Verition Partners Master Fund Ltd. v. Aruba Networks, Inc.

• Court: Court of Chancery of Delaware
• Decided: February 15, 2018
• Docket: CA 11448-VCL
• Status: Published

Opinion

IN THE COURT OF CHANCERY OF THE STATE OF DELAWARE

VERITION PARTNERS MASTER FUND ) LTD. and VERITION MULTI-STRATEGY ) MASTER FUND LTD., ) ) Petitioners, ) v. ) C.A. No. 11448-VCL ) ARUBA NETWORKS, INC., ) ) Respondent. )

MEMORANDUM OPINION

Date Submitted: January 26, 2018 Date Decided: February 15, 2018

Stuart M. Grant, Michael J. Barry, Christine M. Mackintosh, Michael T. Manuel, Rebecca A. Musarra, GRANT & EISENHOFFER P.A., Wilmington, Delaware; Attorneys for Petitioners.

Michael P. Kelly, Steven P. Wood, McCARTER & ENGLISH, LLP, Wilmington, Delaware; Marc J. Sonnenfeld, Karen Pieslak Pohlmann, Laura Hughes McNally, MORGAN, LEWIS & BOCKIUS LLP, Philadelphia, Pennsylvania; Attorneys for Respondent.

LASTER, V.C. In May 2015, Hewlett-Packard Company (“HP”) acquired Aruba Networks, Inc.

(“Aruba” or the “Company”). The transaction was governed by an Agreement and Plan of

Merger by and among Aruba, HP, and Aspen Acquisition Sub., Inc., a wholly owned

subsidiary of HP. Under the merger agreement, each share of Aruba common stock was

converted into the right to receive consideration of $24.67 per share, subject to the holder’s

statutory right to eschew the merger consideration and seek appraisal.1 The petitioners

perfected their appraisal rights and litigated this statutory appraisal proceeding. This is the

court’s post-trial decision on the issue of fair value.

The Delaware Supreme Court’s decisions in Dell2 and DFC3 endorse using the

market price of a widely traded firm as evidence of fair value.4 As in Dell and DFC, the

market for Aruba’s shares exhibited attributes associated with the premises underlying the

efficient capital markets hypothesis. Under Dell and DFC, these attributes provide

sufficient evidence of market efficiency to make Aruba’s stock price “a possible proxy for

fair value.”5 Aruba’s thirty-day average unaffected market price was $17.13 per share.

1 See 8 Del. C. § 262.

Dell, Inc. v. Magnetar Glob. Event Driven Master Fund Ltd, – A.3d –, 2017 WL 2

6375829 (Del. Dec. 14, 2017). 3 DFC Glob. Corp. v. Muirfield Value P’rs, L.P., 172 A.3d 346 (Del. 2017). 4 See Dell, 2017 WL 6375829, at *16; DFC, 172 A.3d at 369-70, 373. 5 Dell, 2017 WL 6375829, at *1; see also id. at *17 (“[T]he price produced by an efficient market is generally a more reliable assessment of fair value than the view of a single analyst, especially an expert witness who caters her valuation to the litigation imperatives of a well-heeled client.”); DFC, 172 A.3d at 369-70 (“Market prices are typically viewed superior to other valuation techniques because, unlike, e.g., a single

1 The Delaware Supreme Court’s decisions in Dell and DFC endorse using the deal

price in a third-party, arm’s-length transaction as evidence of fair value.6 When evaluating

the reliability of the deal price, a trial judge must remember that

the purpose of an appraisal is not to make sure that the petitioners get the highest conceivable value that might have been procured had every domino fallen out of the company’s way; rather, it is to make sure that they receive fair compensation for their shares in the sense that it reflects what they deserve to receive based on what would fairly be given to them in an arm’s- length transaction.7

Put differently, “[t]he issue in an appraisal is not whether a negotiator has extracted the

highest possible bid. Rather, the key inquiry is whether the dissenters got fair value and

were not exploited.”8

In this case, the merger was an arm’s-length transaction that provided stockholders

with consideration of $24.67 per share. By definition, it provided stockholders with “fair

compensation” in the sense of “what would fairly be given to them in an arm’s-length

transaction.”9 The petitioners proved that the Company’s negotiators might have done

better, but there is no reason to believe that they left any of Aruba’s fundamental value on

person’s discounted cash flow model, the market price should distill the collective judgment of the many based on all the publicly available information about a given company and the value of its shares.”) 6 Dell, 2017 WL 6375829, at *22; DFC, 172 A.3d at 367. 7 DFC, 172 A.3d at 370-71. 8 Dell, 2017 WL 6375829, at *24. 9 DFC, 172 A.3d at 371.

2 the bargaining table. When the merger consideration of $24.67 per share is compared to

the unaffected market price of $17.13 per share, it is not possible to say that Aruba’s

stockholders were exploited. The deal price therefore provides reliable evidence of fair

value.

The Dell and DFC decisions recognize that a deal price may include synergies, and

they endorse deriving an indication of fair value by deducting synergies from the deal

price.10 The respondent’s expert cited a study that provides data on the base rates at which

targets successfully extract a share of anticipated synergies from acquirers. Using that data,

this decision arrives at a midpoint valuation indication for Aruba of $18.20 per share. I

personally believe that Aruba’s negotiators did not extract as great a share of the synergies

as they might have, which suggests that deal-price-less-synergies figure is slightly higher.

The Dell and DFC decisions caution against relying on discounted cash flow

analyses prepared by adversarial experts when reliable market indicators are available.11

The decisions teach that discounted cash flow models should be “used in appraisal

proceedings when the respondent company was not public or was not sold in an open

10 Dell, 2017 WL 6375829, at *13; DFC, 172 A.3d at 371. 11 See Dell, 2017 WL 6375829, at *26 (describing the management buy-out in that proceeding and stating that “this appraisal case does not present the classic scenario in which there is reason to suspect that market forces cannot be relied upon to ensure fair treatment of the minority”); DFC, 172 A.3d at 369 n.118 (explaining that discounted cash flow models are “often used in appraisal proceedings when the respondent company was not public or was not sold in an open market check”).

3 market check.”12 When market evidence is available, “the Court of Chancery should be

chary about imposing the hazards that always come when a law-trained judge is forced to

make a point estimate of fair value based on widely divergent partisan expert testimony.”13

In this case, the discounted cash flow analysis prepared by the petitioners’ expert generated

a value of $32.57, which was inconsistent with the market evidence. The discounted cash

flow analysis prepared by the respondent’s expert generated a value of $19.75, nestled

nicely between the unaffected market price and the deal price. Its methodological

underpinnings, however, provided cause for concern, as did the meandering route by which

the expert arrived at his final figure. I do not rely on the discounted cash flow valuations.

The two most probative indications of fair value are Aruba’s unaffected market

price of $17.13 per share and my deal-price-less-synergies figure of approximately $18.20

per share. In the context of this case, the unaffected market price provides the most

persuasive evidence of fair value. My deal-price-less-synergies figure suffers from two

major shortcomings.

First, my deal-price-less-synergies figure is likely tainted by human error.14

Estimating synergies requires exercises of human judgment analogous to those involved in

12 DFC, 172 A.3d at 369 n.118. 13 Dell, 2017 WL 6375829, at *26. 14 To cite just a few possibilities, my specific allocation of synergies to the sell-side in this case, while well intentioned, could be erroneous. The size of the original synergy estimates, while the best available under the circumstances, represented predictions about

4 crafting a discounted cash flow valuation. The Delaware Supreme Court’s preference for

market indications over discounted cash flow valuations counsels in favor of preferring

market indications over the similarly judgment-laden exercise of backing out synergies.15

Second, my deal-price-less-synergies figure continues to incorporate an element of

value derived from the merger itself: the value that the acquirer creates by reducing agency

costs.16 A buyer’s willingness to pay a premium over the market price of a widely held firm

reflects not only the value of anticipated synergies but also the value created by reducing

agency costs.17 The petitioners are not entitled to share in either element of value, because

complex matters and could be wrong. More broadly, the baseline data about the manner in which buyers and sellers allocate synergies could reflect sampling or measurement errors. 15 See Dell, 2017 WL 6375829, at *26; DFC, 172 A.3d at 388; see also Applebaum v. Avaya, Inc., 812 A.2d 880, 890 (Del. 2002) (“[A] well-informed, liquid trading market will provide a measure of fair value superior to any estimate the court could impose.”). 16 See William J. Carney & Mark Heimendinger, Appraising the Nonexistent: The Delaware Court’s Struggle with Control Premiums, 152 U. Pa. L. Rev. 845, 847-48, 857- 58, 861-66 (2003) [hereinafter Control Premiums]; Lawrence A. Hamermesh & Michael L. Wachter, Rationalizing Appraisal Standards in Compulsory Buyouts, 50 B.C. L. Rev. 1021, 1023-24, 1034-35, 1044, 1046-54, 1067 (2009) [hereinafter Rationalizing Appraisal]; Lawrence A. Hamermesh & Michael L. Wachter, The Short and Puzzling Life of the “Implicit Minority Discount” in Delaware Appraisal Law, 156 U. Penn. L. Rev. 1, 30-36, 49, 52, 60 (2007) [hereinafter Implicit Minority Discount]; Lawrence A. Hamermesh & Michael L. Wachter, The Fair Value of Cornfields in Delaware Appraisal Law, 31 J. Corp. L. 119, 128, 132-33, 139-42 (2005) [hereinafter Fair Value of Cornfields]. 17 See Rationalizing Appraisal, supra, at 1023-24, 1038, 1046-54, 1067; Implicit Minority Discount, supra, at 30-36, 52; Fair Value of Cornfields, supra, at 139-41. The concept of reduced agency costs is the flipside of the benefits of control, which includes the ability to make changes in corporate management, strategy, and policy. See DFC, 172 A.3d at 369 n.117 (citing ability to change dividend policy). The key point is that control creates value distinct from synergy value. See Fair Value of Cornfields, supra, at 148 (“[E]xcluded gains [for purposes of appraisal] include, for example, those resulting from

5 both “aris[e] from the accomplishment or expectation of the merger.” 18 The synergy

deduction compensates for the one element of value arising from the merger, but a further

downward adjustment would be necessary to address the other.19

Fortunately for a trial judge, once Delaware law has embraced a traditional

formulation of the efficient capital markets hypothesis, the unaffected market price

provides a direct route to the same endpoint, at least for a company that is widely traded

and lacks a controlling stockholder.20 Adjusting down from the deal price reaches,

economies of scale or increased market share, or those that derive from the acquirer’s plans to operate the post-merger enterprise more efficiently.”); id. at 151 (same); accord Implicit Minority Discount, supra, at 4 (explaining that stockholders in an appraisal proceeding “do not receive value that includes synergies or benefits of control”). 18 8 Del. C. § 262(h); see Rationalizing Appraisal, supra, at 1038 (“[T]hird-party sale value is an inappropriate standard for determining the fair value of dissenting shares because it incorporates elements of value—associated with acquisitions of control by third parties—that do not belong to the acquired enterprise or to shares of stock in that enterprise.”); Implicit Minority Discount, supra, at 30 (“The value of the firm is not its third-party sale value (V3PS). In an arm’s-length transaction, an acquirer will pay a premium to VE in purchasing the firm. The premium largely reflects synergies arising from the merger, but it can also reflect benefits of control.”); Fair Value of Cornfields, supra, at 148 (“[E]xcluded gains [for purposes of appraisal] include, for example, those resulting from economies of scale or increased market share, or those that result from the acquirer’s plans to operate the post-merger enterprise more efficiently.”); id. at 151 (concluding that Section 262(h) excludes value arising from both “synergies dependent on the consummation of an arm’s-length acquisition” and “operating efficiencies that arise from the acquirer’s new business plans”). 19 See Rationalizing Appraisal, supra, at 1055 (discussing an acquisition of a widely held firm and explaining that “the firm’s going concern value can be estimated in this case as the actual purchase price minus synergies minus control value”). 20 See Control Premiums, supra, at 858 (“The basic conclusion of the Efficient Capital Markets Hypothesis (ECMH) is that market values of companies’ shares traded in competitive and open markets are unbiased estimates of the value of the equity of such

6 indirectly, the result that the market price already provides. Aruba’s unaffected market

price provides the most persuasive evidence of fair value.

By awarding fair value based on the unaffected market price, this decision is not

interpreting Dell and DFC to hold that market price is now the standard for fair value.

Rather, Aruba’s unaffected market price provides the best evidence of its going concern

value.21 The fair value of Aruba is $17.13 per share.

I. FACTUAL BACKGROUND

The parties reached agreement on 180 stipulations of fact in the Pre-Trial Order.

Trial took place over three days. The parties submitted 996 exhibits, including eleven

firms.”); Implicit Minority Discount, supra, at 52 (“Take the case of a publicly traded company that has no controller. Efficient market theory states that the shares of this company trade at the pro rata value of the corporation as a going concern.”); id. at 60 (“As a matter of generally accepted financial theory . . . , share prices in liquid and informed markets do generally represent th[e] going concern value . . . .”). Professors Hamermesh and Wachter make the same point indirectly in Rationalizing Appraisal, where they explain that market prices should not be used to determine fair value where there is either no public market price at all, where the shares are illiquid or thinly traded, or where there is a controlling stockholder. See Rationalizing Appraisal, supra, at 1033-34. Outside of these scenarios, they appear to agree that “because financial markets are efficient, one can simply use the market price,” and they note that courts already rely on market price data when using the capital asset pricing model to select a discount rate or when adding a small stock premium. See id. 21 See Control Premiums, supra, at 879 (noting that the appraisal statute requires consideration of all relevant factors and stating that “in an efficient market, absent information about some market failure, market price is the only relevant factor”); id. at 849 (“[W]e urge courts to presume that market value is the best measure of fair value.”).

7 deposition transcripts. Three fact witnesses and three experts testified live. The parties

proved the following facts by a preponderance of the evidence.

A. Aruba

Aruba was a Delaware corporation headquartered in Sunnyvale, California. Aruba

went public in 2007. Until its acquisition by HP, Aruba’s common stock traded on the

NASDAQ under the symbol “ARUN.”22

Aruba principally sold components for enterprise wireless local area networks

(“WLANs”).23 From 2008 until 2014, its market share increased from 8.6% to 12.8%.

Shortly before the merger, Aruba’s market share peaked at 14%.24 Although Aruba was a

significant player in the industry, Cisco Systems, Inc. dominated it.25 For those same years,

Cisco’s market share hovered around 50%.26

During the years leading up to the merger, Dominic Orr served as Aruba’s CEO.27

Before joining Aruba, Orr already had enjoyed a successful career in the technology and

22 Pre-Trial Order ¶¶ 27, 28. This decision cites the Pre-Trial Order as “PTO.” 23 Id. ¶¶ 63-64. 24 Id. ¶ 79. 25 See, e.g., Warmenhoven Tr. 126 (referring to Cisco as “the 800-pound gorilla”); Orr Tr. 472 (suggesting Cisco was “[p]robably more than 800 pounds”). 26 PTO ¶ 80. 27 Id. ¶ 151.

8 telecommunications sectors.28 In 2002, in his early fifties, Orr retired to pursue a range of

personal interests.29

In 2006, the Aruba board of directors (the “Board”) lured Orr out of retirement. Orr

anticipated serving for three years. But when his first term ended in 2009, the Board had

not identified a successor. Orr agreed to stay on for a second three-year term, but when that

term ended, the Board still had not identified a successor. At that point, Orr agreed to stay

on “year by year,” on the condition that Aruba seriously engage in succession planning.30

The Board agreed but did not move rapidly. The Board did not engage an executive search

firm until 2014. Even then, the Board limited the engagement to developing a position

specification for a “CEO succession review.”31 Active recruitment would require a separate

engagement.32 There remained “[n]o firm date” for Orr’s retirement.33

28 See id. 56. 29 Orr Tr. 371-73 (describing activities during retirement, including spending time with his children, learning to cook, traveling to Japan and taking Japanese classes, working with The Philanthropy Workshop, and pursing other philanthropic endeavors including founding a non-profit business that provides low-power, solar-panel-driven phones). 30 Orr Tr. 377. See generally id. at 368-79. 31 JX 248 at 1 (search firm engagement letter); see also JX 1529 (completed position specification). 32 JX 248 at 2. 33 Warmenhoven Tr. 278; see also id. at 286 (agreeing Orr “wouldn’t have retired until the successful candidate had been identified”); id. at 291 (“There’s no date certain.”); accord Orr Tr. 459 (testifying he would remain until suitable replacement found “[t]o a certain limit. Ultimately, you know, I’m not getting younger every year.”).

9 By the time he led the merger negotiations with HP in late 2014 and early 2015, Orr

was ready to return to an active retirement.34 As a responsible and conscientious individual,

he was not about to leave Aruba in the lurch, and he cared deeply about the Company and

its employees.35 But he also had other things that he wanted to do with his life. The sale of

Aruba to HP gave Orr a path to an honorable personal and professional exit.

B. Wall Street Analysts Question Aruba’s Financial Performance.

One of the precipitating events for the merger talks between HP and Aruba was a

negative stock market reaction to Aruba’s results for the third quarter of 2014.36 In May

2014, Aruba announced its quarterly results. Revenue exceeded both Aruba’s own

guidance and the Wall Street consensus estimates.37 But the Company reported a gross

34 See Warmenhoven Tr. 228-31; id. at 273 (Orr’s “going to be 66 in March. He’s told me many times that he did not want to be in a regular employment situation when he turned 65”); id. at 289 (“[H]e was going to leave before his 65th birthday.”). 35 See Orr Tr. 380 (citing a “moral obligation to make sure that the transaction goes through, that things are reasonably integrated”); id. at 458 (agreeing he “wouldn’t suddenly resign [his] position and leave Aruba without a successful, valid succession plan in place”); Warmenhoven Dep. 154 (calling Orr “the cornerstone of the entire company” and “the face to the customers, the face to the partners, and the face to the employees”). 36 Aruba operated on a fiscal year that ended on July 31. For Aruba, fiscal year 2015 began on August 1, 2014. Marcus Tr. 25. 37 See JX 47 (Barclays report); JX 48 (Janney Capital Markets report).

10 margin of 70.5%, below the consensus estimate of 72% and Aruba’s “longstanding target

of 71-73%.”38

The resulting analyst coverage was harsh. Wall Street firms headlined their reports

with titles like “Weak Gross Margins Outweigh Sales Upside; Maintain Market Perform”

and “FY3Q14: Disappointing Gross Margin Offsets Ongoing 802.11ac Ramp.” 39 Aruba’s

stock price dropped 12.11% on the news, from $20.06 to $17.63.40

Internally, Aruba management was disappointed.41 They had not anticipated the

furor over gross margins.42 Orr vented to the Board: “[W]e, as an executive team, are finally

sick of wall st discrediting our tremendous come back in revenue growth because they said

we are not as profitable as Ubiquiti (give me a break!).”43

38 JX 47 at 1. 39 JX 259 at 2-3 (email collecting analyst coverage). 40 The parties originally provided limited data on the stock price. See JX 245. As part of the supplemental briefing process, the respondent provided more extensive data. See Dkt. 188 Ex. A (publicly available stock-price data for Aruba and S&P 500). The trading data is suitable for judicial notice. See Lee v. Pincus, 2014 WL 6066108, at *4 n.11 (Del. Ch. Nov. 14, 2014) (“I take judicial notice of these reported stock prices because they are not subject to reasonable dispute.” (citing D.R.E. 201(b)(2); In re Gen. Motors (Hughes) S’holder Litig., 897 A.2d 162, 169 (Del. 2006))). 41 See JX 260 at 5-6 (email thread among management); Warmenhoven Tr. 235 (“It was very frustrating for the management team, I must say. I mean, in -- in fiscal year 2014, they had 20 percent year-over-year growth, and we were delivering everything to the street, and the stock really didn’t move.”). 42 JX 260 at 1 (email from Aruba’s Vice President of Sales that he was “not sure we had any real insights [into the gross margins] until it was too late.”) 43 JX 267; see also JX 269 at 2 (Orr email to Aruba’s CFO: “Between you and me, if we do all these and our stock price stays $13-20, you and I fail miserably. I don’t know

11 To improve margins, Aruba management developed a cost optimization plan called

“Project Greyhound.”44 It contemplated eliminating approximately 130 employees and

relocating another eighty to “lower-cost geographies.”45 Management undertook the

project because the “Company value [was] not adequately reflected in [the] stock price.”46

In August 2014, Aruba announced its results for the full year of 2014, including

quarterly results for the fourth quarter. Aruba achieved record revenue. Orr told investors

that Aruba had achieved “significant market share gains” and had a “strong platform for

why we are doing all these [sic] tough stuff. I really don’t.”). Ubiquiti Networks, Inc. is one of Aruba’s competitors. 44 PTO ¶ 101. 45 JX 280 at 10-11 (internal Aruba deck on Project Greyhound); JX 285 (email from Orr announcing Project Greyhound internally as part of an effort to “continue to evaluate all aspects of our business to ensure that we achieve our full potential and that we are able to invest in our strategic initiatives and infrastructure”); see also Warmenhoven Tr. 217- 18 (discussing Project Greyhound); Orr Tr. 413-14 (same); id. at 477-78 (reaffirming statements in JX 285). 46 JX 273 at 2 (draft internal talking points); see also Orr Tr. 486 (suggesting Project Greyhound was motivated, in part, by management’s belief that the stock price “does not reflect what we believe is our commitment to deliver to what the analysts and the investors want us to do”); Galvin Tr. 597-98 (acknowledging Aruba undertook Project Greyhound, in part, because “we did have pressure from Wall Street to improve what we were dropping to the bottom line”); id. at 600 (agreeing “Greyhound was initiated because there was a perception that the company value was not adequately reflected in the stock price”).

12 future growth.”47 Aruba simultaneously announced the implementation of Project

Greyhound.48

The analysts’ reactions were mixed. Some were positive.49 Others were more

cautious.50 Traders bid up Aruba’s stock price by 8%, from $20.24 to $22.01.51

C. HP Approaches Aruba.

HP had been monitoring Aruba as an acquisition candidate. HP felt it had strong

offerings in the wired networking space and wanted to combine those products with

47 JX 828 (press release announcing results); see also JX 637 at 3 (earnings call transcript announcing “[o]ur Q4 results exceeded our guidance on all fronts,” “[w]e saw bookings growth in all of our major [sic] with particular sequential strength in North America,” and “[w]e delivered solid year-over-year growth in all of our core verticals plus notable strength in the enterprise markets”); Orr Tr. 477 (confirming revenue for fiscal year 2014 grew 21% from prior year); Galvin Tr. 545-46 (discussing fiscal year 2014 generally and acknowledging it “was a strong year for the company”). 48 JX 828; JX 637 at 4 (announcing “we have identified opportunities to optimize our cost structure by eliminating certain positions”). 49 See JX 59 at 2 (JMP Securities report commenting favorably on revenue); JX 62 at 1 (BMO Capital Markets report commenting favorably on Project Greyhound); see also JX 309 at 7 (Aruba Board deck summarizing analyst reports and noting that fifteen analysts increased their target price). 50 See JX 61 at 1 (Wells Fargo Securities report positing that “[g]ood news mostly priced in” and advising caution “due to valuation and increased competition from Cisco”); JX 64 at 1 (Deutsche Bank report affirming “hold” recommendation in light of risks that “1) unpredictable sales cycles could adversely impact sales 2) highly competitive marketplace could pressure margins”). 51 JX 245.

13 Aruba’s wireless offerings. HP believed the combined product set could compete

effectively with Cisco and take significant market share.52

On August 27, 2014, the day after Aruba announced its full-year earnings, HP

approached Aruba about a deal. Antonio Neri, a Senior Vice President at HP, contacted

Orr.53 Orr promptly notified Daniel Warmenhoven, Aruba’s lead independent director.54

Warmenhoven, in turn, contacted Frank Quattrone, a senior investment banker at Qatalyst

Partners LP.55

The next day, Warmenhoven notified the full Aruba Board of HP’s interest. He also

reported that he had contacted Quattrone and that Orr would contact Stuart Francis, an

investment banker who had recently left Barclays Capital Inc. to join Evercore Group

52 See, e.g., Johansson Dep. 59-60 (“[T]here was a strong desire from customers to have a converged network portfolio and a converged product solution, and we felt that combining the two companies we would really create a unique offering in the market, which today you can only get from Cisco.”); id. at 248 (acknowledging anticipated $1.6 billion in synergies from transaction); Whitman Dep. 54 (“Hewlett-Packard would feel entitled to those synergies because probably our go-to-market, our G&A, would be enabling those synergies.”); see also Orr Tr. 396 (acknowledging “[t]his is a really, really optimal situation to leverage two companies’ assets” because HP had “the switching line” but “a very, very weak wireless” and “Aruba has the mobility software and the security software . . . and the wireless reputation and portfolio”); id. at 434 (confirming “that the HP-Aruba combination offered synergies to both”); JX 345 at 5 (internal HP deck touting deal as providing “complementary go to market and wireless capabilities, bolstering our Enterprise Mobility offering, and uniquely positioning HP as the market moves towards a unified wired/wireless solution”). 53 PTO ¶ 151. 54 See Warmenhoven Tr. 236-37.

JX 291 at 5 (email from Quattrone to Orr: “I just spoke to Dan [Warmenhoven]. 55

Can we chat early tomorrow morning, say around 7am?”).

14 L.L.C.56 Warmenhoven told the other directors that HP wanted to proceed “quickly and

present the proposal to the [HP] board on Sept[ember] 16.”57

Qatalyst jumped at the potential engagement. Immediately after hearing from

Warmenhoven, Quattrone reached out to Orr and sent him a proposed engagement letter.58

Two days later, George Boutros, another senior banker with Qatalyst, sent Orr a “[s]cript

for [d]iscussion with H[P].”59 Meanwhile, Warmenhoven reconsidered reaching out to

Francis. Warmhoven “had no experience with Evercore” and “didn’t know the team.”60

On August 29, 2014, Neri again spoke with Orr. Neri insisted that HP was serious.

Orr tried to “delicately set expectations that this is going to be a high premium deal.”61

On September 1, 2014, Neri and Orr met in person. A talking point that Boutros

prepared for the meeting stressed the importance of price: “[T]here will have to be a very

substantial premium to market, well in excess of the typical m&a premium, in order to fully

56 See JX 288 (email from Warmenhoven to Aruba Board); see also JX 859 (press release announcing Francis’s new position); see also Warmenhoven Tr. 240; id. at 249 (calling Francis “probably the world’s best banker”). 57 JX 288. 58 JX 289 (email from Quattrone to Orr forwarding draft engagement letter); JX 291 at 4-5. 59 JX 291 at 3. 60 Warmenhoven Tr. 241. 61 JX 291 at 2.

