In re Appraisal of Stillwater Mining Company

Opinion

IN THE COURT OF CHANCERY OF THE STATE OF DELAWARE

IN RE APPRAISAL OF STILLWATER ) Consol. C.A. No. MINING COMPANY ) 2017-0385-JTL

MEMORANDUM OPINION

Date Submitted: May 23, 2019 Date Decided: August 21, 2019

Samuel T. Hirzel, II, Elizabeth A. DeFelice, HEYMAN ENERIO GATTUSO & HIRZEL LLP, Wilmington, Delaware; Lawrence M. Rolnick, Steven M. Hecht, Jonathan M. Kass, Glenn McGillivray, LOWENSTEIN SANDLER LLP, New York, New York; Attorneys for Petitioners.

S. Mark Hurd, Lauren Neal Bennett, MORRIS, NICHOLS, ARSHT & TUNNELL LLP, Wilmington, Delaware; James R. Warnot, Jr., Adam S. Lurie, Brenda D. DiLuigi, Nicole E. Jerry, Elizabeth M. Raulston, LINKLATERS LLP, New York, New York; Attorneys for Respondent.

LASTER, V.C.

This post-trial decision determines the fair value of the common stock of Stillwater Mining Company (“Stillwater” or the “Company”) as of May 4, 2017, which is when Sibanye Gold Limited completed its acquisition of Stillwater through a reverse-triangular merger (the “Merger”). Pursuant to an agreement and plan of merger dated December 9, 2016 (the “Merger Agreement”), each share of Stillwater common stock was converted at closing into the right to receive $18.00, subject to the right of each holder to eschew the merger consideration and seek appraisal.

The petitioners perfected their appraisal rights and litigated this appraisal proceeding. They contended that Stillwater’s fair value was $25.91 per share. To justify this outcome, they relied on an expert who valued Stillwater using a discounted cash flow (“DCF”) model.

The respondent in an appraisal proceeding is technically the surviving corporation, but the real party in interest is the acquirer. The petitioners’ true opponent in this proceeding was Sibanye.

Sibanye contended that Stillwater’s fair value was $17.63 per share. To justify this outcome, Sibanye relied on a combination of metrics, including the deal price, Stillwater’s unaffected trading price with an adjustment for a valuation increase between the unaffected date and closing, and an expert valuation based on a DCF model.

Sibanye proved that the sale process was sufficiently reliable to make the deal price a persuasive indicator of fair value. Although Sibanye argued for a deduction from the deal price to account for value arising from the Merger, Sibanye failed to prove that an adjustment was warranted.

The parties engaged in lengthy debate over whether Stillwater’s adjusted trading price could provide a persuasive indicator of fair value. The reliability of the adjusted trading price depended on the reliability of the unaffected trading price, and both sides engaged experts who conducted analyses and offered opinions about the attributes of the market for Stillwater’s common stock. The evidence demonstrated that Stillwater’s trading price could provide a persuasive indicator of value, but that it was a less persuasive indicator than the deal price. This decision therefore does not use a trading price metric.

Neither side proved that its DCF valuation provided a persuasive indicator of fair value. The experts disagreed over too many inputs, and the resulting valuation swings were too great, for this decision to rely on a model when a market-tested indicator is available.

This decision concludes that the deal price is the most persuasive indicator of fair value. Relying on any of the other valuation metrics would introduce error. The fair value of the Stillwater on the valuation date was therefore $18.00 per share.

I. FACTUAL BACKGROUND The parties generated an extensive evidentiary record. They commendably reached agreement on 283 stipulations of fact. During four days of trial, they introduced 909 exhibits and lodged twenty-one depositions in evidence. Three fact witnesses and seven expert witnesses testified live. What follows are the court’s findings based on a preponderance of the evidence.1

1 Citations in the form “PTO ¶ ––” refer to stipulated facts in the pre-trial order. Dkt.

209. Citations in the form “[Name] Tr.” refer to witness testimony from the trial transcript. Citations in the form “[Name] Dep.” refer to witness testimony from a deposition

3

A. The Company At the time of the Merger, Stillwater was a Delaware corporation engaged in the business of extracting, processing, smelting, and refining minerals from an orebody known as the J-M Reef. Located in in the western United States, the J-M Reef contains deposits of palladium, platinum, and rhodium, which are known in the mining industry as “platinum group metals” or “PGMs.” These metals are rare, and the J-M Reef is the only PGM asset in the United States. The other principal sources of PGMs are located in South Africa, Russia, and Zimbabwe, which present significantly greater political risk.

Stillwater was headquartered in Littleton, Colorado, and its common stock traded on the New York Stock Exchange under the symbol “SWC.” Stillwater’s trading price was heavily influenced by commodity prices for palladium and, to a lesser degree, platinum.

At the time of the Merger, Stillwater’s operations consisted of two producing mines in south central Montana: the Stillwater Mine and the East Boulder Mine. Stillwater’s other assets were development projects or exploratory properties that were not yet generating revenue.

At the time of the Merger, Stillwater’s two development projects were Blitz and Lower East Boulder. Blitz expanded the Stillwater Mine eastward. Lower East Boulder was a contemplated expansion of the East Boulder mine. Stillwater’s two exploratory

transcript. Citations in the form “JX –– at ––” refer to a trial exhibit with the page designated by the last three digits of the control or JX number. If a trial exhibit used paragraph or section numbers, then references are by paragraph or section.

4

properties in the J-M Reef were Iron Creek and the Boulder Extension. Outside of the J-M Reef, Stillwater owned two other exploratory properties: (i) Altar, a copper-gold-porphyry deposit in the San Juan province of Argentina, and (ii) Marathon, a copper-PGM deposit in Ontario, Canada.

At the time of the Merger, Michael “Mick” McMullen served as Stillwater’s President and CEO and as a member of its board of directors (the “Board”). The other six members of the Board were independent, outside directors:

 George Bee was a mining engineer who had held senior management positions or served on the boards of other mining companies.

 Patrice Merrin had served as an executive or director for numerous companies and was a director of Glencore plc, a multi-national mining firm. Merrin chaired the Board’s Corporate Governance and Nominating Committee.

 Peter O’Hagan had worked at Goldman Sachs for nearly twenty-three years, including as co-head of its global commodities business.

 Michael Parrett was a Chartered Professional Accountant who had served in senior management positions and as a director for other mining companies.

 Brian Schweitzer had served as Governor of Montana. He was Chairman of the Board.

 Gary Sugar had spent thirty-two years at RBC Capital Markets, where he specialized in the mining sector. He served on the boards of other mining companies.

B. McMullen Convinces The Board To Build A Mid-Cap Mining Company.

McMullen was hired in December 2013 as a “turnaround CEO.” McMullen Tr. 814– 16; see Schweitzer Tr. 170. By early 2015, McMullen had refocused Stillwater’s operations, cut costs, and generally turned the Company around. At this point, McMullen believed that market conditions favored the creation of a mid-cap mining company. He

5

thought Stillwater could achieve this outcome either by growing through acquisitions or by combining with another industry player through a merger of equals.

During a meeting of the Board in June 2015, McMullen gave a lengthy presentation on Company strategy that devoted twenty-six slides to various alternatives. See JX 44 at ‘848 to ‘874. McMullen’s presentation discussed means of increasing earnings, increasing the trading multiple, and optimizing the capital structure, and then turned to the pros and cons of selling some or all of the business. The presentation was particularly negative about the prospect of a sale. See id. at ‘866 to ‘868. In another presentation, McMullen devoted over forty slides to discussing candidates for acquisitions or mergers of equals. See id. at ‘929 to ‘970.

In addition to his own presentation, McMullen provided the Board with presentations from three investment banks. McMullen had a close relationship Dan Vujcic, then an investment banker with Jefferies Financial Group, Inc., and the Jefferies presentation was the most detailed. It analyzed an acquisition of another base metals company, focusing on Sandfire Resources NL, Western Areas Ltd., and Panoramic Resources Ltd. It also analyzed the possible acquisition of a downstream company, a possible spinoff of Stillwater’s processing and trading business, and the option of maintaining the status quo. See id. at ‘014 to ‘080. A presentation from BMO Capital Markets was more of a high-level pitch book, but it identified selected acquisition opportunities. See id. at ‘081 to ‘183. A presentation from Nomura Holdings, Inc. discussed alternatives for refinancing Stillwater’s convertible bonds. See JX 44 at ‘164 to ‘182.

6

Sibanye has argued that this meeting marked the start of the Board’s careful and thoughtful consideration of a sale of the Company, but the purpose of the meeting was not to prepare the Board for a sale. McMullen hoped to convince the Board to back him in creating a mid-cap mining company.2 The Board, however, resisted, recalling unsuccessful acquisitions that had necessitated hiring a turnaround CEO in the first place. During the June 2015 meeting, the Board did not provide McMullen with a mandate to pursue any strategic options. See JX 43.

After the June 2015 meeting, McMullen kept looking for opportunities to build a mid-cap mining company. During the second half of 2015, McMullen worked with Jefferies, BMO, and Citigroup to identify acquisition targets and merger-of-equals candidates.3 McMullen was focused on an acquisition, particularly “something not in the PGM space to diversify risk.” JX 59.

2 The two slides in the management presentations that addressed a sale contained comments like “[f]inding a willing buyer with higher priced currency is difficult,” “[m]uch of the value from Blitz, Lower East Boulder and recycle ramp up yet to be recognized by the market and potential buyers,” and the “[r]ecent downward trend in PGM prices not the right environment in which to be a seller.” Id. at ‘866 to ‘867. Out of the nearly 190 slides in the banker presentations, only one discussed a possible sale. There, BMO opined that selling was “unlikely to be a value maximizing strategy until value has been extracted from all the other alternatives” available to the Company. Id. at ‘108.

3 See, e.g., JX 50 at ‘586 to ‘594; JX 52; JX 53; JX 55; JX 57; JX 58; JX 67; JX 68;

PTO ¶¶ 138–39. Although principally focused on acquisitions, McMullen asked BMO in an October 2015 email for its views about “who would potentially be a buyer of Stillwater in an M+A deal?” JX 57 at ‘920. BMO sent back a list of twenty-one candidates, but warned that “[g]enerally as a whole we would say that we do not believe there is a high level of current interest and capability for an acquisition of Stillwater.” Id. at ‘919.

7

During a meeting of the Board in October 2015, McMullen gave another presentation on the Company’s strategy. See JX 61 at ‘102 to ‘127. He highlighted the risks Stillwater faced because of its dependence on palladium, which was used principally in catalytic converters. His presentation discussed the disruptive threat posed by electric cars, which could displace gasoline-powered cars and render catalytic converters obsolete. See id. at ‘105 (“Know Your Enemy—Electric Cars”). He recommended making a diversifying acquisition from which Stillwater would “emerge as a multi mine, multi commodity and multi jurisdiction mid cap miner with a bullet proof balance sheet.” Id. at ‘127. He then reviewed six possible candidates: Sandfire, Western Areas, Panoramic, Northern Star Resources Ltd., Imperial Metals, and Hecla Mining Co. See id. at ‘128 to ‘179. He also circulated a presentation from Jefferies that discussed an acquisition of Sandfire. See id. at ‘249 to ‘292. During the weeks after the meeting, Jefferies provided McMullen with more detailed analyses of a deal with Northern Star, a large gold producer in Australia. See JX 67; JX 68.

In December 2015, McMullen and a team from Stillwater visited the mining operations of Northern Star, where McMullen had a close relationship with senior management. During the visit, McMullen met with the CEO and CFO of Northern Star and discussed a potential merger of equals. See PTO ¶ 145; JX 73 at ‘867; see also JX 61 at ‘282 to ‘286; JX 67. At this point in time, a merger of equals with Northern Star was McMullen’s top choice among Stillwater’s strategic options.

During meeting of the Board in January 2016, McMullen gave another presentation on the Company’s strategy. See JX 86 at ‘002 to ‘040. As with the meetings in June and

8

October 2015, his goal was to convince the Board to authorize him to build Stillwater into a mid-cap metals company. See JX 78 (McMullen discussing his desire to “come away from [the January] board meeting with a clear mandate”). McMullen recommended a merger of equals with Northern Star as the best option, telling the Board that the transaction “would make a very strong mid cap precious metals miner.” JX 86 at ‘038. If Northern Star would not engage, then he recommended acquiring Sandfire or Western Areas. See id. at ‘039. He also identified some smaller acquisitions that “should be pursued independently” and “[r]egardless of whether Stillwater completes one of the larger deals.” Id. at ‘040. Later in the meeting, he provided additional information about the proposed M&A strategy and further detail about Northern Star, Sandfire, Western Areas, Panoramic, Hecla, and Imperial. See id. at ‘’320 to ‘367. McMullen also distributed a presentation from Jefferies that analyzed mergers with Northern Star and Western Areas. See id. at ‘275 to ‘319 At the conclusion of the January 2016 meeting, the Board gave management a mandate, but it was broad and vague. According to the minutes, “[t]he Board provided management with a sense of the Board for management to continue to pursue the options as discussed, but to return to the Board for any final decision.” JX 90. During this litigation, Sibanye has argued that this mandate authorized management to pursue a sale of the Company, but that is not accurate.4 McMullen put it best when he told a banker at

4 Only one slide in McMullen’s presentation referenced a sale of the Company, and it advised that there was a “[v]ery limited number of potential buyers” and that because “commodity prices and sentiment are low,” the Company “would not realize full value potentially.” JX 86 at ‘025. By contrast, he presented multiple slides discussing positively

9

Blackstone that he had “finally convinced the Stillwater board to go off and buy some things.” JX 93 at ‘628; see Schweitzer Tr. 187. C. The Company’s Stock Price While McMullen was trying to convince the Board to let him “buy some things,”

Stillwater’s stock price was falling. The decline began in June 2016 and continued steadily through December. Over the course of this six month period, Stillwater’s stock price fell by over 40%, dropping from $14.46 per share on June 1 to $8.57 per share on December 31. The market drop did not reflect any problems with Stillwater’s operations. Instead, it reflected a decline in the spot price of palladium, which fell by 27% from $773.70 per ounce on June 1 to $562.98 per ounce on December 31. PTO Exs. A, B.

During the Board meeting in January 2016, McMullen had told the Board that “[d]espite our stock being down 40%, we still have options open to us today.” JX 86 at ‘012. But during the weeks following the January 2016 meeting, the stock price fell further. On January 19, it closed at $5.29 per share, down 38% from its closing price of $8.57 per share on December 31. The drop corresponded with further declines in the spot price of palladium, which closed on January 19 at $494.83 per ounce, down another 12% from its close of $562.98 per ounce on December 31.

The Company’s dismal stock performance caused McMullen to conclude that Stillwater did not have a currency that it could use for either an acquisition or a merger of

how the Company could deploy its “capital and currency” (its stock) to make an acquisition. See id. at ‘026 to ‘035.

10

equals. JX 93 at ‘628 (“[U]nfortunately the stock price has collapsed in the last 2 weeks and I don’t think Stillwater has the currency to do anything anymore. Ce [sic] la vie.”); see McMullen Tr. 826; JX 97 at ‘308 to ‘310, ‘313. He felt Stillwater had missed its opportunity to expand and was now just an “an option play on the P[alladium] price.” JX 93 at ‘628; see JX 97 at ‘313 At this point, McMullen told a banker at Blackstone that “[s]itting around for one or two years waiting for the price to recover” was “not my idea of a job.” McMullen Tr. 828; JX 93 at ‘628. McMullen did not view himself as an “operational CEO.” McMullen Tr. 814–16. He thought he “would become bored.” McMullen Tr. 828. With his contract set to expire at the end of the year, McMullen began thinking about what he would do next, including the possibility of building a mining portfolio company for Blackstone. See McMullen Tr. 828; JX 93 at ‘627 to ‘628. D. Sibanye Contacts McMullen.

On January 30, 2016, Sibanye reached out through BMO to arrange a meeting between McMullen and Sibanye’s CEO, Neal Froneman. Without telling the Board, McMullen accepted.

The meeting took place at an industry conference on March 1, 2016. PTO ¶ 161.

When Froneman broached the subject of buying Stillwater, McMullen was receptive. He asked Froneman to provide “an informal proposal” in writing that included “an idea of valuation” and “transaction structure.” JX 109 at ‘976; see PTO ¶ 164. Froneman had the impression that a deal “was doable if we got the valuation right.” JX 109 at ‘976.

11

After the meeting, Froneman asked McMullen for “specific guidance” about what would be acceptable. JX 110. McMullen indicated that Sibanye’s offer should include “a large cash component.” JX 113 at ‘175. He also told Froneman during these early discussions that an acceptable transaction should be priced at a premium of 30% over Stillwater’s thirty-day volume-weighted average price (“VWAP”). Stewart Dep. 39; see also JX 162 at ‘283. Froneman agreed in principle to this pricing metric, and he began organizing a team to visit Stillwater’s mines. See JX 113 at ‘174 to ‘175. Froneman asked to enter into a confidentiality agreement to facilitate diligence, but McMullen rejected the request, commenting that he wanted “to see some form of indicative, non-binding and highly confidential terms of a transaction before we go too far down the path.” Id. at ‘174.

McMullen took all of these actions without involving the Board. Indeed, he did not even inform the Board about Sibanye’s approach. See Schweitzer Tr. 189–92; Wadman Tr. 657. Instead, on March 25, 2016, he agreed to extend his employment for an additional two years. JX 114 § 4.1. His original employment agreement had been scheduled to terminate on December 31, 2016, and the Board had expected that because McMullen was a short- term, turnaround CEO, he would not stay beyond that date. Wadman Tr. 670–71; see Wadman Dep. at 341; Schweitzer Tr. 170, 193. But with acquisition talks in the offing, McMullen agreed to a new deal. See JX 114.

The new employment agreement permitted McMullen to serve concurrently as a director of Nevada Iron Limited and New Chris Minerals Limited, which later became GT Gold Corp. See JX 114 § 3.1, Ex. A. During 2016, McMullen did more than serve on the boards of these companies. He became Executive Chairman and CEO of Nevada Iron, and

12

he served as Non-Executive Chairman and President of New Chris. See McMullen Tr. 863– 64; McMullen Dep. 45, 553; JX 93 at ‘628. Both companies were Australian resource firms whose equity comprised a significant portion of McMullen’s net worth. JX 157 at ‘315; see McMullen Tr. 709, 863–64. Over the next year, while McMullen was busy selling the Company, he also caused Nevada Iron and New Chris to engage in transformative transactions.5 In May 2016, the Board held its next regular meeting. In connection with that meeting, McMullen did not inform the Board about Sibanye’s approach or his discussions with Sibanye.6 E. Sibanye Submits An Indication Of Interest.

During the first week of June 2016, executives from both Sibanye and Northern Star toured the Company’s mines. PTO ¶¶ 171–72. Sibanye toured as part of their exploration of a potential acquisition of the Company. Northern Star toured separately, ostensibly as part of a mutual benchmarking exercise but really in connection with a potential merger of equals. McMullen and the Company’s CFO, Christopher Bateman, led Sibanye and Northern Star on separate tours and ensured that neither saw one another. McMullen

5 See JX 138; JX 139 at ‘831; JX 154 at ‘087; JX 155; JX 157 at ‘315; McMullen Tr. 709–10; see also JX 349.

6 See Schweitzer Tr. 189–90. McMullen testified that he told Schweitzer and Merrin about Sibanye’s approach after his initial meeting with Froneman. He also claimed that he kept the Board informed as discussions progressed. McMullen’s self-interested testimony conflicted with Schweitzer’s more credible testimony and other record evidence.

13

claimed that despite keeping the two teams separate, each knew that the other was on site because McMullen and Bateman would alternate between the tours and McMullen had them both sign the visitors log. McMullen said he did this as a clever way to create competition between the firms. See McMullen Tr. 726–27.

After the visits, McMullen believed that a deal with Sibanye was more likely than with Northern Star. See JX 140 at ‘048; JX 142. Toward the end of June 2016, Northern Star reported that they were primarily interested in a joint venture involving Blitz. JX 145 at ‘845. That possibility did not interest McMullen. Id. Meanwhile, McMullen pushed Sibanye to provide an indication of interest in advance of the Board’s next meeting, which was scheduled for July 28, 2016.7 Sibanye began working with Citigroup to develop its bid. Two of the Citigroup bankers had previously advised McMullen and Bateman about the Company’s alternatives. As part of its advice, Citigroup had recommended against a sale of the Company because of the limited universe of potential buyers. See JX 32 at ‘829; cf. JX 42 at ‘422.

On July 21, 2016, Sibanye provided McMullen with a non-binding indication of interest to acquire Stillwater at $15.75 per share in cash, which valued the Company at $1.9 billion. PTO ¶ 177; JX 165. The letter described that price as reflecting “a 30% premium

7 See JX 152 at ‘532 ‘533. At trial, McMullen testified that after Sibanye conducted its site visit, the Board told him that they wanted “some sort of written expression of interest” before starting “a data room process.” McMullen Tr. 728–29. That testimony was not credible. The evidence indicates that McMullen did not brief the Board about a potential transaction with Sibanye until the July 2015 board meeting. See Schweitzer Tr. 189–90.

14

to Stillwater’s volume-weighted average share price [(VWAP)] of US$12.12 over the last 20 trading days prior to 20 July 2016.” JX 165 at ‘880; see PTO ¶ 178.

As suggested by Sibanye’s offer, Stillwater’s stock price had mostly recovered, reflecting a recovery in the price of palladium. At the beginning of July 2016, the stock closed at $12.25 per share, up 132% from its low of $5.29 in January. During that same period, the palladium spot price had increased 22% to $605.63 per ounce. PTO Exs. A, B. Despite the stock’s performance, McMullen did not revisit potential acquisitions or a merger of equals. He was now focused on selling the Company. See JX 156 (email from Vujcic to McMullen stating, “[W]e’ll make sure the company gets sold. Don’t worry about that.”). F. McMullen Presents The Indication Of Interest To The Board.

On July 27 and 28, 2016, the Board held a regularly scheduled meeting. At the end of the two-day meeting, the directors held a forty-five minute “executive session” with McMullen, who distributed and walked through a presentation titled “Business Development Update.” JX 151 at ‘551; see Schweitzer Tr. 193; JX 526 at ‘377; Wadman Tr. 657–64. The presentation compared the Company’s recent performance to various potential transaction partners, then described the pros and cons of transactions with Northern Star and Sibanye. After summarizing the terms of Sibanye’s expression of interest, the presentation described the premium as “within the right range for shareholder value” and “broadly within the range of mining transactions.” JX 151 at ‘568. McMullen gave his “strong recommendation . . . to engage with Sibanye and attempt to conclude [due diligence] as quickly as possible (likely to take 2 months) and achieve a higher price.” Id.

15

McMullen added that he would “look to engage with other potential bidders on a low key and informal basis to determine if there are alternative bidders.” Id. He warned: “The list of other potential bidders is short given the commodity, size of transaction and whether [Stillwater’s] shareholders would want their paper. The process of determining if there are alternatives will not be a long process.” Id. He also told the directors that “[t]he market appears to be open for people to carry out M+A, and asset values have risen to a level where you want to be a seller rather than a buyer.” Id.

Brent Wadman, the Company’s General Counsel, became concerned about what took place during the July meeting. He had not been asked to stay for the executive session and was not given access to McMullen’s presentation. See JX 526 at ‘377; Wadman Tr. 657–64. He suspected that McMullen was running a sale process on his own, without Board oversight, and potentially using it as a means of exiting from the Company. Wadman believed that as General Counsel, he should have been involved. After the July meeting, Wadman asked McMullen to include him in the planning process. McMullen rebuffed him, saying that Wadman would be “brought in at a later date” and “offer[ing] no other information.” JX 526 at ‘377; Wadman Tr. 658.

After the July meeting, McMullen told Sibanye to submit its list of due diligence questions so the Company could start pulling the information together. He told Sibanye to direct all inquiries to himself or Bateman. See PTO ¶ 181; JX 183. G. McMullen Remains Committed To Sibanye.

On August 9, 2016, Stillwater and Sibanye entered into a confidentiality agreement, and Sibanye gained access to the data room. PTO ¶ 183; JX 525 at 26; see also JX 194. On

16

August 10, the Board met again. See JX 193. McMullen testified that at this meeting, the Board instructed him “to go out and . . . to sign the NDAs with the likes of Sibanye, and then, also, . . . to get as much interest as possible.” McMullen Tr. 835.

Rather than working closely with an investment bank to develop a process designed to generate “as much interest as possible,” McMullen pressed forward with Sibanye. He interacted with some investment banks, but in a haphazard and unstructured way. For example, back in July 2016, a Macquarie banker had asked McMullen to meet for a market update. See JX 167. On August 10, the same day that the Board met, Macquarie proposed a formal engagement. Five days later, McMullen told Macquarie that it was “a bit early for us I think to be signing anyone up.” JX 196.

One week after the Board meeting, on August 18, 2016, McMullen and Bateman met with Bank of America Merrill Lynch (“BAML”), who had arranged the meeting to pitch Stillwater on possible mergers of equals. See JX 199; see also JX 163; JX 190. The BAML presentation materials did not discuss a sale of the Company or mention Sibanye, and McMullen and Bateman did not use the meeting to identify other possible acquirers. Instead, the BAML bankers got “the sense . . . that a sale was a possibility,” and so they decided on their own to “pivot[] to focus more, as time went on, on that.” Hunt Dep. 35.

