contributed to the result that in fact occurred. This is not a counterfactual, theoretical or abstract(ed) enquiry, but an examination into the historical process by which the relevant result that is the subject of proceedings came about.

THE BASIC CONCEPTS APPLIED AND EXPLAINED

We are now ready to consider how and why these basic concepts apply in circumstances of misleading conduct and failure to disclose.

Misleading conduct

Suppose a plaintiff purchases a box of cereal that is falsely labelled ‘Made in Australia’. Applying the ‘but for’ test to determine the impact of the label on the consumer’s decision-making (in lay terms, did she rely on it?) is deeply problematic in this simplest of cases. This is because of the nature of the ‘but for’ test and of decision-making.

We have seen that the ‘but for’ test requires that all relevant or (at least) main reasons for the plaintiff’s decision be identified in order to establish the hypothetical scenario from which the putative cause (the false label) will be excluded. But, decision-making here is something of a ‘black box’. The plaintiff will be aware of some factors that influenced her decision (for example, price, taste and ‘Made in Australia’). However, there will be a range of other conscious and subconscious factors that may or may not have had a significant impact on that process of decision-making and neither she nor the court will be in a position to weigh their impact. As Professor Birks observed, willpower has no voltage.[15] Thus significant factors such as the (conscious and subconscious) impact of marketing campaigns, use of brand ambassadors, packaging, as well as other reasons such as the product’s position on the shelf, its perceived nutritional value, how busy the shopping centre is, the amount of time the plaintiff has to make a decision, and other factors (including potentially random factors such as what the plaintiff had for breakfast that morning), may all operate in various ways to produce the result that in fact occurred. This means that it will not be possible to construct an accurate hypothetical test scenario for application of the ‘but for’ test in which most causally relevant factors are retained other than the alleged cause (here, ‘Made in Australia’). Any constructed hypothetical will be at best incomplete and at worst wholly inaccurate. In that context, asking whether the plaintiff would have bought the product ‘but for’ the misleading statement is to invite unfounded speculation and to assess the causal potency of the misrepresentation by reference to a factual scenario that never occurred.

Moreover, even if we could somehow construct the requisite test scenario, there is a further problem. In our example, suppose there were a range of reasons (for example, price, taste and the ‘Made in Australia’ label) which individually provided the plaintiff with an independent ‘but for’ reason for her decision (but for ‘Made in Australia’, she would have purchased the cereal anyway because of price or taste). This is a case of ‘over-determined’ causation: there is more than one ‘but for’ cause. Conversely, suppose there were a wide range of lesser reasons (branding, what the plaintiff had for breakfast, the fact that she was in a hurry on the morning in question, or other subconscious matters and biases) that, taken together with ‘Made in Australia’, would have sufficed to produce her decision to purchase, but none of which (including ‘Made in Australia’) was a ‘but for’ cause. This is often called ‘under-determined’ causation: where there are no ‘but for’ causes.

In both the under- and over-determined scenarios, the ‘but for’ test yields the patently ridiculous answer that ‘Made in Australia’ was not causally relevant, and nor were any of the putative factors (price, taste or the lesser factors). But clearly there must have been some causal reasons for the plaintiff’s decision to purchase. Even more problematically, the test invites us to assign wholly arbitrary weights to reasons for a decision in an enquiry that looks objective and scientific, but is entirely unsubstantiated by any scientific theory of decision-making.[16] This is a very poor basis on which to deny or impose liability for misconduct.

By contrast, the ‘a factor’ test places the plaintiff’s decision in its factual, historical context. It highlights the salient factors that appear as a matter of history to have guided the consumer’s exercise of choice, without demanding that the choice was wholly rational, determinate or replicable. Adopting this approach, it will be relatively easy for a court to assess the plaintiff’s claim that the fact (as she supposed it) that the cereal was Australian-made was one of the reasons that she chose that product. This not only is something that we may expect her to be able to give accurate testimony about (where that is permissible), but we can also use sensible rules of thumb that are commonly employed by courts. For example, where a statement is inherently relevant or material to a plaintiff’s decision and is used by the defendant with the intention of inducing reliance, it is likely to be ‘a factor’ in the plaintiff’s decision.[17]

Omissions

Turning to omissions in cases of misleading conduct, suppose in our simple example that the cereal packet omits the information that, although processed and packaged in Australia, the cereal grains were grown overseas. Suppose also that the cereal producer has secretly been engaging in exploitative trade practices by systematically underpaying its employees. Has our consumer’s decision to purchase been relevantly caused by these omissions? Are they causally the same kind of omission and if not, why not?