15 reflect both the substantial upside potential we have as an independent company, and our

strategic value to you and others.”62

On September 2, 2014, Warmenhoven spoke with HP’s CEO, Meg Whitman. She

confirmed HP’s interest.63 Later that day, the Aruba Board met. Orr described the

developments with HP and reported that HP had not yet proposed “any financial terms or

other parameters regarding a possible strategic transaction.”64

During the meeting, members of the Board expressed concern about having Qatalyst

advise the Company in discussions with HP. Qatalyst had represented Autonomy

Corporation PLC when HP bought it for $11 billion in 2011.65 The deal was widely

understood to have been a “disaster”66 for HP that resulted in an $8.8 billion write-down

and protracted litigation.67 HP’s acquisition of Aruba would be its first significant deal

since Autonomy. The directors wondered if HP would balk at working across from

Qatalyst.68

62 JX 292 at 1. 63 JX 909 at 47. This decision cites Aruba’s definitive proxy statement, introduced as JX 909, as the “Proxy.” 64 JX 228 at 1 (minutes). 65 PTO ¶¶ 32-35. 66 JX 731 (Financial Times article). 67 PTO ¶¶ 32-34. 68 Warmenhoven Tr. 243 (“And the question was, does [the Autonomy transaction] leave any residual -- negative residual with Hewlett-Packard. And would it cause a problem if Qatalyst represented it in another transaction to HP.”); Orr Tr. 331-32 (confirming Aruba

16 When Warmenhoven raised these concerns with Quattrone and Boutros, they

reassured him that (i) Qatalyst had worked off of Autonomy’s audited financials and was

as much a victim as anyone else, (ii) HP understood this, (iii) the entire HP M&A team had

turned over since Autonomy, and (iv) Qatalyst had a good relationship with the new team.

The Aruba Board accepted these reassurances.69

With the Autonomy concerns allayed, Qatalyst and Aruba negotiated the terms of

Qatalyst’s engagement letter. Qatalyst projected that a deal price “around $30” per share

was “the most likely outcome” and proposed a fee equal to 1.25% of the transaction value

for a deal at $30 per share or higher.70 Qatalyst proposed a richer fee of 1.5% for a price at

$36 or higher. Qatalyst felt that a price below $30 warranted 1% of deal value.71

Aruba’s CFO, Mike Galvin, pushed back. He pointed out that “it’s universally

thought that [Aruba] is undervalued right now” and, therefore, a price at the higher end of

Board “discussed the fact that Qatalyst had previously represented Autonomy in the sale to Hewlett-Packard” as well as “whether that prior history might ‘put a strain on the relationship’ between . . . Qatalyst and HP”). 69 Warmenhoven Tr. 243-44 (testifying Qatalyst representatives were “candid and direct in their answers” and asserted that “they were representing audited financials . . . so they didn’t think that HP would harbor any grudge,” that “[t]here had been a complete turnover of the M&A team,” and that “[w]e work well with these people”); see also Orr Tr. 332 (confirming “Qatalyst assured the board that [Autonomy] wasn’t going to be a problem”). 70 JX 303 at 2 (internal Aruba email recounting negotiations). 71 Id.

17 the range was a strong possibility.72 Orr agreed with Galvin’s assessment. He stressed that

the “board doesn’t want [a] deal below $30; I want it above $33.”73 He further advised that

he was “not comfy [with] a ramp from 30 to 33,” because “[i]t would not be spicy enough”

to incentivize Qatalyst “to focus on the 33-35 target range we want.”74 After heated

negotiations, the final engagement letter provided for a flat 1% fee.75

D. The Discussions With HP Move Forward.

Orr and Neri met again on September 10, 2014. Orr emphasized Aruba’s strong

results, its willingness to remain independent, and the synergies that a deal would

produce.76 Internal HP analyses confirmed the potential for substantial synergies.77

72 Id. at 1. 73 Id. 74 Id.; see also Orr Tr. 512-14 (discussing JX 303 but stating Boutros had advised that “without . . . a truly competing situation, I think the deal’s going to get done with a 2 in front of it” and that “[s]tarting with a 3 [in front] would be unrealistic”) 75 PTO ¶ 42; JX 305 at 2-3 (signed engagement letter). 76 See JX 308 (talking points for Orr including, “we are not trying to sell the company and both the management team and the board are very bullish about our prospects as an independent company,” “a combination could be very powerful,” and “the price would have to be very compelling to take us off our current path which we know will result in significant shareholder value”). 77 See JX 310 at 1-2 (internal HP email summarizing conversation between Orr and Neri, stating Aruba “would like to have a deeper conversation around how serious we are and our willingness to execute the plan/synergies,” and requesting a write-up on “why [Aruba]+HP is better” and “[o]ur view of the synergies”); JX 312 at 3, 5 (resulting write- up stressing “[w]e believe that there are meaningful revenue synergies” because “[c]ombined, the companies would have a leading converged campus solution” and significant “cross-selling” opportunities).

18 On September 15, 2014, HP kicked off its due diligence. From that point on,

management representatives at various levels met on a series of occasions.78 The speed and

intrusiveness of HP’s diligence surprised Orr.79 In a discussion with Neri on September 18,

Orr stressed that for a transaction to occur, HP would “have to pay a very compelling price

that reflects both the significant upside potential that we have ahead of us, and the strategic

value of Aruba . . . . This means that this deal will not happen at a conventional M&A

premium.”80

E. No Other Strategic Buyers Show Interest.

On September 25, 2014, the Aruba Board authorized Qatalyst to contact other

potential buyers to gauge their interest. Qatalyst developed a preliminary list.81 The Board

instructed Qatalyst to focus on “a limited number of third parties with financial

wherewithal and a strategic interest in mobile technology” that would enjoy “compelling

synergies” so that the buyer could be “competitive with any potential proposal from

[HP].”82 The Board instructed Qatalyst not to contact any private equity firms, believing

78 See JX 319 at 2-5 (emails coordinating meetings between Aruba and HP management); Proxy at 48-49. 79 JX 319 at 6 (Aruba talking points prepared for Orr: “I was frankly surprised by the breadth and scope of the information request, as it was not consistent with the message I thought you conveyed to me when we last spoke.”). 80 Id. 81 JX 229 at 1 (minutes); accord JX 327 at 3 (Qatalyst board deck). 82 JX 229 at 1-2.

19 that “given the Company’s volatile revenues and unpredictable cash flows and the potential

for synergies between Aruba’s business with a strategic acquiror, private equity firms

would not be competitive in their potential valuations.”83

Qatalyst identified thirteen “Selected Potential Partners.”84 Between September 29

and October 4, 2014, Qatalyst approached five of them.85 By October 9, each had

declined.86 Boutros explained that “[i]t was very clear that none of them had any interest

in an acquisition” and that it had “nothing to do with price.” 87 Orr concluded that “[n]ow

our only (but strong) weapon is to say we go alone.” 88 Boutros was “not at all troubled by

that,” observing that it was “what we expected anyway.”89

83 Id. at 2; accord Proxy at 48; Warmenhoven Tr. 259 (testifying Aruba did not contact financial bidders because “they like steady-state businesses, cash-cow businesses, and, you know, something that’s just this volatile doesn’t fit that profile” and because “they’re what I will refer to as bottom-feeders,” describing them as “generally low-ball bidders, very small premiums”). 84 JX 757 at 17 (draft Qatalyst board deck). 85 JX 760 (Qatalyst “Outreach Summary”); see Warmenhoven Tr. 256-58, 329-31 (describing outreach); Orr Tr. 440-42 (same). 86 JX 760 at 2 (chart showing status of each investor as “Declined to Proceed”); Proxy at 48-49. 87 Boutros Dep. 219; accord Warmenhoven Tr. 260 (“[I]t was definitive noninterest.”); id. 329-30 (agreeing potential bidders provided “feedback” that they “had no strategic interest in acquiring Aruba” and that it “[h]ad nothing to do with the price”). 88 JX 341 at 1. 89 Id.

20 F. The Parties’ Initial Valuations

In early October 2014, Aruba provided HP with a set of internal projections based

on figures from Aruba’s revised three-year strategic financial plan, which Aruba had

prepared in the ordinary course of business in June (the “June Plan”).90 In August, Galvin

and his finance team had updated the June Plan to incorporate the effects of Project

Greyhound and to make the plan more conservative. They had reduced the anticipated

revenue growth rate for 2016 and 2017 and adopted more conservative assumptions for

bookings, gross margin, and operating margin.91 In September, the Aruba team reviewed

the numbers with Qatalyst, describing the plan as a “medium” case that was “more

moderate” than the June Plan.92 By early October, the Aruba team and Qatalyst had created

a more bullish set of projections that forecasted revenue consistent with the June Plan (the

90 See Warmenhoven Tr. 219-20; id. at 313-14 (confirming Galvin prepared the June Plan “in the spring of 2014, as part of the normal operations” of the Company); Galvin Tr. 553-54 (discussing development of “top-down” projections incorporating “broader strategic assumptions” discussed by the Aruba Board regularly at the “June meeting”). See generally PTO ¶¶ 110-14. 91 See JX 297 at 2 (email from Galvin to member of the finance team: “I think we need to make this more conservative than the 3 [year] plan.”); see also Warmenhoven Tr. 314-15 (discussing Aruba’s ordinary process of preparing a yearly operating plan in September); Galvin Tr. 586 (acknowledging September figures adjusted margins, reflecting “[t]he overall impacts” of Project Greyhound). 92 JX 315 at 1 (email from Galvin to Qatalyst banker).

21 “October Projections”).93 Using the October Projections and a discounted cash flow

methodology, Qatalyst derived a valuation range for Aruba of $23.50 to $31.08 per share.94

HP’s internal deal team used the October Projections to prepare a discounted cash

flow valuation of its own. HP estimated Aruba’s value as a standalone company at $18.76

per share.95 But HP also estimated that a transaction between Aruba and HP would generate

$1.4 billion in revenue synergies and another $300 million in cost synergies. 96 With

synergies, the team estimated that the pro forma value of Aruba could be as high as $32.05

per share.97

G. HP Begins Recruiting Orr.

While the HP deal team was internally developing its pricing parameters, the senior

members of the team continued their discussions with Orr. Neri and Orr had a “pretty open

dialogue,” and Orr “remain[ed] positive about [HP’s] approach.”98 Neri understood from

Orr that Aruba was “not running a sales process,” and Orr made no effort to “postur[e]

about trying to pin [HP] against someone else.”99

93 Proxy at 64; see also Warmenhoven Tr. 315; Galvin Tr. 587. 94 JX 342 at 4 (internal Qatalyst emails attaching draft discounted cash flow model). 95 JX 345 at 20. 96 Id. at 15-16. 97 Id. at 20. 98 JX 348 (email from Neri to Whitman). 99 Johansson Dep. 112.

22 In early November 2014, HP’s Global Head of Corporate Development, Joakim

Johansson, met with Orr. In an email to Qatalyst and other Aruba executives, Orr recounted

that Johansson let him “know clearly that, post combination, they expect me to run the

whole networking business.”100 Orr said that Johansson wanted “to look me in the eye and

see that I have no objection. I told him I have no objection.”101

HP’s solicitation of Orr violated the terms of a confidentiality agreement that HP

had entered into with Aruba on October 2, 2014. That agreement contained a non-solicit

provision, which stated:

HP hereby agrees that, except to the extent expressly authorized by the board of directors of the Company (or any authorized committee thereof) in advance, neither HP nor any of its Representatives acting on its behalf will directly or indirectly have any formal or informal discussions, or directly or indirectly enter into any agreement, arrangement or understanding (whether or not binding), with any director, officer or other employee of the Company relating to (i) any retention, severance or other compensation, incentives or benefits that may be or become payable to any directors, officers or employees of the Company in connection with the Transaction or following the consummation thereof, or (ii) any directorship, employment, consulting arrangement or other similar association or involvement of any directors, officers or other employees of the Company with HP or any of its businesses or operations following the consummation of a Transaction.102

The Aruba Board had not authorized HP’s solicitation of Orr.103

100 JX 349 at 1. 101 Id.; see also Orr Tr. 462-66 (confirming solicitation and that “part of the reason they are willing to pay a premium for the Aruba price is for my leadership”). 102 JX 3 § 8. 103 Orr Tr. 463-64 (agreeing Aruba Board had not authorized discussion).

23 H. The HP Board Balks.

On November 6, 2014, the HP deal team asked Whitman to approve paying up to

$3 billion for Aruba, or $26.66 per share.104 In its presentation to Whitman, the deal team

scaled back the synergies slightly, with revenue synergies of $1.26 billion and cost

synergies of $295 million.105 The pro forma, with-synergies discounted cash flow value of

Aruba declined to $31.17 per share.106 Whitman backed the acquisition; the next step was

to obtain authority from HP’s board of directors (the “HP Board”).107

On November 20, 2014, Aruba announced its earnings for the first quarter of 2015.

Management described an “outstanding quarter” that included “[r]ecord revenues” that

exceeded “the top end of [their] guidance range.”108 Management also reported that Project

Greyhound had improved margins, “with non-GAAP operating margin growing to

21.8%.”109 But Aruba also announced a range of revenue guidance for the second quarter

of 2015 that was 1% lower at the midpoint than the pre-announcement analyst consensus.110

104 JX 350 at 4 (internal deck prepared for presentation to Whitman). 105 Id. at 14-15. 106 Id. at 20. 107 JX 356 (internal deck prepared for HP Board subcommittee). See generally Johansson Dep. 22-24 (discussing HP approval process). 108 JX 638 at 3 (transcript from Aruba earnings call). 109 Id. at 4. 110 JX 363 (internal HP email transmitting Barclays analyst report).

24 Analysts fixated on the lowered guidance. Although Aruba management explained that

they were trying to be prudent,111 Aruba’s stock dropped by 14%, closing at $18.82.112

The HP deal team saw the price drop as an opportunity to buy Aruba at a discount.113

Internally, HP acknowledged that Aruba’s results were “better than we expected,”114 that

they validated the case that the deal team had presented to the HP Board,115 and that “[t]he

softer guidance did not cause us to change our financial model.”116

111 JX 638 at 13 (Galvin stating Aruba was “just being prudent in a mixed environment” and “[n]ot getting ahead of our skis”); see also JX 355 at 1 (email from Galvin to Aruba Board: “our guide is shaping up to be more cautious than our strong results”). 112 JX 245. 113 See JX 357 at 1 (internal email among members of the HP deal team: “They are down 8% after hours though because of guidance below consensus for this quarter (although it looks like the range of guidance is still within expectations). Good time to pull the trigger and snap them up I would say.”); Hardegree Dep. 193 (observing that the stock price drop “probably did, on balance” put HP “in a more tacti[cally] advantaged position”). 114 JX 362 at 1. 115 JX 357 at 1 (Neri: “We told our story and we were on point.”); see also JX 360 (email from HP Senior Vice President of Corporate Development: “I thought the results were pretty good. . . . They blew past the revenue expectations for the quarter. But they guided soft, below consensus.”). 116 JX 780 at 2 (internal HP deck discussing November earnings announcement).

25 The HP Board, however, remained skittish after the Autonomy fiasco, and it was

not ready to authorize a bid. Neri told Orr that the HP Board had questions about the deal

and that it would take another two or three weeks to answer them.117

Orr felt the process had dragged on long enough, and he recommended that the

Aruba Board terminate discussions.118 With the Aruba Board’s backing, management

conveyed that Aruba was moving on.119 Aruba formally terminated discussions on

November 25.120

I. HP Engages Advisors And Continues Analyzing The Deal.

After Aruba terminated discussions, HP continued working on the deal. In late

November 2014, HP engaged McKinsey & Company to validate its business case for the

JX 328 (email from Orr to Aruba Board reporting that “Antonio [Neri] said that 117

[HP] needs another 2-3 weeks to address the board’s questions”). 118 Id. (“We have been in dialogue with [HP] since August 27, and have not received a proposal in all this time. . . . We cannot continue to wait for them.”); accord Proxy at 48- 49 119 JX 372 at 1 (email from Aruba General Counsel relaying conversation with Johansson during which she explained that “[w]e feel it’s time to suspend discussions” because “it has reached a point where we need to focus on running our business and not be distracted by discussions that did not seem to be progressing”); see also Orr Tr. 448-49 (“So we are basically saying, if you cannot get it done, let’s cool off. And when you are ready, contact me again.”). 120 See JX 328; JX 367 (email from Aruba to Qatalyst advising it had “decided to terminate discussions with [HP] at this time” and that it would “let [Qatalyst] know if discussions resume”).

26 acquisition. McKinsey concluded that HP could expect market share gains and revenue and

cost synergies that were in line with HP management’s estimates.121

In December 2014, HP engaged Barclays as its financial advisor.122 At the time,

Barclays had an existing relationship with Aruba, having worked with Aruba on a potential

convertible debt financing since June 2013.123 The debt financing had been ready to launch

in September 2014, but Aruba declined to move forward. The lead banker at Barclays

inferred that Aruba was considering a major M&A transaction.124 Barclays spent the next

three months trying to get a role representing Aruba, until they secured the engagement for

HP.125 Barclays also was one of two banks executing Aruba’s ongoing share repurchase

JX 383 at 1, 3 (McKinsey report projecting “share gains in line with what is in 121

the business case (2.6-3.1% share gain in wired vs. 3% in the business case and correspondingly 4.3-5.6% share gain in wireless vs. 7% in the business case)”). 122 See JX 387 (internal Barclays email dated December 19: “HP wants to have our views on value on Monday”); JX 389 (email from Barclays to HP on December 22 providing “the deck we plan to review on our 10am conference call”). 123 See PTO ¶¶ 51, 54; JX 224 (Aruba Board subcommittee minutes); Galvin Tr. 601. Evercore was also advising in connection with the potential debt financing. See JX 286 (email from Francis to Orr and Galvin discussing fiscal year 2014 results ahead of “meeting on the convertible market”). 124 JX 325 at 1 (email from Galvin relaying conversation wherein he told Barclays banker Aruba would not be executing the convertible offering and banker responded “in my business, when this happens, either an exec is leaving the company or a major M&A [transaction] is in process”). 125 See, e.g., JX 335 at 2 (email from Barclays banker to Aruba management: “[W]e’re hearing (from non-Aruba sources) that HP has reinitiated dialogue with you” and “[w]e want to be as helpful as we can.”); id. at 1 (describing phone call in which Barclays banker reminded Orr that “we helped you with the Barclay IT people, we walked away from [the] Ruckus IPO when Galvin said it was competitive . . . , [and] we have done a lot [o]f work to support you in the last 8 years”); id. (banker telling Orr that Barclays “would

27 program, which had been in place since June 2012.126 In February 2014, the Aruba Board

had authorized management to repurchase up to $500 million of Aruba’s common stock in

the open market at prices up to $25 per share.127 When approving the repurchases, the

Board made the following determination: “The recent trading price of the Company’s stock

on the NASDAQ Global Select Market has been depressed and the Board believes that the

trading price of the Company’s common stock may be undervalued . . . .” 128 Aruba

management suspended the repurchases in October 2014 because of the discussions with

HP.129 After terminating discussions with HP in November 2014, Orr recommended

resuming the repurchase program and buying shares worth up to $75 million, because “the

stock is underperforming.”130 Aruba resumed its repurchases at up to $25 per share,

believing its shares to be undervalued below that figure.131

be disappointed if the HP rumor is true and you do not work with us”); JX 384 (email between Barclays bankers: “I am trying to get back in front of the Aruba team. Seems like we have given up on the HP side.”). 126 PTO ¶ 53; JX 227 (minutes). See generally Galvin Tr. 594-96 (discussing share repurchase program). 127 JX 226 at 1-2 (minutes). 128 PTO ¶ 133. 129 See id. ¶ 136. 130 JX 328. 131 See JX 373 at 1 (internal Aruba emails discussing share repurchase program); see also Orr Tr. 448 (testifying that, in breaking off discussions, he considered that “[w]e have a stock buy-back program that . . . we have to resume”); Galvin Tr. 595-96 (discussing

28 Using consensus analyst estimates for Aruba’s standalone performance, Barclays

provided HP with a range of discounted cash flow values for Aruba as a standalone

company and compared them with pro forma values for Aruba that incorporated

synergies.132

Discount Rate Perpetuity Growth Rate 2% 3% 4%

Street Synergy Street Synergy Street Synergy

11% $19.31 $29.49 $20.86 $32.29 $22.85 $35.89

12% $17.61 $26.45 $18.80 $28.62 $20.30 $31.32

13% $16.21 $23.98 $17.16 $25.69 $18.32 $27.78

Barclay’s valuation work confirmed the HP deal team’s internal estimates.

J. HP Approaches Aruba Again.

Meanwhile, Aruba’s stock price remained stuck around $18-19 per share, and

analysts continued to criticize the Company. Orr felt that the analysts had soured on Aruba

and were complaining about everything.133 Galvin expressed frustration that analysts

the share repurchase and stating “[w]e had a standard set of language that, when we approached these things, that it may be depressed, it may be undervalued”). 132 JX 389 at 16-17 (Barclays deck). 133 Orr Tr. 454 (comparing analysts to “somebody [who] has decided to get out of a relationship” because “the other party can learn how to cook, clean the floor, you know, be available on the weekend, and so on. Something is always wrong.”); see also JX 406 at 1 (email from Orr: “we simply cannot miss Q2 and have the stock tanked further, risking employee morale and retention getting into an unrepairable state”).

29 seemed unwilling to “acknowledge the 6 very strong [quarters] in a row we’ve had.”134

One analyst suggested to Aruba management that the stock price was low enough to justify

an accelerated buyback, observing that if management did not pursue that option, an

activist might.135 A group of investors met with management and privately criticized the

Company’s performance.136

Against this backdrop, in late December 2014, Whitman and Neri invited Orr to

dinner.137 The meeting took place on January 21, 2015. Whitman told Orr that HP still

wanted to acquire Aruba.138 Orr responded positively and suggested that they try to sign

up a deal by early March. He told Whitman that Aruba had “over 2000 customers and

partners coming to Atmosphere 2015 in Vegas the week of March 1-5,” and “[i]t would be

silly not to announce it there.”139 Whitman “completely agreed,” observing “that, in her

experience, mergers need forcing function and let this be the one.”140

134 JX 378 at 2 (email among Aruba executives). 135 JX 374 at 1 (email from Jason Adler, analyst at William Blair & Company). 136 See JX 377 (email from Galvin recapping meeting). 137 PTO ¶ 154; JX 392 (correspondence between Orr and Neri scheduling dinner). 138 PTO ¶ 155. 139 JX 423 (email from Orr relaying conversation to Aruba management); see also Orr Tr. 524-25. 140 JX 423; see also Orr Tr. 527 (“We’re saying that let’s use this as a forcing function.”).

30 Whitman also told Orr that HP would not work with Qatalyst under any

circumstances. That same night, Orr told Warmenhoven that “Meg [Whitman] spoke with

conviction and emotion over dinner that they [Qatalyst] were guilty. Qatalyst will argue

the reverse, but it does not matter.”141 Orr concluded that “if we don’t insert [a] buffer

person, our negotiation will suffer severely.”142 To resolve the problem, Warmenhoven

doubled back to the solution he had previously dismissed: using Francis. Warmenhoven

reminded Orr that Francis “is in a new firm and not conflicted, and Meg [Whitman] knows

and (I think) trusts him.”143 He offered to call Whitman to see if involving Francis would

be acceptable.144

141 JX 412 at 1. Warmenhoven disputed at trial whether he correctly understood Orr to be saying that Whitman felt that Qatalyst, rather than Autonomy, was guilty. His deposition transcript indicates that his counsel planted this idea. See Warmenhoven Tr. 336-40; Warmenhoven Dep. 141-43 (“[Aruba counsel]: If you are going to move off of this, would you mind if I clarify one point? [Verition counsel]: Okay, that’s fine. [Aruba counsel]: Mr. Warmenhoven, if you don’t mind my interjecting -- The Witness: No, I know where you’re going. . . . That’s how I interpreted it at that time. That’s not right. Meg [Whitman’s] conviction was that Autonomy was guilty -- not Qatalyst -- that Autonomy was guilty of fraud.”). Aruba’s counsel should not have influenced the substance of Warmenhoven’s testimony during his deposition. Nor should counsel have been reshaping Warmenhoven’s recollection of the facts. 142 JX 412 at 1; see also Warmenhoven Tr. 247-49 (discussing JX 412); Orr Tr. 449- 51 (same). 143 JX 412 at 1. 144 Id.; accord Warmenhoven Tr. 249-51; Orr Tr. 451 (“Obviously, the top of mind is the person that I started out with, and that has been a trusted advisor to the company and to me, and that was Stu [Francis].”).

31 On January 23, 2015, Warmenhoven updated the Aruba Board. He reported that

Aruba should receive a formal offer letter soon. He explained that “[w]e do have a bit of

an issue and that is our choice of advisors. [HP] is very anti [Qatalyst].”145 He reminded

the Board that the Company already had a signed engagement letter with Qatalyst, so Aruba

would have to pay Qatalyst regardless, but he suggested hiring Francis to handle the

negotiations. Warmenhoven pointed out that “Evercore is new in the tech sector, so they

may be willing to do a deal at ¼% just to get a deal done that they can brag about

publicly.”146

Warmenhoven spoke with Whitman that same day. She told him that “Qatalyst,

Frank [Quattrone] & George [Boutros] are not welcome in the negotiations. The issue is

bigger than Autonomy and goes back to EBay & Yahoo.”147 Whitman described Boutros

as “evil.”148 But Whitman said that she would happily negotiate with either Francis or

145 JX 413. 146 Id.; accord JX 414 (email from Orr to Warmenhoven: “I think Qatalyst will not be willing to carve out fees for Stu [Francis].”); JX 426 at 2 (internal Aruba email summarizing discussions with Francis concerning fees: “He started at Evercore 5 months ago and is their first tech sector person. They want this deal to establish a presence in tech.”); see also PTO ¶ 46 (“J. Stuart Francis is a Senior Managing Director of Evercore. Mr. Francis joined Evercore in the summer of 2014.”). 147 JX 420 (email from Warmenhoven recounting discussion to Aruba Board and management). 148 Id.

32 Warmhoven, as long as Qatalyst stayed “in the back room.”149 She also wanted Qatalyst

“squeezed a bit” on their fees.150

Aruba retained Evercore that day. Francis reported to his partners that the “[d]eal

timing is to try to sign a merger agreement and announce by mid February.”151 His partners

were thrilled. One replied: “Truly amazing! This is a franchise transaction! Well done!”152

Another responded: “This is franchise defining. Well done, and it shows the power of

loyalty, which you have always eschewed! [sic]”153 A third offered: “Just remarkable,

Stu[.] What a coup! Would be, as you say, a dynamic advance for Evercore in The

Valley.”154

K. Qatalyst Tries To Repair Its Relationship With HP.

Qatalyst was as crushed by the news as Evercore was elated. When Orr told Boutros,

he was “so emotional, defensive AND offensive (to Meg [Whitman]) that he hardly let me

149 Id.; accord Warmenhoven Tr. 252-53 (Whitman told Warmenhoven “I don’t care who you get, but it can’t be Qatalyst” and that “I’m not going to take into my boardroom a deal proposed by Qatalyst”); id. at 341-45; Orr Tr. 530. 150 JX 432 at 1 (email among Aruba management discussing negotiation of Qatalyst’s fees). 151 JX 422 at 1 (email among Evercore senior bankers). 152 JX 427. 153 Id. 154 JX 428; accord JX 439 (congratulatory emails from two additional senior bankers).