Acting on their own, the BAML bankers developed a list of fifteen possible acquirers whom they approached independently, pitching a potential acquisition of Stillwater as “a banker idea.” JX 206 at ‘360. The record does not reveal exactly how many companies BAML contacted, what the BAML bankers said, or how seriously the companies took the pitch. Because BAML did not know that Stillwater was in discussions

17

with Sibanye, they reached out to Sibanye as part of these efforts, ironically describing that a deal for Stillwater would be “[a] little pricey.” JX 207 at ‘093. In the end, five companies expressed interest: Sibanye; Hecla; Coeur Mining, Inc.; CITIC Resources Holdings Limited, and Anemka Resources Ltd. See JX 211; JX 213; JX 214; JX 217 at ‘588 to ‘591.

Having made these calls on their own, the BAML bankers held a follow-up meeting with McMullen and Bateman on September 7, 2016. The pitch book identified the parties contacted and expressing interest. It then described three types of sale processes Stillwater could pursue: a “proprietary process” with a single bidder, a targeted auction involving a limited number of likely buyers, or a broad auction involving outreach to many potentially interested parties. JX 217 at ‘603. BAML recommended against the proprietary process because the absence of competition would minimize Stillwater’s negotiating leverage. BAML also recommended against a broad auction, given the existence of a “narrow list of most likely buyers.” Id. This left a targeted auction as the recommended route.

The pitch book described an illustrative timeline for a sale process. BAML recommended allocating the rest of September 2016 to contact potential buyers. During October and early November, the Company would enter into confidentiality agreements, respond to diligence requests, and then receive and evaluate initial indications of interest. From mid-November through early January 2017, the Company would host site visits, provide additional diligence, and then solicit and receive final bids. JX 217 at ‘605.

Nothing formal came out of the September 7 meeting. McMullen and Bateman did not instruct BAML to proceed, nor did they take BAML’s recommendation to the Board.

18

Instead, McMullen and Bateman asked BAML and Vujcic, the investment banker who had been with Jefferies and was now working on his own, to arrange meetings with potential suitors at an industry conference during the week of September 20, 2016. BAML arranged a meeting with Coeur, and McMullen arranged a meeting with Hecla. See JX 220 at ‘609; JX 222; JX 224; PTO ¶ 190–91. Vujcic set up meetings with Kinross Gold Corporation and Gold Fields Limited, neither of whom had expressed interest. During each meeting, McMullen conducted what he called a “soft sound” regarding potential interest in buying the Company. PTO ¶ 192; see id. ¶¶ 193–97.

On the last night of the conference, McMullen had dinner with Froneman.

McMullen told him that he “remain[ed] committed” to a deal with Sibanye and that “no one else is in the data room,” but cautioned that he was “being flooded by investment banks” pitching ideas for deals with gold-mining companies. JX 231 at ‘711.

After the conference, BAML sent McMullen “a fairly detailed timeline” for a more compressed sale process. JX 225 at ‘629. The new timeline contemplated the process starting during the last week of September and ending during the first week of December. See id. at ‘632. BAML anticipated site visits taking place during November as part of the due diligence phase, but McMullen told BAML that the site visits needed to take place earlier in the process before parties sent their initial indications of interest: “Unless people get to site, they can’t appreciate the scale of it and will not be putting their best foot forward in the indicative, non binding offers.” JX 229 at ‘603. BAML revised the timeline, noting that they were “putting [it] together in a vacuum of info on what’s taken place.” Id. At this

19

point, BAML had not been retained and did not yet know about Sibanye’s bid. They only knew about their own, independent efforts to solicit interest. H. The Board Decides Not To Form A Special Committee.

In anticipation of a board meeting on October 3, 2016, Wadman circulated a “list of potential buyers” to the directors. JX 234. The list identified eighteen companies and the status of Stillwater’s discussions with each. According to the list, Sibanye had completed its first phase of diligence and was working with Citigroup to secure financing. Hecla and Coeur had expressed interest, entered into non-disclosure agreements (“NDAs”), and scheduled site visits. Northern Star was listed as “interested but very foucssed [sic] on a gold deal.” Id. at ‘630. Six other companies were described as “[p]otentially interested” or as having “some interest,” including Anglo American Platinum Limited (“Amplats”). Id. Six candidates were described as “[u]nlikely” and two as “not interested.” Id. The list omitted CITIC and Anemka, even though both had expressed interest when BAML called with its “banker’s idea.”

The list identified a representative who was responsible for interacting with each company. Evidencing the uncoordinated, unstructured nature of the Company’s process, the list identified a hodgepodge of names. Vujcic was the contact for eight companies. BAML was the contact for four companies. Jefferies was the contact for another three. Macquarie was the contact for one company. An executive at New Chris, the company where McMullen served as Non-Executive Chairman and President, was listed as the contact for another company. No one had been formally engaged. Two companies had no contact listed.

20

During the meeting, McMullen reported on the Company’s outreach to the various parties. After his presentation, the directors instructed McMullen to obtain formal proposals from investment banks for a sell-side engagement. The Board also instructed McMullen to create a cash flow model that could be used to value the Company. See JX 246 at ‘308 to ‘309.

Ever since the July 2016 meeting, Wadman had been concerned that McMullen was running a sale process to facilitate his exit from the Company. After McMullen rebuffed him, Wadman had shared his concerns privately with Schweitzer and Merrin. See Wadman Tr. 664–65; Schweitzer Tr. 157–58, 194. Neither took action.

During the October meeting, Wadman presented his concerns to the full Board and recommended the formation of a special committee to oversee the sale process. Lucy Stark of Holland & Hart LLP, the Company’s longstanding outside counsel, disagreed and advised the Board that she did not believe any conflict existed that warranted the creation of a special committee. JX 246 at ‘309; see Schweitzer Tr. 159.

The directors other than McMullen then met in executive session. Schweitzer reported to Wadman that the Board had decided to form a special committee, and Wadman drafted a set of minutes memorializing the decision. See JX 238 at ‘245; Wadman Dep. 134–35; see also Schweitzer Tr. 205–06. But in the meantime, McMullen learned of the decision from two other directors. McMullen Tr. 745–47. The final minutes described the outcome of the executive session as follows:

- No decision was made to pursue or not pursue a potential strategic transaction at this time. The Board further discussed the potential for a committee and agreed that, should the need arise, the committee would

21

consist of the entire Board with the exception of the CEO. It also discussed timing and the potential engagement of an investment banking firm to assist in the assessment process.

JX 246 at ‘310. I. McMullen Continues To Focus on Sibyane.

On October 15, 2016, almost two weeks after the Board directed McMullen to solicit terms from investment bankers, McMullen finally drafted and sent out an email asking bankers to respond “by no later than COB Wednesday Oct 19 2016.” JX 279 at ‘867. Other than Macquarie, the record does not reflect what bankers received the email or whom McMullen solicited, but Macquarie, BMO, BAML, and Jefferies submitted proposals.

On October 17, 2016, Froneman told McMullen that Sibanye’s offer of a “30% premium to VWAP remained unchanged” and that Sibanye’s board of directors unanimously supported the transaction. JX 281 at ‘425. McMullen responded that he remained fully supportive of the deal. He also shared that Stillwater did not yet have a banker, telling Froneman that he had started reaching out to investment banks on a no- names basis. Demonstrating his commitment to the deal, McMullen told Froneman that he would be happy to have Stillwater’s legal advisors start putting together an initial sales agreement. Id.

The Board met again on October 26 and 27, 2016. After reviewing the proposals from the investment banks, the Board narrowed the list to BMO and BAML. JX 295 at ‘790. Vujcic, whom McMullen regarded as his “in house banker,” summarized the state of the Company’s outreach. JX 293 at ‘521. Compare JX 262 at ‘485, with JX 234 at ‘630. He reported that third parties exhibited a general “[l]ack of knowledge around the

22

significant improvement in operations and general performance,” and he reported that a number of parties were either focused on other deals, not considering M&A because of prior bad acquisitions, or not considering PGM companies because of negative associations with risky jurisdictions like South Africa and Russia. JX 293 at ‘522. For the first time, the Board authorized management “to engage in discussions with strategic buyers, financial buyers or any other party interested in consummating a potential strategic transaction with the [Company].” JX 296 at ‘791.

After the meeting, McMullen scheduled a second site visit for Sibanye and discussed the “timelines to and post announcement” with Froneman. JX 315 at ‘291 to ‘292; see PTO ¶ 214. Sibanye convinced McMullen that they needed to announce the deal by mid-December 2016. See JX 281 at ‘425; JX 282 at ‘776; see also PTO ¶ 241. J. BAML Begins An Abbreviated Pre-Signing Market Check.

On November 7, 2016, the Board formally retained BAML. PTO ¶¶ 216–17; see JX 323 at ‘371. The Board also decided to hire “additional legal counsel with substantial experience in advising Delaware publicly traded companies in respect of potential strategic transactions.” JX 323 at ‘372. Four days later, the Board retained Jones Day. PTO ¶ 232.

On November 8, 2016, Bateman sent BAML a package of information that included Sibanye’s indication of interest from July, the non-disclosure agreements with Hecla and Coeur, a cash flow model, and instructions for accessing the data room. See JX 325; JX 326; JX 327; JX 328; JX 329. The next day, BAML sent management a slide deck titled “M&A Process Considerations.” JX 331 at ‘277.

23

BAML understood from management that Sibanye wanted to sign up a deal in December 2016, so BAML proposed to complete its outreach to a list of parties in just two days. That timeframe was drastically shorter than the four weeks that BAML had recommended in September 2016. Anyone who expressed interest would have three weeks to conduct diligence and submit an indication of interest, just half of the six weeks that BAML had recommended in September. At that point, the Board would decide whether to proceed with Sibanye or engage with the other bidders. PTO ¶ 226; see JX 331 at ‘280. Even though McMullen had previously told BAML that it was critical for potential bidders to visit the Company’s mines before making an initial indication of interest, BAML’s compressed timeline did not contemplate that step.

BAML’s presentation identified twenty-eight third parties divided into four categories:

 “Interested Parties”—Sibanye, Coeur, and Hecla.

 “Possibly Interested Parties”—Gold Fields, Independence Group NL, Kinross, MMG Limited, Rio Tinto, and South32 Limited.

 “Additional Parties To Contact”—Alamos Gold Inc., Anemka, CITIC, Fresnillo plc, Goldcorp Inc., IAMGOLD Corporation, Impala Platinum Holdings Limited, New Gold Inc., Northam Platinum Limited, Pan American Silver Corporation, X2 Resources, and Yamana Gold Inc.

 “Not Interested”—Northern Star, Amplats, Eldorado Gold Corporation, Evolution Mining Limited, Newcrest Mining Limited, Newmont Mining Corporation, and OZ Minerals.

JX 331 at ‘279. Anemka and CITIC were listed as “Additional Parties to Contact,” even though they had expressed interest during BAML’s earlier independent outreach.

24

OceanaGold Corporation and Boliden AB, whom Vujcic had included in his review of the Company’s outreach, were omitted from BAML’s list.

BAML’s presentation included scripts for its bankers to use when making their calls.

For “Possibly Interested Parties,” the script stated:

 Announce participants and remind parties of confidentiality;

 BofA Merrill Lynch has been retained by Stillwater Mining Company to explore strategic alternatives;

 We understand you have had some discussions previously with our client;

 We would like to further clarify your potential interest in Stillwater as the process moves forward;

 Do you have any interest to learn more?

 If so, we would suggest you sign an NDA for access to diligence on the company.

PTO ¶ 225 (formatting added); JX 331 at ‘281. For the “Additional Parties To Contact,” the script omitted Stillwater’s name and asked generally about interest in the PGM sector.

 Announce participants and remind parties of confidentiality;

 We are calling to gauge your potential interest in a situation in the PGM sector;

 Our client is a leading player and low cost producer of PGMs and substantial organic production growth;

 Do you have any interest to learn more?

 If yes, disclose that our client is Stillwater and suggest they sign an NDA for access to diligence.

25

PTO ¶ 224 (formatting added); JX 331 at ‘281. For Hecla and Coeur, BAML planned to skip the call and send instructions for submitting an indication of interest by November 23. PTO ¶ 231; JX 336; JX 337.

Because of the expedited timeline, BAML decided not to contact companies in the “Not Interested” category, even though many of those companies had said they were not interested when BAML previously called them with “a banker idea.” The response could have been different with a formal mandate. BAML’s script for “Additional Parties to Contact” was not likely to generate interest because it did not say anything more than “a situation in the PGM sector.” Because almost every other PGM company was located in a politically unstable jurisdiction, additional parties were less likely to have interest without a signal that the company involved was Stillwater. And because Stillwater had been advertising its interest in acquisitions, there was no reason for the additional parties to think that the situation involved Stillwater. See JX 124 at ‘074.

Using its scripts, BAML contacted five of the six possibly interested parties, missing Gold Fields. See JX 351. BAML contacted eight of the twelve additional parties, missing Alamos, Goldcorp, New Gold, and Yamana Gold. See PTO ¶ 230; JX 338; JX 339; JX 340; JX 341; JX 342. BAML contacted Northern Star, even though they were listed as not interested. See JX 351 at ‘953.

Three of the companies expressed interest: Anemka, Northern Star, and X2. BAML sent a confidentiality agreement and an invitation to submit a bid by November 29 to Anemka and Northern Star. BAML sent only a confidentiality agreement to X2, which quickly retracted its interest. See JX 395 at ‘412; see also JX 359 at ‘413.

26

Sibanye learned about BAML’s market check from Bateman. JX 332 at ‘969.

Sibanye perceived that a compressed timeline was its “only real advantage” in the process. Id. K. The Abbreviated Pre-Signing Market Check Continues.

On November 17, 2016, the Board met again, with Jones Day attending for the first time. BAML and McMullen updated the Board on the outreach and “the Board directed management to continue the strategic assessment process.”8 Sibanye had already sent a draft merger agreement to Jones Day.

On November 18, 2016, BAML suggested contacting Norilsk Nickel, a Russian mining company that had owned a majority stake in the Company between 2003 and 2010. JX 367. McMullen decided against it. See McMullen Dep. 476.

On November 20, 2016, the CFO of Northern Star informed McMullen that they were not interested in buying Stillwater but remained interested in a merger of equals. Northern Star asked McMullen to send a proposal. PTO ¶ 242.

On November 22, 2016, the CEO of Independence informed McMullen that they were not interested in buying Stillwater but were interested in a merger of equals. PTO ¶ 246. Independence asked to sign a confidentiality agreement and perform diligence,

8 JX 364 at ‘374. At trial, Schweitzer testified that this was the meeting at which the Board finally decided it did not need a special committee. See Schweitzer Tr. 157–58, 194. The minutes omit any discussion of the matter.

27

explaining that they had trouble reaching BAML. Independence did not receive a confidentiality agreement until November 25. See JX 403; JX 405.

The Board met again on the afternoon of November 23, 2016. McMullen reported that he had told Sibanye that its July proposal of $15.75 per share was not sufficient. He also reported that Sibanye needed the transaction to be “announced by the second week in December”; otherwise, Sibanye would need to delay the deal until the following year so that it could obtain stockholder approval to raise the capital needed to fund the Merger. JX 395 at ‘411. McMullen viewed a December signing as “ambitious given that . . . the Company’s assessment process with other potential parties was ongoing and would need to be concluded prior to proceeding with a transaction with Sibanye.” Id.

By the time of the board meeting, twenty-four parties had received some type of formal or informal contact from BAML or Stillwater management. Four parties—Sibanye, Hela, Coeur, and Anemka—had signed NDAs and accessed the data room. Four parties— Sibanye, Hela, Coeur, and Northern Star—had conducted site visits. Two parties—Coeur and Anemka—had notified BAML that they would not proceed further. PTO ¶ 235; JX 393 at ‘868. Two other parties—Northern Star and Independence—had informed Stillwater that they were only interested in a merger of equals. Hecla had reported that it needed to find a partner and had asked Stillwater to extend its bid deadline from November 23 to November 30. PTO ¶ 247; JX 383. The Board extended Hecla’s deadline to November 28. JX 395 at ‘413. By comparison, the Board had given Sibanye until November 30 to update its expression of interest from July. See JX 359 at ‘414.

28

After receiving these updates, the Board met in executive session, and the minutes reflected for the first time that McMullen did not participate. See JX 395 at ‘413. The Board instructed BAML to evaluate a merger of equals as a potential alternative. Id. When McMullen learned of the decision, he was skeptical, believing that a merger of equals could not compete with “a circa $18/share []all cash offer from S[ibanye].” JX 406 at ‘376. He shared his negative opinion with one of the directors, who replied that a merger of equals was actionable and needed to be explored as an alternative to Sibanye. See JX 401.

McMullen and BAML worked together to update the presentation that McMullen had given the Board in January 2016 on a potential merger of equals. See JX 384; JX 396. McMullen ranked the Company’s options as follows: 1) Sibanye’s acquisition; 2) a merger of equals with Northern Star; and 3) do nothing or a merger of equals with Independence. JX 396 at ‘707.

After the board meeting on November 23, 2016, BAML followed up with Hecla to solicit a specific indication of interest. See JX 394 at ‘214. Hecla did not respond, and the Company treated Hecla as having dropped out of the process.

On November 29, 2016, Northam asked to be included in the process. JX 414.

BAML sent Northam a confidentiality agreement and invited them to submit a bid by December 7. PTO ¶ 258; see JX 423; JX 424. That same day, Independence asked for an extension to the bid deadline since they were still negotiating the confidentiality agreement. JX 411. McMullen decided that meant that Independence was not interested.

29

L. Sibanye Revises Its Price.

As of November 20, 2016, Sibanye anticipated borrowing $2.5 billion to complete the Merger. Of this amount, $1.98 billion would be used to pay for the Company’s stock, with the consideration priced at a 30% premium over the Company’s thirty-day VWAP, just as McMullen and Froneman had agreed in March. See JX 378 at ‘979, ‘009, ‘016, ‘017. The additional $500 million would be used to pay off the Company’s debt, fund change- of-control payments for management, and pay transaction fees.

But on November 30, 2016, Sibanye ran into problems. First, Sibanye realized that the Company’s stock price had increased to a point where the pricing metric would cause the total purchase price to exceed Sibanye’s financing. Using the 30% premium over the thirty-day VWAP, Sibanye would have to pay approximately $18.25 per share, an amount that would require Sibanye to supplement the transaction financing with cash on hand or from its revolving credit line. See JX 420 at ‘876.

Second, Sibanye realized that it had calculated the purchase price in its indication of interest using a twenty-day VWAP rather than a thirty-day VWAP. Id. at ‘874. The Sibanye team recognized that they had agreed in principle to a thirty-day VWAP, but when they sent their initial indication of interest, they used a twenty-day VWAP because the Company’s stock had been in a declining trend, so the shorter period resulted in a lower price. Id. at ‘873.

Citigroup recommended pretending that Sibanye had never agreed to a pricing mechanism and had instead offered a fixed price. Id. The Sibanye team went along and disavowed all of the communications in which they had agreed in principle to a 30%

30

premium over the thirty-day VWAP. See Stewart Dep. 147–48; PTO ¶¶ 243, 245; JX 397 at ‘448; JX 378 at ‘009, ‘016. Going forward, Sibanye would discuss price based on an indication of interest of $15.75 per share. M. Stillwater Negotiates With Sibanye.

On December 1, 2016, the deal teams from the Company and Sibanye met in New York City. Sibanye proposed to acquire the Company for between $17.50 and $17.75 per share in cash. PTO ¶ 261.

On December 2, 2016, the Board met in New York City. See JX 432; JX 430.

McMullen shared Sibanye’s revised offer. The minutes do not reflect any discussion of Sibanye’s departure from the prior agreement in principle on a 30% premium over the thirty-day VWAP or the fact that the agreed-upon pricing metric would have supported a price around $18.25 per share. Even though BAML had worried about Sibanye using precisely this tactic, and even though McMullen had assured BAML that Sibanye would stick to the agreed-upon pricing metric, see JX 343 at ‘740 to ‘741, no one appears to have mentioned the change to the Board. See JX 432 at ‘414.

During the meeting, BAML presented its preliminary financial analysis of the Company. Using a discounted cash flow analysis, BAML valued the Company at between $10.78 and $14.14 per share. Id. at ‘416. That same day, the Company’s stock closed at $15.17 per share. PTO Ex. A.

BAML also reviewed potential merger of equals transactions with Northern Star and Independence. JX 432 at ‘417. According to the minutes, the Board decided not to pursue either transaction because: (i) the lack of synergies; (ii) “the significant disparity in

31

trading multiples”; (iii) “no merger-of-equals or similar transaction appeared to be available to the Company at this time”; (iv) “neither Northern Star nor Independence Mining had signed a confidentiality agreement”; and (v) “a substantial delay in the process to pursue such a possible transaction could result in the loss of a potential transaction with Sibanye.” JX 432 at ‘417; see McMullen Tr. 769. At the time, Northern Star and Independence had both proposed a merger-of-equals transaction and both had signed confidentiality agreements. There was also a meaningful probability that the Sibanye transaction would slip into the following year.

During the meeting, the Board instructed management to seek a higher price from Sibanye. That evening, McMullen and Bateman had dinner with Richard Stewart, Sibanye’s Executive Vice President of Business Development. PTO ¶ 265. After the dinner, Stewart emailed Froneman that “Mick’s number is 18$+ and that he thinks he can get his board across the line on that.” JX 434 at ‘426. Froneman, Stewart, and Citigroup discussed the limits of Sibanye’s financing, which would support a bid up to $18.20 per share. A 30% premium on the twenty-day VWAP for the Company’s common stock was $19.20 per share. Id. The group decided to bid $18.00 per share, observing that “if this is truly not good enough – they will come back but we need to be firm.” JX 434 at ‘425.

On December 3, 2016, Stewart called McMullen and offered $18 per share. PTO ¶ 267. BAML had been expecting $19 per share. See JX 438.

On the evening of December 3, 2016, Bateman had “a very open discussion” with one of Sibanye’s bankers from Citigroup, sharing information about the Board’s internal

32

dynamics, the Company’s lack of other prospects, and his preferences for employment. See JX 444. The Citigroup banker reported on the conversation as follows:

- 1. Value. Didn’t push back, as knows we’re at our limits. Said Mick will recommend our proposal to the Board, [that two directors] are “very commercial”. [Schweitzer] is the one most focused on 30% premium to 20D VWAP. I reiterated that we’ve truly been talking about 30D VWAP internally and with [Stillwater], which he seems to understand.

- ...

- 3. MOE. He seemed quite dismissive of the MOE candidate, but said certain Board members are keen to not shut it down completely (I suspect more from a litigation perspective).

- 4. Chris’ Plans. Said he honestly hasn’t given a lot of thought to what’s next, and he’s generally open minded about it. . . . He could be open to staying with [Sibanye], but depends on the vision and the role. He would have no desire to be a divisional CFO, but potentially interested in an Americas Head position. . . .

Id. Bateman participated in this discussion one day after Jones Day had advised the Board and senior management about the risk of conflicts during the negotiations. In response, Bateman and other members of management had represented to the Board that they had not had any discussions with Sibanye about their roles. See JX 432 at ‘418.

On December 4, 2016, Stewart called McMullen and told him that $18.00 was Sibanye’s best and final offer. PTO ¶ 270. After Bateman’s dinner with the Citigroup banker, Sibanye knew it did not have to bid higher.

Later that afternoon, McMullen shared the offer with the Board. Fearing that the timeline might slip into 2017, the directors instructed management “to progress discussions with Sibanye” and to find out whether Northam remained interested. JX 440 at ‘742.

33

On December 5, 2016, BAML reported that it had not heard anything from Northam. JX 445. That same day, Froneman called McMullen to reiterate that $18.00 per share was the best Sibanye could do given their financing constraints. PTO ¶ 272. N. McMullen Demands His Stock Awards.

On December 7, 2016, McMullen asked Sibanye to “put something into the merger agreement” about his 2017 stock awards. JX 451. According to McMullen, Sibanye had previously agreed to the following terms:

- On Closing of the deal, the value of the awards would be converted to cash based on the metrics of the deal (share price etc) and the amount paid out as per the normal vesting schedule in cash, namely 1/3 of the RSU value at each of the end of 2017, 2018 and 2019, and all the PSU value is paid out at the end of 2019. If any employee leaves for Good Cause (fired or diminution of job role) then the RSU’s accelerate in accordance with our plan docs, but the PSU amount is still paid out at the end of 2019.

Id. McMullen told Sibanye that the Compensation Committee had “decided that the 2015 and 2016 PSU’s would vest at 150% for each series in the event of an $18 bid.” Id. O. The Board Approves The Merger.

On December 8, 2016, the Board met to consider the Merger Agreement and decide whether to proceed with the Merger. McMullen reported that Northam had withdrawn from the process. JX 454 at ‘744; see JX 459. By this point, BAML had interacted with fourteen parties since being formally retained. Five had signed NDAs and conducted diligence. Only Sibanye had made a bid.

BAML rendered its opinion that Sibanye’s offer of $18 per share was fair. The consideration of $18 per share represented a 21% premium to the Company’s then-current stock price, a 21% premium to the 20-day VWAP, and a 25% premium to the 30-day

34

VWAP. JX 453 at ‘260. In its presentation, BAML valued the Company between $10.58 per share and $13.98 per share using a discounted cash flow analysis. Id. at ‘279 to ‘281.

The Merger Agreement contained a no-shop clause with a fiduciary out that permitted the Company to provide information to and negotiate with a third-party bidder if the bidder made an “Acquisition Proposal” that constituted or was reasonably likely to lead to a “Superior Proposal” and the Board concluded that its fiduciary duties required it. See JX 525 Annex A § 6.2.4. The Board had the right to change its recommendation in favor of the Merger if a competing bidder made a superior proposal and the Board concluded that its fiduciary duties required it. The Board did not have the right to terminate the Merger Agreement to pursue the superior proposal. The Company had to proceed through the stockholder meeting and only gained the right to terminate if the stockholders voted down the deal.

