The Share Price Contention
1230 A significant point of contention between the parties concerns the relevance of GetSwift's share price in the materiality evaluation. It is convenient to first set out the competing positions before turning to the appropriate approach.
1231 ASIC accepts that the materiality of information may be assessed by having regard to the actual impact of the information on the share price, which provides a means to "cross-check" the reasonableness of an ex ante judgement about a different hypothetical disclosure: see Fortescue (at 301 [477] per Gilmour J); James Hardie (at 197 [534]-[537] per Spigelman CJ, Beazley and Giles JJA); Lombe (at 98 [277] per Foster J). However, the core of ASIC's case is that the rapid increase in GetSwift's share price from $0.20 at its IPO to $4 on 7 December 2017 (and the ultimate fall to $0.52 by 8 December 2018) in the absence of any other apparent factor demonstrates graphically that investor expectations were engendered by the ASX announcements.1999 This is consistent with Mr Molony's opinion that GetSwift's share price was "event driven" and related to the ASX announcements.2000 Further, as recognised by Mr Molony, while a number of GetSwift announcements had a material short term impact on GetSwift's share price,2001 many of the contracts merely supported market expectations at the relevant times; those expectations being that the benefits of the existing contracts would be captured by GetSwift and that further contract "wins" were forthcoming.2002 It is said that given the significance of investor expectations, assessing the impact of individual GetSwift announcements cannot be limited to a consideration of the short term price reaction following a given announcement.2003
1232 As might be expected, GetSwift's approach to this issue is the opposite. GetSwift submits that unless ASIC can demonstrate that the ASX announcements had a material impact on the GetSwift share price, its continuous disclosure case must fail because any omitted qualifying information could not be material.2004 In this sense, GetSwift's mode of analysis is to use the reaction to that which was disclosed (i.e. the ASX announcements) as a "stepping-stone" to consider the likely impact of that which was omitted.2005 It was argued that, as a "matter of logic", one can infer the reaction to a hypothetical announcement containing the omitted information on the basis of the market reaction to the actual announcement.2006 That is, where the disclosed information is likely to have a positive effect, while the omitted information is likely to have a partially offsetting or negative effect, if the information that was disclosed did not have a positive material effect, it can be concluded that the omitted information could not have had a material effect on the share price.
1233 To ground this argument, GetSwift relies on Gilmour J's observations in Fortescue, which it contends provides "a useful prism for analysis". In that case, while his Honour recognised the approach to assessing materiality was ex ante, he did accept (at 301 [477]) the following:
[E]vidence of the actual effect of the information actually disclosed on [the company's] share price may be relevant to assist the court in its determination of whether s 674(2) has been contravened: Rivkin Financial Services Ltd v Sofcom Ltd (2004) 51 ACSR 486; [2004] FCA 1538 by analogy at [113]-[116]; Jubilee Mines at [33], [130] and [134]. This involves an ex post inquiry. Such evidence may constitute a relevant cross-check as to the reasonableness of an ex ante judgment about a different hypothetical disclosure.
1234 Applying this line of reasoning to the facts of that case, Gilmour J noted (at 315 [549]):
[A]n ex ante analysis of the likely influence of the actual notifications followed by an ex ante analysis of the likely influence of the hypothetical information is not the preferred approach. Whether [the company] ought to have announced the hypothetical information following the actual notifications made in August and November, is better informed by considering, on an ex post basis, the actual effect on the [the company's] share price following the actual announcements and against that market information, and then considering, on an ex ante basis, the s 677 influence, if any, of the hypothetical notifications.
(Emphasis added).
1235 See also James Hardie (at 197 [537] per Spigelman CJ, Beazley and Giles JJA); Australian Securities and Investments Commission (ASIC) v Fortescue Metals Group Ltd (Fortescue (FC)) [2011] FCAFC 19; (2011) 190 FCR 364 (at 424-425 [188] per Keane CJ). Cf. Grant-Taylor (at 737 [64] per Perram J).
1236 Since the vast majority of the ASX announcements did not result in a significant impact on GetSwift's share price, even if GetSwift had, in hypothetical terms, qualified the announcements by watering down the anticipated benefits, this would not have resulted in any meaningful change to the share price. Similarly, it is said that if the announcement of a new customer relationship was met with price indifference such that it was not material, then the hypothetical disclosure of the termination of that relationship would have been treated in the same way and therefore could not have been material.