33 talk.”155 Quattrone asked Warmenhoven to intervene with Whitman on his behalf, and

Warmenhoven asked Whitman to meet personally with Quattrone. 156 Quattrone then sent

Whitman an email of his own:

I was very surprised and disappointed to learn from Dan Warmenhoven today that you recently expressed very strong negative feelings about our firm, some of our people (including me) and our current representation of Aruba. I would greatly appreciate the opportunity to speak or meet with you at your earliest convenience to understand from you directly what your concerns are and give me the opportunity to address them. . . . [W]hile our loyalties are always to our client on any assignment, I am confident we can address your concerns, play a constructive role and engage with your team in a professional manner.157

When Whitman did not respond, Warmenhoven followed up the next day, vouched for

Qattrone, and expressed confidence that “if you two could ‘clear the air’ [then] Frank

[Quattrone] and [Qatalyst] could be constructive participants in getting this deal done.”158

In an email to Aruba management, Warmenhoven explained why the dispute with

Whitman was so important to Quattrone:

The issue is not Aruba. It is about the [Qatalyst] brand . . . . If word spreads that they were tossed from this deal because HP will not engage with them on any M&A transaction, that creates a big issue for them. . . . Frank

155 JX 426 (email from Orr to Aruba management updating on negotiations with Qatalyst). Orr expressed his opinion that Boutros “reacted so strongly cause (1) pride of their brand (2) reaction to Stu [Francis] coming in[to] the picture [and] (3) protecting their fees.” Id. He noted that “[n]one of the above related to the benefits of [A]ruba!” Id. 156 JX 430. 157 JX 434 at 1. 158 JX 436 at 1. See generally Warmenhoven Tr. 344-45 (recounting process of attempting to ingratiate Qatalyst with HP).

34 [Quattrone] wants to save his firm . . . . The relationship, or lack thereof, between [Qatalyst] and HP / Meg [Whitman] is now their focus.159

Orr wondered “how much time we allow” before saying “sorry, [Qatalyst]. We need to

protect our transaction. [W]e cannot worry about your brand!”160

Whitman finally spoke with Quattrone on January 29, 2015.161 She repeated the

concerns she had relayed to Warmenhoven and again emphasized that “we cannot have

[Qatalyst] as the primary representation for Aruba interacting with us and with our

board.”162 On February 1, Aruba formally retained Evercore for a contingent fee equal to

0.25% of the deal value.163

L. HP’s Initial Proposal

By the time Whitman and Neri had dinner with Orr at the end of January 2015, HP

had worked with Barclays to analyze a range of prices from $23 to $26.50 per share. Based

on this analysis, HP anticipated making an “opening bid” of $24.00 and received board

159 JX 437 at 1; see also JX 438 (Quattrone sending Warmenhoven what he described as a cordial rejection by Whitman of a social invitation as evidence that she had no bias against Qatalyst and was bluffing to gain leverage in the negotiations). 160 JX 440 at 1 (email from Orr to Aruba management). 161 JX 447 at 1 (email from Whitman to Quattrone dated January 28, 2015: “Will call you tomorrow.”). 162 Whitman Dep. 127-28. 163 PTO ¶ 47; see also JX 5 (executed engagement letter); JX 232 (Aruba Board minutes authorizing Evercore engagement).

35 approval to go up to $25.00.164 After the dinner and Orr’s enthusiastic response, HP revised

its strategy. On January 31, Johansson called Orr to notify him that HP was preparing a

written offer.165 Orr remained eager. He suggested accelerating the timeline and “getting a

deal announced by [Aruba’s] earnings on Feb 26.”166 He offered that to achieve that

timeline, Aruba “would respond as early as Monday/Tuesday of this week.”167

Later that day, HP sent Aruba a written indication of interest for a cash transaction

at $23.25 per share, for an aggregate valuation of $2.563 billion.168 The price per share

represented a 40.2% premium to Aruba’s closing price on the previous day and a 35.1%

premium to the stock’s thirty-day average price.169 The price exceeded Barclays’ latest

stand-alone discounted cash flow valuations of Aruba based on analyst estimates, which

ranged from $17.47 to $22.61 per share.170 The price was seventy-five cents below the

opening bid of $24.00 per share that HP had been considering at the beginning of the

month. The offer was $3.80 per share below the low end of Barclays’ most recent pro forma

164 JX 398 at 1-2 (cover email for internal deck seeking HP Board approval); JX 805 at 12 (internal HP deck). 165 JX 455 at 1 (email from Orr to Aruba management and Qatalyst recounting call). 166 JX 454 (internal Barclays email summarizing Johansson’s relay of the call). 167 Id. 168 JX 452 at 3 (“Indicative Non-Binding Proposal”). 169 Id. at 3-4. 170 JX 782 at 4 (Barclays deck).

36 discounted cash flow valuations that included synergies, which valued Aruba at between

$27.05 and $37.61 per share.171

M. Aruba Responds.

The Aruba Board initially met to consider HP’s proposal on January 31, 2015. The

directors decided to defer any detailed consideration of the proposal until after management

presented revised projections and Evercore and Qatalyst had an opportunity to analyze

them.172

On February 2, 2015, the Aruba Board met again. Management presented an

updated version of the October Projections, prepared in the ordinary course of business,

that reflected the Company’s performance to date (the “February Projections”).173 Qatalyst

reported that they had spoken with a sixth potential strategic partner who also was not

interested in acquiring Aruba.174

While the Aruba Board was meeting, a new analyst report criticized the Company,

and the stock price fell.175 Aruba’s General Counsel forwarded the report to Evercore and

171 Id. 172 JX 461 at 2 (internal Evercore email relaying what Aruba’s General Counsel anticipated to be the timeline of events). 173 JX 233 (minutes); JX 910 at 3 (Aruba 8-K comparing October Projections and February Projections); Galvin Tr. 558-63 (discussing process of extending ordinary course six-month budgeting projections out to three years to create February Projections). 174 Proxy at 51. 175 JX 466 at 1-2 (internal Aruba email distributing report).

37 Qatalyst, telling them that they should inform HP that “today’s stock price does not reflect

reality.”176 She noted that Aruba was going to beat its guidance for the quarter, but that “no

one knows that yet.”177 Francis was similarly concerned that the negative analyst reports

had depressed the stock price and created a buying opportunity for HP.178 But he also

worried that HP would disengage if Aruba waited to release its quarterly results first and

then negotiated from a place of strength; to get a deal done, it was “now or never.”179

On February 4, 2015, the Aruba Board met again and received a presentation from

Evercore on valuation.180 Using the February Projections, Evercore generated the

following discounted cash flow valuation range for Aruba on a standalone basis:

Perpetuity Growth Rate

4.0% 4.3% 4.5% 4.8% 5.0% 10.5% $23.97 $24.80 $25.71 $26.69 $27.76 11.0% 22.03 22.73 23.49 24.31 25.20 WACC

11.5% 20.35 20.95 21.60 22.29 23.03 12.0% 18.88 19.40 19.95 20.54 21.17 12.5% 17.58 18.03 18.51 19.01 19.56

176 Id. at 1. 177 Id. 178 JX 470 at 1 (email from Francis to other Evercore bankers: “I am thinking about the analyst downgrades and if there [is] anything we can do.”). 179 Id. 180 JX 794 at 14 (Evercore deck).

38 Based on these figures, the Board authorized a counteroffer at $29 per share.181

Evercore conveyed the counteroffer to Barclays. Evercore emphasized that “now is

not an opportune time for a sale, given the stock is at a 52-week low.”182 Evercore also told

Barclays that “[t]he low stock price reflects a misperception in the market that [Aruba] will

miss its quarter. In fact, [Aruba] will beat consensus and have good guide.”183 Barclays

responded that the counteroffer was “not even within the realm of possibility.”184

N. HP Counters At Effectively The Same Price.

After receiving Aruba’s counter, HP management caucused with its advisors. They

prepared talking points which stated, contrary to HP’s synergy-based valuations, that HP

did “not have the ability to reach anywhere near” Aruba’s counteroffer.185 The talking

points also focused on timing, stressing that “[i]f we don’t seize the opportunity now, there

are many external pressure points that impact HP’s ability to do a transaction with [Aruba]

in the foreseeable future.”186

181 JX 234 (minutes); accord Warmenhoven Tr. 261 (testifying the Aruba Board came to the figure because “[w]e wanted to reaffirm that we thought there was great value there, meaning they should bid higher, but we felt like if we put a 3 in the first digit and started at 30, that they might conclude that’s too big a gap to close and stop discussion”). 182 JX 477 at 1 (Barclays email relaying counteroffer to HP). 183 Id. 184 Id. 185 JX 484 at 1. 186 Id.

39 On February 7, 2015, Barclays told Evercore that any price increase would be on

the scale of “quarters, not dollars.”187 Evercore’s reaction “was pretty constructive.”188

Evercore “emphasized that [Aruba would] like to announce deal at or before the [Aruba]

earnings announcement” because Aruba was “afraid stock runs like Ubiquiti’s did which

could make the deal more challenging from the [Aruba] perspective.”189

During the call with Barclays, Evercore also explained that Aruba’s share

repurchases had rendered outdated the share count contained in its public disclosures.

Aruba in fact had 119.1 million shares outstanding, which was 4.5 million fewer than

reported. Evercore told Barclays that HP should increase its price per share to account for

the change.190

On February 9, 2015, Barclays recalculated the deal price based on the new share

count. With fewer shares, the same aggregate consideration of $2.563 billion resulted in a

price per share of $23.89.191 This was still below the opening bid of $24.00 per share that

HP had contemplated before Whitman and Neri had dinner with Orr. That opening bid had

187 Proxy at 51. 188 JX 491 (email from Barclays to HP recounting exchange). 189 Id. Ubiquiti announced financial results on February 5, 2015 that beat the consensus estimates, and its stock rose by 14%. See JX 639 (Ubiquiti earnings transcript); JX 902 (Ubiquiti 8-K announcing results). 190 JX 491; accord JX 488 at 4 (internal Evercore emails confirming change in shares outstanding and suggesting repurchases caused the change). 191 JX 798 at 2 (Barclays slide entitled “Impact of New Share Count and B/S Data on Offer Price”).

40 equated to aggregate deal consideration of approximately $3 billion. With the lower share

count, that same $3 billion enterprise value generated a price of $24.67 per share.192 HP

decided to tell Aruba that its best and final bid was $24.67 per share, a figure that yielded

the same enterprise value that HP originally intended to offer as its opening bid. Internally,

HP described $24.67 as “the new $24.00,” because the price merely adjusted for the change

in Aruba’s public share count.193

On February 9, 2015, Barclays communicated the counter of $24.67 per share to

Evercore. The Aruba Board immediately met and authorized a counter at $25.00.194 Orr

conveyed Aruba’s ask of $25 per share to Neri.195 Orr also spoke to Whitman. Neither

budged on price.196 Later that day, HP sent Aruba a “Revised Indicative Non-Binding

Proposal” that proposed a cash price of $24.67 per share.197 HP described the bid as its

“best and final offer.”198 The proposal represented a 51.6% premium to Aruba’s closing

192 Id. 193 JX 497 (internal HP email from Johansson); accord JX 805 at 12 (timeline showing “02/08 – HP offer $24.00 and present as best and final. [Aruba] provide new shares outstanding and cash information supporting paying $24.67 at the same enterprise value.”). 194 Proxy at 52. 195 Id.; accord JX 494 at 2 (email from Neri to Whitman summarizing conversation). 196 JX 496 at 1 (email from Francis to Aruba management transmitting revised proposal: “They did not move up just as Meg [Whitman] said to you.”). 197 Id. at 3. 198 Id.

41 price on the previous day and a 48.9% premium to Aruba’s thirty-day average trading

price.199

O. Aruba Accepts.

On February 10, 2015, the Aruba Board met to consider HP’s revised proposal.

Evercore, Orr, and Warmenhoven reported that the revised price resulted from considerable

negotiation. They believed it represented HP’s best and final offer. Qatalyst advised that it

was unlikely that any other party would offer a higher price.200 The Board discussed “the

recent weakness” in Aruba’s share price and considered the alternative of continuing as a

standalone company.201 The Board also considered standing firm on its ask for $25 per

share. Ultimately, the Board decided to accept HP’s offer of $24.67 per share.202

On February 18, 2015, the Aruba Board considered whether to permit HP to speak

with Orr and other members of Aruba’s senior management about employment

opportunities.203 No one disclosed to the Board that HP previously had made clear to Orr

that they wanted him to run Aruba after the merger, or that Orr had told HP that he was

199 Id. 200 JX 235 at 1-2 (minutes); accord Proxy at 52; Boutros Dep. 219-20 (describing lack of interest from other strategic bidders); Warmenhoven Dep. 104 (same). 201 JX 235 at 1-2; see also Warmenhoven Tr. 265-66 (testifying prospects of standing alone “were looking extremely challenging” because the “core market” was “going to zero growth” and Aruba was experiencing difficult “market conditions” and “competitive positioning”). 202 JX 235 at 1-2; accord Proxy at 52; Warmenhoven Tr. 263. 203 JX 237 (minutes).

42 willing to do so. The Board consented to the discussions and waived the non-solicitation

provision in the confidentiality agreement with HP.204

With approval from the Aruba Board in hand, HP conducted additional due

diligence. Evercore approached this phase as an extended audition for future work from

HP. In late February 2015, Francis reported to his partners about “a really interesting

negotiating dinner at Meg’s house Thursday night” and that “it was fun to be the only

banker in the room to help both sides think through some issues.”205 A senior Evercore

banker responded “That’s HUGE! Meg is going to be very active. . . . Would be a great

new relationship.”206 Francis bragged about having effectively acted as HP’s advisor:

Agreed… I think we made a pretty good impact from an advisory perspective, and she and I have known each other a long time socially through [P]rinceton events and when our kids were at Menlo school…please pardon the “pat on the back” nature of this comment, but after the meeting one of the people on our side said we had done a “masterful” job of taking [M]eg [Whitman] through the issues as if we were her advisor…let’s hope that can help us get some traction in the future with her…207

Rather than acting as a banker for Aruba, Evercore acted as a banker for the deal.

204 Id.; accord Proxy at 52. See generally Orr Tr. 466-70 (discussing negotiation of new employment agreement with HP). 205 JX 505 at 1 (internal Evercore email). 206 Id. 207 Id. (reproduced in full, ellipses in original); see also Reisenberg Dep. 248-54 (discussing JX 505).

43 P. The Deal Leaks.

On February 25, 2015, one day before Aruba was scheduled to announce its

earnings, Bloomberg News ran a story on the merger.208 Internally at Qatalyst, Boutros

speculated that HP had leaked the news so that Aruba’s “results and subsequent stock price

reaction won’t be easy to measure.”209 Aruba’s stock price jumped from $18.37 to $22.24

on the news.210 An analyst issued a report positing that, in light of the synergies from the

merger, a deal price of “$28 or a premium of 25% from today’s close is reasonable [for

Aruba].”211

On February 26, 2015, Aruba released its second quarter results. The Company beat

analyst expectations and hit management’s guidance.212 Analysts called the results

“impressive”213 and “[b]etter-[t]han-[e]xpected.”214 One praised Aruba for “[a]nother

208 JX 866 (article). 209 JX 510 (internal Qatalyst email); see also Boutros Dep. 206-07 (“[N]ow the stock is going to trade based on takeover speculation and no more based on fundamentals, pure fundamentals. There is a lot of noise.”). 210 Proxy at 53. 211 JX 152; accord JX 155 (JMP Securities report stating “[w]e believe the acquisition would be a logical step for HP” and anticipating “that Aruba’s board would look for an acquisition price in the $23-$27 range”). 212 JX 514 at 1, 3 (internal Barclays email discussing results). 213 JX 156 (Macquarie Research report). 214 JX 159 (William Blair report).

44 [s]trong [q]uarter.”215 An analyst at Citi doubted that Aruba would be willing to sell “at

prices near the current trading level” and suggested that Aruba would not take “less than

$30/share.”216

On February 27, 2015, Aruba’s stock closed at $24.81 per share, above the merger

price.217 That evening, the Aruba Board met to discuss how to respond. Qatalyst advised

the Board that the deal still represented a 28% premium to Aruba’s average trading price

and a 44% premium to its thirty-day average price.218 Qatalyst also explained that the 10%

price bump that Aruba enjoyed after its earnings release matched the price performance of

peer firms after reporting similar results.219

Despite Qatalyst’s report on the market reaction to peer companies reporting similar

results, the Aruba Board concluded that “the higher trading price was primarily being

driven by market speculation of a transaction, and not by changes in the fundamentals of

the business.”220 The Board discussed renewing its request for $25 per share but rejected

the idea in favor of using the stock price as leverage to insist on a lower termination fee

215 JX 157 (Jeffries report). 216 JX 168 at 3-4 (Citi report). 217 Proxy at 53. 218 JX 517 at 3 (Qatalyst board deck). 219 Id. at 6. 220 Proxy at 53.

45 and stronger deal terms.221 After the meeting, Aruba notified HP that it would not “be

asking for a higher price.”222

Q. The Final Board Approvals

On February 28, 2015, the HP Board met to consider the definitive merger

agreement. Barclays prepared a discounted cash flow analysis based on Aruba’s

“management case”—the February Projections. The analysis produced a valuation range

of $26.20 to $33.64 per share for Aruba on a standalone basis.223 With synergies, Barclays

valued Aruba at between $27.53 and $39.69 per share.224 Not surprisingly, Barclays opined

that the deal price of $24.67 per share was fair to HP and its stockholders. The HP Board

approved the merger agreement.225

On March 1, 2015, the Aruba Board convened to consider the definitive merger

agreement. Both Qatalyst and Evercore gave valuation presentations. Qatalyst’s discounted

cash flow valuation ranged from $23.23 to $26.76 per share.226 Evercore’s discounted cash

flow valuation ranged from $21.12 to $29.78 per share.227 Both firms opined that the deal

221 Id. 222 JX 520 (internal Barclays email). 223 JX 806 at 14 (Barclays deck). 224 Id. at 15. 225 JX 238 at 1-3 (HP Board minutes). 226 JX 811 at 19 (Qatalyst deck). 227 JX 812 at 14 (Evercore deck).

46 price of $24.67 per share was fair to Aruba and its stockholders. The Aruba Board approved

the merger agreement.228

Getting the Aruba deal signed was a key step in Qatalyst’s effort to get back in HP’s

good graces. Afterwards, Quattrone reported to his partners that Whitman had “asked [Orr]

to pass along the message to us that there is now a path towards ‘rehabilitation’ of our

relationship.”229 Quattrone proposed to contact Whitman and “get her assurance that if we

don’t send the blast email” announcing Qatalyst’s role in the deal, then Qatalyst would

have “a clean slate going forward” with HP.230 His partners supported this approach, so

Quattrone sent Whitman an email asking for a “clean slate.”231 Quattrone also added “an

appeal to your fairness,” noting that “we have already been embarrassed and our business

damaged by what has already occurred.”232

228 Proxy at 53. 229 JX 521 at 1; see also Boutros Dep. 270 (testifying he understood Whitman to be amenable to “resolution of the issues she had and the feelings she had towards Qatalyst”). 230 JX 521 at 2. 231 JX 522 at 2 (“[W]e do want to have a good relationship with you, and we are willing to explore alternatives to our normal course of business in marketing this deal, if I can have your assurance that by doing so, we will have a ‘clean slate’ with you and HP immediately going forward.”). 232 Id. at 3. Aruba and HP announced the merger before Whitman responded. Qatalyst held off on publicity efforts pending her response. On March 3, 2015, Whitman told Quattrone he should “go ahead and do whatever communication feels right to you under the circumstances.” Id. at 1-2. On March 4, Qatalyst sent out a blast announcing its role in the deal. JX 541 at 2. Whitman also told Quattrone that she was willing to “get together . . . and discuss the path forward.” JX 522 at 2. The meeting eventually took place in October 2015. Afterwards, Quattrone sent Whitman a follow-up email which

47 R. Stockholder Approval

On March 2, 2015, Aruba and HP formally announced the merger.233 The final

merger agreement (i) prohibited Aruba from soliciting competing offers and required the

Aruba Board to continue to support the merger, subject to a fiduciary out and an out for an

unsolicited superior proposal; (ii) included a $90 million termination fee; and (iii) provided

a drop-dead date of February 28, 2016.234 Orr and Keerti Melkote, Aruba’s co-founder and

Chief Technology Officer, entered into voting agreements supporting the merger.235

No competing bidder emerged. On May 1, 2015, Aruba held a meeting of

stockholders to consider the merger.236 Under Delaware law, unless a corporation’s

constitutive documents impose a higher voting requirement, a merger requires the approval

of a majority of the voting power represented by the corporation’s outstanding shares.237

Approximately 80.88% of Aruba’s outstanding shares were represented at the meeting

either in person or by proxy. Approximately 98% of those shares voted in favor of the

emphasized that he was “look[ing] forward to next steps in building a strong long-term relationship between our firms.” JX 572 at 1-2. 233 JX 907 (8-K announcing signing). 234 Id. at 3-4, 73-78. 235 Id. at 4. 236 Proxy at 1. 237 See 8 Del. C. § 251.

48 merger. 238 As a result, approximately 80% of the outstanding shares voted in favor of the

merger, clearing the statutory requirement. The transaction closed on May 18, 2015.239

S. This Litigation

At the effective time of the merger, petitioners Verition Partners Master Fund Ltd.

and Verition Multi-Strategy Master Fund Ltd. (jointly, “Verition”) owned in the aggregate

2,288,234 shares of Aruba’s common stock.240 Verition filed this appraisal proceeding on

August 28, 2015. The parties engaged in discovery for nearly two years. During discovery,

Aruba’s counsel took several aggressive and credibility-impairing positions.

On May 5, 2016, Verition noticed Whitman’s deposition. Aruba promptly filed for

a protective order seeking to limit the deposition to three hours. Aruba asserted that

Whitman had “limited involvement in HP’s acquisition of Aruba,” that “Aruba and HP are

aware of nothing in the vast discovery in this case to suggest there were any

communications between Ms. Whitman and any representative of Aruba negotiating

pricing terms of the Aruba deal,” and that her view towards Qatalyst was immaterial

because it “has no bearing on the fair value of Aruba as a stand-alone company.”241 While

238 Aruba Networks, Inc., Current Report (Form 8-K) (May 1, 2015). Although the parties did not submit this filing as an exhibit, the Delaware Supreme Court has taken judicial notice of public filings in an appraisal case and relied on them for the truth of their contents. See, e.g., DFC, 172 A.3d at 351 n.7. 239 JX 914 at 2 (8-K announcing closing). 240 See PTO ¶¶ 11, 17. 241 Dkt. 56 at 11-13 (motion opening brief).

49 perhaps technically correct, these representations created a misleading picture of

Whitman’s involvement. Verition submitted exhibits in response to the motion that made

the assertions look silly, and the evidence at trial subsequently reinforced their misleading

character. I denied the motion for protective order and required Whitman to testify from

day to day until the deposition was completed.242

The parties next clashed when Verition moved to strike aspects of the report of

Kevin Dages, Aruba’s valuation expert. In his report, Dages relied on an email exchange

between Aruba management and the Company’s lawyers at Wilson Sonsini Goodrich &

Rosati, P.A. for information about stock-based compensation. During Dages’s deposition,

it became evident that Aruba had withheld the communication as privileged. Dages then

attempted, unpersuasively, to suggest the citation had been a typographical error and that

he had actually drawn the information from a different source that Aruba had produced.

Complicating matters further, Verition showed that, during the deposition of another

witness (Galvin), Aruba’s lawyers engaged in substantive discussions about the email with

the witness during a break. At the hearing, I expressed significant concern about Aruba’s

discovery conduct, but I concluded that I could address the matter by weighing the evidence

rather than by striking a portion of Dages’s report.243

242 Dkt. 65. 243 See Dkt. 125 (transcript and rulings on motion to strike).

50 Verition dug deeper. Aruba had served a vast privilege log containing 20,000 entries

and spanning 1,462 pages. Verition confirmed that at least 529 entries asserted privilege

for communications about Aruba’s projections and stock-based compensation. Those

communications took place during the preparation of Aruba’s proxy statement, and Wilson

Sonsini lawyers were copied on the communications as part of the team, but the

communications were not privileged. Aruba had asserted privilege for those

communications simply because a lawyer appeared on the document. After Aruba

produced them, Verition renewed its motion to strike. This time, I found that Verition had

shown a broader pattern of problematic conduct that had prejudiced Verition. It was too

late to remedy the prejudice through other means, and I therefore struck the portions of

Dages’s report relating to stock-based compensation.244

Trial took place from December 13-15, 2016. Through no fault of the parties, the

post-trial proceedings became protracted. The parties initially completed post-trial briefing

by March 30, 2017, and post-trial argument was scheduled for May 17. I postponed the

hearing once it became clear that the Delaware Supreme Court’s forthcoming decision in

DFC likely would have a significant effect on the legal landscape. The Delaware Supreme

Court issued its decision on August 1. Both sides submitted supplemental briefs addressing

the implications of DFC, and post-trial argument took place on September 29.

244 See Dkt. 153 (transcript and rulings on renewed motion to strike).

51 While this matter was under submission, on December 14, 2017, the Delaware

Supreme Court issued its decision in Dell. I invited the parties to provide supplemental

submissions addressing the implications of Dell and the extent to which attributes of the

market for Aruba’s stock resembled the attributes that the Delaware Supreme Court

emphasized in Dell. The parties filed their submissions on January 26, 2018.