If the Company exercised its right to terminate after a negative stockholder vote, then the Company was obligated to pay Sibanye a termination fee of $16.5 million plus reimbursement of Sibanye’s expenses up to $10 million, for a total payment of $26.5 million. The total payment represented approximately 1.2% of equity value, with the termination-fee portion reflecting 0.76% of equity value. The Company had approximately $110 million more cash than debt, resulting in a slightly smaller enterprise value than equity value. The total payment represented approximately 1.3% of enterprise value.

The Board adopted the Merger Agreement and resolved to recommend that the Company’s stockholders approve it. JX 454 at ‘746. On December 9, 2016, Sibanye and

35

the Company announced the Merger. Sibanye’s stock price dropped 18% from $8.20 per share to $6.96 per share.

The last day of unaffected trading in Stillwater’s common stock was December 8, 2016. On that date, the Company’s shares closed at $14.68, equating to a market capitalization of approximately $1.8 billion. The deal price represented a 22.6% premium over the unaffected trading price and a 24.4% premium over the 30-day VWAP. During the previous two years, Stillwater’s stock price had never traded above $15.58, a level it reached on August 1, 2016. P. Vujcic Gets Paid.

After the Merger was signed, McMullen sent Vujcic a retroactive consulting agreement to compensate him for assisting with the Merger. Vujcic had two comments. First, he wanted confirmation that he would “not be named in the proxy.” JX 474 at ‘101. Second, he was disappointed with his compensation, stating:

- I’m a little perplexed as to why you are being so aggressive on the comp, especially when you are exposed to a potentially large claim from Jefferies and when I feel I have been pretty fair all along in (a) not locking you in earlier (trusted your guidance on compensation in July) and (b) in making every effort leading up to the board meetings in late October to give you the comfort to reiterate that Sibanye were the only show in town.

Id. at ‘100.

The petitioners argue that Vujcic’s statement that he made “every effort . . . to give you comfort to reiterate that Sibanye were the only show in town” shows that McMullen and Vujcic had been trying to eliminate the competition for Sibanye. That is a conspiratorial reading, rather than a credible reading. Vujcic was attempting to justify

36

receiving greater compensation by pointing to his efforts to solicit other potential bidders. He showed that Sibanye was “the only show in town” by engaging in outreach and demonstrating that no one else wanted to bid. The record does not support an inference that McMullen and Vujcic deceived the Board. See also McMullen Dep. 442–46.

McMullen and Vujcic agreed on a fixed fee of $20,000 per month beginning on October 24, 2016, plus a discretionary bonus of $100,000. JX 477. Vujcic’s name and compensation arrangement did not appear in the proxy statement. See JX 525. Q. Wadman’s Noisy Withdrawal In February 2017, McMullen and Bateman negotiated the terms of their post-closing employment with Sibanye. As part of those discussions, Sibanye agreed to treat the Merger as triggering McMullen and Bateman’s change-of-control payments, without the need for a second trigger such as termination or a resignation for “Good Reason.” None of the Company’s other employees received this special treatment. For the other employees, the Merger was only the first trigger, and no change-in-control benefits would be paid absent a second trigger.

When McMullen reported on this agreement to the Board during a meeting on February 23, 2017, Wadman objected. He had been concerned since July 2016 that McMullen and Bateman had pursued a sale of the Company in their own interest and had used the deal to advantage themselves. He regarded their special deal on change-in-control benefits as “clearly self-dealing.” JX 526 at ‘376. The Board did not address Wadman’s concerns during the meeting.

37

One month later, Wadman resigned. In his resignation letter, Wadman restated his concerns about how the deal process unfolded. He noted that after the board meeting on February 23, 2017, McMullen and Bateman “removed [me] from all legal conversations and decision-making” and “prohibited me from doing my job.” Id. at ‘377 to ‘378. Quoting his employment agreement, Wadman resigned for “Good Reason” based on a “material diminution” to his “nature of responsibilities, or authority.” Id.

Over the next several days, the Company’s counsel negotiated a settlement with Wadman. On March 30, 2017, the Company released a Form 8-K, which stated:

On March 29, 2017, Brent R. Wadman, our Vice President, Legal Affairs & Corporate Secretary, terminated employment. In connection therewith, we entered into an agreement with Mr. Wadman with respect to his separation pursuant to which we will pay him up to approximately $1.49 million. This amount includes the settlement of Mr. Wadman’s outstanding equity awards, which will continue to vest in accordance with their terms, including in connection with the previously announced merger with Sibanye Gold Limited.

JX 527. The Form 8-K did not mention Wadman’s letter or the reasons for his resignation. R. Stockholder Approval And Closing During Stillwater’s annual meeting on April 26, 2017, the stockholders approved the Merger Agreement. Under Delaware law, a merger requires the approval of holders of a majority of the outstanding shares, making a non-vote the equivalent of a “no” vote. Because stockholders can vote no by not voting, the percentage of the outstanding shares is the appropriate metric for evaluating the level of stockholder support for a merger. The Company had 121,389,213 shares outstanding. Holders of 91,012,990 shares voted in favor of the Merger, representing 75% of the issued and outstanding equity. Holders of

38

103,088,167 shares were present at the meeting in person or by proxy, so the same number of affirmative votes results in a misleadingly higher approval percentage of 88%. See JX 549 at 1.

The Merger closed on May 4, 2017. Between signing and closing, the spot price of palladium increased by 9.2%. The spot price of a weighted basket of Stillwater’s products increased by 5.9%. S. Post-Closing Developments On July 1, 2017, Sibanye entered into employment agreements with Bateman and McMullen. Bateman agreed to serve as Executive Vice President—US Region, reporting directly to Froneman. Bateman waived his change-of-control benefits in return for a higher base salary and additional incentive compensation. See JX 585.

McMullen agreed to serve as a Technical Advisor to Sibanye. His employment agreement permitted him “to perform the functions of that role while residing in the Turks and Caicos.” JX 586 at ‘041. Like Bateman, McMullen waived his change-of-control benefits in return for an annual salary of $712,000 plus incentive compensation. See id.

In November 2017, Sibanye issued a Competent Person’s Report that valued the Company’s operating mines at $2.7 billion as of July 31, 2017. This valuation was 23% greater than the total consideration that Sibanye paid for the Company at closing, just three months before the valuation date for the report. PTO ¶ 102; JX 615 at 205.

39

T. This Appraisal Proceeding Holders of 5,804,523 shares of the Company eschewed the consideration offered in the Merger and pursued appraisal. In August 2018, the holders of 384,000 shares settled their claims. The remaining petitioners litigated their claims through trial.

II. LEGAL ANALYSIS

“An appraisal proceeding is a limited legislative remedy intended to provide shareholders dissenting from a merger on grounds of inadequacy of the offering price with a judicial determination of the intrinsic worth (fair value) of their shareholdings.” Cede & Co. v. Technicolor, Inc. (Technicolor I), 542 A.2d 1182, 1186 (Del. 1988). Section 262(h) of the Delaware General Corporation Law states that

the Court shall determine the fair value of the shares exclusive of any element of value arising from the accomplishment or expectation of the merger or consolidation, together with interest, if any, to be paid upon the amount determined to be the fair value. In determining such fair value, the Court shall take into account all relevant factors.

8 Del. C. § 262(h). The statute thus places the obligation to determine the fair value of the shares squarely on the court. Gonsalves v. Straight Arrow Publ’rs, Inc., 701 A.2d 357, 361 (Del. 1997).

Because of the statutory mandate, the allocation of the burden of proof in an appraisal proceeding differs from a traditional liability proceeding. “In a statutory appraisal proceeding, both sides have the burden of proving their respective valuation positions . . . .” M.G. Bancorp., Inc. v. Le Beau, 737 A.2d 513, 520 (Del. 1999). “No presumption, favorable or unfavorable, attaches to either side’s valuation . . . .” Pinson v. Campbell- Taggart, Inc., 1989 WL 17438, at *6 (Del. Ch. Feb. 28, 1989). “Each party also bears the

40

burden of proving the constituent elements of its valuation position . . . , including the propriety of a particular method, modification, discount, or premium.” Jesse A. Finkelstein & John D. Hendershot, Appraisal Rights in Mergers and Consolidations, Corp. Prac. Series (BNA) No. 38-5th, at A-90 (2010 & 2017 Supp.) [hereinafter Appraisal Rights].

As in other civil cases, the standard of proof in an appraisal proceeding is a preponderance of the evidence. M.G. Bancorp., 737 A.2d at 520. A party is not required to prove its valuation conclusion, the related valuation inputs, or its underlying factual contentions by clear and convincing evidence or to exacting certainty. See Triton Constr. Co. v. E. Shore Elec. Servs., Inc., 2009 WL 1387115, at *6 (Del. Ch. May 18, 2009), aff’d, 2010 WL 376924 (Del. Jan. 14, 2010) (ORDER). “Proof by a preponderance of the evidence means proof that something is more likely than not. It means that certain evidence, when compared to the evidence opposed to it, has the more convincing force and makes you believe that something is more likely true than not.” Agilent Techs., Inc. v. Kirkland, 2010 WL 610725, at *13 (Del. Ch. Feb. 18, 2010) (internal quotation marks omitted).

“In discharging its statutory mandate, the Court of Chancery has discretion to select one of the parties’ valuation models as its general framework or to fashion its own.” M.G. Bancorp., 737 A.2d at 525–26. “[I]t is entirely proper for the Court of Chancery to adopt any one expert’s model, methodology, and mathematical calculations, in toto, if that valuation is supported by credible evidence and withstands a critical judicial analysis on the record.” Id. at 526. Or the court “may evaluate the valuation opinions submitted by the parties, select the most representative analysis, and then make appropriate adjustments to

41

the resulting valuation.” Appraisal Rights, supra, at A-31 (collecting cases). The court may also “make its own independent valuation calculation by . . . adapting or blending the factual assumptions of the parties’ experts.” M.G. Bancorp., 737 A.2d at 524. “If neither party satisfies its burden, however, the court must then use its own independent judgment to determine fair value.” Gholl v. eMachines, Inc., 2004 WL 2847865, at *5 (Del. Ch. Nov. 24, 2004). But the court must also be cautious when adopting an approach that deviates from the parties’ positions. Doing so “late in the proceedings” may “inject[] due process and fairness problems” that are “antithetical to the traditional hallmarks of a Court of Chancery appraisal proceeding,” because the court’s approach will not have been “subjected to the crucible of pretrial discovery, expert depositions, cross-expert rebuttal, expert testimony at trial, and cross examination at trial.” Verition P’rs Master Fund Ltd. v. Aruba Networks, Inc., 210 A.3d 128, 140-41 (Del. 2019).

In Tri-Continental Corporation v. Battye, 74 A.2d 71 (Del. 1950), the Delaware Supreme Court explained in detail the concept of value that the appraisal statute employs:

The basic concept of value under the appraisal statute is that the stockholder is entitled to be paid for that which has been taken from him, viz., his proportionate interest in a going concern. By value of the stockholder’s proportionate interest in the corporate enterprise is meant the true or intrinsic value of his stock which has been taken by the merger. In determining what figure represents the true or intrinsic value, . . . the courts must take into consideration all factors and elements which reasonably might enter into the fixing of value. Thus, market value, asset value, dividends, earning prospects, the nature of the enterprise and any other facts which were known or which could be ascertained as of the date of the merger and which throw any light on future prospects of the merged corporation are not only pertinent

42

to an inquiry as to the value of the dissenting stockholder’s interest, but must be considered . . . .9

Subsequent Delaware Supreme Court decisions have adhered consistently to this definition of value.10 Most recently, the Delaware Supreme Court reiterated that “[f]air value is . . . the value of the company to the stockholder as a going concern,” i.e., the stockholder’s “proportionate interest in a going concern.” Aruba, 210 A.3d at 132–33.

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.” Dell, Inc. v. Magnetar Global Event Driven Master Fund Ltd., 177 A.3d 1, 20 (Del. 2017) (quoting Cavalier Oil, 564 A.2d at 1142–43). 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.” Id. (quoting Cavalier Oil, 564 A.2d at 1144). When doing so, the corporation “must

9 Id. at 72. Although Battye is the seminal Delaware Supreme Court case on point, Chancellor Josiah Wolcott initially established the meaning of “value” under the appraisal statute in Chicago Corporation v. Munds, 172 A. 452 (Del. Ch. 1934). Citing the “material variance” between the Delaware appraisal statute, which used “value,” and the comparable New Jersey statute that served as a model for the Delaware statute, which used “full market value,” Chancellor Wolcott held that the plain language of the statute required “value” to be determined on a “going concern” basis. Id. at 453–55. But see Union Ill. 1995 Inv. Ltd. P’ship v. Union Fin. Gp., Ltd., 847 A.2d 340, 355–56 (Del. Ch. 2004) (“This requirement that the valuation inquiry focus on valuing the entity as a going concern has sometimes been confused as a requirement of § 262’s literal terms. It is not.”).

10 See, e.g., Montgomery Cellular Hldg. Co. v. Dobler, 880 A.2d 206, 222 (Del.

2005); Paskill Corp. v. Alcoma Corp., 747 A.2d 549, 553 (Del. 2000); Rapid-Am. Corp. v. Harris, 603 A.2d 796, 802 (Del. 1992); Cavalier Oil Corp. v. Harnett, 564 A.2d 1137, 1144 (Del. 1989); Bell v. Kirby Lumber Corp., 413 A.2d 137, 141 (Del. 1980); Universal City Studios, Inc. v. Francis I. duPont & Co., 334 A.2d 216, 218 (Del. 1975).

43

be valued as a going concern based upon the ‘operative reality’ of the company as of the time of the merger,” taking into account its particular market position in light of future prospects. M.G. Bancorp., 737 A.2d at 525 (quoting Cede & Co. v. Technicolor, Inc. (Technicolor IV), 684 A.2d 289, 298 (Del. 1996)); accord Dell, 177 A.3d at 20. The concept of the corporation’s “operative reality” is important because “[t]he underlying assumption in an appraisal valuation is that the dissenting shareholders would be willing to maintain their investment position had the merger not occurred.” Technicolor IV, 684 A.2d at 298. Consequently, the trial court must assess “the value of the company . . . as a going concern, rather than its value to a third party as an acquisition.” M.P.M. Enters., Inc. v. Gilbert, 731 A.2d 790, 795 (Del. 1999).

“The time for determining the value of a dissenter’s shares is the point just before the merger transaction ‘on the date of the merger.’” Appraisal Rights, supra, at A-33 (quoting Technicolor I, 542 A.2d at 1187). Put differently, the valuation date is the date on which the merger closes. Technicolor IV, 684 A.2d at 298; accord M.G. Bancorp., 737 A.2d at 525. If the value of the corporation changes between the signing of the merger agreement and the closing, then the fair value determination must be measured by the “operative reality” of the corporation at the effective time of the merger. See Technicolor IV, 684 A.2d at 298.

The statutory obligation to make a single determination of a corporation’s value introduces an impression of false precision into appraisal jurisprudence.

[I]t is one of the conceits of our law that we purport to declare something as elusive as the fair value of an entity on a given date . . . . [V]aluation decisions are impossible to make with anything approaching complete confidence.

44

Valuing an entity is a difficult intellectual exercise, especially when business and financial experts are able to organize data in support of wildly divergent valuations for the same entity. For a judge who is not an expert in corporate finance, one can do little more than try to detect gross distortions in the experts’ opinions. This effort should, therefore, not be understood, as a matter of intellectual honesty, as resulting in the fair value of a corporation on a given date. The value of a corporation is not a point on a line, but a range of reasonable values, and the judge’s task is to assign one particular value within this range as the most reasonable value in light of all the relevant evidence and based on considerations of fairness.11

As the Delaware Supreme Court recently explained, “fair value is just that, ‘fair.’ It does not mean the highest possible price that a company might have sold for had Warren Buffet negotiated for it on his best day and the Lenape who sold Manhattan on their worst.” DFC Glob. Corp. v. Muirfield Value P’rs, 172 A.3d 346, 370 (Del. 2017).

Because the determination of fair value follows a litigated proceeding, the issues that the court considers and the outcome it reaches depend in large part on the arguments advanced and the evidence presented.

An argument may carry the day in a particular case if counsel advance it skillfully and present persuasive evidence to support it. The same argument may not prevail in another case if the proponents fail to generate a similarly persuasive level of probative evidence or if the opponents respond effectively.

11 Cede & Co. v. Technicolor, Inc., 2003 WL 23700218, at *2 (Del. Ch. Dec. 31, 2003, revised July 9, 2004), aff’d in part, rev’d in part on other grounds, 884 A.2d 26 (Del. 2005); accord Finkelstein v. Liberty Dig., Inc., 2005 WL 1074364, at *12 (Del. Ch. Apr. 25, 2005) (“The judges of this court are unremittingly mindful of the fact that a judicially selected determination of fair value is just that, a law-trained judge’s estimate that bears little resemblance to a scientific measurement of a physical reality. Cloaking such estimates in grand terms like ‘intrinsic value’ does not obscure this hard truth from any informed commentator.”).

45

Merion Capital L.P. v. Lender Processing Servs., L.P., 2016 WL 7324170, at *16 (Del. Ch. Dec. 16, 2016). Likewise, the approach that an expert espouses may have met “the approval of this court on prior occasions,” but may be rejected in a later case if not presented persuasively or if “the relevant professional community has mined additional data and pondered the reliability of past practice and come, by a healthy weight of reasoned opinion, to believe that a different practice should become the norm . . . .” Glob. GT LP v. Golden Telecom, Inc. (Golden Telecom Trial), 993 A.2d 497, 517 (Del. Ch.), aff’d, 11 A.3d 214 (Del. 2010). A. The Deal Price Sibanye contends that the deal price of $18.00 per share is a persuasive indicator of fair value if adjusted downward to eliminate elements of value arising from the Merger. The petitioners argue that the deal price should receive no weight. As the proponent of using the deal price, Sibanye bore the burden of establishing its persuasiveness. Sibanye also bore the burden of proving its downward adjustment.

1. The Standard For Evaluating A Sale Process There is no presumption that the deal price reflects fair value. Dell, 177 A.3d at 21;

DFC, 172 A.3d at 366–67. Relying on the statutory requirement that the Court of Chancery must consider “all relevant factors” when determining fair value, the Delaware Supreme Court has rejected “requests for the adoption of a presumption that the deal price reflects fair value if certain preconditions are met, such as when the merger is the product of arm’s- length negotiation and a robust, non-conflicted market check, and where bidders had full

46

information and few, if any, barriers to bid for the deal.” Dell, 177 A.3d at 21. Yet the Delaware Supreme Court has also cautioned that its

refusal to craft a statutory presumption in favor of 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.

DFC, 172 A.3d at 366. The Delaware Supreme Court has likewise 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.” Id. Based on the facts presented in DFC and Dell, the Delaware Supreme Court endorsed using the deal price as a persuasive indicator of fair value in those cases. Based on the facts presented in Aruba, the Delaware Supreme Court used a deal-price-less-synergies metric to make its own fair value determination.

As a general matter, the persuasiveness of the deal price depends on the reliability of the sale process that generated it. When assessing whether a sale process results in fair value, the issue “is not whether a negotiator has extracted the highest possible bid.” Dell, 177 A.3d at 33. “[T]the purpose of an appraisal is . . . to make sure that [the petitioners] 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.” DFC, 172 A.3d at 370–71. “[T]he key inquiry is whether the dissenters got fair value and were not exploited.” Dell, 177 A.3d at 33.

47

Relying on the Delaware Supreme Court’s decision in DFC, the petitioners assert that the deal price “deserves weight only if the merger is the product of a ‘robust market search’ and an arm’s-length third party transaction with ‘no hint of self-interest that compromised the market check.’” Dkt. 210 at 36 [hereinafter PTOB] (quoting DFC, 172 A.3d at 349). That is not what DFC held.

The petitioners have accurately quoted phrases from the decision in DFC, but when the Delaware Supreme Court made those observations, it was describing the trial court’s findings regarding the sale process that took place in that case. The Delaware Supreme Court then determined that given those attributes, “the best evidence of fair value was the deal price.” DFC, 172 A.3d at 349. The high court’s comments in DFC explained why the particular sale process in that case was so good as to make the deal price “the best evidence of fair value.” The decision did not identify minimum characteristics that a sale process must have before a trial court can give it weight. The decision also did not address what makes a sale process sufficiently bad that a trial court cannot give it weight. Technically, the decision did not even delineate when a sale process would be sufficiently good that a trial court should regard it as “the best evidence of fair value.” The Delaware Supreme Court could have believed the sale process in DFC warranted that level of consideration without excluding the possibility that a not-as-good sale process could warrant the same treatment.

The same is true for the Delaware Supreme Court’s comments about the sale process in Dell. There, the Delaware Supreme Court described the sale process as having featured “fair play, low barriers to entry, outreach to all logical buyers, and the chance for any

48

topping bidder to have the support of Mr. Dell’s own votes . . . .” Dell, 177 A.3d at 35. Based on its view of the sale process, the Delaware Supreme Court suggested that “the deal price deserved heavy, if not dispositive weight.” Dell, 177 A.3d at 23. After describing the sale process in greater detail, the Delaware Supreme Court observed, “Overall, the weight of evidence shows that Dell’s deal price has heavy, if not overriding, probative value.” Id. at 30. As in DFC, the Delaware Supreme Court was explaining why it regarded a particular sale process as so good that it deserved “heavy, if not dispositive weight.” The Delaware Supreme Court was not identifying the minimum requirements for a sale process to generate reliable information about fair value, nor was it enumerating qualities which, if absent, would render the outcome of a sale process so unreliable as to provide no insight into fair value.

The Delaware Supreme Court’s decision in Aruba likewise did not address the minimum requirements for a sale process to generate reliable information about fair value. There, the trial court found the sale process to be sufficiently reliable to use the deal price as a valuation indicator, but declined to give it weight. The Delaware Supreme Court accepted that the sale process was sufficiently reliable and used the deal price as the exclusive basis for its own fair value determination. As with Dell and DFC, the Aruba decision did not have to address when a sale process was sufficiently bad that a trial court should decline to rely on the deal price.

The decisions in DFC, Dell, and Aruba are highly informative because they analyze fact patterns in which the Delaware Supreme Court viewed the sale processes as sufficiently reliable to use the deal price as either (i) the exclusive basis for its own fair

49

value determination (Aruba), (ii) as a valuation indicator that “deserved heavy, if not dispositive weight” (Dell), or (iii) as a valuation indicator that provided “the best evidence of fair value” (DFC). But Aruba, Dell, and DFC do not establish legal requirements for a sale process. Whether a sale process is sufficiently good that the deal price should be regarded as persuasive evidence of fair value, or whether a sale process is sufficiently bad that the deal price should not be regarded as persuasive evidence of fair value are invariably fact-specific questions, and the answers depend on the arguments made and the evidence presented in a given case.

2. Objective Indicia Of Reliability In the recent appraisal decisions that have examined the reliability of a sale process, the Delaware Supreme Court has cited certain “objective indicia” that “suggest[] that the deal price was a fair price.” Dell, 177 A.3d at 28; accord DFC, 172 A.3d at 376. The presence of objective indicia do not establish a presumption in favor of the deal price. The indicia are a starting point for analysis, not the end point, and in each of its recent appraisal decisions, the Delaware Supreme Court has determined that a combination of the objective indicia and other evidence outweighed the shortcomings in the sale processes that the petitioners had identified (Aruba) or which the trial court had regarded as undermining the persuasiveness of the deal price (Dell and DFC).

First, the Merger was an arm’s-length transaction with a third party. See DFC, 172 A.3d at 349 (citing fact that “the company was purchased by a third party in an arm’s length sale” as factor supporting fairness of deal price). It was not a transaction involving a controlling stockholder. See Dell, 177 A.3d at 30 (citing fact that “this was not a buyout

50

led by a controlling stockholder” as a factor supporting fairness of deal price). Sibanye was an unaffiliated acquirer with no prior ownership interest in Stillwater.

Second, the Board did not labor under any conflicts of interest. Six of the Board’s seven members were disinterested, outside directors, and they had the statutory authority under the Delaware General Corporation Law to say “no” to any merger. See 8 Del. C. § 251(b) (requiring board adoption and recommendation of a merger agreement); Dell, 177 A.3d at 28 (citing fact that special committee was “composed of independent, experienced directors and armed with the power to say ‘no’” as factor supporting fairness of deal price). Stillwater’s stockholders were widely dispersed, and the petitioners have not identified divergent interests among them. Cf. id. at 11 (citing the fact that “any outside bidder who persuaded stockholders that its bid was better would have access to Mr. Dell’s votes” as a factor supporting fairness of deal price).

Third, Sibanye conducted due diligence and received confidential information about Stillwater’s value. See Aruba, 210 A.3d at 137 (emphasizing that buyer armed with “material nonpublic information about the seller is in a strong position (and is uniquely incentivized) to properly value the seller”). Like the acquirer in Aruba, Sibanye “had signed a confidentiality agreement, done exclusive due diligence, gotten access to material nonpublic information,” and had a “sharp[] incentive to engage in price discovery . . . because it was seeking to acquire all shares.” Id. at 140.

Fourth, Stillwater negotiated with Sibanye and extracted multiple price increases.

See id. at 139 (citing “back and forth over price”); Dell, 177 A.3d at 28 (citing fact that special committee “persuaded Silver Lake to raise its bid six times”). In July 2016, when

51

Sibanye indicated interest in a transaction at $15.75 per share, Stillwater did not rush into a deal. In December 2016, when Sibanye raised its indication of interest to a range of $17.50 to $17.75 per share, Stillwater again did not proceed. With the Board’s backing, McMullen demanded a higher price. When Sibanye offered $18.00 per share, the Board did not immediately accept. Only after Sibanye twice stated that $18.00 per share was its best and final offer did the Board accept that price.