1237 Further, GetSwift submits that for ASIC's claim to hold, it is necessary that a string of agreements similar to those announced were not merely expected, but were expected with a sufficiently high degree of certitude, such that the ASX announcements added nothing of material value to an investor's assessment of the stock and that the subsequent news regarding the actual entry into those agreements did not evoke a material price reaction. Indeed, it was said that not only must investors have been anticipating these partnerships with such a high degree of certitude, they must have also been anticipating any announced partnerships to be already in the revenue generating phase at the time of the announcement. In support of this contention, GetSwift point to how its share price experienced no positive reaction when the May Investor Presentation (which concerned how growth in transactions were consistent with prior estimates of deliveries provided in the CBA Agreement) was released on 9 May 2017.2007
1238 Having considered the detailed submissions on the Share Price Contention, it is clear that ASIC's approach, as outlined above (at [1231]) is to be preferred. A number of reasons support this conclusion. Before turning to these reasons, it is necessary to set out the agreed table which reflects GetSwift's share price following each ASX announcement (which was marked as Exhibit Y):
1239 First, from the outset, ASIC's case was not pleaded nor opened on the basis that share price was critical to establishing materiality; that is, its case was not dependent upon it proving that the ASX announcements each had a readily observable and immediate impact on GetSwift's share price. To the extent that GetSwift contends that ASIC has therefore simply ignored inconvenient primary material, such a contention should not be accepted.
1240 Secondly, GetSwift's Share Price Contention mischaracterises ASIC's acceptance that looking to fluctuations in share price may at times be useful in the materiality assessment, but in this case, such an analysis is not overly helpful, nor in any way conclusive. In the present case, an analysis of share price reaction alone in relation to disclosed information does not provide a proper basis for determining the materiality of omitted information, which was not disclosed to the market. It offends the basic tenet of how materiality should be assessed by the Court in a continuous disclosure case; that is, on the basis of a common sense approach, which gives consideration to context and takes into account all objective circumstances.
1241 That point was made in James Hardie, where in respect of omitted information in that case, the Court of Appeal stated (at 197 [537] per Spigelman CJ, Beazley and Giles JJA):
We accept that evidence of the market's reaction to the release of information may be relevant as a cross-check in the manner suggested by Gilmour J. However, in this case, [the primary judge's] failure to have regard to the market reaction on 15 May 2003 does not affect the ultimate outcome of the appeal. In the first place, although the share price was relatively stable, there was a significant increase in the volume of share trading … Given the proximity in time to the public release of the information, the inference is that the market reacted to the information. Second, there is an issue as to whether such disclosure as was made on 15 May 2003 satisfied [the company's] statutory obligation … if there was no disclosure on 15 May 2003 that satisfied the statutory requirements, the trading results on that day and the immediately succeeding days would not be relevant in any event.
1242 In that case, the materiality of the omitted information was not dependent upon share price reaction, but rather, on an application of a common sense approach to whether the omitted information was material in the context of James Hardie's business at that point in time: James Hardie (at 197-198 [537]-[540]). Similarly, in Vocation, it was found that the omitted information was material without an analysis of the actual impact of the disclosed information on share price, but with the assistance of an expert, much like in this case, who expressed opinions on the types of information that were important to investors: see Vocation (at 286-290 [531]-[549]).
1243 Indeed, while subsequent authorities have generally endorsed the approach of Gilmour J, they have been cognisant of its limitations. For example, in Fortescue (FC), Keane CJ made plain (at 424-425 [188]) that "the terms of s 677 do not invite an inquiry as to whether any change in the price of securities has occurred". In saying this, it is important to highlight that I am not rejecting the "useful prism for analysis" that GetSwift draws upon from Fortescue. In certain circumstances, on the current authorities, the materiality of information may be partly informed by reference to an ex post analysis of the actual impact of the information on the company's share price; but this is one factor, which may or may not be relevant in the overall common sense evaluation.