II. LEGAL ANALYSIS

Delaware’s appraisal statute “allows stockholders who perfect their appraisal rights

to receive ‘fair value’ for their shares as of the merger date instead of the merger

consideration.”245

[T]he purpose of an appraisal is not to make sure that the petitioners get the highest conceivable value that might have been procured had every domino fallen out of the company’s way; rather, it is to make sure that they receive fair compensation for their shares in the sense that it reflects what they deserve to receive based on what would fairly be given to them in an arm’s- length transaction.246

Put differently, “[t]he issue in an appraisal is not whether a negotiator has extracted the

highest possible bid. Rather, the key inquiry is whether the dissenters got fair value and

were not exploited.”247

245 Dell, 2017 WL 6375829, at *1. 246 DFC, 172 A.3d at 370-71. 247 Dell, 2017 WL 6375829, at *24.

52 The trial court’s “ultimate goal in an appraisal proceeding is to determine the ‘fair

or intrinsic value’ of each share on the closing date of the merger.” 248 To accomplish this

task, “the court should first envisage the entire pre-merger company as a ‘going concern,’

as a standalone entity, and assess its value as such.”249 “Because the court ‘strives to value

the corporation itself as distinguished from a specific fraction of its shares as they may

exist in the hands of a particular shareholder,’ the court should not apply a minority

discount when there is a controlling stockholder.” 250 The court should exclude “any

synergies or other value expected from the merger giving rise to the appraisal

proceeding.”251 “[O]nce the total standalone value is determined, the court awards each

petitioning stockholder his pro rata portion of this total . . . plus interest.”252

When seeking to prove fair value, parties may introduce “proof of value by any

techniques or methods which are generally considered acceptable in the financial

248 Id. at *13 (quoting Cavalier Oil Corp. v. Hartnett, 564 A.2d 1137, 1142-43 (Del. 1989)). 249 Id. (quoting Cavalier Oil, 564 A.2d at 1144). 250 Id. (quoting Cavalier Oil, 564 A.2d at 1144). 251 Glob. GT LP v. Golden Telecom, Inc. (Golden Telecom Trial), 993 A.2d 497, 507 (Del. Ch. 2010) (Strine, V.C.), aff’d, 11 A.3d 214 (Del. 2010); accord DFC, 172 A.3d at 368 (“[B]y valuing the company on its value as a ‘going concern,’ the [Delaware Supreme] Court [in Cavalier Oil] seemed to require the excision of any value that might be attributable to expected synergies by a buyer.”); M.P.M. Enters., Inc. v. Gilbert, 731 A.2d 790, 797 (Del. 1999) (“[S]ection 262(h) requires that the Court of Chancery discern the going concern value of the company irrespective of the synergies involved in a merger.”). 252 Dell, 2017 WL 6375829, at *14.

53 community and otherwise admissible in court.”253 “[T]he statute assigns the Court of

Chancery the duty to consider the relevant methods of valuation argued by the parties and

then determine which method (and inputs), or combination of methods, yields the most

reliable determination of value.”254 “But, whatever route it chooses, the trial court must

justify its methodology (or methodologies) according to the facts of the case and relevant,

accepted financial principles.”255 “Although the Court of Chancery has broad discretion to

make findings of fact, those findings of fact have to be grounded in the record and reliable

principles of corporate finance and economics.”256

In this case, “the relevant methods of valuation argued by the parties” are (i) Aruba’s

unaffected market price, (ii) the deal price, and (iii) competing discounted cash flow

analyses. The degree of emphasis that the parties have placed on these methodologies has

evolved. During discovery and at trial, both sides focused on their experts’ discounted cash

flow valuations. As the number of opinions that focused on the deal price mounted, the

respondent placed greater emphasis on that metric, and the petitioners responded by

attacking the process that led to the deal. After DFC, the respondent stressed a combination

of the unaffected market price and the deal price. After Dell, the respondent redoubled its

emphasis on the combination of the unaffected market price and the deal price.

253 Weinberger v. UOP, Inc., 457 A.2d 701, 713 (Del. 1983). 254 Dell, 2017 WL 6375829, at *15 n.105. 255 Id. at *15. 256 DFC, 172 A.3d at 372.

54 A. The Unaffected Market Price

The Delaware Supreme Court’s recent decisions in DFC and Dell teach that if a

company’s shares trade in a market having attributes consistent with the assumptions

underlying a traditional version of the semi-strong form of the efficient capital markets

hypothesis,257 then the unaffected trading price provides evidence of the fair value of a

257 By “traditional,” I mean a framing of the efficient capital markets hypothesis consistent with Eugene Fama’s seminal work and its baseline Chicago-school assumptions. See Eugene F. Fama, Efficient Capital Markets: A Review of Theory and Empirical Work, 25 J. Fin. 383 (1970).

At the trial court level in Dell, I cited some points of entry into a significant and growing body of literature that raises question about the assumptions undergirding the traditional model, which suggest a need for greater nuance. See In re Appraisal of Dell, Inc. (Dell Trial Fair Value), 2016 WL 3186538, at *25 n.16 (Del. Ch. May 31, 2016), rev’d in pertinent part sub nom. Dell, Inc. v. Magnetar Glob. Event Driven Master Fund Ltd, – A.3d –, 2017 WL 6375829 (Del. Dec. 14, 2017). In the legal field, much of this work has responded to the United States Supreme Court’s relatively high-level framing of the efficient capital markets hypothesis as the cornerstone for using the fraud-on-the-market theory to create a presumption of reliance in securities fraud actions. See Basic Inc. v. Levinson, 485 U.S. 224, 241-42, 243-44, 246 (1988). The field of behavioral economics has yielded particularly powerful insights. See, e.g., Donald C. Langevoort, Theories, Assumptions, and Securities Regulation: Market Efficiency Revisited, 140 U. Pa. L. Rev. 851 (1992); Larry E. Ribstein, Fraud on a Noisy Market, 10 Lewis & Clark L. Rev. 137 (2006); Jeff Schwartz, Fairness, Utility, and Market Risk, 89 Or. L. Rev. 175 (2010). Noise trading theory and chaos theory have yielded additional insights. See, e.g., Lawrence A. Cunningham, From Random Walks to Chaotic Crashes: The Linear Genealogy of the Efficient Capital Markets Hypothesis, 62 Geo. Wash. L. Rev. 546 (1994); Lawrence A. Cunningham, Capital Market Theory, Mandatory Disclosure, and Price Discovery, 51 Wash & Lee L. Rev. 843 (1994); Andrei Schleifer & Lawrence H. Summers, The Noise Trader Approach to Finance, 4 J. Econ. Persp. 19 (1990).

Perhaps future appraisal litigants will retain experts on market efficiency, as is common in federal securities actions, and maybe future appraisal decisions will consider subtler aspects of the efficient capital markets hypothesis. This decision does not provide any opportunity for doing so. In its supplemental submissions on the implications of Dell and DFC, the petitioners alluded to potential objections to the Delaware Supreme Court’s

55 proportionate interest in the company as a going concern. That evidence is more reliable

than the single estimate of any one individual, be he a knowledgeable market participant,

corporate insider, valuation professional, or trial judge.258 Under this standard, Aruba’s

unaffected market price provides persuasive evidence of fair value.

1. The Efficient Capital Markets Hypothesis

Both Dell and DFC endorse the efficient capital markets hypothesis and its

predictions about the reliability of market prices. In DFC, the Delaware Supreme Court

stated that “real world transaction prices can be the most probative evidence of fair value

even through appraisal’s particular lens.”259 The high court observed that “[m]arket prices

are typically viewed superior to other valuation techniques because, unlike, e.g., a single

person’s discounted cash flow model, the market price should distill the collective

judgment of the many based on all the publicly available information about a given

company and the value of its shares.”260 The court added that, from the perspective of

economics, when the subject company’s shares are “widely traded on a public market based

upon a rich information base,” then the fair value of a proportionate interest in the company

framing of the efficient capital markets hypothesis, but they did not develop those objections in any meaningful way. Absent a case-specific expert opinion supported by credible evidence and the weight of social-science research, I do not believe a trial judge has the flexibility to disregard the Delaware Supreme Court’s framing of the efficient capital markets hypothesis. 258 Dell, 2017 WL 6375829, at *17; DFC, 172 A.3d at 369-70. 259 DFC, 172 A.3d at 370. 260 Id. at 369-70.

56 as a going concern would “likely be best reflected by the prices at which [the] shares were

trading as of the merger.”261

In Dell, the Delaware Supreme Court stated that “the price produced by an efficient

market is generally a more reliable assessment of fair value than the view of a single

analyst, especially an expert witness who caters her valuation to the litigation imperatives

of a well-heeled client.”262 The court explained that, when the market for a company’s

stock has attributes associated with efficient trading, the stock price

reflects the judgments of many stockholders about the company’s future prospects, based on public filings, industry information, and research conducted by equity analysts. In these circumstances, a mass of investors quickly digests all publicly available information about a company, and in trading the company’s stock, recalibrates its price to reflect the market’s adjusted, consensus valuation of the company.263

The court concluded that, when the market for a company’s shares has the requisite

attributes, the stock price is “likely a possible proxy for fair value.”264

Under Dell and DFC, the critical question is whether the market for the subject

company’s shares has attributes associated with market efficiency. In Dell, the high court

described the relevant attributes as follows: “A market is more likely efficient, or semi-

261 Id. at 367. 262 Dell, 2017 WL 6375829, at *17. 263 Id. (internal quotation marks and footnote omitted). 264 Id. at *1 (reversing trial court’s fair value determination because, among other reasons, “[h]ere, the trial court gave no weight to Dell’s stock price because it found its market to be inefficient. But the evidence suggests that the market for Dell’s shares was actually efficient and, therefore, likely a possible proxy for fair value.”).

57 strong efficient, if it has many stockholders; no controlling stockholder; highly active

trading; and if information about the company is widely available and easily disseminated

to the market.”265

In both Dell and DFC, the Delaware Supreme Court found that the market for the

subject company’s shares had the necessary attributes. The Dell decision described the

market for Dell’s stock as follows:

Dell’s stock traded on the NASDAQ under the ticker symbol DELL. The Company’s market capitalization of more than $20 billion ranked it in the top third of the S&P 500. Dell had a deep public float and was actively traded as more than 5% of Dell’s shares were traded each week. The stock had a bid-ask spread of approximately 0.08%. It was also widely covered by equity analysts, and its share price quickly reflected the market’s view on breaking developments. Based on these metrics, the record suggests the market for Dell stock was semi-strong efficient, meaning that the market’s digestion and assessment of all publicly available information concerning Dell was quickly impounded into the Company’s stock price. For example, on January 14, 2013, Dell’s stock jumped 9.8% within a minute of Bloomberg breaking the news of the Company’s take-private talks, and the stock closed up 13% from the day prior—on a day the S&P as a whole fell 0.1%.266

The DFC decision described the market for DFC’s stock in similar, albeit more

abbreviated, terms:

DFC’s shares were traded on the NASDAQ exchange from 2005 until the merger. Throughout its history as a public company, the record suggests that DFC never had a controlling stockholder, it had a deep public float of 39.6 million shares, and, it had an average daily trading volume just short of one million shares. DFC’s share price moved sharply in reaction to information

265 Id. at *17 (internal quotation marks omitted). 266 Id. at *3 (footnotes omitted).

58 about the company’s performance, the industry, and the overall economy . . . .267

The high court later noted that “DFC’s stock was listed on a major U.S. exchange, traded

actively, and had moved sharply over the years when the company was poised for growth

or facing dimming prospects.”268

In neither case did an expert render an opinion on market efficiency, as is common

in federal securities law actions when a plaintiff seeks to invoke the presumption of reliance

associated with the fraud-on-the-market theory.269 Nor was all of the market evidence part

of the trial record. In DFC, the Delaware Supreme Court cited record evidence for some of

the information about DFC’s stock profile; it drew other information from DFC’s public

filings with the SEC or from an expert report addressing valuation issues.270 In Dell, the

267 DFC, 172 A.3d at 352. 268 Id. at 372. 269 See, e.g., Amgen Inc. v. Conn. Ret. Plans & Tr. Funds, 568 U.S. 455, 471 n.6 (2013) (noting trial court relying on “unchallenged expert report . . . expressly found that the market for Amgen’s stock was efficient”); IBEW Local 98 Pension Fund v. Best Buy Co., Inc., 818 F.3d 775, 779 (8th Cir. 2016) (noting “plaintiffs submitted a report by their expert” to support their “motion for class certification [which] relied on Basic’s fraud-on- the-market presumption”); Schleicher v. Wendt, 618 F.3d 679, 682 (7th Cir. 2010) (“A financial economist concluded, in an expert report that the district judge credited, that the market for Conseco’s shares was efficient . . . and that investors therefore can use the fraud- on-the-market doctrine as a replacement for person-specific proof of reliance and causation.”). See generally 7AA Charles Alan Wright, Arthur R. Miller & Mary Kay Kane, Federal Practice and Procedure § 1781.1 (3d ed. 2005). 270 See DFC, 172 A.3d at 352-53 (citing analyst reports and petitioners’ and respondent’s expert reports on valuation).

59 Delaware Supreme Court similarly drew much of the market-related information from

public filings with the SEC or from an expert report addressing valuation issues.271

In this case, as in Dell and DFC, no expert offered an opinion, pro or con, on whether

the subject company’s shares traded in an efficient market. During trial, the parties did not

emphasize the attributes of the market for Aruba’s common stock. Nevertheless,

information drawn from sources comparable to those the Delaware Supreme Court used in

Dell and DFC indicates that the market for Aruba’s common stock had basic attributes

consistent with what the high court found sufficient in those decisions:

● Aruba’s shares traded on the NASDAQ through the date of the merger under the symbol ARUN.272

● Aruba did not have a controlling stockholder.

● Aruba made public filings in compliance with the disclosure requirements imposed by federal securities laws.

● Thirty-three securities analysts covered Aruba.273

271 See Dell, 2017 WL 6375829, at *3 nn.9-15 (citing public filings with SEC and report of Dell’s valuation expert). 272 See Marcus Tr. 59 (Verition’s expert testifying that “the stock was trading well”); Marcus Dep. 198-99 (Verition’s expert testifying that Aruba was “actively traded” as contrasted with “a company that, you know, is thinly traded” and where “there might be some efficiency issues”); see also Aruba Networks, Inc., Notification of Removal from Listing and/or Registration Under Section 12(b) of the Securities Exchange Act of 1934 (Form 25) (May 19, 2015).

See Marcus Opening Report Ex. 3-1 (“Summary of Analyst Price Targets”); 273

Marcus Tr. 59 (Verition’s expert noting that “[t]here was lots of analyst coverage”); Marcus Dep. 198-99 (Verition’s expert noting that Aruba had “30 plus analysts covering them”).

60 ● Aruba’s weekly trading volume was 9.5 million shares or 8.7% of total shares outstanding.274

● Aruba’s bid-ask spread was 0.055%.275

The following table compares the numerical attributes of Aruba’s common stock

with the comparable attributes for the subject companies in Dell and DFC.

DFC Dell Aruba Market Cap. $375 million $20 billion $2.5 billion Shares in public 37.5 million 1.45 billion 104 million float Public float as % 95% 85% 96% of outstanding Bid-ask spread 0.098% 0.08% 0.055% # of analysts 10 33 33

Given these attributes, Aruba’s stock price is “likely a possible proxy for fair value.” 276

In addition, as in Dell, there is evidence that the Company’s stock price reacted

quickly to the release of news about the Company.277

274 JX 164 (BMO Capital Markets report containing trading statistics); see also Dell, 2017 WL 6375829, at *3 n.10 (citing 5 Alan R. Bromberg et al., Bromberg & Lowenfels on Securities Fraud § 7:484 (2d ed. June 2017 Update) (“Turnover measured by average weekly trading of 2% or more of the outstanding shares would justify a strong presumption that the market for the security is an efficient one.”)).

See Dell, 2017 WL 6375829, at *3 (stating that Dell’s spread of approximately 275

0.08% was indicative of market efficiency). 276 Id. at *1; see also DFC, 172 A.3d 367-68. 277 See, e.g., JX 245; Dkt. 162 at 13 (Petitioners’ Opening Post-Trial Brief containing stock price chart); Dkt. 188 Ex. A (publicly available stock-price data for Aruba & S&P 500).

61  When Aruba announced Project Greyhound after the market closed on August 26, 2014, the stock price rose by 5% the next day, closing at $21.26 on a day when the S&P 500 was stagnant.

 When Aruba announced its first quarter fiscal year 2015 earnings after the market closed on November 20, 2014, Aruba’s stock price dropped by 14% on November 21 on a day when the S&P 500 was up 0.5%.

 When Bloomberg News reported that HP was in talks to buy Aruba on February 25, 2015, Aruba’s stock price rose 21%. The news came out at 3:02 p.m. and, within one minute, Aruba’s stock price had increased 12.7%. By 3:11 p.m., the price had increased to $22.86, before closing at $22.24 at 4 p.m. The same day, the S&P 500 decreased 0.1%.

 When Aruba announced its second quarter fiscal year 2015 earnings after the market closed on February 26, 2015, the stock price increased the next day by 9.7%. That same day, the S&P 500 decreased by 0.3%.

 When the merger was confirmed and the merger price of $24.67 announced on March 2, 2015, the stock price decreased slightly to close at $24.65.

Obviously, these are anecdotal observations and not event studies, but they compare

favorably with the Dell court’s observation that Dell’s share price “quickly reflected the

market’s view on breaking developments,” citing, as an example, that “on January 14,

2013, Dell’s stock jumped 9.8% within a minute of Bloomberg breaking the news of the

Company’s take-private talks, and the stock closed up 13% from the day prior—on a day

the S&P 500 as a whole fell 0.1%.”278 Similar evidence in this case reinforces the

278 Dell, 2017 WL 6375829, at *3.

62 conclusion that Aruba’s stock price leading up to the merger is “likely a possible proxy for

fair value.”279

2. Evidence Of Market Mispricing

The petitioners dispute the reliability of Aruba’s market price in this case,

contending that HP timed its acquisition to take advantage of a trough in the market. They

rely on a range of authorities, including the Delaware Supreme Court’s decision in

Glassman v. Unocal Exploration Corp., which stated that, if an acquisition “was timed to

take advantage of a depressed market, or a low point in the company’s cyclical earnings,

or to precede an anticipated positive development, the appraised value may be adjusted to

account for those factors.”280

The petitioners argue that the market mispricing in this case began after the

Company reported positive quarterly results in May 2014. Revenue exceeded both

management guidance and Wall Street consensus estimates,281 but the Company reported

a gross margin of 70.5%, below the consensus estimate of 72% and Aruba’s “longstanding

target of 71-73%.”282 The miss triggered harsh analyst coverage,283 and Aruba’s stock price

279 Id. at *1; see also DFC, 172 A.3d at 367-68. 280 777 A.2d 242, 248 (Del. 2001). 281 See JX 47 (Barclays report); JX 48 (Janney Capital Markets report). 282 JX 47 at 1. 283 JX 259 at 2-3 (email collecting analyst coverage).

63 dropped 12.11% on the news, declining from $20.06 to $17.63. As evidence of the market

overreaction, the petitioners rely on internal assessments by Aruba management.284

To address the gross margin issue, Aruba management developed Project

Greyhound.285 Management undertook the project because the “Company value [was] not

adequately reflected in [the] stock price.”286 Aruba announced record results in August

2014 and simultaneously announced the implementation of Project Greyhound.287 Analysts

had mixed reactions.288 Aruba’s stock price rose by roughly 9%, from $20.24 to $22.01.289

284 See JX 260 at 5-6 (email thread among management discussing market reaction); JX 267 (Orr reporting to the Aruba Board: “[W]e, as an executive team, are finally sick of wall st discrediting our tremendous come back in revenue growth because they said we are not as profitable as Ubiquiti (give me a break!).”); JX 269 at 2 (Orr email to Galvin: “Between you and me, if we do all these and our stock price stays $13-20, you and I fail miserably. I don’t know why we are doing all these [sic] tough stuff. I really don’t.”); see also Warmenhoven Tr. 235 (“It was very frustrating for the management team, I must say. I mean, in -- in fiscal year 2014, they had 20 percent year-over-year growth, and we were delivering everything to the street, and the stock really didn’t move.”). 285 See JX 280 (internal Aruba deck on Project Greyhound). 286 JX 273 at 2 (draft internal talking points); see also Orr Tr. 486 (suggesting Project Greyhound was motivated, in part, by sense stock price “does not reflect what we believe is our commitment to deliver to what the analysts and the investors want us to do”); Galvin Tr. 597-98 (acknowledging Aruba undertook Project Greyhound, in part, because “we did have pressure from Wall Street to improve what we were dropping to the bottom line”); id. at 600 (agreeing “Greyhound was initiated because there was a perception that the company value was not adequately reflected in the stock price”). 287 JX 828 (press release announcing results). 288 Compare JX 59 at 2 and JX 62 at 1 with JX 61 at 1 and JX 64 at 1. See also JX 309 at 7 (Aruba Board deck summarizing analyst reports and noting that fifteen analysts increased their target price). 289 JX 245.

64 As evidence of continued mispricing, the petitioners rely on Aruba management’s internal

view that it would take “a couple of quarters” after the implementation of Project

Greyhound for Wall Street to credit the results.290

The petitioners contend that matters worsened in November 2014. That month,

Aruba reported on an “outstanding quarter” that included “[r]ecord revenues” that

exceeded “the top end of [its] guidance range” and improved margins thanks to Project

Greyhound.291 But Aruba also announced a range of revenue guidance for the upcoming

quarter that was 1% lower at the midpoint than the pre-announcement analyst consensus.292

Analysts fixated on the lowered guidance. Aruba’s stock dropped by 14%, closing at

$18.82.293 As evidence of the market overreaction, the petitioners rely on the internal

assessments of Aruba management, who explained that they lowered guidance simply to

be prudent and not because of any change in the business dynamics. 294 They also rely on

internal assessments by the HP deal team, who viewed Aruba’s strong results as validating

290 Galvin Tr. 602-03 (testifying that, with respect to Project Greyhound, “there were things that would play out over time”); Galvin Dep. 82 (agreeing “that it would take a couple of quarters after announcement of the initiation of the transformation to see the real results and have the results on Wall Street”). 291 JX 638 at 3 (transcript from Aruba earnings call). 292 JX 363 (internal HP email summarizing the earnings announcement). 293 JX 245. 294 JX 638 at 13 (Galvin stating Aruba was “just being prudent in a mixed environment” and “[n]ot getting ahead of our skis”); see also JX 355 at 1 (email from Galvin to Aruba Board: “our guide is shaping up to be more cautious than our strong results”).

65 their internal business case295 and saw the price drop as an opportunity to buy Aruba at a

discount.296

After Dell and DFC, I do not believe that the petitioners’ evidence provides any

basis to question the integrity of Aruba’s pre-announcement market price as an indicator

of fair value. As a threshold matter, it is not clear that Glassman has continuing relevance

to a widely held, publicly traded entity. Although the Delaware Supreme Court in

Glassman did not limit its comments about the appraisal standard to any particular context,

the case involved a short-form merger in which a controlling stockholder eliminated the

minority.297 Aruba was not a controlled company, and the market for its shares exhibited

the attributes that the Delaware Supreme Court in Dell and DFC found sufficient to give

effect to the implications of the semi-strong form of the efficient capital markets

hypothesis.

295 See JX 357 at 1 (Neri: “We told our story and we were on point.”); JX 360 (email from HP’s Senior Vice President of Corporate Development: “I thought the results were pretty good. . . . They blew past the revenue expectations for the quarter. But they guided soft, below consensus.”); JX 780 at 2 (internal HP deck discussing Aruba’s November earnings announcement and commenting that “[t]he softer guidance did not cause us to change our financial model”) 296 See JX 357 at 1 (internal email among members of the HP deal team: “They are down 8% after hours though because of guidance below consensus for this quarter (although it looks like the range of guidance is still within expectations). Good time to pull the trigger and snap them up I would say.”); Hardegree Dep. 193 (observing that the stock price drop “probably did, on balance” put HP “in a more tacti[cally] advantaged position”). 297 See 777 A.2d at 243 (“In this appeal, we consider the fiduciary duties owed by a parent corporation to the subsidiary’s minority stockholders in the context of a ‘short-form’ merger.”).

66 In Dell, at the trial level, I found “widespread and compelling evidence of a

valuation gap between the market’s perception and the Company’s operative reality.”298

As I viewed the evidence, “[t]he gap was driven by (i) analysts’ focus on short-term,

quarter-by-quarter results and (ii) the Company’s nearly $14 billion investment in its

transformation, which had not yet begun to generate the anticipated results.”299 In making

this finding, I relied on record evidence indicating that (i) Michael Dell and other members

of management valued the company at levels significantly above the market price in light

of its ongoing transformation and a related cost-savings initiative,300 (ii) the financial

advisors to the special committee running the sale process generated valuations implying

values for the company that far exceeded its market price, 301 and (iii) Mr. Dell and the

special committee’s advisors (including two financial advisors and a consulting firm)

believed that the valuation gap existed because the company’s stockholders were focused

on the short-term rather than the long-term.302

On appeal, the Delaware Supreme Court held that, in light of the attributes of the

market for Dell’s shares and the implications of the semi-strong form of the efficient capital

298 Dell Trial Fair Value, 2016 WL 3186538, at *32. 299 Id. 300 See id. at *1-2, *34. 301 See id. at *34. 302 See id. at *34-35 (citing JX 96; JX 110; JX 137; JX 170; JX 226; JX 344; JX 530).