The better answer is that they are not the same and should be treated differently; on the face of it the omission regarding place of production was causal, while the omission of ethical malpractice may have been wholly irrelevant to the plaintiff’s decision. To understand why, we need to return to the nature of the ‘but for’ and ‘a factor’ tests of causation.

We know that the ‘but for’ test involves the creation of a metaphysical or abstracted world where the alleged cause is excluded to test whether there would have been a different result. In non-disclosure cases, this approach requires that the hypothetical world includes the omitted information; the cereal package informed the consumer that its grains were grown overseas and that the producer engaged in exploitative trade practices. Here the ‘but for’ test is likely to yield the answer that the omission was causally relevant in both cases – had the consumer been aware of the information, she would have acted differently.

However, courts are increasingly recognising that the ‘but for’ test is over-inclusive in cases of omissions. It is often the case that, had plaintiffs been aware of some fact or matter of which they were ignorant, they might have acted differently.[18] This is so even where, as a matter of historical fact, the particular plaintiff never turned her mind to the matters informed by the ‘missing’ fact or matter. Suppose we know that the consumer in our example never considered the ‘ethical treatment of employees’ as a reason for purchasing this cereal over others. However, now that she is aware of the misconduct, she firmly and truthfully believes that ‘if she had her time again’ she would never purchase the product. Here, the ‘but for’ test is falsely evaluating a scenario that never actually occurred and which contradicts the reality of the consumer’s actual process of decision-making.

By contrast, on the ‘a factor’ analysis, the omission of information that the cereal was grown overseas, albeit processed and packaged in Australia, would be causally potent in our consumer example, because it informs a positive factor (‘Made in Australia’) on which her decision to purchase actually proceeded. By contrast, the hidden fact that the company has systematically underpaid its workers would not, without more, count as an ‘a factor’ cause, unless the ethical standards of the company was a positive factor in the consumer’s process of decision-making. The ‘a factor’ test correctly distinguishes the causal potency of the two omissions.

On this analysis, both practitioners and courts faced with cases of misleading conduct involving omissions (such as in cases of non-disclosure) arguably should abandon attempts to gauge what a plaintiff would have done had she known of the omitted material. That question is over-inclusive, highly speculative and distracts attention from the key question: what was the role (if any) played by the omission in informing the plaintiff’s positive reasons for the decision? On this approach, an omission will only form part of the process of decision causation where it informed a positive assumption or belief (was ‘a factor’) on which a plaintiff acted.[19]

Knowledge that ‘breaks the chain of causation’?

Can a plaintiff have relied on an omission if she knew, or suspected, the true state of affairs? Here, questions of factual causation are often accompanied by difficult normative questions that go to the defendant’s scope of liability. For example, the customer in our example might have purchased the cereal doubting that the grains were grown in Australia but confident that they must have been processed and packaged in this country. The statement ‘Made in Australia’ is still ‘a factor’ in her decision to purchase, but there is a separate question of whether her decision to proceed in the face of her doubt means that she is an undue risk-taker.

The question then becomes: to what extent should the defendant bear full responsibility in light of the customer’s fault? This normative question must be resolved by considering the scope and protective purpose of the law’s prohibition. The answer will depend on the claim and the circumstances of the case. For example, where the claim is one of deceit, the policy of the law tends to favour the customer’s complete protection, unless the plaintiff’s fault is so extreme that it ‘breaks the chain of causation’.[20] Where the claim is one based on a consumer protection statute or the defendant’s negligence, the consumer’s own failure to take care may feed into the liability enquiry, for example, through an apportionment (or contributory negligence) rule or perhaps through a requirement of reasonable reliance incorporated as an element of the primary claim. Importantly, whether labelled as a ‘break in the chain of causation’ or not, this is not a question of factual causation but involves consideration of quite different factors going to the defendant’s responsibility for the loss.