Most importantly, no bidders emerged during the post-signing phase, which is a factor that the Delaware Supreme Court has stressed when evaluating a sale process.12 The Merger Agreement did not contain any exceptional deal protection features, and the total amounts due via the termination fee and expense reimbursement provision were comparatively low, representing approximately 1.2% of equity value. Excluding the expense reimbursement, the termination fee reflected only 0.76% of equity value. The absence of a topping bid was thus highly significant.

As noted, these are fewer objective indicia of fairness than the Delaware Supreme Court identified when reviewing the sale processes in DFC, Dell, or Aruba, and the presence of these factors does not establish a presumption in favor of the deal price.

12 See Aruba, 210 A.3d at 136 (“It cannot be that an open chance for buyers to bid signals a market failure simply because buyers do not believe the asset on sale is sufficiently valuable for them to engage in a bidding contest against each other.”); Dell, 177 A.3d at 29 (“Fair value entails at a minimum a price some buyer is willing to pay— not a price at which no class of buyers in the market would pay.”); id. at 33 (finding that absence of higher bid meant “that the deal market was already robust and that a topping bid involved a serious risk of overpayment,” which “suggests the price is already at a level that is fair”).

52

Nevertheless, the objective indicia that were present provide a cogent foundation for relying on the deal price as a persuasive indicator of fair value, subject to further review of the evidence.

3. The Challenges To The Pre-Signing Phase The petitioners have advanced a multitude of reasons why they believe the deal price for Stillwater does not provide a persuasive indicator of fair value. The bulk of their objections concern the pre-signing phase.

As a threshold matter, the petitioners argue generally that a reliable sale process requires some degree of pre-signing outreach, citing a comment from the Union Illinois decision in which this court used a deal-price-less-synergies metric to value a privately held company after concluding that the company was “marketed in an effective manner.” Union Ill., 847 A.2d at 350. The petitioners also cite a statement from the AOL decision to the effect that a sale process will provide persuasive evidence of statutory fair value when “(i) information was sufficiently disseminated to potential bidders, so that (ii) an informed sale could take place, (iii) without undue impediments imposed by the deal structure itself.” In re Appraisal of AOL Inc., 2018 WL 1037450, *8 (Del. Ch. Feb. 23, 2018). Neither decision established a rule that pre-signing outreach is invariably required before the deal price can serve as persuasive evidence of fair value. At least for a widely held, publicly traded company, a sale process could justify both sets of observations through the public announcement of a transaction and a sufficiently open post-signing market check.

The petitioners’ myriad arguments about the pre-signing process in this case raise a fundamental question: Would the deal price provide persuasive evidence of fair value if

53

Stillwater had pursued a single-bidder strategy in which it only interacted with Sibanye before signing the Merger Agreement, recognizing that the Merger Agreement was sufficiently open to permit a meaningful post-signing market check? If the deal price would have provided persuasive evidence of fair value under those circumstances, then the additional efforts that Stillwater made before signing, even if disorganized and flawed, should not change the outcome. It is conceivable that a pre-signing process could involve features that undermined the effectiveness of a post-signing market check, such as never- waived standstill agreements containing don’t-ask-don’t-waive provisions, but that was not the case here. At least on the facts presented, Stillwater’s efforts were additive, not subtractive. They might not have added much, but they did not detract from what Stillwater could have achieved through a single-bidder process focused on Sibanye followed by a post-signing market check.

a. The Possibility Of A Single-Bidder Strategy Although the Delaware Supreme Court has not had the opportunity to consider a single-bidder strategy for purposes of determining the persuasiveness of a deal-price metric in an appraisal proceeding, extant precedent suggests that if Stillwater had pursued a single- bidder strategy in which it only interacted with Sibanye before signing the Merger Agreement, then the deal price would provide persuasive evidence of fair value because the Merger Agreement was sufficiently open to permit a meaningful post-signing market check. The reasoning that leads to this endpoint starts not with the recent triumvirate of appraisal cases, but rather with an important Delaware Supreme Court decision that restated the high court’s enhanced scrutiny jurisprudence for purposes of applying that

54

standard of review in a breach of fiduciary duty case. C & J Energy Servs., Inc. v. City of Miami Gen. Empls.’ & Sanitation Empls.’ Ret. Tr., 107 A.3d 1049 (Del. 2014). The Delaware Supreme Court’s enhanced scrutiny jurisprudence becomes pertinent to appraisal proceedings because, as commentators have perceived, the deal price will provide persuasive evidence of fair value in an appraisal proceeding involving a publicly traded firm if the sale process would satisfy enhanced scrutiny in a breach of fiduciary duty case.13 In C & J Energy, the Delaware Supreme Court held that plaintiffs who challenged a transaction involving only a passive, post-signing market check had not shown a reasonable likelihood that the director defendants had breached their fiduciary duties under the enhanced scrutiny standard of review. The transaction in C & J Energy was a stock- for-stock merger between C & J Energy Services, Inc. and a subsidiary of Nabors Industries Ltd. Although C & J Energy was nominally the acquirer, it would emerge from the transaction with a controlling stockholder, and the Delaware Supreme Court therefore

13 See Lawrence A. Hamermesh & Michael L. Wachter, Finding the Right Balance in Appraisal Litigation: Deal Price, Deal Process, and Synergies, 73 Bus. Law. 961, 962 (2018) (commending outcomes in Dell and DFC and arguing that “the Delaware courts’ treatment of the use of the deal price to determine fair value does and should mirror the treatment of shareholder class action fiduciary duty litigation”); id. at 982–83 (citing Dell and DFC in observing, “What we discern from the case law, however, is a tendency to rely on deal price to measure fair value where the transaction would survive enhanced judicial scrutiny . . . . Thus, in order to determine whether to use the deal price to establish fair value, the Delaware courts are engaging in the same sort of scrutiny they would have applied under Revlon if the case were one challenging the merger as in breach of the directors’ fiduciary duties.” (footnote omitted)); Charles Korsmo & Minor Myers, The Flawed Corporate Finance of Dell and DFC Global, 68 Emory L.J. 221, 269 (2018) (explaining that Dell and DFC “conflate questions of fiduciary duty liability with the valuation questions central to appraisal disputes”).

55

examined whether the directors had fulfilled their situationally specific duty to seek the best transaction reasonably available. See C & J Energy, 107 A.3d at 1067.

The merger in C & J Energy resulted from a CEO-driven process. Joshua Comstock, the founder, chairman, and CEO of C & J Energy, spearheaded the discussions. Talks between Comstock and the CEO of Nabors started in January 2014, and although Comstock discussed the deal with some of C & J Energy’s directors, he did not receive formal board approval to negotiate until April. Later in the process, he made a revised offer to Nabors without board approval. The plaintiffs argued that Comstock acted without authority and misled the board about key issues. The Delaware Supreme Court found “at least some support for the plaintiffs’ contention that Comstock at times proceeded on an ‘ask for forgiveness rather than permission’ basis.” Id. at 1059.

There was evidence in C & J Energy that Comstock had personal reasons to favor a deal with Nabors. The Nabors CEO “assured Comstock throughout the process that he would be aggressive in protecting Comstock’s financial interests if a deal was consummated.” Id. at 1064. After the key terms of the transaction had been negotiated, but before it was formally approved, Comstock asked for a side letter “affirming that C & J’s management would run the surviving entity and endorsing a generous compensation package.” Id. When the Nabors CEO balked, Comstock threatened to not sign or announce the deal. The Nabors CEO gave in, and the deal was announced as planned. Id. at 1064– 65. In addition, there was evidence that C & J Energy’s primary financial advisor was less than optimally effective and seemed to be advocating for the deal rather than advocating for C & J Energy. See id. at 1056. The banker also had divergent interests because of its

56

role as a financing source for the deal. Id. at 1057. There were thus reasons to think that the two principal negotiators for C & J—its CEO and its banker—had personal reasons to favor a transaction with Nabors and to push for that outcome.

The merger agreement in C & J Energy included a no-shop clause subject to a fiduciary out and a termination fee equal to 2.27% of the deal value. Id. at 1063. The period between the announcement of the deal on June 25, 2014, and the trial court’s issuance of the injunction on November 25, 2014, lasted 153 days. No competing bidder emerged during that period.

On these facts, the Delaware Supreme Court found no grounds for a potential breach of duty, explaining that “[w]hen a board exercises its judgment in good faith, tests the transaction through a viable passive market check, and gives its stockholders a fully informed, uncoerced opportunity to vote to accept the deal, we cannot conclude that the board likely violated its Revlon duties.” Id. at 1053. Elaborating, the senior tribunal explained that a board may pursue a single transaction partner, “so long as the transaction is subject to an effective market check under circumstances in which any bidder interested in paying more has a reasonable opportunity to do so.” Id. at 1067. The high court emphasized that “[s]uch a market check does not have to involve an active solicitation, so long as interested bidders have a fair opportunity to present a higher-value alternative, and the board has the flexibility to eschew the original transaction and accept the higher-value deal.” Id. at 1067–68. The transaction in C & J Energy satisfied this test. Describing the suite of deal protections, the Delaware Supreme Court observed that “a potential competing bidder faced only modest deal protection barriers.” Id. at 1052. Later, the court reiterated

57

that “there were no material barriers that would have prevented a rival bidder from making a superior offer.” Id. at 1070; accord id. (“But in this case, there was no barrier to the emergence of another bidder and more than adequate time for such a bidder to emerge.”). The Delaware Supreme Court also cited with approval precedents in which a sell-side board had engaged exclusively with a single buyer, had not conducted a pre-signing market check, then agreed to a merger agreement containing a no-shop clause, a matching right, and a termination fee, and the resulting combination was found sufficient to permit an effective post-signing market check that satisfied the directors’ duties under enhanced scrutiny.14 Procedurally, the Delaware Supreme Court’s decision in C & J Energy vacated an injunction that the trial court had entered in advance of the stockholder vote. In holding that the trial court had issued the injunction improvidently, the high court noted that “[t]he ability of the stockholders themselves to freely accept or reject the board’s preferred course of action is also of great importance in this context.” Id. at 1068. The role of the vote, however, should not detract from the high court’s observations about the adequacy of the

14 See id. at 1068 n.87 (citing cases including In re Dollar Thrifty S’holders Litig., 14 A.3d 573, 612–13, 615 (Del. Ch. 2010) (finding that the target board’s use of no-shop, matching rights, and termination fee provisions were reasonable even though the company had agreed to deal exclusively with the buyer without conducting a pre-signing market check); and In re MONY Gp. Inc. S’holders Litig., 852 A.2d 9 (Del. Ch. 2004) (finding that the board acted reasonably even though it did not actively shop the company because the board was financially sophisticated, had knowledge of the relevant industry, and there was a “substantial opportunity for an effective market check” after the agreement was announced)); id. at 1069 (citing Lyondell Chem. Co. v. Ryan, 970 A.2d 235, 243 (Del. 2009)).

58

single-bidder process. Underscoring that point, the Delaware Supreme Court cited the trial court’s apparent belief “that Revlon required C & J’s board to conduct a pre-signing active solicitation process in order to satisfy its contextual fiduciary duties,” then explicitly rejected that understanding of the enhanced scrutiny standard. Id. at 1068. As a result, the Delaware Supreme Court’s decision in C & J Energy has implications that go beyond the injunction context.

One area where its implications subsequently became manifest was in a post-closing liability action where plaintiffs sought to recover from an alleged aider-and-abettor under a quasi-appraisal theory of damages. See In re PLX Tech. Inc. S’holders Litig., --- A.3d -- -, 2019 WL 2144476 (Del. May 16, 2019) (TABLE). The PLX litigation challenged a merger agreement in which the acquirer (Avago) purchased the target (PLX) for cash. As in C & J Energy, the sale process was not pristine. The trial court found that a key director and the company’s investment banker had divergent interests that caused them to favor a sale over having PLX remain independent, that Avago tipped the director and the banker about the timing and pricing of a deal, that the director and the banker failed to disclose the tip to the board while using the information to help them position PLX to be sold, and that the proxy statement failed to disclose these issues. See In re PLX Tech. Inc. S’holders Litig. (PLX Trial), 2018 WL 5018535, at *32–35, *44–47 (Del. Ch. Oct. 16, 2018) (subsequent history omitted). Based on these findings, the trial court found a predicate breach of fiduciary duty under the enhanced scrutiny standard. The trial court also found that the sole remaining defendant—an activist stockholder affiliated with the key director—had participated knowingly in the breach. See id. at *48–50.

59

The plaintiffs’ claim foundered, however, at the damages stage. The plaintiffs sought to recover compensatory damages on behalf of a class of stockholders based on the theory that PLX should have remained independent rather than being sold. Under this theory, the plaintiffs sought “out-of-pocket (i.e., compensatory) money damages equal to the ‘fair’ or ‘intrinsic’ value of their stock at the time of the merger, less the price per share that they actually received,” with “[t]he ‘fair’ or ‘intrinsic’ value of the shares . . . determined using the same methodologies employed in an appraisal.” Id. at *50 (internal quotation marks omitted) (collecting cases). The plaintiffs’ expert used a DCF methodology to value PLX at $9.86 per share, well above the deal price of $6.50 per share. See id. at *51.

Although PLX’s pre-signing process was marred by breaches of fiduciary duty resulting from Avago’s tip to the key director and the company’s banker, the trial court found that the sale process as a whole was sufficiently reliable to warrant rejecting the plaintiffs’ valuation. The trial court explained that “[m]ore important than the pre-signing process was the post-signing market check.” Id. at *55. After discussing the outcome in C & J Energy, the trial court reasoned that “the structure of the Merger Agreement satisfied the Delaware Supreme Court’s standard for a passive, post-signing market check.” Id. The merger agreement (i) contained a no-shop with a fiduciary subject an unlimited match right that gave Avago four days to match the first superior proposal and two days to match any subsequent increase, and (ii) required PLX to pay Avago a termination fee of $10.85 million, representing 3.5% of equity value ($309 million) and 3.7% of enterprise value ($293 million). See id. at *26, *44. Avago launched its first step-tender offer on July 8,

60

2014. No competing bidder intervened, and the merger closed thirty-five days later on August 12. Id. at *27. This time period compared favorably with other passive, post-signing market checks that Delaware decisions had approved.15 On appeal, the Delaware Supreme Court affirmed the judgment based solely on the trial court’s damages ruling and without reaching or expressing a view on any of the other issues raised by the case. See PLX, 2019 WL 2144476, at *1. For present purposes, the damages issue is the important one, because the trial court had determined that the suite of defensive measures in the merger agreement, together with the absence of a topping bid, provided a more reliable indication of value than the plaintiffs’ discounted cash flow model. See PLX Trial, 2018 WL 5018535, at *44, *54–56. Notably for present purposes, although the burden of proof rested solely with the plaintiffs, the trial court in PLX made its determination using the same valuation standard that would apply in an appraisal proceeding. Id. at *50.

To reiterate, in its appraisal jurisprudence, the Delaware Supreme Court has not yet been asked to rule on the reliability of a sale process involving a single-bidder strategy, no pre-signing outreach, and a passive post-signing market check. The closest precedent is

15 See id. at *44. The PLX Trial decision included an appendix that collected decisions approving a passive market check. The table somehow swapped the details of the passive market check in Braunschweiger v. American Home Shield Corporation, 1989 WL 128571 (Del. Ch. Oct. 26, 1989), with the details from In re Formica Corporation Shareholders Litigation, 1989 WL 25812 (Del. Ch. Mar. 22, 1989). A corrected version is attached to this decision as an appendix.

61

Aruba, where the dynamics of the sale during the pre-closing phase resembled a single- bidder strategy, although the company’s banker did engage in some minimal outreach.

The pre-signing phase of the sale process in Aruba had two stages. See Verition P’rs Master Fund Ltd. v. Aruba Networks, Inc. (Aruba Trial), 2018 WL 922139, at *7–8 (Del. Ch. Feb. 15, 2018) (subsequent history omitted). The first stage began in late August 2014, when HP approached Aruba about a deal. Aruba hired an investment banker (Qatalyst), who identified thirteen potential partners and approached five of them. For reasons having “nothing to do with price,” no one was interested. Id. at *10. Aruba and HP entered into an NDA that restricted HP from speaking with Aruba management about post-transaction employment, and HP began conducting due diligence. Id. at *11. Despite the restriction in the NDA, HP asked Aruba’s CEO, Dominic Orr, if he would take on a key role with the combined entity. Orr replied that he had no objection. Id.

The parties seemed to be making progress towards a deal, but the HP board of directors balked at making a bid without further analysis, recalling the fallout from a disastrous acquisition in 2011. In November 2014, Aruba terminated discussions, bringing the first stage of the pre-signing process to a close. Id. at *12.

For its part, HP continued to evaluate an acquisition of Aruba. In December 2014, HP tapped Barclays Capital Inc. as its financial advisor. That firm had worked for Aruba and had been trying to secure the sell-side mandate. Id. at *13. On January 21, 2015, HP’s CEO met with Orr for dinner. During the meeting, when HP’s CEO proposed resuming merger talks, Orr responded with enthusiasm and suggested trying to announce a deal by early March. But HP’s CEO also told Orr that because Qatalyst had represented the seller

62

in HP’s disastrous acquisition from 2011, HP would not proceed if Aruba used Qatalyst. Id. at *14.

The Aruba board decided to move forward with the deal and informed Qatalyst about HP’s ukase. Aruba was obligated to pay Qatalyst a fee in the event of a successful transaction, so it kept Qatalyst on as a behind-the-scenes advisor. From then on, Qatalyst’s primary goal was to repair its relationship with HP, and Qatalyst regarded a successful sale of Aruba to HP as a key step in the right direction. Aruba also needed a new HP-facing banker. It hired Evercore, a firm that was trying to establish a presence in Silicon Valley. During the sale process, Evercore likewise sought to please HP, viewing HP as a major source of future business. See id. at *9, *15–16, *19, *21.

The ensuing negotiations proceeded quickly. HP had anticipated making an opening bid of $24 per share, but after Orr’s enthusiastic response, HP opened at $23.25 per share. Id. at *16–17. Qatalyst reached out to a sixth potential strategic partner, but it was not interested. Id. at *17. The Aruba board decided to counter at $29 per share. Evercore conveyed the number to Barclays, but when Barclays dismissed it, Evercore emphasized Aruba’s desire to announce a deal quickly. Id. at *17–18. On February 10, 2015, twenty days after HP resumed discussions with Orr, the Aruba board agreed to a price of $24.67 per share. Id. at *19. The parties negotiated a merger agreement, and on March 1, 2015, the Aruba board approved it.

The post-signing phase was uneventful. On March 2, 2015, Aruba and HP announced the merger. The merger agreement (i) contained a no-shop clause subject to a fiduciary out, (ii) conditioned the out for an unsolicited superior proposal on compliance

63

with an unlimited match right that gave HP five days to match the first superior proposal and two days to match any subsequent increase, and (iii) required Aruba to pay HP a termination fee of $90 million, representing 3% of Aruba’s equity value. No competing bidder emerged, and on May 1, 2015, Aruba’s stockholders approved the merger. Id. at *21–22.

Although the sale process in Aruba had flaws, the trial court found that it was sufficiently reliable to make the deal price a persuasive indicator of fair value. Overall, the trial court viewed the HP-Aruba merger as “a run-of-the-mill, third party-deal,” where “[n]othing about it appear[ed] exploitive.” Id. at *38. The petitioners argued that the deal price resulted from a closed-off sale process in which HP had not faced a meaningful threat of competition. Id. at *39. The trial court rejected that contention, noting that the petitioners failed “to point to a likely bidder and make a persuasive showing that increased competition would have led to a better result.” Id. (citing Dell, 177 A.3d at 28–29, 32, 34).

The petitioners also argued that the negotiators’ incentives undermined the pre-

signing phase, citing the desire of Aruba’s bankers to cater to HP and the more subtly divergent interests of Aruba’s CEO. The trial court found that although the petitioners proved that Aruba could have negotiated more aggressively, they did not prove that “the bankers, [the CEO], 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.” Id. at *44.

In other portions of the decision, the trial court found that Aruba’s unaffected trading price was a reliable indicator of fair value and rejected the parties’ DCF valuations

64

as unreliable. These holdings left the trial court with two reliable valuation indicators: the unaffected trading price and the deal price. The trial court determined that the unaffected trading price was the better measure of the fair value of Aruba’s shares. See id. at *53–55.

On appeal, the Delaware Supreme Court reversed. The high court found that the trial court had incorrectly relied on the unaffected trading price, but it accepted the trial court’s finding that the deal price was a reliable indicator of fair value. Aruba, 210 A.3d at 141–42.

Addressing the petitioners’ claim that the pre-signing phase of the sale process was insufficient to establish a competitive bidding dynamic, the Delaware Supreme Court emphasized that

when there is an open opportunity for many buyers to buy and only a few bid (or even just one bids), that does not necessarily mean that there is a failure of competition; it may just mean that the target’s value is not sufficiently enticing to buyers to engender a bidding war above the winning price.

Id. at 136. Applying this principle to the facts in Aruba, the high court explained:

Aruba approached other logical strategic buyers prior to signing the deal with HP, and none of those potential buyers were interested. Then, after signing and the announcement of the deal, still no other buyer emerged even though the merger agreement allowed for superior bids. It cannot be that an open chance for buyers to bid signals a market failure simply because buyers do not believe the asset on sale is sufficiently valuable for them to engage in a bidding contest against each other. If that were the jurisprudential conclusion, then the judiciary would itself infuse assets with extra value by virtue of the fact that no actual market participants saw enough value to pay a higher price. That sort of alchemy has no rational basis in economics.

Id. On the facts presented, the level of competition in Aruba was sufficient to support the reliability of the deal price.

65

The Delaware Supreme Court also explained that the negotiations between Aruba and HP over price had important implications for the reliability of the deal price:

[A] buyer in possession of material nonpublic information about the seller is in a strong position (and is uniquely incentivized) to properly value the seller when agreeing to buy the company at a particular deal price, and that view of value should be given considerable weight by the Court of Chancery absent deficiencies in the deal process.

Id. at 137. The high court noted that HP and Aruba went “back and forth over price” and that HP had “access to nonpublic information to supplement its consideration of the public information available to stock market buyers . . . .” Id. at 139. The Delaware Supreme Court elsewhere emphasized that “HP had signed a confidentiality agreement, done exclusive due diligence, gotten access to material nonpublic information,” and “had a much sharper incentive to engage in price discovery than an ordinary trader because it was seeking to acquire all shares.” Id. at 140. On the facts presented, the extent of the negotiations in Aruba was sufficient to support the reliability of the deal price.

The high court ultimately concluded that Aruba’s sale process was sufficiently reliable to render the deal price the best measure of fair value. The Delaware Supreme Court declined to use the trial court’s estimate of the deal price minus synergies, instead adopting HP’s contemporaneous synergies estimate and remanding with instructions that “final judgment be entered for the petitioners in the amount of $19.10 per share plus any interest to which the petitioners are entitled.” Id. at 142.

The Aruba decision technically did not involve a single-bidder process, but the dynamics closely resembled one. Although Qatalyst reached out to five bidders at the beginning of the first phase of the pre-signing process, none of those parties had any interest

66

in Aruba. After this development, both Qatalyst and Aruba’s CEO concluded that Aruba’s “only (but strong) weapon is to say we go alone.” Aruba Trial, 2018 WL 922139, at *10. Later, Aruba’s CEO had a “pretty open dialogue” with HP during which he informed HP that Aruba was “not running a sales process” and did not attempt to posture about pitting HP against anyone else. Id. at *40 (internal quotation marks omitted). During the second phase of the pre-signing process, after HP re-engaged, HP understood that Aruba was not pursuing other options. Id. at *41. The negotiations unfolded in a manner consistent with a single-bidder dynamic. See id.

In concluding that the deal price was a reliable indicator of fair value, the trial court considered a number of factors, including that “HP and Aruba agreed to terms for the merger agreement that the petitioners have not meaningfully challenged.” Id. at *38. After describing the suite of defensive measures in the merger agreement, the trial court noted that “[t]his combination of defensive provisions would not have supported a claim for breach of fiduciary duty.” Id. The petitioners had argued about a lack of competition during the pre-signing phase, and the trial court had discussed that factor at length, ultimately rejecting the objection. See id. at *39–41. On appeal, the Delaware Supreme Court emphasized that a failure of competition does not result simply because a limited number of parties bid, “or even just one bids.” Aruba, 210 A.3d at 136. The Delaware Supreme Court also emphasized the reliability of the price that resulted from the “back and forth” between Aruba and HP. Id. at 139.

Given these precedents, I cannot agree that a reliable sale process must invariably involve some level of active outreach during the pre-signing phase. By making this

67

observation, I am not suggesting that the Delaware Supreme Court has ever endorsed a single-bidder process for purposes of appraisal, nor that any of the precedents that this decision has discussed are squarely on point. Nor am I claiming to have any privileged insight into how the Delaware Supreme Court would or should evaluate the persuasiveness of a single-bidder strategy on the facts of any particular case. It nevertheless seems to me that if the proponent of a single-bidder process could show that the merger agreement allowed for a passive post-signing market check in line with what decisions have held is sufficient to satisfy enhanced scrutiny, and if there were no other factors that undermined the sale process, then the deal price would provide persuasive evidence of fair value.