67 markets hypothesis, my reliance on the views of these knowledgeable insiders constituted

an abuse of discretion.303 I had cited various analyst reports as evidence of the contrast

between external views and the insiders’ assessments. The high court found that the analyst

reports showed “just the opposite: analysts scrutinized Dell’s long-range outlook when

evaluating the Company and setting price targets, and the market was capable of

accounting for Dell’s recent mergers and acquisitions and their prospects in its valuation

of the Company.”304 More broadly, the Delaware Supreme Court held that my finding

“ignored the efficient market hypothesis long endorsed by this court.”305 The high court

303 Dell, 2017 WL 6375829, at *19 (“In short, the record does not adequately support the Court of Chancery’s conclusion that the market for Dell’s stock was inefficient and that a valuation gap in the Company’s market trading price existed in advance of the lengthy market check, an error that contributed to the trial court’s decision to disregard the deal price.”). 304 Id. at *16 & n.112 (citing analyst reports discussing Dell’s M&A activity). 305 Id. at *17 (citing DFC, 172 A.3d 346). Legal historians can debate how longstanding that endorsement had been. See Cede & Co. v. Technicolor, Inc. (Technicolor II), 684 A.2d 289, 301 (Del. 1996) (observing, in context of appraisal of publicly traded company following arm’s-length deal, that “the market price of shares may not be representative of fair value” (internal quotation marks omitted)); Rapid-Am. Corp. v. Harris, 603 A.2d 796, 806 (Del. 1992) (describing the Court of Chancery’s rejection of market value in Chicago Corp. v. Munds, 172 A. 452 (Del. Ch. 1934), and observing that “Munds’ succinct evaluation of the market has lost none of its luster”); see also Cede & Co. v. Technicolor, Inc. (Technicolor I), 542 A.2d 1182, 1187 n.8 (Del. 1988) (“Information and insight not communicated to the market may not be reflected in stock prices; thus, minority shareholders being cashed out may be deprived of part of the true investment value of their shares. The issue we are addressing is not the manipulation of the transaction, nor the suppression or misstatement of material information by insiders defrauding the market. Instead, we recognize that the majority may have insight into their company’s future based primarily on bits and pieces of nonmaterial information that have value as a totality.” (citations omitted)). See generally Implicit Minority Discount, supra, at 8 (“Delaware appraisal law has never been particularly friendly to the idea that stock

68 found that “[t]he apparent efficiency of Dell’s pre-signing stock market and the long-term

approach of its analysts undermine concerns of a ‘valuation gap.’”306

market prices always accurately represent a proportional share of the value of the enterprise as a going concern.”). If the lens is broadened to take in fiduciary duty cases, the longstanding nature of the endorsement becomes even more debatable. See, e.g., Unitrin, Inc. v. Am. Gen. Corp., 651 A.2d 1361, 1384 (Del. 1995) (reaffirming the Delaware Supreme Court’s recognition of the threat of substantive coercion, defined as “the risk that shareholders will mistakenly accept an underpriced offer because they disbelieve management’s representations of intrinsic value” (internal quotation marks and citation omitted)); Paramount Commc’ns, Inc. v. Time Inc., 571 A.2d 1140, 1150 n.12 (Del. 1989) (“Thus, we endorse the Chancellor’s conclusion that it is not a breach of faith for directors to determine that the present stock market price of shares is not representative of true value or that there may indeed be several market values for any corporation’s stock.”); Smith v. Van Gorkom, 488 A.2d 858, 875-76 (Del. 1985) (“[I]n the absence of other sound valuation information, the fact of a premium [over market price] alone does not provide an adequate basis upon which to assess the fairness of an offering price. . . . Using market price as a basis for concluding that the premium adequately reflected the true value of the Company was a clearly faulty, indeed fallacious, premise, as the defendants’ own evidence demonstrates.”), overruled on other grounds by Gantler v. Stephens, 965 A.2d 695 (Del. 2009). See generally Bernard Black & Reinier Kraakman, Delaware’s Takeover Law: The Uncertain Search for Hidden Value, 96 Nw. U. L. Rev. 521 (2002) (surveying Delaware takeover decisions and explaining their reliance on a theory of hidden value inconsistent with the efficient capital markets hypothesis). If the management-buyout in Dell had instead been an unsolicited, all-cash, all-shares offer at the deal price, I have no doubt that the Dell board of directors could have defended against that offer based on management’s belief in the considerably greater long-term value of the company. See Air Prods. & Chems., Inc. v. Airgas, Inc., 16 A.3d 48, 108-13 (Del. Ch. 2011).

Setting temporal characterizations aside, I do not question the authority of the Delaware Supreme Court to endorse a traditional framing of the efficient capital markets hypothesis as a method of assessing the reliability of market prices in appraisal proceedings. Once the Delaware Supreme Court has done so, the obligation of a trial judge is to adhere to that endorsement and its implications. 306 Dell, 2017 WL 6375829, at *25.

69 In this case, I regard the petitioners’ evidence of market mispricing as considerably

weaker than what I abused my discretion by crediting in Dell. The evidence in Dell

involved the company’s likelihood of successfully completing a corporate transformation

after spending $14 billion to acquire eleven businesses over three years. The evidence in

this case concerns revenue guidance for an upcoming quarter and the implications of a cost-

cutting effort (Project Greyhound). As in Dell, the analyst reports show that market

observers were assessing these variables. As in Dell, there is no indication that management

did not try to communicate forthrightly with the market. In contrast to Dell, the internal

concerns that Orr and other members of management expressed in this case lacked the

degree of analytical and valuation-based support that accompanied the critiques by Mr.

Dell, his management team, and the special committee’s advisors. In this case, the internal

concerns seem more like reactive expressions of frustration. To reiterate, the evidence in

Dell was insufficient to support my finding regarding the existence of a valuation gap.

Indeed, the Delaware Supreme Court regarded it as the equivalent of no evidence at all.307

307 See id. at *17 (“The record before us provides no rational, factual basis for such a ‘valuation gap.’”); id. (“There is also no evidence in the record that investors were ‘myopic’ or shortsighted.”). The senior tribunal believed that, without any evidence, I “presumed” that “‘investor myopia’ and hangover from the Company’s ‘nearly $14 billion investment in its transformation, which had not yet begun to generate the anticipated results,’ produced a ‘valuation gap’” and this “presumption contributed to the trial court’s decision to assign no weight to Dell’s stock price or deal price.” Id. at *16 (quoting Dell Trial Fair Value, 2016 WL 3186538, at *32, *34).

I did not use the term “presumption,” and I regret that poor drafting on my part seemingly created the impression that I had applied a presumption of some sort. I personally thought I was relying on record evidence that took the form of contemporaneous assessments by knowledgeable insiders about investors’ growing short-term focus and a

70 divergence between the trading price and indications of fundamental value. See Dell Trial Fair Value, 2016 WL 3186538, at *2 (discussing divergence between management’s internal valuation and the market price, including Mr. Dell’s testimony that the market just “didn’t get” Dell (Tr. 409) and his belief that, in spite of Dell’s transformation, “Dell [was] still seen as a PC business” (JX 44 at 1)); id. (discussing management’s hiring consultants to address market misperception (JX 46 at 1)); id. at *4 (discussing Mr. Dell’s assessment of divergence between short-term results and long-term value (JX 109 at 7; JX 110 at 1)); id. (discussing management’s reports to board regarding short-term challenges and need to “sacrifice short term results” to create long-term value (JX 96 at 2; JX 97 at 16)); id. at *5 (discussing JP Morgan report explaining that Dell’s recent earnings misses “have put investors in a ‘wait and see mode’ with increased focus on quarter-by-quarter execution and improved visibility” (JX 137 at 7)); id. at *7 (discussing views of JPMorgan and Goldman Sachs regarding factors contributing to low market valuation that appeared disconnected from fundamentals (JX 170)); id. at *9 (discussing Mr. Dell’s reasons for proposing a going-private transaction including that his initiatives would be “poorly received by the public markets” and that they “could best be accomplished in an environment without quarterly earnings pressure” (JX 231 at 2)); id. at *19 (discussing buyout group presentations to financing sources and ratings agencies, including management’s explanation that poor short-term performance had resulted from “conscious trade-offs to reposition and transform the company” (JX 660 at 32)); id. at *17 (comparing analyst response to results which focused on quarterly earnings miss with Boston Consulting Group’s internal analysis concluding that Dell’s long-term earning power would not be adversely affected (JX 536)); id. at *19 (comparing International Data Corporation’s response to quarterly results with buyout group’s long-term projections to lenders and related assessment of company (JX 669; JX 678)); id. at *34-35 (summarizing evidence of valuation gap, including JX 96; JX 109; JX 110; JX 137; JX 170; JX 226; JX 344; JX 530).

That said, I do not question the Delaware Supreme Court’s authority to elevate the importance for a fair value determination of a trading price generated by a market with attributes associated with semi-strong form efficiency. I also do not question the Delaware Supreme Court’s authority to hold that a trial court should require evidence sufficient to make findings undermining market efficiency before considering and regarding as persuasive case-specific evidence of the views of knowledgeable insiders. See Dell, 2017 WL 6375829, at *17 (observing that I did not make any findings inconsistent with the premises of market efficiency, such as findings that “Dell lacked a vast and diffuse base of public stockholders, that information about the Company was sparse or restricted, that there was not an active trading market for Dell’s shares, or that Dell had a controlling stockholder—or that the market for its stock lacked any of the hallmarks of an efficient market”).

71 The weaker evidence here is insufficient to undermine the reliability of Aruba’s unaffected

market price.

The DFC decision points to the same conclusion. There, the Court of Chancery

found DFC’s performance “appeared to be in a trough, with future performance depending

on the outcome of regulatory decision-making that was largely out of the company’s

control.”308 The trial court relied on record evidence that the acquirer “was aware of DFC’s

trough performance and uncertain outlook” and that “these attributes were at the core of

[the acquirer’s] investment thesis to obtain assets with potential upside at a favorable

price.”309 On appeal, the Delaware Supreme Court explained that “the market’s assessment

of the future cash flows necessarily takes regulatory risk into account as it does with all the

other reasonable uncertain factors that affect a company’s future.”310 The senior tribunal

found that “the record reveals that equity analysts, equity buyers, debt analysts, debt

providers and others were in fact attuned to the regulatory risks facing DFC.”311 The same

reasoning applies here in terms of the ability of equity analysts and other market

participants to assess the risk associated with Project Greyhound and Aruba’s ability to

meet management guidance.

308 In re Appraisal of DFC Glob. Corp. (DFC Trial), 2016 WL 3753123, at *22 (Del. Ch. July 8, 2016), rev’d sub nom. DFC Glob. Corp. v. Muirfield Value P’rs, L.P., 172 A.3d 346 (Del. 2017). 309 Id. (citing Barner Tr. 533-37 and JX 428 at 16). 310 DFC, 172 A.3d at 372. 311 Id. at 373.

72 The Delaware Supreme Court in DFC also questioned whether a trial court should

have relied on evidence that the buyer thought it was getting a good deal to support the

possibility of underpricing:

One would expect a buyer to think it made a wise decision with an upside, and, to be candid, it is in tension with the statute itself to argue that the subjective view of post-merger value of the acquirer can be used to value the respondent company in an appraisal, as the statute’s exclusion of transaction- specific value seems to be directed at the concern a buyer who pays fair value should not have its economic upside for taking that risk expropriated in the appraisal process, a result that if it were the law, would discourage sales transactions valuable to selling stockholders. That a buyer views itself as having struck a good deal is far from reliable evidence that the resulting price from a competitive bidding process is an unreliable indicator of fair value. . . . [O]ne would think that the buyer who paid the highest price in a competitive process had the most confidence there was an upside and must think that post-purchase gains would justify its purchase; otherwise, no sale would ever occur in the world. That [the acquirer] expected to profit does not mean that the collective view of value that results from the deal price is not a reliable indicator of fair value; to hold otherwise, is to adopt a non-binary view of fair value in which only the upside view of what could happen in the future is taken into account.312

312 Id. at 374 n.145. To the extent a buyer bases its assessment on synergistic gains or other aspects of value that would not exist but for the transaction, the Delaware Supreme Court’s “candid” observation regarding tension with the statutory standard makes sense to me. To the extent that a buyer is assessing the target’s standalone value in its pre-deal configuration, it is not clear to me why any tension would exist. I personally would regard the beliefs of knowledgeable market participants, including the buyer, as relevant evidence of fair value. See In re Appraisal of Dole Food Co., Inc., 114 A.3d 541 (Del. Ch. 2014) (ordering production of valuation materials prepared by appraisal petitioners when deciding whether to purchase or sell shares of the subject company’s stock). The market price is, after all, merely an aggregation of the views of knowledgeable market participants. Id. at 559 (“The market price . . . represents an aggregation of the views that many lay people hold about the value of a stock.”); see also DFC, 172 A.3d at 367 (discussing “the collective judgment of value embodied in a market price”). Nor would one need to exclude the possibility of downside risk when evaluating the buyer’s valuation, to the extent that the buyer had not already taken that risk into account. A court could consider evidence showing that the buyer’s assessment was overly optimistic. In the past, courts deciding

73 This passage cautions against regarding HP’s belief that it had seized upon an opportune

time to purchase Aruba as sufficient to undercut the reliability of Aruba’s market price.

3. Bundling Aruba’s Earnings Release With The Merger Announcement

So far, the petitioners’ evidence of a market trough or other mispricing is

conceptually similar to the types of evidence that the Delaware Supreme Court rejected in

Dell and DFC. But the petitioners advanced another argument that falls into a slightly

different category because it involved Aruba and HP making conscious decisions about

when to release information. At the end of January 2015, HP offered to acquire Aruba for

$23.25 per share. During the first week of February, while Aruba was considering its

response, another analyst report criticized the Company, and the stock price fell again,

closing around $16.07 the day after the report.313 Contrary to the market’s perception,

appraisal cases have considered conceptually similar valuation indications. Chancellor Allen considered the fact that “knowledgeable officers and directors all sold their stock” at the transaction price. Cede & Co. v. Technicolor, Inc., 1990 WL 161084, at *32 (Del. Ch. Oct. 19, 1990) (Allen, C.), aff’d in part and rev’d in part on other grounds, 634 A.2d 345 (Del. 1993). Chancellor Chandler considered third-party offers to purchase corporate assets, such as a wholly owned subsidiary whose operations were not affected by the merger giving rise to appraisal rights. See Ryan v. Tad’s Enters., Inc., 709 A.2d 682, 702 (Del. Ch. 1996) (considering offers to purchase one of corporation’s two remaining businesses four months after the merger, one year after the merger, and two years after the merger).

That said, I do not question the Delaware Supreme Court’s authority to instruct the trial courts that when determining fair value, they should give less weight to the views of individual market participants, such as the buyer, and more weight to the collective views of many market participants, aggregated through trading, or the information generated by a sale process. 313 JX 466 at 1-2 (internal Aruba email distributing report).

74 Aruba management knew internally that Aruba was having an excellent quarter and would

beat its guidance.314 But, rather than correcting the market’s perception, Aruba

management proposed to time the announcement of the merger to coincide with the

announcement of Aruba’s February 2015 earnings.315 Companies often announce

significant items as part of an earnings release, particularly if the earnings are bad and the

news is good (or vice versa).316 In this case, Aruba management believed that an increase

Id. at 1 (email from Aruba General Counsel suggesting the stock price “reflects 314

the misconception that we missed. We actually beat guidance but no one knows that yet”). 315 The idea of announcing the merger along with Aruba’s strong quarterly results came from Aruba management, not HP. When Whitman and Neri first re-engaged with Orr at dinner on January 21, 2015, Orr suggested announcing the deal at an industry conference at the beginning of March. JX 423 (email from Orr relaying exchange to Aruba management). When Johansson contacted Orr at the end of January to notify him HP intended to bid imminently, Orr accelerated the timeline, proposing “getting a deal announced by [Aruba’s] earnings on Feb 26.” JX 454 (internal Barclays email summarizing Johansson’s relay of the call). When Evercore later attempted to negotiate with HP’s banker, Evercore again “emphasized that [Aruba would] like to announce [the] deal at or before the [Aruba] earnings announcement” because Aruba was “afraid stock runs like Ubiquiti’s did which could make the deal more challenging from the [Aruba] perspective.” JX 491 (email from Barclays to HP recounting exchange). 316 See, e.g., DFC Trial, 2016 WL 3753123, at *13 (“[D]iscouraging financial results [were] issued on April 2, 2014, the same day the transaction was announced.”); In re Dollar Thrifty S’holder Litig., 14 A.3d 573, 591 (Del. Ch. Sept. 8, 2010) (Strine, V.C.) (“Dollar Thrifty’s advisors were suspicious that Hertz wanted to use the deal to cover up what would be an otherwise disappointing earnings announcement, but eventually concluded that Frissora’s eagerness was attributable to a normal desire to announce the deal in conjunction with an earnings release.”); In re Trans World Airlines, Inc. S’holders Litig., 1988 WL 111271, at *4 (Del. Ch. Oct. 21, 1988) (“The proposal now under consideration was announced . . . the same day on which the Company announced very favorable financial results for the first quarter of 1988.”), abrogated on other grounds by Kahn v. Lynch Commc’n Sys., Inc., 638 A.2d 1110 (Del. 1994); see also Boutros Dep. 203 (“[I]t’s very customary, very, very customary, if you’re pursuing an M&A transaction and it’s close to being done, . . . and you have an earnings release, to make the two concurrent,

75 in the stock price would hurt their chances of getting the deal approved. Providing both

pieces of information simultaneously would blur the market’s reaction to Aruba’s strong

quarterly results and help get the deal approved.317

In Dell, the Delaware Supreme Court implied that a petitioner might be able to call

into question the integrity of the market price if they proved that management had withheld

information from the market or misled investors. As one of its several reasons for holding

that I abused my discretion, the high court noted that I “expressly found no evidence that

information failed to flow freely or that management purposefully tempered investors’

expectations for the Company so that it could eventually take over the Company at a fire-

sale price.”318 My prediction of the law before the Delaware Supreme Court’s decision in

Dell would have been that scienter did not matter for an appraisal case where the sole

litigable question is valuation rather than culpability. But this passage indicates that

because these are two material events that will impact the stock price, and the last thing you want to do is release material information piecemeal to your shareholders.”); cf. In re Smurfit-Stone Container Corp. S’holder Litig., 2011 WL 2028076, at *8 (Del. Ch. May 20, 2011) (“[The bidder] threatened to suspend the merger discussions if the proposed transaction could not be finalized before the end of the weekend and the release of both companies’ earnings announcements the following week.”). 317 HP agreed to this course of action, which favored HP’s interests. HP also may have leaked news of the deal to further mask the significance of Aruba’s strong earnings. On February 25, 2015, one day before Aruba was scheduled to announce its earnings, Bloomberg News ran a story on the merger. JX 866 (article). Qatalyst speculated internally that HP had leaked the news so that Aruba’s “results and subsequent stock price reaction won’t be easy to measure.” JX 510. Aruba’s stock price jumped from $18.37 to $22.24. Proxy at 53. 318 Dell, 2017 WL 6375829, at *18

76 whether management causes an informational distortion is pertinent not only for a breach

of fiduciary duty claim or fraud action, but for an appraisal proceeding as well.

In this case, the petitioners contend that Aruba and HP manipulated the timing of

announcing Aruba’s strong quarterly results and the merger to interfere with investors’

ability to perceive Aruba’s standalone value. The petitioners do not contend that

management never provided the quarterly results or falsified the quarterly results, only that

they bundled them together with the announcement of the merger.

As framed by the Delaware Supreme Court in Dell and DFC, the semi-strong form

of the efficient capital markets hypothesis does not contemplate that directional error will

arise from the order in which information is released or from bundling information

together. Releasing information simultaneously or in close proximity might make it

difficult for an expert to disentangle the price reaction for purposes of an event study, but

the market still would have the information and would respond. As the high court stated in

Dell, when a market is efficient, “a mass of investors quickly digests all publicly available

information about a company and, in trading the company’s stock, recalibrates its price to

reflect the market’s adjusted, consensus valuation of the company.”319 And as the high

court observed in DFC, “in an efficient market, you can trust prices, for they impound all

available information about the value of each security.”320 Aruba’s stock traded briefly

319 Dell, 2017 WL 6375829, at *17. 320 DFC, 172 A.3d at 370 (citation and internal quotation marks omitted).

77 above the deal price, indicating the market took into account both the announcement of the

deal and Aruba’s strong results. Viewed within the framework established by DFC and

Dell, the record does not provide a persuasive reason to question the reliability of Aruba’s

trading price based on the decision by Aruba management to bundle together two pieces of

information.321

4. The Conclusion Regarding The Market Price Evidence

Aruba’s thirty-day average unaffected market price was $17.13.322 Viewed within

the framework established by DFC and Dell, Aruba’s market price provides reliable

evidence of the going concern value of the firm.

B. The Deal Price

The Delaware Supreme Court’s recent decisions in DFC and Dell hold that when a

widely held, publicly traded company has been sold in an arm’s-length transaction, the deal

price has “heavy, if not overriding, probative value.”323 Applying that standard in this case,

321 But cf. Jonathan R. Macey & Geoffrey P. Miller, Good Finance, Bad Economics: An Analysis of the Fraud-on-the-Market Theory, 42 Stan. L. Rev. 1059, 1084-85 (1990) (explaining that efficient markets may process different types of information at different rates and with different effects). Perhaps in an appropriate case an expert could opine that the order in which information was released or the fact that information was bundled together had some meaningful effect. No one made that argument or offered that opinion here. Absent a case-specific expert opinion supported by credible evidence and the weight of social-science research, I do not believe a trial judge has the flexibility to disregard the Delaware Supreme Court’s framing of the implications of the efficient capital markets hypothesis. 322 JX 659. 323 Dell, 2017 WL 6375829, at *22.

78 the merger price carries heavy weight, although the inclusion of elements of value arising

from the merger requires adjustments to generate an indication of fair value.

1. The Role Of The Deal Price

On three occasions, the Delaware Supreme Court has declined to establish a

presumption regarding the relationship between the deal price and fair value. In Golden

Telecom, the high court explained that “Section 262(h) neither dictates nor even

contemplates that the Court of Chancery should consider the transactional market price of

the underlying company. Rather, in determining ‘fair value,’ the statute instructs that the

court ‘shall take into account all relevant factors.’”324 The court reasoned that “[r]equiring

the Court of Chancery to defer—conclusively or presumptively—to the merger price, even

in the face of a pristine, unchallenged transactional process, would contravene the

unambiguous language of the statute and the reasoned holdings of our precedent.”325

In DFC, the Delaware Supreme Court again rejected a request to establish a

presumption that the deal price reflects fair value, seeing “no license in the statue for

creating a presumption” and expressing doubt about “our ability to craft, on a general basis,

the precise pre-conditions that would be necessary to invoke a presumption of that kind.”326

At the same time, the Delaware Supreme Court cautioned that its

324 Golden Telecom, Inc. v. Glob. GT LP, 11 A.3d 214, 217-18 (Del. 2010) (quoting 8 Del. C. § 262(h)). 325 Id. at 218. 326 DFC, 172 A.3d at 366.

79 refusal to craft a statutory presumption in favor the deal price when certain conditions pertain does not in any way signal our ignorance to the economic reality that the sale value resulting from a robust market check will often be the most reliable evidence of fair value, and that second-guessing the value arrived upon by the collective views of many sophisticated parties with a real stake in the matter is hazardous.327

The Delaware Supreme Court also cautioned that “we have little quibble with the economic

argument that the price of a merger that results from a robust market check, against the

back drop of a rich information base and a welcoming environment for potential buyers, is

probative of the company’s fair value.”328

The DFC court reversed a decision by this court to give only one-third weight to the

deal price. The high court noted that the trial court had made the following post-trial

findings of fact:

i) the transaction resulted from a robust market search that lasted approximately two years in which financial and strategic buyers had an open opportunity to buy without inhibition of deal protections;

ii) the company was purchased by a third party in an arm’s length sale; and

iii) there was no hint of self-interest that compromised the market check.329

The high court further observed that

[a]lthough there is no presumption in favor of the deal price, under the conditions found by the Court of Chancery, economic principles suggest that the best evidence of fair value was the deal price, as it resulted from an open process, informed by robust public information, and easy access to deeper,

327 Id. 328 Id. 329 Id. at 349 (formatting added).

80 non-public information, in which many parties with an incentive to make a profit had a chance to bid.330

The Delaware Supreme Court determined that the Court of Chancery’s “decision to give

one-third weight to each metric was unexplained and in tension with the Court of

Chancery’s own findings about the robustness of the market check.”331 The senior tribunal

therefore reversed and remanded the case so the trial court could “reassess [its] conclusion

as to fair value in light of our decision.”332

Most recently, in Dell, the Delaware Supreme Court reiterated that “there is no

requirement that the court assign some mathematical weight to the deal price.” 333 On the

facts presented, however, the high court held that I “erred in not assigning any

mathematical weight to the deal price” under circumstances suggesting that “the deal price

deserved heavy, if not dispositive weight.”334 Those circumstances included (i) stock

market attributes associated with efficient trading335 and (ii) a sale process that involved

330 Id. 331 Id. at 388. 332 Id. at 388-89. 333 Dell, 2017 WL 6375829, at *16. 334 Id.; accord id. at *22 (“Overall, the weight of evidence shows that Dell’s deal price has heavy, if not overriding, probative value.”). 335 Id. at *17.

81 “fair play, low barriers to entry, outreach to all logical buyers, and the chance for any

topping bidder to have the support of Mr. Dell’s own votes.”336

The decisions in DFC and Dell identify factors that make the deal price so probative

that a trial court abuses its discretion by failing to give it enough weight, but they provide

less guidance for determining when a process is sufficiently bad to warrant discounting the

deal price. One passage in the DFC decision suggests an answer to the latter inquiry by

stating that

the purpose of an appraisal is not to make sure that the petitioners get the highest conceivable value that might have been procured had every domino fallen out of the company’s way; rather, it is to make sure that they receive fair compensation for their shares in the sense that it reflects what they deserve to receive based on what would fairly be given to them in an arm’s- length transaction.337

This test focuses on whether the deal in question was an arm’s-length transaction, and it

appears to rule out inquiry into whether a different transaction process might have achieved

a superior result. A passage from Dell points in a similar direction, where the high court

stated: “The issue in an appraisal is not whether a negotiator has extracted the highest

possible bid. Rather, the key inquiry is whether the dissenters got fair value and were not

exploited.”338 As with the passage from DFC, the passage from Dell appears to discount

whether a different approach might have done better. The Dell test turns on exploitation.

336 Id. at *26. 337 DFC, 172 A.3d at 370-71. 338 Dell, 2017 WL 6375829, at *24. The reference to “dissenters” in this sentence strikes me as odd because the dissenters have opted not to receive the merger consideration.

82 2. The Deal Price In This Case

In this case, the HP-Aruba transaction was a third-party, arm’s-length merger. HP

was not a controller engaged in squeezing out the minority. Nor was the transaction a

management buyout where insiders’ informational advantages might have raised concerns.

The transaction did not involve particular stockholders, such as members of management

or a large blockholder, rolling over their shares or otherwise receiving differential

treatment. Nothing about the deal structure could be considered exploitive.

The ultimate decision makers for Aruba—the Board and the stockholders—did not

labor under any conflicts of interest. The Board was disinterested and independent. Of its

eight members, six were experienced, outside directors. Aruba’s stockholder base was

widely dispersed. No one identified any stockholders with a dominant position or divergent

interests.

Aruba negotiated with HP over the price. On January 31, 2015, HP sent Aruba a

written indication of interest for a cash transaction at $23.25 per share, for an aggregate

valuation of $2.563 billion.339 Aruba countered at $29 per share.340 While HP considered

By seeking appraisal, they avoided the possibility of being “exploited” by the deal. The larger point seems to be for the trial judge to assess whether the deal price is somehow exploitive such that it would exploit the dissenters for the court to use it as the basis for awarding fair value. 339 JX 452 at 3 (“Indicative Non-Binding Proposal”). 340 JX 234 (minutes); accord Warmenhoven Tr. 261 (testifying the Aruba Board came to the figure because “[w]e wanted to reaffirm that we thought there was great value there, meaning they should bid higher, but we felt like if we put a 3 in the first digit and started at 30, that they might conclude that’s too big a gap to close and stop discussion”).