Moreover, not all cases of non-disclosure will involve direct reliance or decision causation on the part the plaintiff, yet the plaintiff’s knowledge may still be relevant to the defendant’s scope of liability. Returning to our earlier market-based causation example, suppose a plaintiff purchases shares at an undervalue because of the company’s failure to disclose some salient fact or matter to the market, with the result that the share price has been artificially inflated. When the fact or matter becomes public, the market corrects and the plaintiff suffers loss or damage. Should the success of a claim based on misleading conduct (assuming that quality is established) or on non-disclosure in breach of statute depend on whether the plaintiff, personally, knew of the true circumstances?

Here, again, statements that the plaintiff’s knowledge or fault may operate to ‘break the chain of causation’ must be treated with caution. In the scenario in question, there is no break in the historical chain of causation, as causation is well and truly established. Rather, the question is whether there are other reasons why the plaintiff should not succeed, or succeed in full.[21] This is a normative question that is dependent on statutory interpretation and is likely to raise issues of statutory or public policy.[22] As a matter of interpretation, it may be arguable (depending on the terms of the statute) that the fact that the plaintiff was not misled is neither here nor there; regulators regularly seek and obtain remedies where the defendant’s conduct is ‘likely to mislead’, even though no one has (yet) been misled. Supporting this expansive view of liability is that, so far as private rights of redress are concerned, many statutes do not explicitly adopt remoteness or other limiting rules but simply adopt language of factual causation (‘by’, ‘because of’, ‘as a result of’).[23] On the other hand, it is increasingly recognised that the broad causal language adopted in most statutes likely captures the second, normative enquiry into ‘responsibility’.[24] If this is correct, the plaintiff’s knowledge may be argued to be a significant barring factor.

Ultimately, where the language of the statute remains ambiguous, the protective policy of the statute may be determinative and lead to subtly different outcomes. For example, where the aim of a statute is to ensure the integrity of the market and only derivatively those who rely on it, courts may distinguish between different types of knowledge. A person who knows the share price was inflated but proceeded to buy nonetheless may fall outside the scope of protection offered by the statute, even though as a matter of historical fact the subversion of the market contributed to her loss. By contrast, the statute may continue to protect a person who was aware of the omitted information but did not realise its impact on market price.

While these are complex questions, the important point for our purposes is to observe that these are ultimately normative enquiries into the defendant’s scope of liability, not issues of factual causation. Seeing this can help us to understand the kinds of questions that the court must ask and answer – and those which have already been determined.

CONCLUSION

The purpose of this article has been to clarify some basic truths and distinctions about the complex issues of decision causation, misleading conduct and omissions. It does not seek to provide answers to the complex questions that face practitioners and courts seeking to navigate these fields. However, the hope is that it provides a road map which identifies the correct paths of reasoning required to lead to a principled result.

Professor Elise Bant is Professor of Private Law and Commercial Regulation at The University of Western Australia, Professorial Fellow (Melbourne Law School) at the University of Melbourne. This article is informed by research conducted with Professor Jeannie Marie Paterson, Melbourne Law School, pursuant to Australian Research Council Discovery Projects DP180100932 ‘Developing a Rational Law of Misleading Conduct’ and ARC DP140100767 ‘Remedies under the Australian Consumer Law and Common Law: Evolution and Revolution’. PHONE (08) 6488 2740 EMAIL elise.bant@uwa.edu.au.


[1] See eg, In the matter of HIH Insurance Limited (In Liquidation) (ACN 008 636 575) v McGrath (in his capacity as Liquidator of HIH Insurance Limited (in liquidation)) [2016] NSWSC 482; Caason Investments Pty Ltd v Cao [2015] FCAFC 94 (Caason); Brosnan v Katke [2016] FCAFC 1, [121]; and TPT Patrol Pty Ltd as trustee for Amies Superannuation Fund v Myer Holdings Limited [2019] FCA 1747 (TPT Patrol).