This decision has already found that the sale process exhibited objective indicia of reliability. As noted and as discussed in greater detail below, the petitioners have not raised a meaningful challenge to the post-signing market check. The operative question for purposes of examining the pre-signing phase is not whether Stillwater’s process fell short of what would have been optimal, but rather whether the pre-signing process sufficiently impaired the sale process as a whole, including the post-signing phase, so as to prevent the deal price from serving as a persuasive indicator of fair value.

b. The Relative Involvement Of McMullen And The Board In The Pre-Signing Phase

In their initial challenge to the pre-signing phase, the petitioners attack McMullen’s role in the pre-signing process. They contend that McMullen acted improperly by pursuing Sibanye’s indication of interest without authorization from the Board and contrary to its direction to pursue acquisitions or a merger of equals. See PTOB at 37. They also criticize

68

McMullen for starting to engage with Sibanye in January 2016, but failing to inform the Board until after receiving an expression of interest from Sibanye in July. During this period, McMullen met with Sibanye’s senior executives at least twice to discuss a sale of Stillwater, reached an understanding with Sibanye’s CEO on pricing the deal at a 30% premium over Stillwater’s thirty-day VWAP, and arranged a multi-day site visit for Sibanye personnel.

The petitioners also contend that after the Board learned of Sibanye’s expression of interest in July 2016, the Board did not exercise meaningful oversight over the sale process. They accurately observe that the record lacks any evidence of meaningful engagement by the Board until October 3, 2016, two months before signing, when the Board received a report on the Company’s outreach to various parties, instructed McMullen to obtain formal proposals for retaining an investment bank, instructed McMullen to create a cash flow model that could be used to value the Company, and decided not to form a special committee. See JX 246.

The petitioners correctly contend that these facts could have contributed to findings that McMullen and the directors breached their duty of care under the enhanced scrutiny standard of review.16 But the enhanced scrutiny analysis would not have ended there. The

16 See Citron v. Fairchild Camera & Instr. Corp., 569 A.2d 53, 66 (Del. 1989) (“[I]n change of control situations, sole reliance on hired experts and management can taint[] the design and execution of the transaction. Thus, we look particularly for evidence of a board’s active and direct role in the sale process.” (internal quotation marks omitted)); Mills Acq. Co. v. Macmillan, Inc., 559 A.2d 1261, 1281 (Del. 1989) (“[A] board of directors . . . may not avoid its active and direct duty of oversight in a matter as significant as the sale of corporate control.”); Cede & Co. v. Technicolor, Inc. (Technicolor II), 634 A.2d

69

C & J Energy decision likewise involved a CEO that began deal discussions without formal board authorization, engaged for months without formally reporting to the board, made a revised offer without board approval, and generally proceeded by asking for forgiveness rather than by getting permission. See C & J Energy, 107 A.3d at 1059. After considering the totality of the sale process, the Delaware Supreme Court concluded that the facts would not support a fiduciary breach, placing heavy reliance on the directors’ decision to “test[] the transaction through a viable passive market check . . . .” Id. at 1053.

The outcome in PLX likewise shows that the existence of problems during the pre-

signing process does not necessarily undermine the reliability of the deal price. The trial court in PLX found that the directors had breached their fiduciary duties under the enhanced scrutiny standard because of an undisclosed tip from the eventual buyer to a key director and the company’s banker. PLX Trial, 2018 WL 5018535, at *15–16, *32–35, *44–47. Despite this defect, the sale process provided reliable evidence of the company’s value based primarily on the adequacy of the company’s post-signing market check. See id. at *55 (“More important than the pre-signing process was the post-signing market check.”).

345, 368 (Del. 1993) (explaining that directors must maintain “an active and direct role in the context of a sale of a company from beginning to end”); In re Rural Metro Corp. S’holders Litig., 88 A.3d 54, 91 (Del. Ch. 2014) (“As a threshold matter, the decision to initiate a sale process falls short under enhanced scrutiny because it was not made by an authorized corporate decisionmaker. The Board did not make the decision to launch a sale process, nor did it authorize the Special Committee to start one.”), aff’d sub nom. RBC Capital Mkts., LLC v. Jervis, 129 A.3d 816 (Del. 2015); id. (“One of the Delaware Supreme Court’s clearest teachings is that ‘directors cannot be passive instrumentalities during merger proceedings.’” (quoting Technicolor II, 634 A.2d at 368)).

70

Applying the same damages standard that would govern in an appraisal proceeding, the trial court found that the sale process was sufficiently reliable to render the plaintiffs’ damages calculation unpersuasive, resulting in a failure of proof. Id. at *50–55. The Delaware Supreme Court affirmed the judgment based solely on the trial court’s damages ruling. See PLX, 2019 WL 2144476, at *1.

McMullen’s unsupervised activities and the Board’s failure to engage in meaningful oversight until October 2016 represent flaws in the pre-signing process. They are factors that must be taken into account, but they do not inherently disqualify the sale process from generating reliable evidence of fair value.

In this case, McMullen’s unsupervised activities did not comprise the entirety of the Company’s sale process. Ultimately, after the Board engaged, Stillwater formally retained BAML, conducted an expedited pre-signing canvass, and entered into the Merger Agreement. The terms of the Merger Agreement facilitated a meaningful post-signing market check, and no other buyer emerged even though the merger agreement allowed for superior bids. As in Dell, the petitioners did not point to any evidence that another party was interested in proceeding and would have bid if McMullen and the Board had acted differently. See Dell, 177 A.3d at 29.

c. McMullen’s Personal Interest In A Transaction In their next challenge to the pre-signing process, the petitioners contend that McMullen undermined the sale process because he planned to leave Stillwater, and “he wanted the benefit of a strategic transaction (i) to boost the Company’s stock price prior to his departure and (ii) to maximize his payout upon stepping down as CEO.” PTOB at 39.

71

The petitioners correctly observe that by leaving after a transaction, McMullen would be entitled to unvested equity awards and accelerated retention payments that he could not obtain if he left without a transaction.

The petitioners also point out that McMullen devoted considerable time to developing and selling his personal investments outside of Stillwater. They cite McMullen’s contemporaneous service in 2016 as CEO of Nevada Iron and as President of New Chris, even though McMullen’s employment agreement with Stillwater limited McMullen’s outside activities to board service and otherwise required him to devote his full efforts to Stillwater. See JX 114. During 2016, McMullen raised money for the successor company to New Chris and sold Nevada Iron. See McMullen Tr. 709, 863–64. The petitioners cite McMullen’s activities (i) to show that McMullen was trying to maximize his personal wealth before retiring to Turks & Caicos, (ii) to suggest that McMullen might have done a better job with the sale process if he had not been pursuing his other investments, and (iii) as further evidence that the Board failed to provide active oversight.

Sibanye takes the extreme position that “there is no evidence to suggest that Mr.

McMullen was motivated by anything other than maximizing stockholder value.” Dkt. 211 at 59. Sibanye points to McMullen’s decision in March 2016 to extend his employment by two years, claiming simplistically that if “McMullen’s intention was truly to do a quick sale and leave the company, there would have been no need for him to renew his employment agreement since his prior contract did not expire until December 31, 2016 and contained essentially the same termination benefits as the new contract.” Id. at 60. To the

72

contrary, McMullen understood that completing a sale to Sibanye or another buyer might extend past December 31. Extending his employment agreement was the smart play for McMullen personally. Although Sibanye has not argued this point, it was also likely good for Stillwater, because it avoided the prospect of a near-term issue with CEO succession.

Sibanye has no meaningful response to McMullen’s pursuit of his other activities.

Sibanye says they were permitted, but the petitioners have correctly described McMullen’s employment agreement as only authorizing board service, not his more active roles. Sibanye also contends that his outside interests were disclosed in public filings, but that is not the point. The issue is whether the interests undermined the sale process, not whether they were disclosed. On this final point, Sibanye asserts that the petitioners “have pointed to no evidence that these outside interests presented an actual conflict, that these interests competed with or were adverse to Stillwater’s interests, or that they otherwise interfered with Mr. McMullen’s ability to carry out his duties as CEO of Stillwater.” Id.

Sibanye has focused on the critical question: whether McMullen’s personal interests undermined the sale process. Senior executives almost invariably have divergent incentives during a sale process, often because of change-in-control agreements, and equally often because the transaction will have implications for their personal employment situations.

Two Delaware appraisal precedents provide insight into factual scenarios involving divergent incentives of this type. The Aruba decision involved a sale process where the top executive and the company’s investment bankers had conflicting incentives. The CEO wanted to retire, but he cared deeply about the company and its employees. When HP proposed to acquire Aruba and keep the CEO on to integrate the companies, it offered the

73

perfect path “to an honorable personal and professional exit.” Aruba Trial, 2018 WL 922139, at *5; see id. at *43 (analyzing CEO’s conflict). Aruba’s investment bankers both wanted to curry favor with HP. Qatalyst was desperate to save its Silicon Valley franchise, and Evercore was auditioning for future business. Id. at *43. The trial court acknowledged the petitioners’ concerns, but found that the conflicting incentives did not undermine the deal price as an indicator of fair value:

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.

Id. at *44. On appeal, the Delaware Supreme Court accepted the reliability of the deal price as a valuation indicator and used it when making its own fair value determination. Aruba, 210 A.3d at 141–42.

The Dell decision also involved a conflict: Mr. Dell, the company’s founder and top executive, was a buy-side participant in the management buyout and would emerge from the transaction with a controlling stake. A special committee negotiated the terms of the transaction with the financial sponsor backing the deal, but the trial court regarded Mr. Dell’s involvement on the buy side as a factor cutting against the reliability of the deal price. For example, the trial court found that Mr. Dell gave the buyout group a leg-up given

74

his relationships within the company and his knowledge of its business, and the trial court accepted the testimony of a sale-process expert that if bidders competed to pay more than what Mr. Dell’s group would pay, then they risked overpaying and suffering the winner’s curse. In re Appraisal of Dell Inc. (Dell Trial), 2016 WL 3186538, at *42–43 (Del. Ch. May 31, 2016) (subsequent history omitted). Equally important, Mr. Dell was a net purchaser of shares in the buyout, so any increase in the deal price cost him money.

If Mr. Dell kept the size of his investment constant as the deal value increased, then Silver Lake would have to pay more and would demand a greater ownership stake in the post-transaction entity. [The petitioners’ sale-

process expert] showed that if Mr. Dell wanted to maintain 75% ownership of the post-transaction entity, then he would have to contribute an additional $250 million for each $1 increase in the deal price. If Mr. Dell did not contribute any additional equity and relied on Silver Lake to fund the increase, then he would lose control of the post-transaction entity at a deal price above $15.73 per share. Because Mr. Dell was a net buyer, any party considering an overbid would understand that a higher price would not be well received by the most important person at the Company.

Id. at *43 (footnote omitted). The trial court found that for purposes of price discovery in an appraisal case, Mr. Dell’s involvement and incentives undermined the reliability of the sale process and the persuasiveness of the deal price. Id. at *44.

On appeal, the Delaware Supreme Court held that Mr. Dell’s involvement in the buyout group had not undermined the sale process. See Dell, 177 A.3d at 32–33. The high court noted that “the [trial court] did not identify any possible bidders that were actually deterred because of Mr. Dell’s status.” Id. at 34. The Delaware Supreme Court also emphasized Mr. Dell’s willingness to work with rival bidders during due diligence and the absence of evidence that Mr. Dell would have left the company if a rival bidder prevailed. Id. at 32–34. The high court concluded that the lack of a higher bid did not call into question

75

the sale process, because “[i]f a deal price is at a level where the next upward move by a topping bidder has a material risk of being a self-destructive curse, that suggests the price is already at a level that is fair.” Id. at 33.

The facts of C & J Energy are also relevant. The merger in C & J Energy resulted from a CEO-driven process, and there was evidence that the sell-side CEO had personal reasons to favor the deal because he would be in charge of the combined company and receive significantly greater compensation. See C & J Energy, 107 A.3d at 1064. After the key terms of the transaction had been negotiated, but before it was formally approved, the CEO went so far as to demand a side letter “affirming that C & J’s management would run the surviving entity and endorsing a generous compensation package.” Id. When the acquirer balked, the CEO threatened to terminate the discussions. He got his way, and the deal was announced as planned. Id. at 1065. There was also evidence that C & J Energy’s primary financial advisor acted as a banker for the deal rather than for C & J Energy, and the banker had divergent interests as a source of financing for the deal. See id. at 1056–57. The Delaware Supreme Court held that the facts could not support a reasonable probability that the defendants had failed to obtain the best transaction reasonably available, relying heavily on the post-signing market check. See id. at 1053, 1067–68.

In this case, McMullen’s personal interests are not as serious as the buy-side conflict that failed to undermine the sale process in Dell. They more closely resembled the divergent sell-side interests that affected the negotiators in Aruba and C & J Energy. Like the CEOs and bankers in those cases, McMullen’s change-of-control benefits gave him a personal reason to secure a deal under circumstances where a disinterested participant

76

might prefer a standalone option. McMullen appears to have been motivated by his desire to maximize his personal wealth and retire to a greater degree than the negotiators in Aruba. Stillwater’s general counsel (Wadman) recognized McMullen’s conflict, voiced his concerns to the Board, and ultimately resigned when McMullen secured more favorable treatment in the Merger for his own change-in-control benefits and for his CFO. See JX 526. As a result, McMullen’s motivations most closely resembled the incentives of the CEO in C & J Energy, who held up the entire transaction until the acquirer agreed to a side letter “affirming that C & J’s management would run the surviving entity and endorsing a generous compensation package.” C & J Energy, 107 A.3d at 1064. The Delaware Supreme Court held that the facts in C & J Energy did not provide reasonable grounds for a breach of fiduciary duty under the enhanced scrutiny standard of review.

At the same time, McMullen had ample reason to pursue the best deal possible for Stillwater. From his testimony and demeanor, McMullen seems like someone who took considerable satisfaction in his ability to achieve outcomes. As a matter of professional pride, he wanted to sell Stillwater for the best price he could. He also had economic reasons to extract a higher price. As disclosed in the proxy statement for the Merger, McMullen held 131,248 common shares, 155,891 restricted stock unit awards, and 222,556 performance based restricted stock unit awards, for a total of 509,695 common shares or share equivalents. See JX 525 at 78. At the deal price, these common shares and share equivalents had a value of $9,174,510. To state the obvious, every $1 increment in the deal price generated another half-a-million dollars for McMullen.

77

When directors or their affiliates own “material” amounts of common stock, it aligns their interests with other stockholders by giving them a “motivation to seek the highest price” and the “personal incentive as stockholders to think about the trade off between selling now and the risks of not doing so.”

Chen v. Howard-Anderson, 87 A.3d 648, 670–71 (Del. Ch. 2014) (quoting Dollar Thrifty, 14 A.3d at 600); see also Lender Processing, 2016 WL 7324170, at *22 (discussing incentive to maximize deal price where target managers were net sellers and would not retain jobs post-merger).

Consistent with his personal desire to obtain a good price for Stillwater, McMullen negotiated with Sibanye to increase the consideration. When Sibanye indicated interest at $15.75 per share in July 2016, McMullen did not rush to sign up a deal. When Sibanye raised indication of interest to $17.50 to $17.75 per share in December 2016, McMullen and the Board demanded a higher price. Even after Sibanye offered $18.00 per share, McMullen wanted more. Only after Sibanye twice said that $18.00 per share was its best and final offer did McMullen and the Board finally agree to transact.

As with McMullen’s initiation of the sale process and the Board’s failure to engage in meaningful oversight of his activities until October 2016, McMullen’s personal motivation to exit from Stillwater and maximize his personal wealth represents a flaw in the sale process. Although Wadman’s noisy withdrawal highlighted these issues, McMullen’s personal interests as a whole do not appear materially different from interests that have not been sufficient in other cases to undermine the reliability of sale processes. On balance, the evidence does not convince me that McMullen’s divergent interests led either McMullen or the Board to accept a deal price that left a portion of Stillwater’s

78

fundamental value on the table, particularly in light of the effective post-signing market check that Stillwater conducted.

d. The “Soft-Sell”

Turning to the details of the pre-signing phase, the petitioners contend that Stillwater’s pre-signing market check fell short because until BAML was formally retained, McMullen relied on a “soft sell” approach that provided potential buyers with insufficient information to conclude that Stillwater was for sale and used unauthorized agents who could not formally engage on Stillwater’s behalf. See PTOB at 44–45.

The evidence demonstrates that on the facts of this case, the “soft sell” strategy was not an effective means of generating interest in the Company. At the same time, the “soft sell” effort did not do anything to harm either BAML’s abbreviated pre-signing process or the post-signing market check. The soft sell strategy was not a positive feature of the sale process, and it does not help support the persuasiveness of the deal price, but it does not detract from it either.

e. BAML’s Compressed Pre-Signing Market Check In a further criticism of the pre-signing phase, the petitioners contend that after BAML was formally retained, BAML did not have time to run an organized and meaningful process. The petitioners complain that BAML hastily called a list of potentially interested parties, who then were given only days after signing an NDA to prepare an expression of interest. Contrary to McMullen’s strong recommendation in September 2016 that any bidder visit the Company’s mines before providing an expression of interest, the November timeline did not accommodate site visits until after a party made an expression

79

of interest. See JX 229 at ‘603. At trial, the petitioners introduced testimony from a sale process expert who questioned the effectiveness of BAML’s abbreviated pre-signing process. See Gray Tr. 567–68. Even Sibanye’s sale process expert questioned the effectiveness of the type of condensed outreach that BAML attempted to conduct. See Stowell Tr. 947–48.

The petitioners have made a persuasive case that the BAML’s pre-signing process was suboptimal, but they have not shown that it was worthless, nor that it was harmful. To the contrary, when evaluated against Delaware precedents, the pre-signing efforts, while rushed, were a positive factor for the sale process.

BAML received its formal mandate on November 7, 2016. The next day, BAML received a package of information from the Company, including Sibanye’s indication of interest from July, the non-disclosure agreements with Hecla and Coeur, a cash flow model, and instructions for accessing the data room. BAML understood that Sibanye was pushing to close a deal by December and swung into action to do what it could. By November 9, BAML had generated a plan for an expedited market check that contemplated reaching out to twenty parties over the next two days, working with parties who expressed interest for the rest of the month, and then receiving expressions of interest at the end of the month. At that point, the Board would decide how to proceed.

In accordance with its expedited plan, BAML engaged directly with Sibanye, Coeur, and Hecla. BAML contacted five of the six parties that BAML regarded as “Possibly Interested,” missing one. BAML contacted eight of the twelve additional parties that

80

BAML had identified, missing four. BAML also contacted Northern Star, even though they originally had been listed as not interested.

Ten of the fourteen parties had no interest, but four engaged. One quickly withdrew, two ultimately expressed interest in a merger of equals, and the fourth dropped out by late November. Coeur also dropped out, and Hecla indicated that it needed to find a partner to pursue a transaction. Although the Board extended Hecla’s deadline for submitting an indication of interest, and BAML followed up with Hecla, Hecla did not respond. At the end of November, an additional party—Northam—asked to be included in the Company’s process.

During a meeting on December 2, 2016, the Board considered the status of the Company’s process. At that point, the Board’s only definitive expression of interest was a proposal that Sibanye had submitted on December 1 to acquire the Company for between $17.50 and $17.75 per share in cash. The Board decided to focus on Sibanye, which later raised its offer to $18 per share. Northam decided to withdraw, and on December 8, the Board approved the Merger Agreement.

Although compressed and expedited, BAML’s outreach resulted in fourteen other parties hearing about Stillwater. In addition to Sibanye, a total of seven parties engaged to some degree. Ultimately, no one other than Sibanye submitted an indication of interest. The plaintiffs have criticized the timing, pacing, and scope of the pre-signing process, but it resulted in BAML contacting the “logical strategic buyers” before Stillwater signed up its deal with Sibanye. Cf. Aruba, 210 A.3d at 136 (observing that “Aruba approached other logical strategic buyers prior to signing the deal with HP, and none of those potential buyers

81

were interested.”). The number of meaningful contacts compares favorably with or is similar to the facts in the Delaware Supreme Court precedents.17 When considering whether a deal price provides persuasive evidence of fair value, it is pertinent that the parties contacted failed to pursue a merger when they had a free chance to do so. See DFC, 172 A.3d at 376 (citing “failure of other buyers to pursue the company when they had a free chance to do so” as factor supporting fairness of deal price).

On balance, BAML’s pre-signing efforts were helpful. At a minimum, the abbreviated process generated incremental interest in Stillwater and gave those parties who engaged a leg up for the post-signing market check. Even the parties who were contacted but did not engage had the benefit of knowing that a transaction potentially was afoot. As with the “soft sell” strategy, there is no evidence that BAML’s abbreviated process did anything to harm the sale process. The bidders who participated in the abbreviated pre- signing phase were free to bid during the post-signing phase. There is no evidence that any were alienated or put off by the Company’s pre-signing efforts.

17 See Aruba, 210 A.3d at 136–39, 142 (adopting deal price less synergies as fair value where company’s banker contacted five potential buyers after HP’s initial outreach, none were interested, sale process terminated, and sale process later resumed as single- bidder engagement with HP, with only one quick contact to a sixth party); Dell, 177 A.3d at 28 (finding competitive pre-signing process where Silver Lake competed one-at-a-time with interested parties); DFC, 172 A.3d at 350, 355, 376 (finding “competitive process of bidding” where company’s banker contacted “every logical buyer,” three expressed interest, and two named a preliminary price with one dropping out before serious negotiations commenced).

82

BAML’s abbreviated pre-signing process was not ideal. Nevertheless, contrary to the petitioners’ contentions, it was a positive factor for the reliability of the sale process.

f. The Negotiations With Sibanye In their penultimate objection to Stillwater’s pre-signing process, the petitioners contend that Sibanye pressured Stillwater to sign a merger agreement before the Company’s rising stock price made what Sibanye was willing to pay look inadequate. The evidence demonstrates that early in his discussions with Sibanye, McMullen and Froneman recognized that any transaction would require a premium over Stillwater’s trading price and agreed in principle on a 30% premium over the thirty-day VWAP. On October 17, 2016, Froneman told McMullen that Sibanye’s offer of a “30% premium to VWAP remained unchanged” and that Sibanye’s board of directors unanimously supported the transaction. JX 281 at ‘425. Another Sibanye executive repeated this message on November 22. PTO ¶ 243.

Sibanye, however, needed to borrow the funds to acquire Stillwater, and by November 30, 2016, Stillwater’s share price had recovered to a point where a 30% premium over the thirty-day VWAP equaled $18.25 per share. Sibanye could not pay more than $18 per share without supplementing the consideration with cash on hand or a draw from its revolving credit line, which Sibanye did not want to do. Rather than sticking with the concept of a 30% premium over a thirty-day VWAP, Sibanye disavowed that concept, instead treating its prior indication of interest from July 2016 as a fixed price of $15.75 per share. On December 1, 2016, Sibanye proposed a transaction in a range of $17.50 to $17.75

83

per share, below what the 30% premium to the thirty-day VWAP would have contemplated.

The petitioners object that rather than breaking off discussions or continuing the sale process, the Board negotiated a price of $18.00 per share, representing the maximum that Sibanye could pay under its financing arrangements. They argue that the highest price a bidder is willing to pay is not the same as fair value. See, e.g., M.P.M. Enters., 731 A.2d at 797 (cautioning that the merger price must be supported “by evidence tending to show that it represents the going concern value of the company rather than just the value of the company to one specific buyer”); In re Appraisal of Orchard Enters., Inc., 2012 WL 2923305, at *5 (Del. Ch. July 18, 2012) (“[A]lthough I have little reason to doubt Orchard’s assertion that no buyer was willing to pay Dimensional $25 million for the preferred stock and an attractive price for Orchard’s common stock in 2009, an appraisal must be focused on Orchard’s going concern value.”).

The petitioners’ objection resembles similar arguments that the Delaware Supreme Court rejected in Dell and DFC. In Dell, the trial court found that the price negotiations during the pre-signing phase were limited by what the financial sponsors could pay based on their leverage-buyout pricing models. The respondent had conceded that the LBO model was not “oriented toward solving for enterprise value,” and the special committee’s financial advisors had briefed the committee about the LBO model and how financial sponsors would use it. Dell Trial, 2016 WL 3186538, at *29 (internal quotation marks omitted). The committee’s financial advisors used a similar model to calculate the maximum prices that a financial sponsor could pay. See id. at *30. The evidence indicated

84

that the financial sponsors bid consistently with the results of an LBO model, and their negotiations with the committee proceeded within that framework. See id. at *30–32. In addition to the record evidence, the trial court relied on treatises which explained how the price generated by an LBO model can diverge from fair value.18 Based on this evidence, the trial court found that the original merger consideration “was dictated by what a financial sponsor could pay and still generate outsized returns,” rather than Dell’s value as a going concern. Id. at *32.

Three months later, the trial court in DFC reached a similar conclusion when evaluating the deal price paid by a financial sponsor (Lone Star) to acquire the company (DFC) that was the subject of the appraisal proceeding. Although the trial court regarded the deal price as sufficiently reliable to use as a valuation input, the court expressed concern that “Lone Star’s status as a financial sponsor . . . focused its attention on achieving a

18 See Dell Trial, 2016 WL 3186538, at *29 & n.24 (citing Joshua Rosenbaum & Joshua Pearl, Investment Banking: Valuation, Leveraged Buyouts, and Mergers & Acquisitions 195–96 (2009) (“[An LBO model] is used . . . to determine an implied valuation range for a given target in a potential LBO sale based on achieving acceptable returns. . . .”); and Donald M. DePamphilis, Mergers, Acquisitions, and Other Restructuring Activities 506 (7th ed. 2014) (“[T]he DCF analysis solves for the present value of the firm, while the LBO model solves for the internal rate of return.”)); id. at *29 nn. 25, 26 (citing Rosenbaum & Pearl, supra, at 195–96 (“In an M&A sell-side advisory context, the banker conducts LBO analysis to assess valuation from the perspective of a financial sponsor. This provides the ability to set sale price expectations for the seller and guide negotiations with buyers accordingly . . . .” (emphasis added)); id. at 235–36 (“Traditionally, the valuation implied by LBO analysis is toward the lower end of a comprehensive analysis when compared to other methodologies, particularly precedent transactions and DCF analysis. This is largely due to the constraints imposed by an LBO, including leverage capacity, credit market conditions, and the sponsor’s own IRR hurdles.”)).