83 Aruba’s counter, it learned that Aruba in fact had fewer shares outstanding than HP had

believed.341 HP had based the $23.25 per share price on Aruba’s old share count. When HP

recalculated its offer to reflect the correct share count, the same aggregate consideration of

$2.563 billion resulted in a price per share of $23.89.342 Based on that calculation, HP

raised its bid to $24.67 per share, or just over 3% on an as-adjusted basis.343 Aruba asked

for $25.00 per share, but HP held firm.344

There is evidence that the price credited Aruba with a portion of the substantial

synergies that the transaction would generate from combining Aruba’s strength in wireless

networking with HP’s strength in wired systems.345 HP’s final internal analysis, reflecting

341 JX 491 (email from Barclays to HP relaying Evercore’s counter); accord JX 488 at 4 (internal Evercore emails confirming change in shares outstanding and suggesting repurchases caused the change). 342 JX 798 at 2 (Barclays slide entitled “Impact of New Share Count and B/S Data on Offer Price”). 343 Proxy at 52. 344 Id. 345 See Marcus Tr. 151; Warmenhoven Tr. 264 (describing HP and Aruba as a “perfect match” because “HP had all the switches and routers” but “no WiFi to speak of”); id. at 276-77 (calling the synergies “[v]ery substantial” and noting the acquisition “was part of not just acquiring share in networking, but a bigger, broader enterprise strategy”); Orr Tr. 434-35 (discussing the benefits the transaction offered to both Aruba and HP); Galvin Tr. 578 (“[T]he win-win is HP provided us the muscle that we never had to really stand up to Cisco in the high end of the corporate stack, in the big deals. And particularly when you start to talk about a wired-wireless convergence, they provided a tremendous amount of muscle for us to do that.”); DePuy Tr. 682 (“[W]hen you put those two together, they could attack customers in a way they could neither do individually.”); Dages Tr. 753- 54, 783; see also JX 805 at 2 (late-stage internal HP approval deck noting the acquisition “will provide complementary go to market and wireless capabilities and uniquely position[] HP as the market moves towards a unified wired/wireless solution”); id. at 7-8 (estimating

84 independent research and validation by McKinsey, anticipated total synergies of $1.41

billion, consisting of revenue synergies of $1.175 billion and cost synergies of $235

million.346

approximately $1.4 billion in synergies); id. at 26 (slide titled “[Aruba] Fills Key Product Gaps Across HPN’s WLAN Product Portfolio”). 346 JX 805; see also JX 350 (similar deck for earlier approval); JX 356 (same); JX 383 (McKinsey synergy analysis). The petitioners argued that the court cannot rely on HP’s analyses because they are hearsay. The petitioners raised their objections in the Pre-Trial Order and again at trial. See PTO ¶¶ 585-86 & Ex. A (objecting to JX 350 and JX 805); Warmenhoven Tr. 273-74 (objecting to JX 350 as “hearsay posed” to Warmenhoven because “[h]e’s never seen the document”).

Hearsay “is a statement, other than one made by the declarant while testifying at the trial or hearing, offered in evidence to prove the truth of the matter asserted.” D.R.E. 801(c). “Hearsay is not admissible except as provided by law or by [the Delaware Uniform Rules of Evidence].” D.R.E. 802. Rule of Evidence 803(6) recognizes an exception for documents

made at or near the time by, or from information transmitted by, a person with knowledge, if kept in the course of a regularly conducted business activity, and if it was the regular practice of that business activity to make the [document], all as shown by the testimony of the custodian or other qualified witness . . . unless the source of information or the method or circumstances of preparation indicate lack of trustworthiness.

D.R.E. 803(6). “The business records exception to the hearsay rule permits the admission of hearsay documents that are likely to be trustworthy because a business regularly maintains and relies on them.” Brown v. Liberty Mut. Ins. Co., 774 A.2d 232, 239 (Del. 2001). “The principal precondition to admission of documents as a business record . . . is that the records have sufficient indicia of trustworthiness to be considered reliable.” Id. (quoting Saks Int’l, Inc. v. M/V Exp. Champion, 817 F.2d 1011, 1013 (2d Cir. 1987)).

Johansson testified extensively by deposition about HP’s thorough, routine approval process for reviewing and approving transactions of this nature. See PTO ¶ 575 (stipulating Johansson would testify “by deposition”). He testified that the process regularly includes an “Approval to Negotiate,” such as the one contained in JX 350, and an “Approval to Sign,” such as the one contained in JX 805. See Johansson Dep. 22-25. Although Johansson’s testimony did not explicitly address those two exhibits, his testimony satisfies

85 HP and Aruba agreed to terms for the merger agreement that the petitioners have

not meaningfully challenged. The merger agreement contained a no-shop clause that

prevented Aruba from communicating with third parties about an acquisition proposal

unless both the Aruba Board’s fiduciary duties required it and the acquisition proposal was

reasonably likely to lead to a superior proposal.347 The merger agreement granted HP an

unlimited match right, with five days to match the first superior proposal and two days to

match any subsequent increase,348 and during the match period Aruba had to negotiate

exclusively and in good faith with HP.349 The merger agreement provided that if the Aruba

Board complied with the no-solicitation provision, including the match right, then the

Aruba Board could terminate the merger agreement to accept a superior proposal after first

paying HP a termination fee of $90 million, or 3% of equity value.350 This combination of

defensive provisions would not have supported a claim for breach of fiduciary duty.351

me that HP prepared and maintained them in the ordinary course of its business. They therefore have the indicia of trustworthiness contemplated by Rule of Evidence 803(6). The petitioners’ selective introduction of similar, and in some cases nearly identical, documents bolsters this conclusion. See, e.g., JX 356 (draft Approval to Negotiate deck); JX 398 at 6 (internal HP “discussion materials” deck); JX 780 (later-dated internal HP “discussion materials” deck). 347 See JX 907 § 5.2(b)(ii). 348 Id. at § 5.2(b)(ii)(A), (B). 349 Id. at § 5.2(b)(ii)(B)(3). 350 Id. at § 7.2(b)(iv). 351 See, e.g., Dent v. Ramtron Int’l Corp., 2014 WL 2931180, at *8-10 (Del. Ch. June 30, 2014) (rejecting fiduciary challenge to “(1) a no-solicitation provision; (2) a standstill provision; (3) a change in recommendation provision; (4) information rights for

86 Considering these factors as a whole, the HP-Aruba merger looks like a run-of-the-

mill, third-party deal. Nothing about it appears exploitive. Particularly given the inclusion

of synergies, there is good reason to think that the deal price exceeded fair value and, if

anything, should establish a ceiling for fair value.352

[the acquirer]; and (5) a $5 million termination fee” where the termination fee represented 4.5% of equity value and the change of recommendation provision included an unlimited match right); In re Novell, Inc. S’holder Litig., 2013 WL 322560, at *10 (Del. Ch. Jan. 3, 2013) (describing “the no solicitation provision, the matching rights provision, and the termination fee” as “customary and well within the range permitted under Delaware law” and observing that “[t]he mere inclusion of such routine terms does not amount to a breach of fiduciary duty”); In re Answers Corp. S’holders Litig., 2011 WL 1366780, at *4 & n.47 (Del. Ch. Apr. 11, 2011) (describing “a termination fee plus expense reimbursement of 4.4% of the Proposed Transaction’s equity value, a no solicitation clause, a ‘no-talk’ provision limiting the Board’s ability to discuss an alternative transaction with an unsolicited bidder, a matching rights provision, and a force-the-vote requirement” as “standard merger terms” that “do not alone constitute breaches of fiduciary duty” (citation omitted)); In re Atheros Commc’ns, Inc. S’holder Litig., 2011 WL 864928, at *7 n.61 (Del. Ch. Mar. 4, 2011) (characterizing a no-solicitation provision, a matching right, and a termination fee as “standard merger terms” that “do not alone constitute breaches of fiduciary duty” (citation omitted)); In re 3Com S’holders Litig., 2009 WL 5173804, at *7 (Del. Ch. Dec. 18, 2009) (describing “the no solicitation provision, the matching rights provision, and the termination fee” as “standard merger terms” that “do not alone constitute breaches of fiduciary duty”). 352 See DFC, 172 A.3d at 371 (“[I]t is widely assumed that the sale price in many M&A deals includes a portion of the buyer’s expected synergy gains, which is part of the premium the winning buyer must pay to prevail and obtain control.”); Merion Capital L.P. v. Lender Processing Servs., Inc., 2016 WL 7324170, at *11, *26 (Del. Ch. Dec. 16, 2016) (noting that evidence supported the view that the merger consideration “included a portion of the value that [the acquirers] expected to generate from synergies” and that “[t]he existence of combinatorial synergies provides an additional reason to think that” the merger consideration “exceeded the fair value of the Company”); see also Olson v. EV3, Inc., 2011 WL 704409, at *10 (Del. Ch. Feb. 21, 2011) (“In an arm’s-length, synergistic transaction, the deal price generally will exceed fair value because target fiduciaries bargain for a premium that includes . . . a share of the anticipated synergies . . . .”); Union Ill. 1995 Inv. Ltd. P’ship v. Union Fin. Gp., Ltd., 847 A.2d 340, 356 (Del. Ch. 2003) (Strine, V.C.) (“[A]cquirers typically share a portion of synergies with sellers in sales transactions and

87 a. The Absence Of Competition

In an effort to undermine the probative value of the deal price, the petitioners argue

that HP did not face a meaningful threat of competition. They note that the recent decisions

in Dell and DFC cited with approval the open nature of the deal processes in those cases.353

The Dell and DFC decisions did not hold that a deal price would be rendered

unreliable in the absence of competition. Instead, the high court indicated that, for an

that portion is value that would be left wholly in the hands of the selling company’s stockholders, as a price that the buyer was willing to pay to capture the selling company and the rest of the synergies.”). 353 See Dell, 2017 WL 6375829, at *20-22 (describing company’s efforts to generate competition for the buy-out group); DFC, 172 A.3d at 349 (relying on the Court of Chancery’s finding that “the transaction resulted from a robust market search that lasted approximately two years in which financial and strategic buyers had an open opportunity to buy without inhibition of deal protections”). Before Dell and DFC, a series of Court of Chancery decisions had stressed the importance of competition during the sale process. See In re PetSmart, Inc., 2017 WL 2303599, at *40 n.439 (Del. Ch. May 26, 2017) (giving exclusive weight to merger price where “negotiated at arm’s-length, in real time, after a well-run pre-signing auction that takes place in the midst of a fully functioning market”); Merion Capital LP v. BMC Software, Inc., 2015 WL 6164771, at *14-15 (Del. Ch. Oct. 21, 2015) (giving exclusive weight to merger price where the company conducted “a robust, arm’s-length sales process” that involved “two auctions over a period of several months,” and “was able to and did engage multiple potential buyers during these periods,” and where the lone remaining bidder “raised its bid multiple times because it believed the auction was still competitive”); LongPath Capital, LLC v. Ramtron Int’l Corp., 2015 WL 4540443, at *9 (Del. Ch. June 30, 2015) (relying on “thorough” sale process initiated in response to “a well-publicized hostile bid and a target actively seeking a white knight”); In re Appraisal of Ancestry.com, Inc., 2015 WL 399726, at *1 (Del. Ch. Jan. 20, 2015) (giving exclusive weight to the deal price where the transaction resulted from an “auction process, which process itself involved a market canvas and uncovered a motivated buyer”); Merlin P’rs LP v. AutoInfo, Inc., 2015 WL 2069417, at *12 (Del. Ch. Jan. 9, 2015) (giving exclusive weight to merger price that “was negotiated at arm’s length, without compulsion, and with adequate information” and it was “the result of competition among many potential acquirers”); Huff Fund Inv. P’ship v. CKx, Inc., 2013 WL 5878807, at *14 (Del. Ch. Nov. 1, 2013) (evaluating sale process and concluding that “the bidders were in fact engaged in

88 appraisal petitioner to call into question a deal process based on lack of competition, the

petitioner should be able to point to a likely bidder and make a persuasive showing that

increased competition would have led to a better result.354 The Dell decision stressed that

“[f]air value entails at minimum a price some buyer is willing to pay—not a price at which

no class of buyers in the market would pay.”355 “[I]f a company is one that no strategic

buyer is interested in buying, it does not suggest a higher value, but a lower one.” 356

Other aspects of the Dell and DFC decisions similarly discounted the importance of

competition. The DFC decision stressed that the purpose of an appraisal “is not to make

sure that the petitioners get the highest conceivable value,”357 and the Dell decision

a process resembling the English ascending-bid auction” involving direct competition between bidders); Union Ill., 847 A.2d at 359 (using merger price as “best indicator of value” where the merger “resulted from a competitive and fair auction” in which “several buyers with a profit motive” were able to evaluate the company and “make bids with actual money behind them”). 354 See Dell, 2017 WL 6375829, at *20 (“Nothing in the record suggests that increased competition would have produced a better result.”); id. at *21 (“The Court of Chancery stressed its view that the lack of competition from a strategic buyer lowered the relevance of the deal price. But its assessment that more bidders—both strategic and financial—should have been involved assumes there was some party interested in proceeding. Nothing in the record indicates that was the case.”); see also id. at *23 (“[A]side from the theoretical, the Court of Chancery did not point to any bidder who actually shied away from exploring an acquisition out of fear of the winner’s curse phenomenon.”); id. at *25 (“[T]he court did not identify any possible bidders that were actually deterred because of Mr. Dell’s status.”). 355 Id.

Id.; see also DFC, 172 A.3d at 375 n.154 (“[T]he absence of synergistic buyers 356

for a company is itself relevant to its value.”). 357 DFC, 172 A.3d at 370.

89 cautioned that “[t]he issue . . . is not whether a negotiator has extracted the highest possible

bid.”358 Competition might help a seller extract a higher price, but that is not the focus of

the inquiry under Dell and DFC.

The role of competition also must be evaluated in light of the Delaware Supreme

Court’s endorsement of the efficient capital markets hypothesis. At the trial level,

Chancellor Bouchard found in DFC and I found in Dell that the market prices of the

acquired firms were depressed and had not been representative of fair value. From that

factual starting point, we examined the sale processes for evidence of competition or a

meaningful threat of competition that would be sufficient to overcome the market

mispricing and generate fair value. On appeal in DFC and Dell, the Delaware Supreme

Court relied on the efficient capital markets hypothesis to hold that the factual findings

about market troughs constituted abuses of discretion. From that different factual starting

point, there is less need for competition among bidders to drive a meaningful sale process,

and less need for a court to delve into the details. With a reliable market price as the base

line, an arm’s-length deal at a premium is non-exploitive. By definition, it provides

stockholders with “fair compensation for their shares,” defined as “what they deserve to

receive based on what would fairly be given to them in an arm’s-length transaction.”359

358 Dell, 2017 WL 6375829, at *24. 359 DFC, 172 A.3d at 371.

90 In this case, the petitioners proved that HP knew it did not face a meaningful threat

of competition. In October 2014, during its first pass at Aruba, HP’s executives established

a “pretty open dialogue” with Orr,360 and he informed HP that Aruba was “not running a

sales process.”361 Orr did not make any effort to create the impression of competition by

“posturing about trying to pin [HP] against someone else.”362 HP consequently did not feel

any pressure to bid. After three full months of discussions and due diligence, HP still had

not put a number on the table.

Aruba has pointed out that, at the end of November 2014, Orr decided that the

discussion had dragged on long enough, and he terminated them with the Aruba Board’s

backing.363 Under different circumstances, this move might have given Aruba some

bargaining leverage by signaling that Aruba was prepared to pursue its standalone plan.364

360 See JX 348 (email from Neri to Whitman). 361 Johansson Dep. 112. 362 Id. 363 See Proxy at 48-49; see also JX 328 (“We have been in dialogue with [HP] since August 27, and have not received a proposal in all this time. . . . We cannot continue to wait for them.”); JX 367 (email from Aruba to Qatalyst advising it had “decided to terminate discussions with [HP] at this time” and that it would “let [Qatalyst] know if discussions resume”); JX 372 at 1-2 (email from Aruba General Counsel relaying conversation with Johansson during which she explained that “[w]e feel it’s time to suspend discussions” because “it has reached a point where we need to focus on running our business and not be distracted by discussions that did not seem to be progressing”). 364 See Lender Processing, 2016 WL 7324170, at *19 (“The Board’s track record of saying ‘no’ gave [the acquirers] a credible reason to believe that the Board would not sell below its internal reserve price.”).

91 But in this case, Orr undercut that implication when he had dinner with Whitman and Neri

on January 21, 2015. After Whitman told Orr that HP still wanted to acquire Aruba,365 Orr

responded enthusiastically and proposed to announce the deal at an industry conference

during the first week of March.366 Days later, when an HP executive called Orr to say that

HP would be sending over its proposal, Orr suggested getting a deal announced by the end

of February.367

HP’s bidding tactics suggest that HP knew it did not face competition. Before the

dinner with Orr, HP planned to open with $24.00 per share and negotiate up to $25.00.368

After Orr’s response, HP lowered its opening bid to $23.25. 369 When Aruba countered at

$29, Barclays told Evercore that any price increase would be on the scale of “quarters, not

dollars.”370 Based on a new share count that Evercore provided, HP recalculated its opening

bid as equating to $23.89 per share.371 This was still below the planned opening bid of

$24.00 per share that HP contemplated before Whitman and Neri had dinner with Orr. With

365 PTO ¶ 155. 366 JX 423 (email from Orr relaying exchange to Aruba management). 367 JX 454 (internal Barclays email summarizing executive’s relay of the call). 368 JX 398 at 1-2 (cover email for deck seeking board approval); JX 805 at 12 (internal HP deck). 369 JX 452 at 3 (“Indicative Non-Binding Proposal”). 370 Proxy at 51. 371 JX 798 at 2 (Barclays slide entitled “Impact of New Share Count and B/S Data on Offer Price”).

92 the lower share count, the same enterprise value generated a price of $24.67 per share.372

HP told Aruba that its best and final bid was $24.67, which HP internally called “the new

$24.00” because the price merely adjusted for the change in Aruba’s public share count.373

HP increased its bid to $24.67 and refused to budge. The deal ended up at the number HP

had planned to use for its opening bid.374

So far, so good for the petitioners. But although they proved that HP knew it did not

face a meaningful threat of competition, they failed to identify any other likely bidder who

would have paid more for Aruba. The Dell decision teaches that “[f]air value entails at

minimum a price some buyer is willing to pay.”375 Elaborating, the court emphasized that,

if no one else is interested in buying, “it does not suggest a higher value, but a lower one.”376

In this case, Aruba (through Qatalyst) contacted six potential strategic partners; none were

interested.377 Nor did anyone come forward after the deal announcement. Under Dell and

372 Id. 373 JX 497 (internal HP email from Johansson); accord JX 805 at 12 (timeline showing “02/08 – HP offer $24.00 and present as best and final. [Aruba] provide new shares outstanding and cash information supporting paying $24.67 at the same enterprise value.”). 374 JX 235 at 1-2 (minutes); Proxy at 52. 375 Dell, 2017 WL 6375829, at *21.

Id.; see also DFC, 172 A.3d at 375 n.154 (“[T]he absence of synergistic buyers 376

for a company is itself relevant to its value.”). 377 See Warmenhoven Tr. 260 (“[I]t was definitive noninterest.”); id. 329-30 (agreeing potential bidders provided “feedback” that they “had no strategic interest in acquiring Aruba” and that it “[h]ad nothing to do with the price”).

93 DFC, the petitioners failed to undermine the deal price by showing a lack of competition.

Instead, the lack of competition supports the reliability of the deal price.

Under Dell and DFC, the test instead is whether the Aruba-HP transaction was

exploitive. “[T]he purpose of an appraisal is not to make sure that the petitioners get the

highest conceivable value . . . .”378 This decision already has found that Aruba’s stock price

exhibited attributes associated with the premises of the efficient capital markets hypothesis.

The merger consideration of $24.67 per share provided Aruba’s stockholders with a

significant premium over a reliable market price. As a result, the Aruba-HP transaction

provided stockholders with “fair compensation for their shares,” defined as “what they

deserve to receive based on what would fairly be given to them in an arm’s-length

transaction.”379

b. The Negotiators’ Incentives

The petitioners next contend that the deal price in this case is unreliable because

Aruba’s negotiators were compromised. The petitioners argue that Aruba’s bankers catered

to HP, and that Orr faced divergent interests of his own. Citing DFC and various trial court

378 DFC, 172 A.3d at 370. 379 Id. at 371.

94 rulings, they argue that the deal price should be discounted because Aruba lacked

representatives who engaged in vigorous, arm’s-length bargaining on its behalf.380

As with the element of competition, the petitioners regard the negotiators’

incentives as a thread which, if pulled, could unravel the sweater. But like competition, the

concept of negotiation cannot be excised from the broader framework that the DFC and

Dell decisions established. The Dell opinion cautioned explicitly that “[t]he issue . . . is not

whether a negotiator has extracted the highest possible bid.”381 If this were a case where

the market price was depressed or unreliable, then perhaps a detailed inquiry into issues

like competition or negotiation might become important in assessing whether the deal

process achieved fair value. In a scenario where the underlying market price is reliable,

competition and negotiation become secondary. Under those circumstances, an arm’s-

380 See, e.g., DFC, 172 A.3d at 349 (relying on the Court of Chancery’s findings that “the company was purchased by a third party in an arm’s length sale” and that “there was no hint of self-interest” in the transaction); PetSmart, 2017 WL 2303599, at *40 n.439 (acknowledging the persuasiveness of a merger price that was “negotiated at arm’s- length”); BMC, 2015 WL 6164771, at *14 (giving exclusive weight to merger price where “[t]he record here demonstrates that the Company conducted a robust, arm’s-length sales process”); AutoInfo, 2015 WL 2069417, at *12 (relying on the merger price where “[t]he Merger was negotiated at arm’s length, without compulsion, and with adequate information”); CKx, 2013 WL 5878807, at *13 (giving exclusive weight to merger price where “the process by which CKx was marketed to potential buyers was thorough, effective, and free from any specter of self-interest or disloyalty”); Highfields Capital, Ltd. v. AXA Fin., Inc., 939 A.2d 34, 42 (Del. Ch. 2007) (stating that a reviewing court should give “substantial evidentiary weight” to the deal price when “the transaction giving rise to the appraisal resulted from an arm’s-length process between two independent parties”). 381 Dell, 2017 WL 6375829, at *24.

95 length deal at a premium over the market price is non-exploitive. By definition, it gives

stockholders “what would fairly be given to them in an arm’s-length transaction.”382

In this case, the petitioners proved that the Aruba’s bankers catered to HP. Once

Whitman refused to work opposite Qatalyst,383 Quattrone and Boutros perceived HP’s

stance as an existential threat to their technology-centered franchise.384 They wanted and

382 DFC, 172 A.3d at 371. 383 JX 412 at 1 (“Meg [Whitman] spoke with conviction and emotion over dinner that they [Qatalyst] were guilty.” (referring to alleged fraud involving sale of Autonomy to HP)); JX 420 (Warmenhoven reporting that Whitman said “Qatalyst, Frank [Quattrone] & George [Boutros] are not welcome in the negotiations. The issue is bigger than Autonomy and goes back to EBay & Yahoo.”); JX 426 (email from Orr to Aruba management updating them on Qatalyst situation and reporting that, when Boutros learned of HP’s position, he was “so emotional, defensive AND offensive (to Meg [Whitman]) that he hardly let me talk”); accord Warmenhoven Tr. 252-53 (testifying that Whitman told him, “I don’t care who you get, but it can’t be Qatalyst” and that “I’m not going to take into my boardroom a deal proposed by Qatalyst”). Aruba has argued that, although Whitman refused to work with Qatalyst, she did not pick Aruba’s banker. To my mind, these are questions of degree, with vetoing a banker to reduce the other side’s choice set operating as a less extreme version of picking the other side’s banker. In this case, however, vetoing effectively meant picking. Warmenhoven testified that, when considering bankers, Francis and Qatalyst were the only names who came up and that “there was no third name mentioned.” Warmenhoven Tr. 238. 384 JX 437 at 1 (Warmenhoven explaining that, in Qatalyst’s view, “[t]he issue is not Aruba. It is about the [Qatalyst] brand . . . . If word spreads that they were tossed from this deal because HP will not engage with them on any M&A transaction, that creates a big issue for them. . . . Frank wants to save his firm . . . . The relationship, or lack thereof, between [Qatalyst] and HP / Meg [Whitman] is now their focus.”); see also JX 521 (Quattrone expressing concern that “Evercore will for sure be beating its chest about its role in this deal and could very well be telling the world that HP wouldn’t do the deal if we were the advisor”).

96 needed to get back on HP’s good side. Their primary goal from that point on was to

rehabilitate their relationship with HP.385

Evercore also wanted to get on HP’s good side. The firm was a new entrant in the

Silicon Valley market and had recently hired Francis as “their first tech sector person.”386

Evercore understood the value of completing a highly visible deal as their first Silicon

Valley transaction,387 and they saw the sale process as an extended audition for HP’s

385 See JX 430 (Quattrone asking Warmenhoven to intervene with Whitman on his behalf); JX 434 at 1 (Quattrone asking Whitman for “the opportunity to speak or meet with you at your earliest convenience to understand from you directly what your concerns are and give me the opportunity to address them” and promising that “while our loyalties are always to our client on any assignment, I am confident we can address your concerns, play a constructive role and engage with your team in a professional manner”); JX 440 at 1 (email from Orr to Aruba management asking “how much time we allow” before saying “sorry, [Qatalyst]. We need to protect our transaction. [W]e cannot worry about your brand!”). At Quattrone’s behest, Warmenhoven sent at least three emails to Whitman in a forty-eight-hour period in which he vouched for Qatalyst and asked Whitman to meet with Quattrone. See JX 430 (asking Whitman to meet personally with Quattrone); JX 436 (vouching for Quattrone and expressing confidence that “if you two could ‘clear the air’ [then] Frank and [Qatalyst] could be constructive participants in getting this deal done”); id. (following-up with Whitman and asking whether her objection was to Qatalyst or only to Boutros). 386 JX 426 at 2 (internal Aruba email); see also JX 5 (executed engagement letter with Evercore showing fee of ¼%); JX 413 (Warmenhoven observing that “Evercore is new in the tech sector, so they may be willing to do a deal at ¼% just to get a deal done that they can brag about publicly”); Warmenhoven Tr. 241 (explaining that, when HP first approached Aruba in August 2014, he considered contacting Francis, but passed because he had no prior experience with Evercore and “didn’t know the team”). 387 See, e.g., JX 427 (“Truly amazing! This is a franchise transaction! Well done!”); id. (“This is franchise defining. Well done, and it shows the power of loyalty, which you have always eschewed! [sic]”); JX 428 (“Just remarkable, Stu [Francis.] What a coup! Would be, as you say, a dynamic advance for Evercore in The Valley.”); JX 439 (congratulatory emails from two additional senior bankers).