1244 Thirdly, while GetSwift places an intense focus on share price for the most part of its case, in respect of those announcements that elicited a significant share price reaction, it simply contends that a "further analysis" is required. This "have your cake and eat it too" approach, or as ASIC puts it, "idiosyncratic and selective approach" - focussing on share price as essentially definitive to then requiring consideration of other circumstances when the share price has altered following an announcement - is inadequate.2008 Two examples demonstrate this shortcoming: first, in respect of the NAW Announcements, GetSwift acknowledges that these announcements "had an immediate effect on the share price",2009 but nonetheless proceed to focus the enquiry on the share price five trading days after the announcement,2010 a seemingly arbitrary and selective cut-off, particularly in the absence of any expert analysis to demonstrate that this timeframe is appropriate; and secondly, in respect of the CBA Announcement, which had a significant impact on share price, GetSwift contended that the qualifications and limitations to the CBA Deliveries Projection (indicating that there were to be over 257,400,000 deliveries) and the CBA Value Projection (an aggregated transaction value of $9 billion) would not have been material; a conclusion which appears to be based on its own subjective assessment.2011
1245 Fourthly, in relation to the dramatic fall in GetSwift's share price in February 2018, GetSwift contends that nothing can be inferred in relation to materiality and that there is no "link" between the omitted information and the share price decline. Support for this contention is drawn from the fact that during the month-long trading halt that began on 22 January 2018, the Australian Financial Review (AFR) published two articles regarding GetSwift: one on 2 February 2018 concerning the class action (Class Action Article); and another which "estimated measurable damages could be about $300 million" (Damages Article). GetSwift contends that these articles caused the substantial impact on its share price. Further, GetSwift highlight that there was negative news that people were openly shorting the stock (although GetSwift did not cite any evidence in support of this contention beyond the two articles) and that GetSwift released numerous ASX announcements unrelated to the issues in this case, including an Appendix 4C (although GetSwift did not identify significance of these announcements relative to the information that was known beforehand). In the light of this body of confounding information, GetSwift says that it is not possible to establish any link between the share price decline that followed after GetSwift came out of a month-long trading halt and the materiality of any information at issue in this case.2012
1246 With respect, these submissions are wholly unpersuasive. The dramatic fall in GetSwift's share price is consistent with Mr Molony's compelling opinion that GetSwift's share price was "event driven".2013 Indeed, as Mr Molony opined, and I accept:
a major driver of this decline in the GetSwift share price was due to many investors significantly discounting the likelihood that GetSwift would capture the stated benefits of the previously announced contracts with Fruit Box, CBA, Pizza Pan, APT, CITO, Hungry Harvest, Fantastic Furniture, Betta Homes, NAW, Johnny Rockets, Yum and Amazon.2014
1247 Of course, news that a company might be on the hook for damages of about $300 million is likely to impact upon its share price adversely, particularly in the wake of a month long trading halt. But, as Mr Molony highlights, there were far greater underlying concerns: the idea that GetSwift was simply not as it once seemed, GetSwift agreements with Enterprise Clients were marred with uncertainty, and the promises GetSwift presented to the market were looking somewhat empty.
1248 Fifthly, ASIC demonstrated that the share price data combined with the traded volume data indicates that the ASX announcements had a twofold impact: first, they each caused an increase in the GetSwift share price (although it concedes that this was of very different magnitudes); and secondly, they each produced an increase in the volume of shares traded. On the final day of hearing, the parties agreed on a document summarising different calculations concerning trading volumes (which I marked Exhibit X), which appears at the Annexure to these reasons. 2015 Without descending into the detail of the calculations, what the table demonstrates, in the broad, is that in the majority of cases, there was an increase in the volume of shares traded on the day of an ASX announcements. ASIC submits that volume data can therefore provide some assistance in assessing materiality, particularly when looked at "as a whole".2016
1249 GetSwift contends, unsurprisingly, that volume data is not useful and not helpful in drawing any conclusion that the expected benefits from the relevant announcements were material.2017 The following submissions are advanced:
(1) ASIC's approach to volume data is "arbitrary", given that it disregards the volume of trading that occurred on days after the day of the relevant announcement. GetSwift contends that such data should be contrasted with the average and median volumes calculated across the whole of the relevant period relevant to this proceeding (i.e. until 19 February 2018), from which GetSwift contends that the announcement day trading volumes generally did not exceed the average volume over that period (although it continues to advance a selective approach to the issue by saying that one must leave aside the days of the CBA, NA Williams, Yum and Amazon Announcements).2018
(2) On days when there were no announcements made, there were significant fluctuations in the volume of trading in GetSwift's shares of the same magnitude to those that occurred on the days of the relevant announcements.2019 For example, between 28 and 29 March 2017, there was an increase of 791% in the volume of GetSwift shares traded, although no announcement was made on either day. Between 15 and 16 November 2017, there was an increase of 396% in the volume of GetSwift shares traded, despite no announcement made on either day. To similar effect, GetSwift highlights that there were a number of days when, despite not releasing an ASX announcement, the volume of trading in GetSwift shares was high. On 24 July 2017, there were 2,363,867 GetSwift shares traded; on 16 November 2017, there were 2,874,088 GetSwift shares traded.
(3) Given the number of GetSwift shares increased over the relevant period, the inclusion of trading days prior to when the number of shares on issue increased has the effect of lowering the average and median trading volumes.
(4) Being a speculative stock, GetSwift is subject to a greater than usual proportion of high frequency traders, speculators and day traders, who had the effect of affecting trading volume figures significantly. It was said that these traders are not relevant to the materiality test: see Vocation (at 291 [553] per Nicholas J).
(5) Given a "median is a statistical measurement that is useful for showing what the typical observation in a sample is, unaffected by the weighting effect that outliers can have on the average", and in circumstances where GetSwift's daily trading volumes were volatile, any assessment should use average volume to incorporate the volatility of the data and the randomness of high and low volume days.