[2] Compare Corporations Act 2001 (Cth), s729(1); Australian Consumer Law, s237; Australian Securities and Investments Commission Act 2001 (Cth), s12GF; National Consumer Credit Protection Act 2009 (Cth), ss154, 179U.

[3] See eg, Caason, above note 1, [156], [167] per Edelman J; TPT Patrol, above note 1, [1523].

[4] March v Stramare (E & MH) Pty Ltd [1991] HCA 12; (1991) 171 CLR 506, 515–16 per Mason CJ.

[5] See discussion in Agricultural Land Management Ltd v Jackson [No 2] [2014] WASC 102, [392]–[396].

[6] Caason, above note 1, [163] per Edelman J, noting the same division in the state Civil Liability Acts: eg, Civil Liability Act 2002 (NSW), s5D(1); Wrongs Act 1958 (Vic), s51(1); Civil Liability Act 2002 (WA), s5C(1).

[7] See eg, J Stapleton, ‘Occam’s razor reveals an orthodox basis for Chester v Afshar’, LQR, Vol. 122, 2006, 426; J Stapleton, ‘Cause-in-fact and scope of liability for consequences’, LQR, Vol, 119, 2003, 388.

[8] David Securities Pty Ltd v Commonwealth Bank of Australia [1992] HCA 48; (1992) 175 CLR 353.

[9] These causation and scope of liability questions are then usually followed by a final, separate liability question, namely whether there are any defences that operate to protect the defendant from what would otherwise be his prima facie liability. See generally J Edelman and E Bant, Unjust Enrichment, 2nd ed, Hart Publishing, Oxford, 2016, chs 8, 14 and 15.

[10] Confirmed in TPT Patrol, above note 1, [1518]–[1521].

[11] Janssen-Cilag Pty Ltd v Pfizer Pty Ltd [1992] FCA 437; Caason, above note 1, [155] per Edelman J.

[12] See eg, Caason, above note 1, [154] per Edelman J.

[13] See TPT Patrol, above note 1, [1526]

[14] See further discussion in E Bant and J Paterson, ‘Statutory causation in cases of misleading conduct: Lessons from and for the common law’, Torts L J, Vol. 2(1), 2017, 1–31, 11–12.

[15] P Birks, An Introduction to the Law of Restitution, revised ed, Clarendon Press, 1989, 157.

[16] Compare the consumer behavioural studies on decision-making: R Korobkin, ‘Bounded rationality, standard form contracts, and unconscionability’, University of Chicago Law Review, Vol. 70, 2003, 1203; MA Eisenberg, ‘The limits of cognition and the limits of contract’, Stanford Law Review, Vol. 47, 1995, 211.

[17] Australian Competition and Consumer Commission v TPG Internet Pty Ltd [2013] HCA 54; (2013) 250 CLR 640, [55] per French CJ, Crennan, Bell and Keane JJ.

[18] Pitt v Holt [2013] UKSC 26; [2013] 2 AC 108, [105]–[108] per Lord Walker, delivering the judgment of the Court. See discussion in Edelman and Bant, above note 9, 173–74.

[19] A conclusion consistent with TPT Patrol, above note 1, [1523] .

[20] I & L Securities Pty Ltd v HTW Valuers (Brisbane) Pty Ltd (2002) 210 CLR 109 (I & L Securities).

[21] TPT Patrol, above note 1, [1531].

[22] Ibid, see the arguments referred to at [1522].

[23] Marks v GIO Australia Holdings Ltd [1998] HCA 69; (1998) 196 CLR 494, 509 [34] per McHugh, Hayne and Callinan JJ.

[24] I & L Securities, above note 20, 119, [25]–[26] per Gleeson CJ.


AustLII: Copyright Policy | Disclaimers | Privacy Policy | Feedback
URL: http://www.austlii.edu.au/au/journals/PrecedentAULA/2020/14.html