97 business.388 During meetings between Aruba and HP, Evercore positioned itself as the

banker to the deal. After one session, Francis reported to his colleagues that “it was fun to

be the only banker in the room and help both sides think through some issues.”389 He even

bragged about having done “a ‘masterful’ job of taking [M]eg [Whitman] through the

issues as if we were her advisor.”390 Even accepting that investment bankers are always on

the lookout for new clients, Evercore’s eagerness in this case went far enough to undermine

its role as Aruba’s advisor.

Warmenhoven testified that the bankers’ relationships with HP did not negatively

affect the negotiations and that having two bankers meant Aruba had “two star players on

the same team.”391 Orr testified similarly.392 Notwithstanding this testimony, I credit that

the bankers’ interests made them less effective negotiators than they might have been

The petitioners likewise proved that Orr had divergent interests, although his

motivations were subtler and less openly mercenary. The sale to HP helped Orr achieve a

combination of personal and professional goals that included hastening his return to a

388 JX 505 (Evercore bankers noting that that “[Whitman] is going to be very active” and that HP “[w]ould be a great new relationship.”). 389 Id. 390 Id. 391 Warmenhoven Tr. 255. 392 Orr Tr. 452.

98 personally fulfilling retirement.393 That said, he was not about to leave Aruba under

circumstances that would hurt the Company or its employees. 394 From his standpoint,

selling Aruba to HP was the perfect solution.395

393 See Warmenhoven Tr. 228-31; id. at 273 (Orr’s “going to be 66 in March. He’s told me many times that he did not want to be in a regular employment situation when he turned 65”); id. at 289 (“[H]e was going to leave before his 65th birthday.”); Orr Tr. 371- 73 (describing activities during retirement, including spending time with his children, learning to cook, traveling to Japan and taking Japanese classes, working with The Philanthropy Workshop, and pursing other philanthropic endeavors including founding a non-profit business that provides low-power, solar-panel-driven phones); id. at 377 (“So every year I say, you know, it’s time. There’s a lot of things I need to do. I need to go back to my philanthropic work. My kids in Africa keep yelling at me to go back and visit. . . . So I was just getting really anxious to do my own things.”); id. at 378 (“I honestly am having fun building this kind of transforming company against the incumbent, and enjoying kind of stirring up a new way of doing things. But it’s getting to the point that I have to tell myself there are other things I want to do in life.”).

Ironically, it was Aruba that stressed Orr’s desire to retire in its pre-trial brief and at trial as a means of undermining the reliability of management’s projections, which Aruba argued were too aggressive because they assumed that Orr would stay with the Company. See Dkt. 138 at 20-21. Aruba also sought to use Orr’s desire to retire to blunt the petitioners’ argument that HP’s unauthorized discussions with Orr about post-transaction employment undermined the negotiations. As Aruba pitched it, Orr agreed to remain reluctantly. See id. at 44 (“Orr will testify that he planned to retire, but he was motivated by loyalty to his team and a request from HP to stay on.”). Once Aruba introduced the retirement theory, the petitioners embraced it as an additional factor that undermined the course of the negotiations, while disputing that Aruba’s ability to achieve its projections depended on Orr.

Aruba also helped prove the petitioners’ case on this point by focusing at trial on an email exchange from March 7, 2015, between Orr and an Aruba director that discussed why the merger benefitted Aruba’s customers and employees, including its salesforce in the field. See JX 535; see also Warmenhoven Tr. 177-83 (discussing JX 535); Orr Tr. 381- 97 (same). Aruba argued that this email showed that the HP deal was a good one. Whether the deal was good for these corporate constituencies is a different question than whether it provided fair value for stockholders. What the email did show is that Orr pursued the HP deal (at least in part) because of loyalties to constituencies beyond the stockholders. In the

99 As with the issue of competition, the answer on negotiation is that the petitioners

proved what they sought to prove, but that is not enough to call into question the deal price

for purposes of appraisal. Once again, “[t]he issue in an appraisal is not whether a

negotiator has extracted the highest possible bid. Rather, the key inquiry is whether the

dissenters got fair value and were not exploited.”396 “[F]air value is just that, ‘fair.’ It does

not mean the highest possible price that a company might have sold for had Warren Buffett

negotiated for it on his best day and the Lenape who sold Manhattan on their worst.”397

grander scheme of life, I find that commendable. For the narrow purpose of Delaware corporate law, those competing loyalties are factors that a court has to weigh. 394 See Warmenhoven Tr. 286 (“Dominic [Orr] guaranteed to us that he would go through an orderly transition, that if we found the right person, he would step aside, and we would just hire the person as CEO. And if we couldn’t find somebody of that experience, he would stay on for a year and help groom him.”); id. at 290 (“Dom [Orr] was a very committed CEO, and independent of the financial implications, he would not have left abruptly. He was a founder of this company. And a founder is . . . it’s like having a child. You don’t abandon it.”); Orr Tr. 458 (testifying that he would not have left without a suitable replacement); JX 326 at 1 (Orr explaining that he wanted to sell at a premium, take care of the employees, and ensure that “the organization and structure is [sic] set up for success and maintaining fun and pride and minimiz[ing] large company pain”). 395 Aruba has argued that Orr could not have been motivated by a desire to retire because he agreed to run the legacy business for HP for a period of time after the acquisition. I think Orr was planning several moves ahead, and he realized that committing to stay on for a period of time after the acquisition would help get the deal done. That, in turn, would bring him closer to returning to a retirement during which he had engaged in a variety of rewarding and commendable pursuits. 396 Dell, 2017 WL 6375829, at *24. 397 DFC, 172 A.3d at 370.

100 The evidence does not convince me that the bankers, Orr, the Aruba Board, and the

stockholders who approved the transaction all accepted a deal price that left a portion of

Aruba’s fundamental value on the table. Perhaps different negotiators could have extracted

a greater share of the synergies from HP in the form of a higher deal price. Maybe if Orr

had been less eager, or if Qatalyst had not been relegated to the back room, then HP would

have opened at $24 per share. Perhaps with a brash Qatalyst banker leading the

negotiations, unhampered by the Autonomy incident, Aruba might have negotiated more

effectively and gotten HP above $25 per share. An outcome along these lines would have

resulted in HP sharing a greater portion of the anticipated synergies with Aruba’s

stockholders. It would not have changed Aruba’s standalone value. Hence, it would not

have affected Aruba’s fair value for purposes of an appraisal.

3. Deducting Synergies

Under Dell and DFC, the deal price in this case has substantial probative value. But

the evidence shows that the deal generated significant synergies. Under the DFC decision,

it is to be assumed that HP shared some of those with Aruba’s stockholders.398 To derive

398 DFC, 172 A.3d at 372; see also Dunmire v. Farmers & Merchs. Bancorp of W. Penn., Inc., 2016 WL 6651411, at *8 n.95 (Del. Ch. Nov. 10, 2016) (surveying academic literature); EV3, 2011 WL 704409, at *10 (“In an arm’s-length, synergistic transaction, the deal price generally will exceed fair value because target fiduciaries bargain for a premium that includes . . . a share of the anticipated synergies . . . .”).

101 an estimate of fair value, the court must exclude “any synergies or other value expected

from the merger giving rise to the appraisal proceeding itself.”399

The parties agree that it is not possible to determine with precision what portion of

the final deal price reflects synergy value. The respondent’s expert conceded that “[t]he

percentage of synergies actually paid by HP to Aruba cannot be accurately measured.” 400

Delaware decisions have recognized the difficulties inherent in quantifying

synergies.401 Despite these difficulties, this court has used a deal-price-less-synergies

metric. In Union Illinois, Chief Justice Strine, then a Vice Chancellor, started with the deal

consideration of $10.20 per share.402 He then discounted that figure by 13% to reflect the

synergies captured by the seller, basing that figure on the opinion of the respondent’s

valuation expert, and also citing the fairness opinion of the seller’s financial advisor, which

399 Golden Telecom Trial, 993 A.2d at 507. 400 Dages Opening Report ¶ 68. 401 See, e.g., Montgomery Cellular Hldg. Co., Inc. v. Dobler, 880 A.2d 206, 221 (Del. 2005) (“The Court of Chancery was unable precisely to quantify those ‘deal-making’ synergies, because [the respondent] did not present any reliable evidence at trial of what those synergies were worth.”); ACP Master, Ltd. v. Sprint Corp., 2017 WL 3421142, at *31 (Del. Ch. July 21, 2017) (recounting respondent synergy estimates, noting that “[o]ther synergy estimates were higher still,” and concluding that, “[i]f the court relied on Clearwire’s deal price, it would have to determine the value of those synergies and back them out”); Ramtron, 2015 WL 4540443, at *26 (acknowledging that the “[p]etitioner’s approach may understate the net synergies” but nonetheless adopting it because “it better conforms to the evidence adduced at trial than [the respondent’s] position”). 402 Union Ill., 847 A.2d at 353 n.27, 364.

102 “had mid-range synergy assumptions of 15%-20% for the synergy value that would be

shared” with the seller.403

In the Highfields case, Vice Chancellor Lamb gave 75% weight to a deal-price-less-

synergies metric.404 The transaction price was $31 per share. The respondent’s expert

opined that the deal price incorporated shared synergies equal to 25% of the deal price, or

$7.75 per share. The court rejected this estimate because it relied, in part, on a discounted

cash flow analysis that the expert had declined to rely on when rendering his other valuation

opinions. The court found more credible an analysis prepared by the acquirer, which

estimated the lower end of shared synergies at $9.54 per share. Vice Chancellor Lamb

regarded this estimate as too high, because it undervalued the acquired company in certain

respects. After correcting the acquirer’s estimate to account for the undervaluation, Vice

Chancellor Lamb concluded that the deal price incorporated synergies of $4.12 per share.405

That figure worked out to a deduction of 13%, the same number used in Union Illinois.

In this case, the HP deal team anticipated $1.41 billion of synergies. McKinsey

projected $1.555 billion in synergies. Barclays’ figure was $1.5 billion.406 McKinsey was

an outside consulting firm hired to vet the HP deal team’s calculation, adding some

403 Id. at 353 n.26. 404 Highfields, 939 A.2d at 61. 405 Id. at 60-62. 406 See Dages Opening Report ¶¶ 64-67.

103 credence to its view. This decision therefore uses McKinsey’s figure.407 Aruba’s expert

drew on a March 2013 study by the Boston Consulting Group which suggested that, on

average, sellers collect 31% of the capitalized value of synergies, with the seller’s share

varying widely from 6% to 51%.408 Using these figures, the range of synergy value shared

in the deal could run from $93 million at the low end to $793 million at the high end. The

deal price implied a value for Aruba of $2.651 billion.409 Using the low-end synergy

deduction of $93 million implies a standalone value of $2.558 billion, or $21.08 per share.

Using the high-end synergy deduction of $793 million implies a standalone value of $1.858

billion, or $15.32 per share. The midpoint is a standalone value of $2.208 billion or $18.20

per share. Recognizing that it would be arbitrary to import the 13% synergy figure used in

both Union Illinois and Highfields, that percentage nevertheless implies a standalone value

of $2.306 billion or $19.06 per share.

Because I am inclined to think that Aruba’s representatives bargained less

effectively than they might have, I tend to think that they obtained a relatively low share

of the synergies from HP. This would indicate that they obtained fewer synergies than the

407 See Highfields, 939 A.2d at 61 & n.87 (crediting synergy estimations included in a study that “was not of the typically skewed, buy-side variety: rather, it was an objective study created by a team of actuaries whose professional standards require neutrality” and noting the study “stands in contrast to the often biased valuation work presented to opposing boards by investment bankers representing a particular company”). 408 Dages Opening Report ¶ 68. 409 Id. ¶ 67.

104 midpoint range and imply a standalone value north of $18.20 per share. Having no way to

gauge the marginal impact of their ineffectiveness, this decision uses $18.20 per share as

the valuation indication for the deal price less synergies.

C. The Experts’ Analyses

Both sides submitted opinions from valuation experts. Both experts used the

discounted cash flow methodology to value Aruba. Both experts believed that the

discounted cash flow methodology provided the best approach for determining the fair

value of the Company. The respondent’s expert, Kevin Dages, said so explicitly: “It is my

opinion that Aruba’s standalone fair value is most accurately measured using a [discounted

cash flow] analysis based on the Management Projections.”410 The petitioners’ expert, Paul

Marcus, expressed his view implicitly by relying exclusively on the discounted cash flow

approach.411

The discounted cash flow methodology is a valuation technique that the financial

community generally accepts and that this court frequently uses in appraisal

proceedings.412 “While the particular assumptions underlying its application may always

410 Dages Opening Report ¶ 129; accord Dages Tr. 782 (“Q. You are all in on a DCF. Right? A. I think that’s fair.”); id. (“I looked at other values, but the opinion was always stated, and has always been stated, at the beginning as being based on DCF.”). 411 Marcus Opening Report ¶ 9 (“I determined the fair value of Aruba’s stock as of May 18, 2015 (the ‘valuation date’) by performing a discounted cash flow (‘DCF’) valuation.”). 412 See generally Jesse A. Finkelstein & John D. Hendershot, Appraisal Rights in Mergers & Consolidations, 38–5th C.P.S. §§ IV(H)(3), V.E. (BNA) [hereinafter Appraisal Rights].

105 be challenged in any particular case, the validity of [the discounted cash flow] technique

qua valuation methodology is no longer open to question.”413 It is a “standard” method that

“gives life to the finance principle that firms should be valued based on the expected value

of their future cash flows, discounted to present value in a manner that accounts for risk.”414

The DCF model entails three basic components: an estimation of net cash flows that the firm will generate and when, over some period; a terminal or residual value equal to the future value, as of the end of the projection period, of the firm’s cash flows beyond the projection period; and finally a cost of capital with which to discount to a present value both the projected net cash flows and the estimated terminal or residual value.415

The Delaware Supreme Court has recently cautioned that “[a]lthough widely considered

the best tool for valuing companies when there is no credible market information and no

market check, [discounted cash flow] valuations involve many inputs—all subject to

disagreement by well-compensated and highly credentialed experts—and even slight

differences in these inputs can produce large valuation gaps.”416

1. Marcus’s Valuation Opinion

Marcus used a discounted cash flow analysis to opine that the Company’s fair value

at closing was $32.57 per share. His model generally adhered to the valuation literature

413 Pinson v. Campbell-Taggart, Inc., 1989 WL 17438, at *8 n.11 (Del. Ch. Feb. 28, 1989). 414 Andaloro v. PFPC Worldwide, Inc., 2005 WL 2045640, at *9 (Del. Ch. Aug. 19, 2005) (Strine, V.C.). 415 In re Radiology Assocs., Inc. Litig., 611 A.2d 485, 490 (Del. Ch. 1991) (quoting Technicolor, 1990 WL 161084, at *7). 416 Dell, 2017 WL 6375829, at *28

106 and the teachings of the Delaware courts. From a methodological perspective, his model

appears sound.

As a source of estimated future cash flows, “Delaware law clearly prefers valuations

based on contemporaneously prepared management projections because management

ordinarily has the best first-hand knowledge of a company’s operations.”417 Marcus used

the February Projections, which covered the fiscal years 2015-2017. The February

Projections had their roots in management’s three-year plan, prepared in the ordinary

course of business and with input from the Aruba Board.418 Management completed an

iteration of its three-year plan in summer 2014.419 Management updated the plan in October

417 Doft & Co. v. Travelocity.com Inc., 2004 WL 1152338, at *5 (Del. Ch. May 20, 2004); see also Cede & Co. v. Technicolor, Inc., 2003 WL 23700218, at *7 (Del. Ch. Dec. 31, 2003) (“When management projections are made in the ordinary course of business, they are generally deemed reliable. Experts who then vary from management forecasts should proffer legitimate reasons for such variance.”) (footnote omitted)), rev’d in part, aff’d in part, 884 A.2d 26 (Del. 2005); Gray v. Cytokine Pharmasciences, Inc., 2002 WL 853549, at *8 (Del. Ch. Apr. 25, 2002) (finding “litigation-driven projections to be unreliable” because “[a]ny other result would condone allowing a company’s management or board of directors to disavow their own data in order to justify a lower valuation in an appraisal proceeding”). 418 JX 233 at 1 (minutes from meeting where projections presented showing finance team accompanied presentation with “third party analyst data regarding the enterprise wireless LAN market growth rates for the last three years,” “compared the Company’s revenue growth rates to the market growth rate,” and “presented financial scenarios and explained the assumptions underlying each scenario” during which “Board members asked questions, provided feedback and discussion ensued”); see also Warmenhoven Tr. 223-24 (discussing process by which Aruba Board reviewed and “chose to accept” the February Projections). 419 See Warmenhoven Tr. 219-20; id. at 313-14 (confirming Galvin prepared the June Plan “in the spring of 2014, as part of the normal operations” of the Company); Galvin

107 2014.420 In February 2015, management revised the plan to reflect intervening results and

to adopt more conservative assumptions.421 To cover the final two years of his projection

period, Marcus used an extension of the February Projections that Qatalyst prepared

working in conjunction with Aruba management.422 Qatalyst used the same projections as

the basis for the fairness opinion that it delivered to the Aruba Board.423 Aruba also used

Tr. 553-54 (discussing development of “top-down” projections incorporating “broader strategic assumptions” discussed by the Aruba Board regularly at the “June meeting”). 420 See JX 297 at 2 (email from Galvin to member of the finance team); JX 315 at 1 (email from Galvin to Qatalyst banker conveying early draft of the October Projections); see also Warmenhoven Tr. 314-15 (discussing Aruba’s ordinary process of preparing a yearly operating plan in September); Galvin Tr. 586 (acknowledging the September figures adjusted margins, reflecting “[t]he overall impacts” of Project Greyhound). 421 See Marcus Opening Report ¶¶ 209-217 (explaining reliance on February Projections); see also Warmenhoven Tr. 220-22 (discussing how and why the finance team made the projections more conservative); Galvin Tr. 558-563 (discussing process of extending ordinary course, six-month budgeting projections out to three years to create February Projections). 422 See Marcus Opening Report ¶ 217; Warmenhoven Tr. 358 (referring to the added two years as “a linear extrapolation”); Galvin Tr. 558-59 (testifying the February Projections initially went out three years “[a]nd then ultimately until 2020, yes”); Dages Tr. 734 (discussing “the extensions that were done with Qatalyst to take it out to 2020”). 423 See Marcus Opening Report ¶ 221; Galvin Tr. 560 (confirming February Projections used by bankers).

108 the same projections in the proxy statement for the deal.424 Marcus adopted management’s

estimates for the cost of stock-based compensation425 and Aruba’s tax rates.426

The projections resulted in Aruba having a high compound annual growth rate

(“CAGR”) of 10% at the end of the projection period. To normalize Aruba’s high growth

and transition the Company into a steady state, Marcus added a second, five-year stage to

create a three-step discounted cash flow model. During the added second stage, he stepped

down the growth rate to reach his terminal, third-stage growth rate of 3.5% per year.427

424 Proxy at 63-64; Orr Tr. 529 (confirming approved projections to be included in the proxy and shared with Qatalyst). 425 See Marcus Opening Report ¶¶ 218-220. 426 The Company had projected a tax rate of 4% for 2015 and 2016 and 25% thereafter. See JX 475 at 1 (email from Galvin to Qatalyst suggesting “I would do 4% thr[ough] 17; then do 25% thereafter”). Management attributed the rate to the Company’s stockpile of valuable net-operating loss credits or “NOLs” from its early, pre-profit days. Due to those credits, the Company had a cash tax rate of only 3.2% and 3.1% in 2013 and 2014, respectively. The Company anticipated it had enough credits remaining to continue paying low taxes through at least 2016. JX 506 at 1 (internal email summarizing available net-operating loss credits and approximate use rates as of February 2015); see also JX 895 at 93 (2014 10-K: “As of July 1, 2014, the Company’s federal loss carryforwards for income tax purposes were approximately $131 million with expiration dates starting in 2028.”). Based on this evidence, Marcus adopted the Company’s estimates. See Marcus Opening Report Ex. 7-1.

Aruba instructed Qatalyst to use the same tax figures that Marcus ultimately adopted. See JX 654 (Qatalyst working spreadsheet indicating Galvin provided tax rates); Marcus Tr. 46-47; Galvin Tr. 622. 427 Marcus Opening Report ¶ 222.

109 Delaware decisions and the valuation literature support this approach.428 Like Marcus, HP

used a three-stage discounted cash flow method when valuing Aruba.429

Marcus calculated Aruba’s weighted average cost of capital (“WACC”) using the

capital asset pricing model (“CAPM”).430 “Under CAPM, the cost of equity capital is the

428 See, e.g., DFC, 172 A.3d at 380 (“Indeed, if the record unambiguously supported the proposition that DFC was to continue a new spurt of growth past 2018, it would have been more appropriate to project out to a point where steady-state growth began. By doing that, the appraiser would have to assess with discipline the next period after the projections end and also the potential that the period might be negative, as well as that another period of above-market growth might be followed by a terminal growth rate more like inflation than the risk-free rate.” (footnote omitted)); Prescott Gp. Small Cap, L.P. v. Coleman Co., Inc., 2004 WL 2059515, at *29 (Del. Ch. Sept. 8, 2004) (Jacobs, J.) (“At the time of the merger, Coleman was projecting a 16% growth in sales for year 2002, which represented a return to Coleman’s prior operating levels. Dr. Kursh utilized a three stage model because he did not believe a 16% growth rate was sustainable long-term.” (footnote omitted)); Robert W. Holthausen & Mark E. Zmijewski, Corporate Valuation Theory, Evidence & Practice 216 (2014) (“We would prepare detailed year-by-year forecasts for the company until the company reaches steady state. You may need to value a company’s cash flows for five years, ten years, or longer if the company is far from becoming a stable mature company as of the valuation date.”); Shannon P. Pratt & Alina V. Niculita, Valuing a Business 219 (5th ed. 2008) (“The appropriate length of the forecast period should be until that variability stops; at the point in time that the company expects normalized or level growth, the terminal value is calculated.”). 429 See Johansson Dep. 40 (testifying that, in valuing a company, HP “want[s] to create a model that gets you to like a steady state . . . . [T]ypically, the companies we look at, ten years is kind of appropriate . . . . So we tend to do a ten-year DCF just to get to that steady state.”). 430 In calculating his WACC, Marcus used an all-equity capital structure. He noted, however, that evidence in the record suggested that Aruba would have issued $300 million in convertible debt if HP had not made its approach. See, e.g., PTO ¶ 51, JX 224 (Aruba Board subcommittee minutes); JX 325 (email from Galvin relaying conversation with Barclays banker wherein banker suggested executing the convertible offering). The debt would have reduced Aruba’s WACC and been a positive signal to the equity markets. Marcus Opening Report ¶ 223. The decision to maintain Aruba’s all-equity capital structure could be seen as a valuation consequence that resulted from the expectation of the merger,

110 risk-free rate of return plus the subject company’s risk. The subject company’s risk is

determined by multiplying the equity risk premium for the market by the company’s

beta.”431 Marcus used a risk-free rate of 2.75%, based on the twenty-year U.S. Treasury

maturity rate, and a supply-side equity risk premium of 6.19%.432 Marcus drew these

figures from reliable sources, and Dages used the same risk-free rate and a virtually

identical supply-side equity risk premium.433 Marcus calculated a beta for Aruba of 0.91,

which he derived by giving one-third weight to Aruba’s two-year, weekly, raw beta (0.81)

and two-thirds weight to the two-year, weekly, raw, unlevered betas of a group of peer

companies (1.11).434 Court of Chancery precedent supports the blended approach,435 and

although one that had a negative effect on Aruba’s stockholders by depriving them of the value generated by a lower-cost capital structure. Cf. 8 Del. C. § 262(h) (instructing Court of Chancery to exclude “any element of value arising from the accomplishment or expectation of the merger”). The petitioners have not made this argument, so this decision does not consider it.

In re Emerging Commc’ns, Inc. S’holders Litig., 2004 WL 1305745, at *16 (Del. 431

Ch. May 3, 2004) (Jacobs, J.). 432 Marcus Opening Report ¶ 225. 433 See Dages Opening Report ¶¶ 117, 124. 434 Marcus Opening Report ¶¶ 226-231. 435 See Merion Capital, L.P. v. 3M Cogent, Inc., 2013 WL 3793896, at *18 (Del. Ch. July 8, 2013) (“[O]ne can ‘smooth’ beta by adjusting historical beta by a market beta of 1, using a 1/3 weighting factor for the market and a 2/3 weighting for the subject company’s beta . . . .”); Golden Telecom Trial, 993 A.2d at 524 (“I find that a beta that gives 2/3 weight to the Bloomberg historic raw beta of 1.32 and 1/3 weight to the 1.24 industry beta is the best approach to this DCF analysis.”).

111 the valuation literature supports the selection of a two-year period for Aruba.436 He then

added the fifth-decline size premium.437 These calculations resulted in a 10% WACC.438

To calculate value for the terminal period, Marcus used the Gordon Growth

Model.439 “To calculate terminal value using the Gordon Growth Model, the Court must

436 Holthausen & Zmijewski, supra, at 300 (“Using more recent data might better reflect a company’s current (and more forward-looking) systematic risk.”); Tim Koller, Marc Goedhart & David Wessels, Valuation: Measuring and Managing the Value of Companies 247 (5th ed. 2010) (noting that “changes in corporate strategy or capital structure often lead to changes in risk for stockholders” and that, where that occurs, “a long estimation period would place too much weight on irrelevant data”); Shannon P. Pratt & Roger J. Grabowski, Cost of Capital: Applications and Examples 208 (5th ed. 2014) (noting that five-years is more common but where “business characteristics change during the sampling period . . . it may be more appropriate to use a shorter sampling period. However, as the sampling period used is reduced, the accuracy of the estimate is generally reduced.”); id. at 224 (recommending that “[i]f the underlying fundamentals of the business have changed, a more recent period should be used in developing a beta estimate”). Aruba had grown significantly during the years preceding the merger. See PTO ¶ 77. Aruba’s expert agreed that he typically uses a two-year weakly raw beta when calculating WACC. Dages Tr. 793; Dages Dep. 432. 437 Marcus Opening Report ¶ 230. “In addition to the equity risk premium, an equity size premium generally is added to the company’s cost of equity in the valuation of smaller companies to account for the higher rate of return demanded by investors to compensate for the greater risk associated with small company equity.” Gearreald v. Just Care, Inc., 2012 WL 1569818, at *10 (Del. Ch. Apr. 30, 2012). Dages disputed the applicability of a size premium at all because, “in [his] experience, a size premium is rarely applied to mid- or larger-cap companies” and “Aruba did not share the characteristics that researchers have hypothesized for returns in excess of what is predicted by the CAPM.” Dages Rebuttal Report ¶ 32. He further argued that Aruba properly belonged in the sixth rather than fifth decile. Id. 438 Dages used a WACC of 11%. Dages observed that all three deal advisors and two research analysts used higher WACCs. Dages Opening Report Ex. 18; Dages Rebuttal Report ¶ 31. 439 Marcus Opening Report ¶ 235.