85

certain internal rate of return and on reaching a deal within its financing constraints, rather than on DFC’s fair value.” In re Appraisal of DFC Glob. Corp. (DFC Trial), 2016 WL 3753123, at *22 (Del. Ch. July 8, 2016) (subsequent history omitted).

The appeal from the trial-level ruling in DFC reached the Delaware Supreme Court before the appeal in Dell. The Delaware Supreme Court rejected the trial court’s finding that the buyer’s financial constraints limited the price it could pay and caused the deal price to diverge from fair value, stating:

To be candid, we do not understand the logic of this finding. Any rational purchaser of a business should have a targeted rate of return that justifies the substantial risks and costs of buying a business. That is true for both strategic and financial buyers. It is, of course, natural for all buyers to consider how likely a company’s cash flows are to deliver sufficient value to pay back the company’s creditors and provide a return on equity that justifies the high costs and risks of an acquisition. But, the fact that a financial buyer may demand a certain rate of return on its investment in exchange for undertaking the risk of an acquisition does not mean that the price it is willing to pay is not a meaningful indication of fair value. That is especially true here, where the financial buyer was subjected to a competitive process of bidding, the company tried but was unable to refinance its public debt in the period leading up to the transaction, and the company had its existing debt placed on negative credit watch within one week of the transaction being announced. The “private equity carve out” that the Court of Chancery seemed to recognize, in which the deal price resulting in a transaction won by a private equity buyer is not a reliable indication of fair value, is not one grounded in economic literature or this record.

DFC, 172 A.3d at 349–50. When the Delaware Supreme Court subsequently ruled on the discussion of the LBO model in the appeal from the trial-level ruling in Dell, the high court relied on its decision in DFC, explaining:

[W]e rejected this view [in DFC] and do so again here given we see “no rational connection” between a buyer’s status as a financial sponsor and the question of whether the deal price is a fair price. After all, “all disciplined buyers, both strategic and financial, have internal rates of return that they

86

expect in exchange for taking on the large risk of a merger, or for that matter, any sizeable investment of its capital.”

Dell, 177 A.3d at 28 (quoting DFC, 172 A.3d at 374–76).

The reasoning that led the Delaware Supreme Court to reject the implications of the LBO model for deal pricing indicates that comparable constraints on a prevailing bidder’s ability or willingness to pay—whether resulting from IRR hurdles, a comparatively higher cost of capital, or limits on the availability of financing—should not undermine the deal price as an indicator of fair value if the sale process was otherwise sufficiently open. Both Dell and DFC suggest that a post-signing market test can be the predominant source of price competition. In Dell, the only participants during the pre-signing phase were the two financial sponsors, whom the committee permitted to participate at any one time and each of whom priced their deals using an LBO model. See Dell Trial, 2016 WL 3186538, at *9– 10, *30–31, *37. In DFC, although the company initially engaged in a broad solicitation, the only bidders who engaged and submitted indications of interest during the pre-signing phase were two financial sponsors, one of whom soon dropped out. See DFC Trial, 2016 WL 3753123, at *4.

On the facts of this case, Sibanye had the ability to pay more. Although it had not secured transactional financing that would have supported a price greater than $18.00 per share, Sibanye could have deployed cash on hand or drawn on its revolving line of credit. As a rational bidder for Stillwater, Sibanye understandably had a targeted rate of return that it needed to satisfy to justify the substantial risks and high costs of the acquisition. That Sibanye did not bid higher does not mean that the price it agreed to pay did not reflect

87

fair value when its bid prevailed. See Aruba, 210 A.3d at 136; Dell, 177 A.3d at 28; DFC, 172 A.3d at 349–50, 374–76.

The negotiations between Stillwater and Sibanye over price, together with Sibanye’s refusal to pay more, provides strong evidence of fair value. In Aruba, the Delaware Supreme Court explained that

a buyer in possession of material nonpublic information about the seller is in a strong position (and is uniquely incentivized) to properly value the seller when agreeing to buy the company at a particular deal price, and that view of value should be given considerable weight by the Court of Chancery absent deficiencies in the deal process.

Id. at 137. The high court observed that HP and Aruba went “back and forth over price” and that HP had “access to nonpublic information to supplement its consideration of the public information available to stock market buyers . . . .” Id. at 139. The Delaware Supreme Court elsewhere emphasized that “HP had signed a confidentiality agreement, done exclusive due diligence, gotten access to material nonpublic information” and “had a much sharper incentive to engage in price discovery than an ordinary trader because it was seeking to acquire all shares.” Id. at 140. Given these facts, the extent of the negotiations in Aruba supported the reliability of the deal price. The same observations apply to Sibanye on the facts of this case. Sibanye entered into an NDA with Stillwater, conducted extensive due diligence, obtained access to material nonpublic information, and was “in a strong position (and is uniquely incentivized) to properly value the seller when agreeing to buy the company at a particular deal price.”

The fact that Stillwater and Sibanye reached agreement at $18.00 per share is entitled to considerable weight. Although the petitioners perceive it to be a weakness of

88

the pre-sale process, the Delaware Supreme Court’s precedents indicate that it was a strength.

4. The Challenges To The Post-Signing Phase In contrast to their many objections to the pre-signing phase, the petitioners have relatively few disagreements with the post-signing phase. They advance perfunctory challenges to the terms of the Merger Agreement, claiming that it prevented the stockholders from capturing the value of an increasing palladium price and foreclosed other bids. They also contend that the proxy statement contained disclosure violations.

a. The Merger Agreement And The Price Of Palladium The petitioners observe that the price of palladium increased between signing and closing. They then object that the Merger Agreement “provided no practical way for Stillwater’s stockholders to receive that additional value.” PTOB at 51. In cursory fashion, they criticize the Board for not asserting the existence of a Company Material Adverse Effect or invoking the fiduciary-out clause. Id. at 52. This objection is not really a criticism of the sale process, but so be it.

The petitioners never engage with the terms of the Merger Agreement and how it uses the concept of a Company Material Adverse Effect. The definition of a Company Material Adverse Effect turns on any “facts, circumstance, condition, event, change, development, occurrence, result, or effect” that is materially adverse to the Company. JX 575, Annex A, at A-3. The arising of a Company Material Adverse Effect does not mean that something good has happened to Stillwater, like an increase in value due to rising commodity prices. It means something very bad has happened to Stillwater. In the Merger

89

Agreement, Stillwater represented that it had not suffered a Company Material Adverse Effect, and the Merger Agreement made the accuracy of this representation a condition to Sibanye’s obligation to close. See id. §§ 4.10.2, 7.2.1. The Merger Agreement also made the absence of a Company Material Adverse Effect a separate condition to Sibanye’s obligation to close. See id. § 7.2.3. Stillwater did not obtain the right to declare something akin to a Company Material Beneficial Effect and terminate the Merger Agreement on that basis. The petitioners’ criticism that the Board did not declare a Company Material Adverse Effect is a turn down a blind alley.

The petitioners likewise never engage with the terms of the Merger Agreement and the scope of the fiduciary out. The Board had the right to change its recommendation in favor of the Merger based on (i) its receipt of a “Superior Proposal” or (ii) the occurrence of an “Intervening Event.” See JX 525, Annex A, § 6.2.4. As permitted by Delaware law, see 8 Del. C. § 146, the Merger Agreement contained a force-the-vote provision that obligated Stillwater to take the Merger to a stockholder vote even if the Board changed its recommendation, but the stockholders would have the benefit of the Board’s negative recommendation when voting. See id. § 6.17.2 (“Without limiting the generality of the foregoing, the Company shall submit this Agreement for the adoption by its stockholders . . . whether or not a Company Adverse Recommendation Change shall have occurred or an Acquisition Proposal shall have been publicly announced or otherwise made . . . .”). If the Company’s stockholders voted down the Merger, or under other defined circumstances, then the Company had the ability to terminate the Merger Agreement. See id. § 8.1.2(ii). The Board’s ability to change its recommendation for an Intervening Event, however, did

90

not include changes in commodity prices. The Merger Agreement defined the concept of an “Intervening Event” as

any material change, event, effect, occurrence, consequence or development with respect to the Company or Parent, as applicable, that (i) is unknown and not reasonably foreseeable as of the date hereof, (ii) does not relate to any Acquisition Proposals, and (iii) does not arise out of or result from changes after the date of this Agreement in respect of prices or demand for products.

Id. at A-6; cf. R. Franklin Balotti & A. Gilchrist Sparks, III, Deal Protection Measures and the Merger Recommendation, 96 Nw. U. L. Rev. 467, 468 (2002) (explaining the importance of an intervening event provision for the target who “discover[s] the world’s largest deposit of gold under its headquarters, causing the value of the target to increase dramatically”). Post-signing changes “in respect of prices or demand” for palladium thus would not qualify as an Intervening Event and would not support a change of recommendation. The petitioners’ criticism that the Board did not exercise its fiduciary out based on changes in commodity prices is another wrong turn.

The record reflects that Stillwater did not want the merger consideration to float with the price of palladium. McMullen testified that “we wanted to know with certainty what was the number that we were taking to shareholders as the value proposition.” McMullen Tr. 770. That was a legitimate goal.

The petitioners may well take these explanations and run with them, claiming that the situation was even worse than they thought because the Board lacked the power to do things that the petitioners previously believed the Board had merely failed to consider. Regardless, the petitioners’ bottom-line criticism of the Merger Agreement misses the point of what the contract was trying to accomplish. The Merger Agreement was not

91

attempting to give the stockholders the benefit of a transaction that included the potential upside or downside that would result from changes in the price of palladium after signing. The Merger Agreement was trying to provide stockholders with the ability to opt for the comparative certainty of deal consideration equal to $18.00 per share.

More broadly, the petitioners are mistaken when they claim that there was no practical way for Stillwater’s stockholders to receive the additional value that the increased commodity price could generate. If Stillwater’s stockholders had wanted to capture the increased value of palladium, then they could have voted down the Merger and kept their shares. The spot price of palladium was readily available public information that Stillwater’s stockholders could take into account when deciding how to vote.

b. The Merger Agreement And Competing Bids In conclusory fashion, the petitioners object that the Merger Agreement “contained a no solicitation provision and 5-day matching rights,” which the petitioners characterize as “more buyer friendly than the protections provided in AOL that this Court described as creating ‘structural disadvantages dissuading any prospective bidder.’” PTOB at 51–52 (quoting AOL, 2018 WL 1037450, at *9, and noting that the decision “describe[ed] a no- shop provision with a 3.5% termination fee and unlimited 3-day matching rights”). The petitioners argue that Sibanye’s matching rights deterred interested buyers from making a topping bid because Sibanye could simply match any competing proposal.

The AOL decision was a fact-specific ruling that turned on the court’s view of the sale process in that case, after hearing the witnesses at trial and considering the evidentiary record. The Dell and DFC decisions issued while the matter was pending, and the trial

92

court requested supplemental briefing on the effect of those decisions. Both sides continued to argue for determining fair value based on financial metrics rather than by relying on the deal price. AOL, 2018 WL 1037450, at *1. The court nevertheless examined the sale process and regarded the persuasiveness of the deal price as “a close question.” Id. On balance, the court decided not to rely on the deal price, except as cross check to a DCF valuation. In reaching this outcome, the court placed heavy weight on a comment made by AOL’s CEO, shortly after the signing of the deal, in which he said he was “committed to doing the deal with Verizon” and emphasized that he “gave the team at Verizon my word that . . . this deal is going to happen.” Id. at *9. The court found that the comment “could reasonably cause potential bidders to pause when combined with the deal protections here.” Id. A trial court’s job is to make that type of decision and determine when the evidence warrants a case-specific departure from a general rule.

The broader Delaware corpus supports the general principle that the package of defensive measures found in the Merger Agreement in this case is sufficient to permit an effective post-signing market check, even when matching rights are present. As noted, commentators have perceived that under the Delaware Supreme Court’s recent appraisal decisions, a sale process involving a publicly traded firm will function as a reliable indicator of fair value as long as it would pass muster if reviewed under enhanced scrutiny in a breach of fiduciary duty case. See Hamermesh & Wachter, supra, at 962, 982–83; Korsmo & Myers, supra, at 269. Based on numerous trial court precedents, the suite of deal protection measures in the Merger Agreement would not have supported a claim for

93

breach of fiduciary duty.19 The suite of deal protections in the Merger Agreement compared favorably with the deal protections in C & J Energy and PLX, which this decision has discussed at length.

19 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 [the acquirer]; and (5) a $5 million termination fee” where termination fee represented 4.5% of equity value and change-of-recommendation provision included unlimited matching right); In re BJ’s Wholesale Club, Inc. S’holders Litig., 2013 WL 396202, at *13 (Del. Ch. Jan. 31, 2013) (rejecting fiduciary challenge to a merger agreement with a no-shop provision, matching and information rights, a termination fee representing 3.1% of deal value, and a force-the-vote provision; observing that “under Delaware law, these deal protection measures, individually or cumulatively, have routinely been upheld as reasonable”); 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 Synthes, Inc. S'holder Litig., 50 A.3d 1022, 1049 (Del. Ch. 2012) (finding that a termination fee of 3.05% of equity value, a no-solicitation provision with a fiduciary out and matching rights, a force-the-vote provision, and a voting agreement that locked up at least 33% of the company shares in favor of the merger were not unreasonable deal protection devices); 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” (quoting In re 3Com S’holders Litig., 2009 WL 5173804, at *7 (Del. Ch. Dec. 18, 2009))); In re Atheros Commc’ns, Inc. S’holder Litig., 2011 WL 864928, at *7 n.61 (Del. Ch. Mar. 4, 2011) (same analysis for no-solicitation provision, matching right, and termination fee); In re 3Com, 2009 WL 5173804, at *7 & n.37 (rejecting challenge to merger agreement with a no-solicitation provision, matching rights, and a termination fee in excess of 4% of equity value; describing provisions as having been “repeatedly” upheld by this court and collecting authorities).

94

The Aruba decision involved a similar suite of deal protections. The merger agreement in that case “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” and included a termination fee equal to 3% of the merger’s equity value. Aruba Trial, 2018 WL 922139, at *21, *38. The matching rights were similar too: HP had “an unlimited match right, with five days to match the first superior proposal and two days to match any subsequent increase, and during the match period Aruba had to negotiate exclusively and in good faith with HP.” Id. at *38 (footnote omitted). Viewing the deal protections holistically, the Delaware Supreme Court found that potential buyers had an open chance to bid, which supported the high court’s use of a deal-price-less- synergies metric to establish fair value. See Aruba, 210 A.3d at 136.

The Delaware Supreme Court has explained that a post-signing market check is effective as long as “interested bidders have a fair opportunity to present a higher-value alternative, and the board has the flexibility to eschew the original transaction and accept the higher-value deal.” C & J Energy, 107 A.3d at 1068. This description comports with guidance from a frequently cited treatise, which identifies “critical aspects” of a merger agreement that does not “preclude or impermissibly impede a post-signing market check.” 1 Lou R. Kling & Eileen T. Nugent, Negotiated Acquisitions of Companies, Subsidiaries and Divisions § 4.04[6][b], at 4-89 to -90 (1992 & Supp. 2019).

First, the economics of the executed agreement must be such that it does not unduly impede the ability of third parties to make competing bids. Types of arrangements that might raise questions in this regard include asset lock-ups, stock lock-ups, no-shops, force-the-vote provisions, and termination fees.

95

The operative word is “unduly;” the impact will vary depending upon the actual type of device involved and its specific terms.

***

Second, the target should be permitted to disclose confidential information to any third party who has on its own (i.e., not been solicited) “shown up” in the sense that it has submitted a proposal or, at a minimum, an indication of interest which is, or which the target believes is, reasonably likely to lead to (and who is capable of consummating) a higher competing bid. In this regard, the target should also be able to negotiate with such third parties. This removes any informational advantage that the initial (anointed) purchaser may have.

***

Finally, the target board of directors should have the contractual right, without violating the acquisition agreement, to withdraw or modify its recommendation to shareholders with respect to the transaction provided for in the executed acquisition agreement.

Id. at 4-90 to -94.1 (footnotes omitted).

Using this framework, the deal protections did not preclude or impermissibly impede a post-signing market check. For starters, any party could submit a bona fide written Acquisition Proposal. If the Board determined that the Acquisition Proposal “constitutes, or could reasonably be expected to result in, a Superior Proposal” and entered into an “Acceptable Confidentiality Agreement” with the party making the proposal, then the Board could “engage in negotiations or discussions with, or furnish any information to,” the party making the Acquisition Proposal. JX 545, Annex A, § 6.2.2. Additional requirements included that the Company notify Sibanye within twenty-four hours of its determination, furnish Sibanye “substantially concurrently” with any information provided to the third party, and not share any of Sibanye’s confidential information unless required

96

by law. Id. The Company also had to notify Sibanye of the terms of the Acquisition Proposal and the identity of the third party making it, then keep Sibanye informed of any developments on a reasonably prompt basis. Id. § 6.2.3.

After that point, if the Board determined that the Acquisition Proposal constituted a Superior Proposal and that its fiduciary duties required it, then the Board could change its recommendation in favor of the Merger, provided that before doing so, the Board gave Sibanye five days in which to match the Superior Proposal or otherwise offer changes to the Merger Agreement to avoid the change of recommendation. The Board could also withdraw or modify its recommendation for an Intervening Event, again conditioned on giving Sibanye five days in which to propose changes to the Merger Agreement to avoid the change of recommendation. If the stockholders voted down the deal, then Stillwater could terminate the Merger Agreement, subject only to paying a termination fee and expense reimbursement equal to 1.2% of the Merger’s equity value.

The post-signing market check began on December 9, 2016, when Sibanye and the Company announced the Merger. It ended on April 26, 2017, when the Company’s stockholders approved the Merger Agreement. The resulting passive market check lasted 138 days, close to the 153 days in C & J Energy and far longer than many of the passive, post-signing market checks that the Delaware courts have approved. See App.

During the post-singing market check, no one bid. The failure of any other party to come forward provides significant evidence of fairness, because “[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.” Dell, 177 A.3d at 29; see id. at 32, 34. The absence of a higher bid

97

indicates “that the deal market was already robust and that a topping bid involved a serious risk of overpayment,” which in turn “suggests the price is already at a level that is fair.” Id. at 33. As in Aruba, “[i]t cannot be that an open chance for buyers to bid signals a market failure simply because buyers do not believe the asset on sale is sufficiently valuable for them to engage in a bidding contest against each other.” Aruba, 210 A.3d at 136. Instead it suggests that “the target’s value is not sufficiently enticing to buyers to engender a bidding war above the winning price.” Id.

c. The Stockholder Vote In their last challenge to the post-signing phase, the petitioners assert that the stockholders approved the Merger based on incomplete and misleading information. They devote only two pages in their opening brief to this argument, the bulk of which describes the legal principles that apply in fiduciary duty cases. See PTOB at 53–54 (citing Morrison v. Berry, 191 A.3d 268, 282–83 (Del. 2018); and Corwin v. KKR Fin. Hldgs., LLC, 125 A.3d 304, 312 (Del. 2015)). The factual description of their disclosure theory appears in just three sentences:

Stillwater’s stockholders were told McMullen led the sale process, but they were never informed that he was preparing to leave the Company or the scope of his outside business ventures. In addition, Stillwater stockholders were told that Wadman left the Company prior to closing, but they were never informed of the context of his departure or his “noisy exit.” Stillwater’s stockholders were also provided no information regarding the Company’s exploration zones.

PTOB at 53–54. They devote the same amount of space to this theory in their reply brief, although the text extends over three pages. Dkt. 228 at 26–28. In their reply brief, they argue that stockholders should have been told that Wadman raised concerns about

98

McMullen’s conflicts of interest and “his manner of soliciting interest from third parties,” and that Wadman was “retaliated against for doing so.” Id. at 27. They also argue that stockholders should have been told that McMullen “was in violation of his 2016 employment agreement” while running the sale process because of his roles with Nevada Iron and New Chris. Id.

The petitioners’ argument about Stillwater’s exploration zones does not appear to hold up under their own understanding of the law. The petitioners elsewhere argued persuasively that under Industry Guide 7, promulgated by the Securities and Exchange Commission, Stillwater was not permitted to disclose information about the value of the Company’s exploration zones. See, infra, Pt. II.B.3.a.

The disclosure theories about McMullen and Wadman would likely have some merit if the petitioners had done more to articulate them, support them with case law, and explain their relationship to a determination of fair value. Presumably the petitioners’ believe that if stockholders had been told that McMullen was pursuing a sale in part because of his personal interest in exiting the Company and that Wadman resigned because of disputes over how McMullen handled the sale process, then some stockholders might have questioned whether the deal price reflected fair value.

These contentions would have to overcome the doctrine against self-flagellation.

See, e.g., Loudon v. Archer-Daniels-Midland Co., 700 A.2d 135, 143 (Del. 1997). That said, the proxy statement should have disclosed McMullen’s interest in retiring, his roles with GT Gold and New Chris, and their implications for his employment agreement.

99

Stockholders also should have been told that Wadman resigned because of disputes with senior management about the conduct of the sale process.

Although I have tried to give the petitioners the benefit of the doubt by crediting their conclusory assertions in this fashion, I am not convinced that their arguments are sufficient to undermine the stockholder vote as an expression of the preference of a supermajority of Stillwater’s stockholders for a sale rather than having the Company continue as a standalone entity. The Delaware Supreme Court has explained that “[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.” Dell, 177 A.3d at 33. The disclosures that the petitioners say the Company should have made could have affected stockholders’ views about whether their negotiators had extracted the highest possible bid. If stockholders had been provided with information about McMullen’s interests and Wadman’s withdrawal, then perhaps some stockholders would have inferred that a different negotiator might have pushed for more from Sibanye or worked harder during the pre-signing phase to find a bidder who could have paid a higher price (an inference undercut by the absence of any topping bid during the post- signing phase). They would not have had any reason to revise their assessment of the Company’s prospects as a standalone entity or to vote down the Merger in the belief that the Company was more valuable as a going concern in its operative reality as a widely held, publicly traded firm.

Because of the disclosure issues, this decision does not give heavy weight to the stockholder vote. Nevertheless, the vote remains a positive factor when evaluating whether

100

the deal price reflected fair value. If stockholders believed that the Company was worth more, they could have voted down the Merger and retained their proportionate share of the Company as a going concern. By approving the Merger at $18.00 per share, they evidenced their belief that the deal price provided fair value and was not exploitive.

5. The Sale Process Was Reliable.

Sibanye proved by a preponderance of the evidence that the sale process made the deal price a persuasive indicator of fair value. The sale process was not perfect, and the petitioners highlighted its flaws, but the facts of this case, when viewed as a whole, compare favorably or are on par with the facts in C & J Energy, PLX, DFC, Dell, and Aruba.

The sale process that led to the Merger bore objective indicia of fairness that rendered the deal price a reliable indicator of fair value. To reiterate, it was an arm’s-length transaction. It was approved by an unconflicted Board and by Stillwater’s stockholders. And it resulted from adversarial price negotiations between Stillwater and Sibanye. Most significantly, no bidders emerged during the post-signing phase, despite a Merger Agreement that contained a suite of deal protections that would pass muster under enhanced scrutiny.

The petitioners pointed to problems during the early phases of the sale process before the Board began exercising serious oversight and before BAML was retained. Those flaws are factors to consider, but they do not undermine the reliability of the sale price given what happened later. BAML’s pre-signing canvass was a positive factor. The negotiations with Sibanye were also a positive factor. And the process culminated in an

101

effective, albeit passive, post-signing market check. If Stillwater had pursued a single- bidder strategy and only engaged with Sibanye, then the terms of the Merger Agreement would have facilitated a sufficiently reliable post-signing market check to validate the deal price. Stillwater did more than what would have been sufficient under a single-bidder scenario.

It is theoretically possible that a more thorough pre-signing process or more vigorous negotiations might have generated a higher transaction price for Stillwater’s stockholders, but the issue in an appraisal “is not whether a negotiator has extracted the highest possible bid.” Dell, 177 A.3d at 33.

Capitalism is rough and ready, and the purpose of an appraisal is not to make sure that the petitioners get the highest conceivable value that might have been procedure 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.

DFC, 172 A.3d at 370–71. “[T]he key inquiry is whether the dissenters got fair value and were not exploited.” Dell, 177 A.3d at 33.

The Merger in this case was rough and ready. McMullen and the Board did not adhere to the best practices and transactional niceties that an advisor steeped in Delaware decisions would recommend. Nevertheless, given the arm’s-length nature of the Merger, the premium over market, and the substance of what took place during the sale process, it is not possible to say that an award at the deal price would result in the petitioners being exploited.

102

6. The Adjustment For Value Arising From The Merger Section 262 provides that “the Court shall determine the fair value of the shares exclusive of any element of value arising from the accomplishment or expectation of the merger or consolidation . . . .” 8 Del. C. § 262(h). “[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.” DFC, 172 A.3d at 371. “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 . . . .” Olson v. ev3, Inc., 2011 WL 704409, at *10 (Del. Ch. Feb. 21, 2011). “[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.” M.P.M. Enters., 731 A.2d at 797. To derive 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 . . . .” Golden Telecom Trial, 993 A.2d at 507. This means the trial court “must exclude . . . the amount of any value that the selling company’s shareholders would receive because a buyer intends to operate the subject company, not as a stand-alone going concern, but as part of a larger enterprise, from which synergistic gains can be extracted.” Aruba, 210 A.3d at 133 (internal quotation marks omitted).