(6) While trading volume data may be of some relevance, from the wording of s 674(2) of the Corporations Act and Listing Rule 3.1 indicates the focus of the materiality test is on share price.
1250 Based on these submissions, GetSwift says that any volume data on which ASIC relies is not capable of establishing materiality. While there are limitations in the data as identified by GetSwift, it ultimately does not really matter, given trading volume data are but an indication, like price reaction. Indeed, ASIC does not seem to rely on volume data in their submissions. What this analysis of trading data coincident with an ASX announcement does serve to reinforce, however, is that it is not inconsistent with the conclusions I would otherwise have drawn in relation to the materiality of the announcements (being conclusions which are, in turn, of some assistance in assessing the materiality of the omitted information). In this sense the trading data operate as no more than a cross-check of the ex ante common sense analysis required as to the omitted information. If the data were manifestly inconsistent with any suggestion of materiality of the announcements, it would give reason for pause and further consideration, but that is not the case (and in some respects, such as the CBA, NA Williams, Yum and Amazon Announcements, it is plainly what one would expect to find if the relevant announcement was highly material).
1251 Sixthly, GetSwift's claim that ASIC's expectation thesis requires a finding that a string of announcements were expected with a high degree of certitude, such that the announcements added nothing of material value to an investor's assessment of the stock, falls down because that is precisely what was engendered by the Prospectus, the Quantifiable Announcement Representations, and GetSwift's other statements as to impending contract announcements (as set out at [25]-[27] and [31]-[33]). Further, all this occurred in the context of a well-publicised high growth and expansion strategy. It is therefore unremarkable that GetSwift's share price was not directly correlative with the announcements of new agreements: investors had an expectation of actual concluded contracts being announced because that is what they were told would happen, and they were told that these would be announced imminently. As Mr Molony opined:
[A]ssessing the impact of individual GetSwift announcements cannot be limited to a consideration of the short-term share price reaction following the given announcement. In simple terms many of GetSwift new contract announcements merely supported market expectations at the relevant time i.e. that the benefits of the existing contracts would be captured by GetSwift and that further contract win announcements were forthcoming.2020
1252 Within this context, if GetSwift released a subsequent announcement that told investors the true position was that, in reality, the contract to be announced was a mere term sheet, that the client had not yet commenced a trial (or had not yet concluded a trial), that if the agreement was terminated during the trial period GetSwift lost exclusivity, and that no revenue would be generated until the trial was competed, as a matter of common sense, of course such information would have altered the engendered expectation of investors in deciding whether to acquire or dispose GetSwift shares. It may well have also undermined the degree of confidence investors held in the accuracy and reliability of GetSwift's other announcements. But this is not what happened.
1253 Further, I reject GetSwift's submission that its share price experienced no positive reaction when the May Investor Presentation (which concerned how growth in transactions were consistent with prior estimates of deliveries provided in the CBA Agreement) was released on 9 May 2017.2021 First, there was in fact a 32% increase in GetSwift's share price in the three days following its release.2022 Secondly, and in any event, the May Investor Presentation informed investors of the number of deliveries that were then occurring, reinforcing what investors had been told and leaving them to believe that those contracts which had been announced to date were generating the level of deliveries estimated.
1254 Investors were also told what was expected in the future pipeline for deliveries in the next 18 to 24 months, engendering an expectation that any announced contracts would convert projections into actuality. Indeed, within this context, a subsequent contract announcement would have been in line with that expectation and, accordingly, if there was no positive movement in the share price following such an announcement, it could suggest the announcement accorded with existing expectations. In saying this, as GetSwift points out, if an announcement was made that significantly added to expectations as to projected volumes, it is likely that it would alter expectations and potentially have an impact on share price.2023 The CBA Announcement did just that (see below at [1357]), which is understandable given the exponential rate of growth that was expected as a result of the projections made in that announcement.
1255 All of these considerations show the limitations in using share price as a singular and determinative factor to assess the materiality of omitted information.2024
1256 Finally, and importantly, there are a number of other fundamental issues that exist in relation to GetSwift's approach to share price which, from a common sense market analysis point of view, was somewhat superficial. For example, GetSwift has: (a) not accounted for any macro-market or economic factors on each applicable trading day; (b) only looked to share price, without regard to liquidity, trading volumes, pre-open bid, ask and sell prices or average closing prices; (c) focussed on the closing price, as opposed to intra-day highs in the share price; (d) ignored the potential effect on the share price on other price sensitive announcements on the day of, or closely proceeding or following the announcements (i.e. Appendix 4C & Quarterly Review Announcement); and (e) did not assess market capitalisation.2025