112 select a long-term growth rate, i.e., the expected growth rate of free cash flows into

perpetuity.”440 As noted, Marcus selected a perpetuity growth rate of 3.5%. He believed it

was reasonable to assume that Aruba would grow at the rate of the overall economy, but to

be conservative he selected a growth rate approximately at the midpoint of the risk-free

rate (2.75%) and nominal GDP growth rate, as predicted by reliable, oft-cited studies

(4.3%).441

Marcus sensitized his valuation for discount rates of 9.5% to 10.5% and terminal

growth rates of 3.0% to 4.0%, generating a valuation range for $29.16 to $36.93. The

midpoint, based on a discount rate of 10% and a terminal growth rate of 3.5%, was

$32.57.442

My primary concerns with Marcus’s opinion are his beta and the contrast between

his valuation and market indicators. Marcus’s raw and blended betas were both lower than

440 3M, 2013 WL 3793896, at *21. 441 Marcus Opening Report ¶¶ 232-235. Marcus did not clarify why he adopted the risk-free rate rather than the projected rate of inflation as the floor for his terminal growth rate. As discussed in addressing Dages’s report below, some of this court’s precedent suggests adopting the risk-free rate as a ceiling for a company’s long-term sustainable growth rate. This court’s precedents support adopting the rate of inflation as a floor for a company’s long-term growth rate. See Golden Telecom Trial, 993 A.2d at 511-12 (“A viable company should grow at least at the rate of inflation and . . . the rate of inflation is the floor for a terminal value estimate for a solidly profitable company that does not have an identifiable risk of insolvency.”); see also Owen v. Cannon, 2015 WL 3819204, at *25 (Del. Ch. June 17, 2015) (“I find that it is appropriate under Golden Telecom to calculate the terminal growth rate as a premium to inflation.”). The distinction does not alter the outcome in this case, and this decision expresses no view on the issue. 442 Marcus Opening Report ¶ 236.

113 one, indicating that Aruba, a relatively young and growing technology company, exhibited

less volatility than the market as a whole.443 Although the data supported the low beta, no

one could offer a good explanation as to why the number was so low.444 Marcus’s beta of

0.91 also fell roughly 20% below the median two-year adjusted beta of companies in

Aruba’s peer group and approximately 35% below Aruba’s five-year adjusted weekly

beta.445 That said, Aruba’s low beta was not unique. The bankers’ fairness presentations

identified other networking and WiFi companies that had betas of less than 1.446

Marcus’s valuation outcome diverged significantly from market indications. His

valuation of $32.57 is

 approximately 32% higher than the deal price of $24.67;

 approximately 39% higher than the mean of the last batch of unaffected analyst price targets at $23.4;447

443 Dages Tr. 747; see also Pratt & Grabowski, supra, at 194 (“Many high-tech companies are good examples of stocks with high betas. . . . The classic example of a low- beta stock would be a utility that has not diversified into riskier activities.”). 444 See, e.g., Dages Tr. 790-91 (“I don’t see a basis for getting comfortable with a beta that is that low given this company, its position in the industry, and what I’ve heard about the challenges it’s facing. Especially when I look at the peers and they are all up over 1. And I look at them, and they are well over 1 in a longer time period.”). 445 See Dages Opening Report ¶¶ 118-19. See generally Dages Rebuttal Report ¶¶ 26-31 (criticizing Marcus’s beta). 446 See JX 802 at 25; JX 809 at 31. 447 Marcus Opening Report Ex. 3-1.

114  approximately 21% above the mean of the midpoints of the final valuations prepared by all three advisors at $26.57;448 and

 nearly double Aruba’s thirty-day average unaffected market price of $17.13 per share.

Despite its seemingly sound methodology, these market indicators combine to create

significant doubt regarding the reliability of the Marcus discounted cash flow analysis and

its resulting valuation.

2. Dages’s Valuation Opinion

Dages rendered several different valuation opinions. They produced relatively

stable outputs but changed substantially in their inputs. Dages also made a significant

judgment call by selecting a WACC from a menu of possibilities, rather than calculating a

beta to generate a WACC as contemplated by CAPM.

In his opening report, Dages opined that the standalone fair value of Aruba was

$19.85 per share, which he derived using a discounted cash flow methodology.449 Like

Marcus, Dages used the February Projections with the two-year extension prepared by

Qatalyst with management’s input.450 Unlike Marcus, who used management estimates for

stock-based compensation and tax rates, Dages used a stock-based compensation figure

448 See JX 806 at 14 (final Barclays valuation); JX 811 at 19 (final Qatalyst valuation); JX 812 at 14 (final Evercore valuation). 449 Dages Opening Report ¶¶ 7, 56. 450 See id. ¶¶ 89-96; Dages Rebuttal Report ¶ 4; see also Dages Dep. 311 (describing the February projections with the two-year extension as the “best projections” for Aruba); accord Dages Tr. 733.

115 from Barclays,451 and his own estimate of Aruba’s effective tax rate.452 Despite recognizing

the issue raised by Aruba’s high growth rate at the end of the projection period, 453 Dages

used a traditional two-stage model rather than a three-stage model. For his terminal value,

Dages explained the principles used when selecting a long-term growth rate in much the

same terms as Marcus,454 but then chose the risk-free rate (2.75%) because “some financial

economists caution that the risk-free rate . . . should serve as the ceiling for a stable, long-

term growth rate” and this court had used that rate in “a recent opinion.”455

451 See Dages Opening Report ¶¶ 98-106 (describing the stock-based compensation figures he generated). 452 See id. ¶ 115 (using tax rate of 30%). Dages stated in his report that “[t]he 30.0 percent tax rate is based on the effective tax rate used by Aruba in the Management Projections.” Id. At trial, he admitted that this was an error. See Dages Tr. 732-33; 812-15. 453 Dages Tr. 802-05. 454 See Dages Opening Report ¶¶ 108-109. 455 See id. ¶ 110. Dages’s report did not cite the financial economists or the opinion. Presumably, he was referring to the DFC trial-level opinion, where he also served as an expert. There, Chancellor Bouchard adopted the risk-free rate as a ceiling in reliance on Dages’s identical suggestion “that some financial economists view the risk-free rate as the ceiling for a stable, longterm growth rate.” DFC Trial, 2016 WL 3753123, at *17. In that case, Dages had also acknowledged that “one suggested ceiling for a company’s perpetuity growth rate is nominal GDP.” Id. at *18; see also Golden Telecom Trial, 993 A.2d at 511 (“Generally, once an industry has matured, a company will grow at a steady rate that is roughly equal to the rate of nominal GDP growth.”). Dages conceded that a 4.5% perpetuity growth rate, substantially above the 3.14% risk-free rate calculated in that case, was “at the high end of the reasonable range of long-term growth rates.” DFC Trial, 2016 WL 3753123, at *18. On appeal, the Delaware Supreme Court stated that the risk-free rate “is viewed to be the ceiling for a stable, long-term growth rate.” DFC, 172 A.3d at 383. The idea that a company in a steady state will grow more or less in line with the average rate of the broader economy has intuitive appeal. See 3M, 2013 WL 3793896, at *21 (quoting Golden Telecom for the proposition that “the rate of inflation is the floor for a terminal value” and noting that “a terminal growth rate should not be greater than the nominal

116 For his discount rate, Dages started out using CAPM to develop a WACC. He used

the same risk-free rate as Marcus (2.75%)456 and a supply-side equity risk premium that

was substantially similar to Marcus’s (6.21%).457 On the issues of a beta and size premium,

however, Dages punted. He described a variety of possible betas, including (i) raw and

adjusted betas for Aruba derived using two years of weekly measurements, five years of

weekly measurements, and five years of monthly measurements, and (ii) raw and adjusted

betas for peer companies derived using the same measuring periods.458 Rather than

selecting a beta, Dages used the various candidates to generate nine possible WACCs. He

then added into the mix the WACCs used by the three financial advisors and WACCs from

two analysts, for a total of fourteen possibilities.459 After surveying these, he chose a

WACC of 11%.460 His WAAC implied a beta of 1.33.461 This court has criticized similarly

unstructured approaches to valuation inputs.462

growth rate for the United States economy”). Because the experts did not develop the issue further, and because resolving it is not necessary to decide this case, this decision expresses no opinion on the matter. 456 See Dages Opening Report ¶ 117. 457 See id. ¶ 124. 458 See id. ¶¶ 118-122. 459 Id. Ex. 18. 460 Id. ¶ 127. 461 Dages Rebuttal Report ¶ 4. 462 See In re Orchard Enters., Inc., 2012 WL 2923305, at *17 (Del. Ch. July 18, 2012) (Strine, C.) (expressing the court’s preference for “the more academically and

117 After I issued an evidentiary ruling precluding Dages from rendering an opinion on

stock-based compensation, Dages changed course and relied at trial on a set of projections

that he had created himself using industry growth rates and referenced in a footnote in his

opening report.463 When he ran his discounted cash flow model with those projections and

management’s figures for stock-based compensation expense, his model generated a value

of $19.45 per share, forty cents below his original opinion. At trial, Dages revised his view

on Aruba’s tax expenses and agreed with management’s use of a 4% tax rate for 2015 and

2016, although he continued to endorse the use of a 30% tax rate for subsequent years

empirically-driven CAPM model when that can be applied responsibly” and noting that it involves “less (but still more than comfortable) amounts of subjectivity”); Del. Open MRI, 898 A.2d at 338 (questioning the use of the build-up method with its concept of “company- specific risk” and observing “[t]he calculation of a company specific risk is highly subjective and often is justified as a way of taking into account competitive and other factors that endanger the subject company’s ability to achieve its projected cash flows. In other words, it is often a back-door method of reducing estimated cash flows rather than adjusting them directly. To judges, the company specific risk premium often seems like the device experts employ to bring their final results into line with their clients’ objectives, when other valuation inputs fail to do the trick.”); Andaloro, 2005 WL 2045640, at *12 n.49 (criticizing expert who “spiraled” into a terminal growth rate “through an incomprehensible ‘iterative process’” and finding that “[r]ather than a reasoned exercise in applied social science, [the expert] appears to have channeled inspiration, more like a great songwriter than a valuation expert”). 463 Dages Tr. 760-61 (“Q. So to be clear, your opinion, when you originally opined, was the February revenue projections; right? A. Correct.”); id. at 767 (“Q. So you didn’t just swap out the Dell’Oro projected growth rates for the industry for management’s. You created your own industry projections. A. Correct.”); id. at 772 (“Q. Now, in fact, you don’t have any expertise that would allow you to determine whether Dell’Oro’s industrywide growth rates are a reasonable proxy for Aruba’s expected future performance, do you? A. No. No independent expertise, no.”).

118 rather than management’s rate of 25%.464 This modification added thirty cents per share to

his valuation, resulting in a figure of $19.75 per share. Serendipitously, that result fell just

ten cents below the valuation in his opening report, although reached using substantially

different inputs. This is the fair value figure that Aruba endorsed at post-trial argument.

Dages’s final opinion of $19.75 per share comported with market evidence by

falling between the unaffected market price and the deal price. Its methodological

underpinnings, however, provided cause for concern, as did the meandering route by which

Dages arrived at this figure.

D. Weighing the Valuation Methodologies

This decision has discussed each of the relevant methods of valuation that the parties

presented. Under the statute, the court must make a point estimate of fair value measured

in dollars and cents. When determining fair value, “[t]he Court of Chancery must exercise

its considerable discretion while also explaining, with reference to the economic facts

before it and corporate finance principles, why it is according a certain weight to a certain

indicator of value.”465

464 Dages Tr. 751, 813. In light of other evidence in the record, the cash tax rate is more persuasive. See Galvin Dep. 296 (stating that management provided the cash tax rate); JX 548 (Qatalyst spreadsheet showing management’s cash tax rates); JX 654 (Qatalyst projections using management’s cash tax rate); see also Dages Tr. 815 (testifying that cash tax rate is typically more accurate than effective tax rate). 465 DFC, 172 A.3d at 388.

119 The forceful discussion of the efficient capital markets hypothesis in Dell and DFC

indicates that Aruba’s unaffected market price is entitled to substantial weight.

[C]orporate finance theory reflects a belief that if an asset—such as the value of a company as reflected in the trading value of its stock—can be subject to close examination and bidding by many humans with an incentive to estimate its future cash flows value, the resulting collective judgment as to value is likely to be highly informative . . . .466

“Market prices are typically viewed superior to other valuation techniques because, unlike,

e.g., a single person’s discounted cash flow model, the market price should distill the

collective judgment of the many based on all the publicly available information about a

given company and the value of its shares.”467 “[I]n many circumstances a property interest

is best valued by the amount a buyer will pay for it” and “a well-informed, liquid trading

market will provide a measure of fair value superior to any estimate the court could

impose.”468

466 DFC, 172 A.3d at 370; accord Dell, 2017 WL 6375829, at *17 (explaining that, when a market is efficient, “a company’s stock price reflects the judgments of many stockholders about the company’s future prospects, based on public filings, industry information, and research conducted by equity analysts. In these circumstances, a mass of investors quickly digests all publicly available information about a company, and in trading the company’s stock, recalibrates its price to reflect the market’s adjusted, consensus valuation of the company” (internal quotation marks and footnotes omitted)). 467 DFC, 172 A.3d at 369-70; see also Dell, 2017 WL 6375829, at *17 (“[T]he price produced by an efficient market is generally a more reliable assessment of fair value than the view of a single analyst, especially an expert witness who caters her valuation to the litigation imperatives of a well-heeled client.”). 468 Applebaum, 812 A.2d at 890 (Del. 2002); see also Dell, 2017 WL 6375829, at *15 n.108 (citing Applebaum); DFC, 172 A.3d at 369 & n.116 (quoting Applebaum).

120 In this case, because Aruba’s shares “were widely traded on a public market based

upon a rich information basis,” the fair value of the petitioners’ shares “would, to an

economist, likely be best reflected by the prices at which their shares were trading as of the

merger.”469 Aruba had “a deep base of public shareholders” and “highly active trading,” so

“the price at which its shares trade is informative of fair value.”470 The unaffected thirty-

day average market price of Aruba’s stock was $17.13 per share.

Dell and DFC teach that the deal price is also entitled to substantial weight. “In

economics, the value of something is what it will fetch in the market. That is true of

corporations, just as it is true of gold.”471 For a court to give weight to the deal price, it

need not be the most reliable evidence of the Company’s value as a going concern.472 This

court has authority “to determine, in its discretion, that the deal price is the most reliable

evidence of fair value . . . , and that’s especially so in cases . . . where things like synergy

gains or minority stockholder discounts are not contested.”473

The deal price in this case resulted from an arm’s-length transaction involving a

publicly traded company without a controlling stockholder. The deal price in this case

469 DFC, 172 A.3d at 367. 470 Id. at 373. 471 Id. at 368-69 (footnote omitted). 472 Id. at 364. 473 Id. at 367; see also Dell, 2017 WL 6375829, at *16 (“In fact, the record as distilled by the trial court suggests that the deal price deserved heavy, if not dispositive, weight.”).

121 contained synergies, so it logically exceeded fair value. There is also the fact that the

petitioners failed to identify a bidder who would pay more than HP. “Fair value entails at

minimum a price some buyer is willing to pay . . . .”474 Taken together, these propositions

indicate that the deal price in this case operates as a ceiling for fair value.

The Dell and DFC decisions recognize that a deal price may include synergies and

endorse deriving an indication of fair value from the deal price by deducting synergies.475

In this case, the evidence shows that the deal generated significant synergies. Using the

low-end synergy range implies a standalone value of $21.08 per share. Using the high-end

synergy range implies a standalone value of $15.32 per share. This decision has adopted

the midpoint of $18.20 per share as its deal-price-less-synergies value.

This decision does not give any weight to the discounted cash flow analyses. As in

Dell, “this appraisal case does not present the classic scenario in which there is reason to

suspect that market forces cannot be relied upon to ensure fair treatment of the minority.”476

Discounted cash flow models are “often used in appraisal proceedings when the respondent

company was not public or was not sold in an open market check.”477

The reason for that is not that an economist wouldn’t consider the best estimate of a private company’s value to be the price it sold at in an open sale process of which all logical buyers were given full information and an equal opportunity to compete. Rather, the reason is that if such a process did not

474 Dell, 2017 WL 6375829, at *21. 475 Dell, 2017 WL 6375829, at *13; DFC, 172 A.3d at 371. 476 Dell, 2017 WL 6375829, at *26. 477 DFC, 172 A.3d at 369 n.118.

122 occur, corporate finance instructs that the value of the company to potential buyers should be reflected in its ability to generate future cash flows.478

“But, a single person’s own estimate of the cash flows are just that, a good faith estimate

by a single, reasonably informed person to predict the future. Thus, a singular discounted

cash flow model is often most helpful when there isn’t an observable market price.”479

When market evidence is available, “the Court of Chancery should be chary about

imposing the hazards that always come when a law-trained judge is forced to make a point

estimate of fair value based on widely divergent partisan expert testimony.”480

Marcus’s discounted cash flow valuation of $32.57 per share diverged substantially

from market indications. His figure is nearly double Aruba’s thirty-day average unaffected

market price of $17.13. It is approximately 32% higher than the deal price of $24.67 per

share. In a transaction involving a financial buyer that could be expected to generate few

if any combinatorial synergies, the Delaware Supreme Court recently emphasized the lack

of reliability of a discounted cash flow analysis that yielded a result that was 40% over the

deal price.481 The transaction in this case generated substantial synergies.

478 Id. 479 Id. at 370. 480 Dell, 2017 WL 6375829, at *26. 481 DFC, 172 A.3d at 362; cf. Lender Processing, 2016 WL 7324170, at *33 (“The proximity between [the discounted cash flow] outcome and the result of the sale process is comforting.”); Ancestry.com, 2015 WL 399726, at *23 (“The DCF valuation I have described is close to the market, and gives me comfort that no undetected factor skewed the sales process.”).

123 Dages’s initial discounted cash flow valuation of $19.85 and revised discounted

cash flow valuation of $19.75 fell nicely between the unaffected market price and the deal

price. His figures also landed close to HP’s standalone discounted cash flow valuation of

$18.98 and Barclay’s standalone discounted cash flow valuation of $19.93. The relative

lack of methodological rigor in the analysis, however, creates cause for concern about the

strategic selection of inputs to channel the result into this range.

The two probative indications of value in this case are the unaffected market price

of $17.13 and the deal-price-less-synergies value of approximately $18.20 per share. Using

these indicators nevertheless carries conceptual difficulties because “[t]he time for

determining the value of a dissenter’s shares is the point just before the merger transaction

‘on the date of the merger.’”482 If the value of the corporation changes between the signing

of the merger and the closing, the fair value determination must be measured by the

“operative reality” of the corporation at the effective time of the merger.483

The unaffected market price provides direct evidence of the collective view of

market participants as to Aruba’s fair value as a going concern during the period before the

announcement of the transaction, which could be different than Aruba’s fair value as of

closing. The same disconnect exists for the deal price, which provides evidence of how the

parties to the merger agreement valued Aruba during the price negotiations, which could

482 Appraisal Rights, supra, at A-33 (quoting Technicolor I, 542 A.2d at 1187). 483 Technicolor II, 684 A.2d at 298.

124 be different than Aruba’s fair value as of closing. Addressing a similar issue in the Union

Illinois case, Chief Justice Strine described the temporal gap as a “quibble” and “not a

forceful objection,” noting that “[t]he negotiation of merger terms always and necessarily

precedes consummation.”484 Observing that “[n]othing in the record persuades me that [the

company] was more valuable by [closing] than it was when the Merger terms were set,” he

continued to use the deal price as an indicator of value.485 Similarly in this case, neither

side proved that Aruba’s value had changed materially by closing, so this decision sticks

with the unaffected market price and the deal price less synergies.

The difficult question is how to choose between, weigh, or otherwise exercise my

discretion non-abusively when evaluating the two probative valuation indications. The

unaffected market price provides a direct measure of the collective judgment of numerous

market participants about Aruba’s value as a going concern. The deal price less synergies

provides an indirect measure with two significant sources of uncertainty.

One is the problem of measurement error. Under the traditional view of the efficient

capital markets hypothesis, errors are randomly distributed and cancel out.486 My deal-

484 Union Ill., 847 A.2d at 358. 485 Id. 486 See Ronald J. Gilson & Reinier H. Kraakman, The Mechanisms of Market Efficiency, 70 Va. L. Rev. 549, 581 (1984). Behavioral economics, noise theory, and chaos theory may provide reasons to question this assumption, but for the reasons already stated, I do not believe that a trial court has the flexibility to disregard the Delaware Supreme Court’s framing of the efficient capital markets hypothesis.

125 price-less-synergies figure could have errors at multiple levels. To cite just a few, I may

have erred when making my case-specific allocation of synergies to the sell-side. I might

have misinterpreted the information that Aruba’s expert cited, or that data itself could

contain sampling and measurement errors. The size of the original synergy estimates might

also be off, as could any number of individual estimates that added up to the overarching

estimates. After all, they were necessarily predictions about complex matters. Perhaps

errors at one level might counterbalance errors at another, but there is no way to know, and

the smaller number of judgments involved (compared to the number of trades generating

the market price) makes it more likely that the errors could skew the figure, just like a small

and undiversified portfolio can produce extreme results. The Delaware Supreme Court’s

expressed preference in Dell and DFC for market indicators over discounted cash flow

valuations counsels in favor of preferring market indicators over the output of a similarly

judgment-laden exercise of backing out synergies.487

The other difficulty is that my deal-price-less-synergies figure continues to

incorporate an element of value resulting from the merger. When an acquirer purchases a

widely traded firm, the premium that an acquirer is willing to pay for the entire firm

anticipates incremental value both from synergies and from the reduced agency costs that

result from unitary (or controlling) ownership.488 Like synergies, the value created by

487 Cf. Dell, 2017 WL 6375829, at *26; DFC, 172 A.3d at 388. 488 See Rationalizing Appraisal, supra, at 1038, 1049.

126 reduced agency costs results from the transaction and is not part of the going concern value

of the firm.489 The value belongs to the buyer, although the seller may extract a portion of

it through negotiations.490 Eliminating shared synergies therefore only goes part of the way

towards eliminating “any element of value arising from the accomplishment or expectation

of the merger.”491 A court also must eliminate the share of value that accrues from the

reduced agency costs.492

For Aruba, using its unaffected market price provides the more straightforward and

reliable method for estimating the value of the entity as a going concern. I could strive to

reach the same endpoint by backing out shared synergies and a share of value for reduced

agency costs, but both steps are messy and provide ample opportunities for error. For

489 See Rationalizing Appraisal, supra, at 1023-24, 1038, 1046-54, 1067; Implicit Minority Discount, supra, at 30-36, 52; Fair Value of Cornfields, supra, at 139-41. 490 See Control Premiums, supra, at 866-71; Rationalizing Appraisal, supra, at 1052-53; Implicit Minority Discount, supra, at 35, 52. 491 8 Del. C. § 262(h). 492 See Rationalizing Appraisal, supra, at 1055 (explaining that, for an acquisition of a widely held firm, “the firm’s going concern value can be estimated . . . as the actual purchase price minus synergies minus control value”). Failing to make this adjustment would treat the value of the firm as greater than the aggregated value of individual shares, which is the same analytical misstep reflected in the concept of the implicit minority discount. See Control Premiums, supra, 854-59 (explaining conflict between efficient capital markets hypothesis and implicit minority discount); Implicit Minority Discount, supra, 53 (explaining logical equivalence between correcting for a non-existent implicit minority discount and introducing a “‘third-party sale value lite’ standard in lieu of the traditional ‘proportionate share of going concern value’ standard”).

127 Aruba, the unaffected market price provides a direct estimate of the same endpoint.493

Rather than representing my own fallible determination, it distills “the collective judgment

of the many based on all the publicly available information about a given company and the

value of its shares.”494 “[T]he price produced by an efficient market is generally a more

reliable assessment of fair value than the view of a single analyst,” particularly when a trial

judge is playing the analyst’s role.495

This approach does not elevate “market value” to the governing standard under the

appraisal statute. The governing standard for fair value under the appraisal statute remains

the entity’s value as a going concern. For Aruba, the unaffected public market price

provides the best evidence of its value as a going concern.

In this case, the best evidence of Aruba’s fair value as a going concern, exclusive of

any value derived from the merger, is its thirty-day average unaffected market price of

$17.13 per share. I recognize that no one argued for this result. I also recognize that the

493 See Control Premiums, supra, at 858-59 (“The basic conclusion of the Efficient Capital Markets Hypothesis (ECMH) is that market values of companies’ shares traded in competitive and open markets are unbiased estimates of the value of the equity of such firms.”); Implicit Minority Discount, supra, at 52 (“Take the case of a publicly traded company that has no controller. Efficient market theory states that the shares of this company trade at the pro rata value of the corporation as a going concern.”); id. at 60 (“As a matter of generally accepted financial theory . . . , share prices in liquid and informed markets do generally represent that going concern value . . . .”); see also Rationalizing Appraisal, supra, at 1033-34. 494 DFC, 172 A.3d at 369-70. 495 Dell, 2017 WL 6375829, at *17

128 resulting award is lower than Aruba’s proposed figure of $19.75 per share. That figure

relied on its expert’s discounted cash flow analysis, which this decision has found

unpersuasive.

“When . . . none of the parties establishes a value that is persuasive, the Court must

make a determination based on its own analysis.”496 The appraisal statute requires that “the

Court shall determine the fair value of the shares.”497 This means that I must reach my own,

independent determination of fair value.498 That determination is $17.13 per share.

III. CONCLUSION

The petitioners are awarded $17.13 per share. The legal rate of interest, compounded

quarterly, shall accrue on this amount from the date of closing until the date of payment.

The parties shall cooperate in preparing a final order. If the parties identify additional issues

that need to be resolved, they shall submit a joint letter within two weeks that explains the

issues and recommends a schedule for bringing this case to conclusion, at least at the trial

court level.

496 Cooper v. Pabst Brewing Co., 1993 WL 208763, at *8 (Del. Ch. June 8, 1993) (citing In re Shell Oil Co., 607 A.2d 1213 (Del. 1992)); accord Del. Open MRI Radiology Assocs. P.A. v. Kessler, 898 A.2d 290, 310-11 (Del. Ch. 2006). See generally Appraisal Rights, supra, at A-89 to A-90 (“If both parties fail to meet the preponderance standard on the ultimate question of fair value, the Court is required under the statute to make its own determination.”). 497 8 Del. C. § 262(h). 498 Dell, 2017 WL 6375829, at *13 (“In reality, the burden falls on the trial judge to determine fair value, using all relevant factors.” (internal quotation marks and alterations omitted)).