Sibanye’s valuation expert was Mark Zmijewski, an emeritus professor of finance at the University of Chicago and a consultant at Charles River Associates. Zmijewski opined that the evidence he reviewed did “not indicate that the Transaction resulted in quantifiable synergies.” JX 652 ¶ 66 [hereinafter Zmijewski Rep.]; see Zmijewski Tr. 1146.

103

Sibanye told its stockholders that the price did not reflect any synergies. JX 421 at ‘224. McMullen testified at trial that he did not believe there were any synergies arising from the Merger. McMullen Tr. 801. There is accordingly no reason to exclude any value from the deal price based on synergies.

In this proceeding, Sibanye argued that despite the absence of quantifiable cost synergies or revenue synergies, it willingly paid more than fair value for Stillwater, resulting in a portion of the consideration reflecting value “arising from the accomplishment or expectation of the merger or consolidation . . . .” 8 Del. C. § 262(h). In its opening brief, Sibanye argued that it paid a premium for two strategic reasons: (i) to facilitate entry into the United States and (ii) to expand its share of the PGM market. Sibanye also argued that it could pay a premium in the Merger because after the Merger, it could obtain a better rating on its debt. See also Zmijewski Tr. 1120–22; JX 397 at ‘452; JX 498 at 20; JX 486 at 1; Rosen Tr. 407–08. Each of these reasons identifies a valuable aspect of Stillwater based on its operative reality as a going concern. Stillwater was the only PGM producer located in the United States, and it generated significant cash flow. None of these features represented a source of value “arising from the accomplishment or expectation of the merger or consolidation.”

Sibanye failed to meet its burden of proof to establish a quantifiable amount that the court should deduct from the deal price. This decision does not make any downward adjustment to the deal price to compensate for combinatorial value.

104

7. The Adjustment For Changes In Value Between Signing And Closing

Under Section 262, the time for determining the value of a dissenter’s shares is the point just before the merger closes. See Appraisal Rights, supra, at A-33. The deal price provides a data point for the value of the company as of the date of signing, but the valuation date for an appraisal is the date of closing. Consequently, 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. Technicolor II, 684 A.2d at 298.

In a merger involving a widely held, publicly traded company, some gap between signing and closing will usually exist. The customary need to prepare and disseminate disclosure documents, then complete a first-step tender offer or obtain a stockholder vote will typically result in several months elapsing between signing and closing. See Robert T. Miller, The Economics of Deal Risk: Allocating Risk Through MAC Clauses in Business Combination Agreements, 50 Wm. & Mary L. Rev. 2007, 2018–19 (2009) (discussing timelines for various transaction structures). If regulatory approvals are required, the temporal gap can expand. Id. at 2020–23. During this period, the value of the company could rise or fall.

Despite the customary existence of a temporal gap between signing and closing, Delaware appraisal decisions have typically not made adjustments to the deal price to reflect a valuation change during the post-signing period. In Union Illinois, this court relied for the first time on a deal-price-less-synergies metric when determining the fair value of

105

a privately held bank (UFG). See Union Ill., 847 A.2d at 343. Six months elapsed between signing and closing, and the petitioners objected to using the deal price because of the temporal gap. The trial court described this argument as a “quibble” and as “not a forceful objection,” because “[t]he negotiation of merger terms always and necessarily precedes consummation.” Id. at 358. Turning to the facts of the case, the court found that the petitioners were not able “to cite any rational explanatory factor that indicates why an investor would perceive UFG’s future more optimistically on New Year’s Eve 2001 than they did on the preceding Fourth of July.” Id. UFG had experienced “a modest upward adjustment in its [net income margin] in the second half of 2001,” but the court saw no evidence that the increase was sustainable or would alleviate UFG’s problems complying with capital adequacy standards. Id. Although UFG had refinanced its debt, the loan came from the acquirer, and UFG was not in a position to either service that debt or refinance it completing the merger. Id. The court concluded that “[c]onsidered fairly, the record does not support the idea that UFG was more valuable at the end of 2001 than it was when the Merger Agreement was signed.” Id.

In PetSmart, this court awarded fair value based on the deal price in a case involving a publicly traded firm. See In re PetSmart, Inc., 2017 WL 2303599, at *2 (Del. Ch. May 26, 2017). The court regarded the petitioners’ argument that the merger price “was stale by the time of closing” as “at best speculative.” Id. at *31. Citing Union Illinois, the court explained that “[m]ergers are consummated after the consideration is set. That temporal separation, however, does not in and of itself suggest that the merger consideration does

106

not accurately reflect the company’s going concern value as of the closing date.” Id. The court then turned to the petitioners’ case-specific arguments:

Petitioners would have me conclude that the Merger Price was stale because, in the gap between signing and closing, PetSmart’s fortunes took a miraculous turn for the better. While the record indicates that the Company did enjoy some favorable results in Q4 2014, such as an uptick in comparable store sales growth, I am not convinced that these short-term improvements were indicative of a long-term trend. In fact, all testimony at trial was to the contrary—the Board, as well as Teffner, believed that the Q4 results were temporary and provided no basis to alter their view of the Company’s long-

term prospects. These perceptions were born out in Q1 2015 (when the Merger closed) during which PetSmart’s comparable store sales dropped to 1.7%. At year end, PetSmart reported comparable store sales growth of 0.9%, a 40% miss from the Management Projections in just the first projection year.

Id. (footnotes omitted). The petitioners in PetSmart thus failed to carry their burden of proving that the value of the company had changed.

Most recently, in Columbia, this court awarded fair value based on the deal price in another case involving a publicly traded firm. See In re Appraisal of Columbia Pipeline Gp., Inc., 2019 WL 3778370, at *1 (Del. Ch. Aug. 12, 2019). The company developed, owned, and operated natural gas pipelines, storage facilities, and other midstream assets, and it had a business plan that called for raising large amounts of equity financing through a master limited partnership (“MLP”). Before agreeing to be acquired, the company had been unable to use the MLP structure to raise capital because of adverse trends in the MLP financing market. The merger agreement was signed on March 17, 2016, and the transaction closed on July 1, 2016. The petitioners argued that in the interim, the market for MLP equity had improved and prices for energy commodities had increased. See id. at *45. The court found that the petitioners had not carried their burden of proving how to

107

quantify the alleged improvements in the form of a higher deal price. Id. The court also found that the improvement in two MLP indices did not persuasively support the claim that the company would have been able to raise capital efficiently through its MLP. The court similarly rejected any valuation increase based on the prices of energy related commodities, because everyone agreed that the company’s value did not depend on commodities. As a midstream company, it did not own, buy, or sell the commodities that it transported or stored. Id.

The one arguable exception is Lender Processing, where this court awarded fair value based on the value of the deal price at closing, rather than at signing, where the deal consideration consisted of 50% cash and 50% stock. See Lender Processing, 2016 WL 7324170, at *1, *8. Because of the stock component, the value of the merger consideration increased from $33.25 per share at signing to $37.14 per share at closing. Id. The petitioners pointed to the existence of the temporal gap as a reason not to rely on the deal price or other market-based metrics associated with the signing of the deal. The respondent pointed to the absence of a topping bid as validating the deal price. After reviewing the evidence, the court concluded that the final merger consideration “was a reliable indicator of fair value as of the closing” and that “because of synergies and a post-signing decline in the Company’s performance, the fair value of the Company as of the closing date did not exceed” that amount. Id. at *23. The acquirer’s expert had not tried to quantify the synergies or the amount of the post-signing valuation decline, and the court concluded that the respondent had failed to carry its burden of proof on those issues. Id. at *33. By using the deal price as measured at closing rather than at signing, the Lender Processing decision

108

accounted for changes in value between signing and closing, but without making an explicit adjustment.

All four precedents considered whether the deal-price metric needed to be adjusted to reflect changes in value between signing and closing. The decisions thus indicate that an adjustment to the deal price can be warranted. But the decisions also show that the proponent of the adjustment must carry its burden by identifying a persuasive reason for the change and proving the amount.

At a minimum, it would seem to make sense to adjust the deal price for inflation.

When the parties agreed to the deal price on December 8, 2016, they reached agreement on a price measured in dollars valued as of that date. Between that date and the closing on May 4, 2017, the purchasing power of those dollars declined. If Stillwater had precisely the same value in the abstract on May 4, 2017, as it did on December 8, 2016, it would still be necessary to adjust the number of dollars used to express that value to reflect the intervening decline in what the value of a dollar represented. Adjusting the deal price for inflation would achieve this result.20

20 The pop-culture illustration of this principle is J. Wellington Wimpy’s offer to “gladly pay you Tuesday for a hamburger today.” See J. Wellington Wimpy, Wikipedia, https://en.wikipedia.org/wiki/J._Wellington_Wimpy (last visited Aug. 20, 2019). Setting aside credit risk, dollars paid next Tuesday are worth less than dollars paid today, so the same price paid next Tuesday is a pleasant deal for Wimpy. The same is true for Sibanye in an appraisal. Valuing Stillwater at $18.00 per share based on an agreement reached on December 8, 2016, then using that figure to determine value as of May 4, 2017, lets Sibanye use December’s dollars for a valuation in May. The statutory interest award is measured from closing, so that aspect of the appraisal remedy does not pick up the decline in the purchasing power of dollars used to measure the deal-price metric. In this respect, the petitioners are differently situated than stockholders who did not pursue their appraisal

109

As their valuation expert, the petitioners relied on Howard Rosen, a senior managing director at FTI Consulting. When adjusting the unaffected trading price, Rosen used an inflation rate of 2% per annum to account for the decrease in the value of dollars between signing and closing, then made further adjustments. See JX 728 ¶¶ 5.19, 5.25 to 5.28. A similar inflation-based adjustment could be made to the deal price, generating a value on the closing date of $18.14 per share, but no one argued for it.

The nature of Stillwater’s business makes this case a plausible one for an upward adjustment that goes beyond inflation. Stillwater was a mining concern that primarily produced palladium and platinum. Stillwater’s cash flows depended on the prices of those metals, so when the prices of those metals increased or decreased materially, the value of the Company increased or decreased materially as well. The Company’s annual report for 2016 explained the relationship as follows:

The Company’s earnings and cash flows are sensitive to changes in PGM prices – based on 2016 revenue and costs, a 1% (or approximately $7 per ounce) change in the Company’s average combined realized price for palladium and platinum would result in approximately a $7.1 million change to before-tax net income and a change to cash flows from operations of approximately $3.9 million.

rights. They accepted the $18.00 per share and received it, without interest, shortly after May 4, 2017, once the merger consideration payouts were processed through the clearing system. The appraisal petitioners did not accept that outcome. They opted for appraisal and sought a determination of Stillwater’s fair value as of May 4, 2017. Sibanye can argue legitimately that the deal price of $18.00 per share provides the best evidence of fair value, but that is a price calculated in December 2016 dollars, not May 2017 dollars.

110

JX 728 ¶ 5.21 (quoting Stillwater Mining Company, Annual Report (Form 10-K) (Feb. 16, 2017)). The Merger was signed on December 9, 2016. The Merger closed on May 4, 2017. Between signing and closing, the prices of palladium and platinum increased materially, with a direct effect on Stillwater’s value. Id. ¶ 5.20.

Rosen determined that the sales-weighted price of Stillwater’s commodities increased by 5.9% between signing and closing. Using the formula in Stillwater’s annual report, Rosen calculated the valuation impact of the additional cash flow as ranging from $248 million (using a 11.2% WACC) to $285 million (using a 10% WACC), which equated to an increase of between $2.00 to $2.30 per share. Id. He regarded his estimate as conservative because he kept production constant and did not account for new sources, such as Blitz, coming on line. Id. ¶¶ 5.23 to 5.25. Rosen used this figure to make adjustments to the unaffected trading price. In theory, he could have made similar adjustments to the merger price.

As this discussion indicates, the petitioners never argued for an adjustment to the deal price based on an increase in value between signing and closing. As discussed in the next section, Sibanye argued that the court could make an adjustment to the unaffected trading price and use the adjusted trading price as an indicator of fair value. The petitioners countered that argument by proposing an adjustment of their own that resulted in the adjusted trading price exceeding the deal price. Those arguments addressed the trading price, not the deal price. There could be considerable conceptual overlap between the approaches, but there could also be significant differences.

111

A petitioner seeking to make valuation-based adjustments to a reliable deal price also would need to confront the implications of the post-signing market check. As in Lender Processing, a respondent in an appraisal case could easily argue that if a company’s value increased between signing and closing, then a competing bidder would have perceived that value and offered more than the deal price. The respondent would argue that if no one bid, then that fact would call for rejecting the petitioners’ evidence of a valuation increase. There are several possible responses to this argument.

One response is a relatively small point from a valuation perspective: the termination fee. Using this case as an example, if a topping bidder made a Superior Proposal, and if the Board changed its recommendation, and if the stockholders voted down the Merger, then Stillwater would have to pay Sibanye a termination fee of $16.5 million plus reimbursement of Sibanye’s expenses up to $10 million, for a total payment of $26.5 million or 21.6 cents per share. Those amounts would reduce Stillwater’s value to the acquirer, making the acquirer neutral as to any increase in Stillwater’s value that did not clear that level. The point of indifference is actually higher, because a competing bidder would incur expenses of its own to make the competing bid. Ignoring those incremental expenses and focusing only on the sell-side fees, Stillwater’s value could increase by up to $26.5 million without a rational acquirer having any reason to bid. The absence of a topping bid could not rule out a valuation change of this magnitude, but an award above the deal price that fell within the range permitted by the termination fee would likely be cold comfort to the typical appraisal petitioner.

112

A more significant counterargument would focus on the timing of the valuation change. A premise that underlies the effectiveness of the post-signing market check is that other bidders learn that the target is for sale when the deal is announced, can examine the target for themselves, and if they value the target more highly (taking into account synergies and other sources of bidder-specific value), then they can intervene. Under this theoretical framework, competing bidders can begin work shortly after the announcement, giving them the full timeline between the signing and the vote in which to intervene. When the potential overbid would be induced by a change in the value of the target company, the time for the competing bidder to act does not begin with the announcement of the deal, but rather when the bidder learns of the valuation change. The delayed signal shortens the amount of time for the bidder to intervene. As the date of the stockholder vote approaches, it becomes less likely (all else equal) that a bidder will intervene, if only because less time is available in which to do so. Because of this effect, a failure to bid during the post-signing phase provides a much noisier signal about changes in the target’s value than it does about the absence of higher-valuing bidders. In this case, the increase in value that resulted from changes in the spot price did not really begin until February 2017, two months after signing. It dropped in March, then picked up again in April, when the stockholder vote took place.

A third counterargument would examine the possibility of changes in value after the stockholder vote but before closing. As this case illustrates, a competing bidder’s only meaningful opportunity to intervene is before the stockholders approve the transaction. In a case where closing is delayed significantly after the stockholder vote because of issues such as the need for regulatory approvals, the post-vote temporal gap would matter more.

113

Perhaps the most significant problem with relying on a post-signing market check to rule out an increase in the target’s standalone value is that the resulting valuation improvement would be available to any bidder. The competition for the incremental value would likely operate as a common value auction, defined as an auction in which “every bidder has the same value for the auctioned object.” Peter Cramton & Alan Schwartz, Using Auction Theory to Inform Takeover Regulation, 75 L. Econ. & Org. 27, 28–29 (1991). In a competition for that incremental value, the incumbent bidder’s matching right would loom large. To make it worthwhile to bid, a potential deal jumper would not only have to perceive that the value of the target had increased above the level set by the deal price plus the termination fee and fee reimbursement plus the deal jumper’s likely transaction costs, but also perceive a pathway to success that was sufficiently realistic to warrant becoming involved, taking into account the potential reputational damage that could result from being unsuccessful. Unless the competitor had a unique reason to value the increased cash flows more highly than the incumbent, the competitor should expect the incumbent to match any incremental bid.21 In a case like this one, where the valuation

21 See Fernán Restrepo & Guhan Subramanian, The New Look of Deal Protection, 69 Stan. L. Rev. 1013, 1058–63 (2017) (analyzing implications of matching rights); Brian JM Quinn, Bulletproof: Mandatory Rules for Deal Protection, 32 J. Corp. L. 865, 870 (2007) (analyzing matching rights as the functional equivalent of a right of first refusal and explaining that “[t]he presence of rights of first refusal can be a strong deterrent against subsequent bids” because “[s]uccess under these circumstances may involve paying too much and suffering the ‘winner’s curse’”); see also Marcel Kahan & Rangarajan K. Sundaram, First-Purchase Rights: Rights of First Refusal and Rights of First Offer, 12 Am. L. & Econ. Rev. 331, 331 (2012) (finding “that a right of first refusal transfers value from other buyers to the right-holder, but may also force the seller to make suboptimal offers”); Frank Aquila & Melissa Sawyer, Diary of a Wary Market: 2010 in Review and What to

114

increment would result from improved commodity prices that would be available to all bidders, a strong argument can be made that a competitor would not think that it had the ability to outbid the incumbent and would not try.

The respondent in an appraisal proceeding could make similar arguments about the stockholder vote. If the reasons for the valuation increase were public, and stockholders still voted for the deal, then their behavior would provide contrary market evidence undermining the claim of increased value. In this case, the increase in commodity prices was publicly available information, and Stillwater’s stockholders had the ability to vote down the deal if they thought the increased value from improving commodity prices changed matters. One obvious response to this argument is that to vote down the deal, stockholders would have had to prefer returning to Stillwater in its operative reality as a widely traded firm, where their only the options for liquidity were either to sell into the market or hold out for a higher-priced takeover down the road. Given these choices, stockholders might well have preferred the surer option of the deal price, even if they believed that the Company’s value had increased between signing and closing such that the deal price no longer reflected fair value.

Expect in 2011, 12 M & A Law. Nov.-Dec. 2010, at 1 (“Match rights can result in the first bidder ‘nickel bidding’ to match an interloper’s offer, with repetitive rounds of incremental increases in the offer price. . . . [M]atch rights are just one more factor that may dissuade a potential competing bidder from stepping in the middle of an already-announced transaction.”); David I. Walker, Rethinking Rights of First Refusal, 5 Stan. J.L. Bus. & Fin. 1, 20–21 (1999) (discussing how a right of first refusal affects bidders).

115

As this discussion shows, whether to adjust the deal price for an increase in value between signing and closing presents numerous difficult questions. In this case, the petitioners did not argue for an adjustment to the deal price, and so the parties did not have the opportunity to address these interesting issues. The court will not take them up at this late stage in the proceeding. The petitioners accordingly failed to prove that the deal price should be adjusted upward to reflect a change in value between signing and closing. See Columbia, 2019 WL 3778370, at *45. This decision finds that the deal price of $18.00 per share provides reliable evidence of fair value. B. The Adjusted Trading Price Sibanye contended that Stillwater’s adjusted trading price is a reliable indicator of the fair value of the Company. Sibanye generates the adjusted trading price by making adjustments to the unaffected trading price, so the reliability of the adjusted trading price depends on the reliability of the unaffected trading price. As the proponent of using this valuation indicator, Sibanye bore the burden of establishing its reliability and persuasiveness.

Assessing the reliability of the trading price for Stillwater’s common stock means getting “deep into the weeds of economics and corporate finance.” In re Appraisal of Jarden Corp., 2019 WL 3244085, at *1 (Del. Ch. July 19, 2019). The thicket of market efficiency is one such place where “law-trained judges should not go without the guidance of experts trained in these disciplines.” Id. In this case, both sides retained financial experts who tried to lead the court through the undergrowth. Zmijewski addressed these issues for

116

Sibanye. Israel Shaked, a professor of economics and finance at Boston University, addressed these issues for petitioners.

1. Informational Efficiency and Fundamental-Value Efficiency The experts agreed on the difference between informational efficiency and fundamental-value efficiency. See Zmijewski Tr. 1087; JX 651 ¶¶ 13–27, 33–41 [hereinafter Shaked Rep.]. “[I]nformational Efficiency . . . is concerned with how rapidly security prices reflect or impound new information that arrives to the market.” Shaked Rep. ¶ 33 (quoting Alex Frino et al., Introduction to Corporate Finance 305 (5th ed. 2013)). There are three recognized types of informational efficiency:

 Weak: a company’s stock price reflects all historical price information.

 Semi-Strong: a company’s stock price reflects all publicly available information.

 Strong: a company’s stock price reflects both publicly available information and inside information.

No one claimed that the market for Stillwater’s common stock could be informationally efficient in the strong sense. Everyone focused on whether the market for Stillwater’s common stock was informationally efficient in the semi-strong sense. All of the references in this decision to informational efficiency as it relates to Stillwater’s common stock therefore contemplate informational efficiency in the semi-strong sense.

“While informational efficiency is a function of speed and how quickly new material information is incorporated into a stock’s price, fundamental value efficiency is an incremental function of how accurately a market in which a stock trades discretely incorporates new material information.” Shaked Rep. ¶ 42. The price of a security in a

117

market that is fundamental-value efficient should reflect its intrinsic value, defined as “the present value of all cash payments to the investor in the stock, including dividends as well as the proceeds from the ultimate sale of the stock, discounted at the appropriate risk adjusted rate.” Shaked Rep. ¶ 40. (internal quotation marks omitted). In other words, a stock trading in a market that is fundamental-value efficient is one in which the trading price “fully reflects all estimates, guidance and other public, material information that portray the risks and returns of a stock accurately, including all key drivers.” Id. ¶ 41.

The experts agreed that it is impossible to observe whether a stock trades in a market that is fundamental-value efficient. See Zmijewski Tr. 1088, 1153–54; Shaked Report ¶ 41. According to the petitioners, this concession means that Sibanye cannot meet its burden of proof.

While theoretically valid, the petitioners’ argument goes too far. Whether called fundamental value, true value, intrinsic value, or fair value, the really-real value of something is always an unobservable concept. No valuation methodology provides direct access to it. Fundamental value is like a Platonic form, and the various valuation methodologies only cutouts casting shadows on the wall of the cave. The real issue is not whether a particular method generates a shadow (they all do), but rather whether the shadow is more or less distinct than what other methods produce.

Reliance on the trading price of a widely held stock is generally accepted in the financial community, and the trading price or metrics derived from it are regularly used to estimate the value of a publicly held firm based on its operative reality in that configuration. For purposes of determining fair value in an appraisal proceeding, therefore, the trading

118

price has a lot going for it.22 Like democracy, the trading price may be imperfect, but it often will serve better than the other metrics that have been tried. Cf. Winston Churchill, Churchill by Himself 574 (Richard Langworth ed., 2008). The petitioners’ admittedly valid objection that it is impossible to prove that a trading price reflects fundamental value is thus not one that automatically disqualifies the use of the trading price as a valuation indicator in an appraisal.

In this regard, it is important to recognize that informational efficiency and fundamental-value efficiency are not all-or-nothing concepts. See Bradford Cornell & John Haut, How Efficient Is Sufficient: Applying the Concept of Market Efficiency in Litigation,

22 See, e.g., Richard A. Booth, Minority Discounts and Control Premiums in Appraisal Proceedings, 57 Bus. Law. 127, 151 n.130 (2001) (“[M]arket price should ordinarily equal going concern value if the market is efficient.”); 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 (2003) (“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.”); id. 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”); Lawrence A. Hamermesh & Michael L. Wachter, The Short and Puzzling Life of the “Implicit Minority Discount” in Delaware Appraisal Law, 156 U. Pa. L. Rev. 1, 52 (2007) (“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 . . . .”); see also Lawrence A. Hamermesh & Michael L. Wachter, Rationalizing Appraisal Standards in Compulsory Buyouts, 50 B.C. L. Rev. 1021, 1033–34 (2009) (positing trading prices should not be used to determine fair value if there is either no public market price at all, if the shares are illiquid or thinly traded, or if there is a controlling stockholder, implying that outside of these scenarios, “because financial markets are efficient, one can simply use the market value of the shares”).

119

74 Bus. Law. 417, 418 (2019). A stock trading in a national market like the New York Stock Exchange will have more attributes of informational efficiency than a stock trading over the counter, but a party might be able to show that the particular over-the-counter market had sufficient attributes to regard the trading price as informationally efficient. The attributes of the over-the-counter market are likely to be consistent with a greater degree of informational efficiency than thinner and chunkier markets, such as markets for houses or entire companies.

Fundamental-value efficiency is likewise a matter of degree. A market could be precisely fundamental-value efficient in that it accurately prices the asset at exactly its true value. Or it might be nearly fundamental-value efficient in that it accurately prices the asset within some percentage, say plus or minus 3%, of its true value. Or it might be approximately fundamental-value efficient in that it accurately prices the asset within some wider range of its true value, such as a factor of two. See id. at 422 (“We might define an efficient market as one in which price is within a factor of 2 of value, i.e., the price is more than half of value and less than twice value.” (quoting Fischer Black, Noise, 41 J. Fin. 553 (1986))).

Although it is impossible to test for fundamental value, there are indicators of fundamental-value efficiency. One indicator is directional consistency, in which the market for a security reacts positively to new material information that is positive, and negatively to new material information that is negative. See Shaked Rep. ¶¶ 43–44. Another indicator is proportionality, which examines not only whether the direction of the reaction to new material information is consistent with its content, but also whether the extent of the

120

reaction corresponds with the informational content. See id. ¶ 45. In simplified terms, if a company announces a positive earnings surprise and its stock price increases, then that outcome is directionally consistent. If the stock price increases by an amount generally proportionate to the present value of the earnings surprise, then that outcome is proportionally consistent. A market that evidences directionality and proportionality is more likely to be fundamental-value efficient. A market that lacks evidence of directionality and proportionality is less likely to be fundamental-value efficient. See id. ¶ 46.

The question in this case is thus not whether the market for Stillwater’s common stock was or was not informationally efficient. Nor is it whether the market for Stillwater’s common stock was or was not fundamental-value efficient. The question is whether the market for Stillwater’s common stock was informationally efficient enough, and fundamental-value efficient enough, to warrant considering the trading price as a valuation indicator when determining fair value. Put differently, the operative question in this case is whether Sibanye proved that Stillwater’s common stock traded in a market having attributes that made the trading price a sufficiently reliable valuation indicator to be taken into account when determining fair value, either in conjunction with other metrics, or even as the sole metric, with the answer turning on both the attributes of the market for Stillwater’s common stock, and also on the relative reliability of the trading price compared to other metrics like the deal price and the outputs of DCF models. See, e.g., Jarden, 2019 WL 3244085, at *4, *27–31 (determining fair value based on the unaffected trading price after concluding that it was comparatively the most reliable valuation indicator); Cornell

121

& Haut, supra, at 425 (“What is important in legal applications is not some abstract notion of market efficiency. Rather, what is important is whether the market is sufficiently efficient in any particular situation.”).

2. Evidence Of Market Efficiency The experts disagreed about the extent to which the market for Stillwater’s shares was efficient. The experts discussed factors that courts have considered as indicative of informational efficiency. The experts also conducted event studies and opined on their implications for informational efficiency, directionality, and proportionality.

a. The Cammer And Krogman Factors Zmijewski examined whether the market for Stillwater’s shares exhibited attributes that courts have associated with informational efficiency. He relied on an instruction from Sibanye’s counsel that “Delaware Courts cite as attributes of market efficiency characteristics such as market capitalization, public float, weekly trading volume, bid-ask spread, analyst following, and market reaction to breaking news and information.” Zmijewski Rep. ¶ 49. He also analyzed the existence of market makers, eligibility to file SEC Form S-3, institutional ownership, and autocorrelation of stock returns, noting that these additional factors were considered in Cammer v. Bloom, 711 F. Supp. 1264 (D.N.J. 1989), and in Krogman v. Sterritt, 202 F.R.D. 467 (N.D. Tex. 2001). Zmijewski Rep. ¶ 51. For simplicity, and following the parties’ lead, this decision refers to these attributes as the “Cammer and Krogman factors,” even though not all of them were considered in those two cases.

122

Based on his review of the record, Zmijewski reached the following conclusions about these attributes:

 Market Capitalization: Zmijewski opined that “firms with a larger market capitalization tend to have larger institutional ownership,” “tend to be listed on the New York Stock Exchange,” and are therefore more likely to have shares that trade in markets that are informationally efficient. Zmijewski Rep. App. C ¶ 35 (citing Randall S. Thomas & James F. Cotter, Measuring Securities Market Efficiency in the Regulatory Setting, 63 L. & Contemp. Probs. 105, 115 (2000) (JX 896)). The Company’s market capitalization averaged approximately $1.3 billion, exceeding roughly 60% of the combined equities of companies listed on the New York Stock Exchange and NASDAQ. Id.

 Public Float: Zmijewski opined that having a large percentage of shares in the public float is indicative of a trading market that is informationally efficient. Id. ¶¶ 42–43 (noting that the Delaware Supreme Court in Dell cited a public float of 1.5 billion shares representing 84.29% of the outstanding stock, and in DFC cited a public float of 37.5 million shares representing 95% of the outstanding stock). The Company’s public float consisted of 106 million shares representing 87.4% of the outstanding stock. Id. ¶ 44.

 Weekly Trading Volume: Zmijewski opined that an average weekly trading volume of at least 2% warrants a “strong presumption” of informational efficiency.

Id. ¶ 2 (quoting Cammer, 711 F. Supp. at 1286). The average weekly turnover for Stillwater was 6.8%. Id. ¶ 3.

 Bid-Ask Spread: Zmijewski opined that a bid-ask spread of less than 2.5% is indicative of a trading market that is informationally efficient. Id. ¶¶ 37–38 (citing DFC, 172 A.3d at 352; Dell, 177 A.3d at 1, 5–6, 24–27, 41; In re Sci.-Atlanta, Inc.

Sec. Litig., 571 F. Supp. 2d 1315, 1340 (N.D. Ga. 2007); Cheney v. Cyberguard Corp., 213 F.R.D. 484, 501 (S.D. Fla. 2003); and Krogman, 202 F.R.D. at 478). The Company’s average daily bid-ask spread was 0.10%. Id. ¶ 39.

 Analyst Coverage: Zmijewski opined that the presence of at least five analysts following a company is indicative of a trading market that is informationally efficient. Id. ¶¶ 4–6 (relying on Thomas & Cotter, supra, at 115). Seven analysts followed the Company. Id. ¶ 7.

 Market Makers: Zmijewski opined that the presence of at least nineteen market makers is indicative of a trading market that is informationally efficient and that the same inference can be drawn when a company’s shares trade on a centralized auction market like the New York Stock Exchange. Id. ¶¶ 8–9 (citing Cammer, 711 F. Supp. at 1293; Cheney, 213 F.R.D. at 499–500; In re Dynex Capital, Inc. Sec.

Litig., 2011 WL 781215, at *5 (S.D.N.Y. Mar. 7, 2011); and Zvi Bodie et al.,

123

Investments 62–70 (12th ed. 2018)). The Company’s stock traded on the New York Stock Exchange and had eighty-two market makers. Id. ¶ 10.

 SEC Form S-3 Eligibility: Zmijewski opined that a company’s eligibility to register shares using SEC Form S-3 eligibility is indicative of a trading market that is informationally efficient. Id. ¶ 11 (citing Cammer, 711 F. Supp. at 1284). A company is eligible for Form S-3 if it, among other things, has been subject to the Securities Exchange Act of 1934 reporting requirements for more than one year, filed documents in a timely manner, and shown that it has not failed to pay certain obligations. Id. The Company filed Forms S-3 in 1996, 1998, 2001, 2009, and 2010.

Id. ¶ 12.

 Institutional Ownership: Zmijewski opined that having a significant percentage of stock owned by institutional investors is indicative of a trading market that is informationally efficient. Id. ¶ 46 (citing Thomas & Cotter, supra, at 106, 119). As of September 30, 2016, institutions held approximately 90% of the Company’s outstanding stock. Id. ¶ 47.

 Autocorrelation: Zmijewski opined that a lack of autocorrelation in a company’s stock return is indicative of a trading market that is informationally efficient. Id. ¶ 48. Autocorrelation measures the extent to which the next day’s stock price movement can be predicted based on the current day’s stock price. Zmijewski found no evidence of statistically significant autocorrelation during the 254 trading days preceding the announcement of the Merger. Id.

 Cause And Effect: Zmijewski opined that market reactions to significant events are indicative of informational efficiency. Id. ¶ 13 (citing Cammer, 711 F. Supp. at 1287). Zmijewski found that after the Merger announcement, there was a quick and significant increase in trading volume. Id. ¶ 17. The first news of the Merger was released at 1:04 a.m on December 9, 2016. Pre-market trading opened at 4:00 a.m.

The first trade occurred at 4:01 a.m. at $17.50. The Company’s stock closed that day at $17.32 per share, with 38 million shares having traded. The day before, the Company’s stock closed at $14.68 per share, and only 3.2 million shares were traded. Id. ¶¶ 15–16; see Zmijewski Tr. 1096.

Having considering the Cammer and Krogman factors, Zmijewski opined that “[t]he

evidence indicates that Stillwater’s common stock traded in a semi-strong efficient market.” Zmijewski Rep. ¶ 49.

In response, Shaked disputed whether the Cammer and Krogman factors established informational efficiency to a sufficiently reliable degree. He opined that “the Cammer and Krogman factors have not been academically tested and are not truly conclusive in judging

124

a market as semi-strong form efficient, but merely an indicator that a market is likely semi- strong form efficient.” Shaked Rep. ¶ 23. The petitioners did not cite any academic studies or provide other forms of evidence that would undermine the use of the Cammer and Krogman factors, at least as a starting point for assessing informational efficiency. Zmijewski did not engage on this issue. He analyzed the factors because he understood that courts considered them.

b. The Event Studies

The experts also conducted event studies. Zmijewski’s event study tested for a cause-and-effect relationship between new information and a trading price reaction, which would provide evidence of informational efficiency. He examined five events—the four quarterly earnings releases leading up to the announcement of the Merger plus the announcement itself. Zmijewski characterized the events as positive or negative, and examined the market evidence to determine if the observations resulted in statistically significant abnormal returns. Three of the five did, but one of those was the reaction to the announcement of the Merger. Shaked persuasively observed that finding a statistically significant relationship between the trading price and the announcement of the Merger was trivial. See Shaked Tr. 468–69.

For the remaining four observations, Zmijewski found that only two resulted in statistically significant abnormal returns, and he admitted that he would have expected the rate of statistically significant results in an informationally efficient market to be higher. Zmijewski Tr. 1101. The events themselves do not suggest any reason why the market would have reacted in one instance and not the other. For example, for both the fourth

125

quarter of 2015 and the third quarter of 2016, Stillwater announced higher earnings per share, yet only the former resulted in a statistically significant abnormal return.

Shaked conducted three event studies, and he analyzed the results not only for evidence of a cause-and-effect relationship consistent with informational efficiency, but also for evidence of directionality and proportionality that would provide indications of fundamental value efficiency. In his first study, Shaked looked at eleven quarterly earnings releases during the three-year period leading up to the announcement of the Merger and characterized their informational content as positive or negative. He then examined whether the announcement resulted in abnormal returns consistent with the direction of the news. Shaked observed that only six of the eleven releases resulted in a directionally consistent reaction; five of the eleven did not.

In his second study, Shaked examined articles, analyst reports and SEC filings during the same three-year period, yielding a total of 181 events that he believed contained material new information. News of the 181 events was published on a total of fifty-six days, resulting in fifty-six observations. Although there are reasons to question some of Shaked’s events, on the whole, his identification appears credible. Of these fifty-six observations, only twelve resulted in statistically significant abnormal returns that were consistent with the directional content of the information. Moreover, there were thirty-eight days in the study period when there was a statistically significant abnormal return but no material news announcement.

In his third study, Shaked tested for proportionality by examining the reaction of the Company’s stock to the announcement of a significant increase in the expansion of its

126

mining operations in its earnings announcement for the third quarter of 2016. In the prior quarterly earnings releases, the Company forecast that the expansion would produce between 150,000 and 200,000 PGM ounces per year. JX 134 at 13; JX 187 at 17. In the earnings announcement for the third quarter of 2016, the Company increased the projection to between 270,000 and 330,000 PGM ounces per year. JX 309 at 16; see JX 306. Shaked estimated the pre-tax net income that would result from the increased output, taking into account the additional costs. He then prepared a discounted-cash-flow model that assumed production would ramp up by 25,000 ounces per year until 2022, continue at 125,000 ounces per year until 2031, then stop with no terminal value. Based on this model, Shaked calculated a net present value of $111.6 million for the increased production, which should have equated to a 7.08% abnormal return. Although the stock reacted positively, the observed abnormal return was only 0.39%. Shaked concluded that the Company’s stock did not react in a proportionate manner, further undermining the claim of informational efficiency.

c. The Assessment Of Market Efficiency Absent any countervailing evidence, Zmijewski’s analysis of the Cammer and Krogman factors would support a finding that the trading market for Stillwater’s common stock had sufficient attributes to be regarded as informationally efficient. Shaked pointed out that the Cammer and Krogman factors have not been shown to provide a reliable

127

indication of informational efficiency, but given the weight of authority on this issue, an absence of evidence on this point is no longer enough.23 The event studies, however, cut in the opposite direction. Courts applying the Cammer and Krogman factors have generally given greater weight to event studies compared to the other factors.24 Based on his studies, Shaked opined that Stillwater’s stock

23 The experts’ exploration of the Cammer and Krogman factors has left me with significantly less confidence in them than I had before this litigation. There appears to be substantial overlap among the factors, such that a single attribute, like a New York Stock Exchange listing, would correlate with and lead to the satisfaction of multiple factors. For an issuer to satisfy multiple Cammer and Krogman factors is thus likely less significant than it might seem. It is also striking how many of the Cammer and Krogman, at least based on Zmijewski’s report, stem from judicial opinions or law review articles, rather than from financial or economic papers. I am left with the concern that the Cammer and Krogman factors may be a convenient heuristic that law-trained judges deploy as a matter of routine, rather than because they have support in reliable research. That said, the absence of evidence is not necessarily evidence of absence, and the record in this case does not provide grounds to call the Cammer and Krogman factors into doubt.

24 See, e.g., In re DVI, Inc. Sec. Litig., 639 F.3d 623, 634 (3d Cir. 2011) (“[B]ecause an efficient market is one in which information important to reasonable investors . . . is immediately incorporated into stock prices, the cause-and-effect relationship between a company’s material disclosures and the security price is normally the most important factor in an efficiency analysis.” (internal quotation marks omitted)), abrogated on other grounds by Amgen Inc. v. Conn. Ret. Plans & Tr. Funds, 568 U.S. 455 (2013); In re Xcelera.com Sec. Litig., 430 F.3d 503, 512 (1st Cir. 2005) (describing the cause and effect prong of Cammer as “in many ways the most important” and explaining that “[i]n the absence of such a relationship, there is little assurance that information is being absorbed into the market and reflected in its price”); Cammer, 711 F. Supp. at 1287 (“[S]howing a cause and effect relationship between unexpected corporate events or financial releases and an immediate response in the stock price” is “the essence of an efficient market and the foundation for the fraud on the market theory.”); see also Teamsters Local 445 Freight Div. Pension, Fund v. Bombardier Inc., 546 F.3d 196, 207 (2d Cir. 2008) (quoting Xcelera.com for the import of the cause and effect prong of Cammer). That said, the cause- and-effect factor is not dispositive. Beaver Cty. Empls.’ Ret. Fund v. Tile Shop Hldgs., Inc., 2016 WL 4098741, at *10–11 (D. Minn. July 28, 2016) (collecting cases and explaining

128

did not trade in a manner consistent with informational efficiency, and Zmijewski’s event study generated relatively unconvincing results. Given this evidence, it is difficult to conclude that Stillwater’s stock was informationally efficient to a degree sufficient to use the trading price as an indicator of fair value when a superior market-based metric, like the deal price, is available. That does not mean that Stillwater’s stock was not informationally efficient, only that the deal price is a superior market-based metric for purposes of determining fair value.

Shaked’s event studies also raised questions about the degree of directionality and proportionality exhibited by the market for Stillwater’s common stock. This evidence does not mean that Stillwater’s stock price was unreliable, but it does make it difficult to conclude that Stillwater’s stock was fundamental-value efficient to a degree sufficient to use the trading price as an indicator of fair value when a superior market-based metric like the deal price is available.

3. Evidence Of Information Gaps The petitioners advance two other challenges to the reliability of Stillwater’s trading price. Because everyone agrees that the market for Stillwater’s common stock could only be informationally efficient in the semi-strong sense, the trading price could only account for publicly available information. The petitioners argue that material information about Stillwater’s inferred reserves was not publicly available, meaning that the trading price

that “[t]he weight of authority on this issue favors” a finding of market efficiency without a favorable resolution of the cause and effect factor).

129

could not be a reliable indicator of fundamental value. They also cite evidence indicating that the parties themselves did not trust the market’s estimation of the Company’s value. The former point is another strike against the trading price; the latter is not.

a. Industry Guide 7

The petitioners argue that Stillwater’s trading price is not a reliable indicator of value because the market did not have access to material information related to the Company’s value. On this issue, the petitioners relied on another expert: Thomas Matthews, a Principal Resource Geologist at Gustavson Associates. Matthews discussed the constraints imposed by Industry Guide 7, which specifies what the United States Securities and Exchange Commission permits a mining company to disclose. See JX 843 [hereinafter Industry Guide 7]; 17 C.F.R. 229.801(g).

To oversimplify a significantly more complex area, Industry Guide 7 only permits a mining company to disclose information about proven reserves or probable reserves. A proven reserve is a mineral deposit where (i) “quantity is computed from dimensions revealed in outcrops, trenches, workings or drill holes,” (ii) “grade or quality are computed from the results of detailed sampling,” and (iii) “the sites for inspection, sampling and measurement are spaced so closely and the geologic character is so well defined that size, shape, depth and mineral content of reserves are well-established.” Industry Guide 7 ¶ (a)(2). A probable reserve is a mineral deposit where “the sites for inspection, sampling, and measurement are farther apart or are otherwise less adequately spaced,” resulting in a “degree of assurance” that is “lower than that for proven reserves” but still “high enough to assume continuity between points of observation.” Id. ¶ (a)(3). Industry Guide 7 does

130

not permit a mining company to disclose information about inferred resources, which are mineral deposits where the quantity, grade, and quality “can be estimated” based on “geological evidence,” “limited sampling,” and “reasonably assumed, but not verified, geological and grade continuity.” JX 7 at 4; see Industry Guide 7 ¶ (b)(5), Instruction 3.

Since at least 2012, the Society for Mining, Metallurgy and Exploration, Inc. has criticized this aspect of Industry Guide 7, complaining that the restrictions on reporting “limits the completeness and relevance of SEC reports for investors.” JX 15 at 1. The Society contrasted Industry Guide 7 with the standards applied in other countries, which permit this disclosure. Id. at 2. In 2016, the SEC acknowledged the issue and proposed revisions to Industry Guide 7, but the new rules did not go into effect until 2018, long after the Merger closed. See Modernization of Property Disclosures for Mining Registrants, Exchange Act Release No. 34-84509, 2018 WL 5668900 (Oct. 31, 2018).

Under Industry Guide 7 as it existed during the period leading up to the Merger, Stillwater could disclose information about the Stillwater Mine and East Bolder Mine, but could not disclose information about the inferred resources at Blitz, Lower East Boulder, Iron Creek, Altar, and Marathon. See JX 727 ¶ 13 (Matthews Reb. Rep.). For Blitz, the Company possessed but could not disclose “a resource estimation, a conceptual mine plan, material movement schedules, a capital and operating cost review, and a preliminary economic analysis for the Blitz expansion.” Id. ¶ 12. The Company could only disclose certain drill data and briefly describe production target ranges, estimated capital spend, and timeframes. See id. ¶ 14.

131

The parties disagreed about whether disclosure of this information would cause investors to place a higher or lower valuation on the Company, but they agreed that it created an information gap for purposes of trading in the Company’s stock. See id. ¶ 16; Zmijewski Tr. 1151. Zmijewski argued that because the effect of the information was unknowable, the court should assume that the absence of the information did not bias the trading price up or down. Sibanye also pointed out that some of the information was available in a filing that Stillwater made in March 2011 under the laws of Canada. See JX 9 at ‘055; cf. JX 501 at ‘345.

Stillwater’s inability to disclose information about inferred resources under Industry Guide 7, combined with its partial disclosure of some of this information in a Canadian filing from 2011, are negative factors for purposes of using the Company’s trading price as a valuation indicator. They are not dispositive in their own right, but they undermine the relative persuasiveness of the trading price.

b. Contemporaneous Evidence Of A Valuation Gap The petitioners also cite contemporaneous evidence in the record in which knowledgeable insiders affiliated with Stillwater, its advisors, or Sibanye regarded the trading price as an unreliable indicator of value. For example:

 In May 2015, Stillwater management told the Board that “[m]uch of the value from Blitz, Lower East Boulder and recycle ramp up yet to be recognized by the market and potential buyers.” JX 41 at ‘715.

 In January 2016, the Board thought that “the stock had been forced down significantly and . . . didn’t feel it really was reflective of what was going on in the business.” McMullen Dep. 145.

132

 In June 2016, Froneman described the markets as “a bit all over the place lately.”

JX 152 at ‘532.

 In their second and third quarter 2016 reports, BMO analysts thought the Company’s stock price did not reflect the value of Blitz. See JX 766 (stating in October 2016 that “[e]ven with arguably conservative assumptions, we maintain our opinion that the magnitude of the growth potential at Blitz is not factored into SWC shares”); Shaked Rep. ¶ 124 (quoting a June 2016 BMO report stating that “Blitz remains an underappreciated growth opportunity”).

 In October 2016, Vujcic told the Board that the market perceived PGMs as “exposed to irrational producer behaviour in both South Africa and Russia.” JX 293 at ‘522.

 During October, November, and December 2016, Stewart repeatedly stated that “[a]t an offer price of ~US$2bn (30% premium to 30 day VWAP) we are effectively paying a full price for the existing operations, 50% of Blitz and getting the remaining upside optionality for free.” JX 282 at ‘775; see JX 410; JX 378 at ‘009;

JX 447 at ‘981. He did not believe the market was “really considering Blitz.” JX 397 at ‘451; see JX 280 at ‘279 (describing the Company’s underperformance as “unlikely to remain as market recognises improvements are sustainable and Blitz comes on line”).

 In late November 2016, two weeks before signing, Stewart stated that the market was “currently at or near the bottom of the PGM cycle,” suggesting a depressed stock price. JX 410 at ‘099; see PTO ¶ 257; see also JX 280 at ‘279; JX 399 at ‘407.

 In early December 2016, days before signing, McMullen commented on how the price of palladium had been artificially depressed. See JX 437 at ‘471 (noting that palladium was “finally starting to reflect the fundamentals”)

 After announcing the Merger, Sibanye received two “deal of the year” awards and commented in both instances that the Merger was signed “at an opportune time in the commodity price cycle.” JX 511; JX 641 at 1.

 At trial, Schweitzer testified that “[t]he company’s stock price was all over the place from 2013 to 2016” and that he and “McMullen both believed there was a disconnect between the price of metals and the share price for Stillwater stock.” Schweitzer Tr.

173.

This evidence as a whole is less extensive and persuasive than what the record demonstrated about the contemporaneous views of knowledgeable insiders regarding the existence of a valuation gap in Dell, and the Delaware Supreme Court in that case found

133

that the trial court erred by giving weight to that evidence. See Dell, 177 A.3d 25–26; cf. Dell Trial, 2016 WL 3186538, at *33–36. This decision therefore does not give any weight to the petitioners’ weaker showing in this case.

4. The Comparative Reliability Of The Trading Price Through Zmijewski’s analysis of the Cammer and Krogman factors, Sibanye made an initial showing that would be sufficient to support the reliability of the trading price as a valuation indicator absent contrary evidence. The results of the experts’ event studies and the limitations imposed by Industry Guide 7 provided contrary evidence. Based on the parties’ showings, the trading price is a less persuasive and less reliable valuation indicator in this case than the deal price. The lack of a reliable trading price does not undermine a court’s ability to rely on the deal price, where the persuasiveness of the deal price has been established by analyzing the sufficiency of the sale process. See Columbia, 2019 WL 3778370, at *49.

This decision does not find that the trading price was so unreliable that it could not be used as a valuation indicator. If a market-tested indicator like the deal price was unavailable, then this decision might well have given weight to the trading price. Had this decision been forced to take that route, it would not have relied on the unaffected trading price, because Sibanye did not argue for its use, but instead would have taken into account the adjusted trading price.

Based on the record that the parties generated, Sibanye did not carry its burden to establish that the adjusted trading price was a sufficiently reliable valuation indicator for the court to use in determining fair value. The reliability of the adjusted trading price

134

depended on the reliability of the unaffected trading price, and the record provides sufficient reason for concern about incorporating a trading price metric. This decision therefore does not give any weight to the adjusted trading price. C. The Discounted Cash Flow Models The petitioners and Sibanye each introduced a DCF valuation prepared by an expert.

The petitioners relied on Rosen, whose DCF model generated a value of $25.91 per share. Sibanye relied on Zmijewski, whose DCF model generated a value of $17.03 per share. The difference amounts to approximately $1 billion in value.

The DCF method is a technique that is generally accepted in the financial community. “While the particular assumptions underlying its application may always be challenged in any particular case, the validity of [the DCF] technique qua valuation methodology is no longer open to question.” Pinson, 1989 WL 17438, at *8 n.11. 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.” Andaloro v. PFPC Worldwide, Inc., 2005 WL 2045640, at *9 (Del. Ch. Aug. 19, 2005).

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.

In re Radiology Assocs., Inc. Litig., 611 A.2d 485, 490 (Del. Ch. 1991) (internal quotation marks omitted).

135

In Dell and DFC, the Delaware Supreme Court cautioned against using the DCF methodology when market-based indicators are available. In Dell, the high court explained that “[a]lthough widely considered the best tool for valuing companies when there is no credible market information and no market check, DCF 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.” Dell, 177 A.3d at 37–38. The high court warned that 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.” Id. at 35. Making the same point conversely in DFC, the Delaware Supreme Court advised that a DCF model should be used in appraisal proceedings “when the respondent company was not public or was not sold in an open market check . . . .” DFC, 172 A.3d at 369 n.118. The high court commented that “a singular discounted cash flow model is often most helpful when there isn’t an observable market price.” Id. at 370.

This case illustrates the problems that the Delaware Supreme Court identified. The experts disagreed over many inputs, with small changes producing large swings in value. The briefing focused on eight inputs, with four generating the bulk of the difference.

First, the experts debated whether to apply a small-company risk premium, otherwise known as a size premium. Zmijewski applied a size premium of 1.66%, relying on Duff & Phelps, 2017 Valuation Handbook – U.S. Guide to Cost of Capital (2017). Rosen did not apply one, arguing that it was not warranted. To the extent the court disagreed, he argued for using a premium of 1.5% drawn from Ibbotson Associates, SBBI 2013 Valuation

136

Yearbook (2013). The scholarly literature on whether and how to apply a size premium is less than enlightening. The same respected scholars have found different results depending on the data set,25 and others have engaged in vigorous debate about how to interpret the data and what inferences to draw.26 This one dispute results in a valuation swing of $2.13 per share, accounting for approximately 24% of the difference between the two models.

25 Compare Eugene F. Fama & Kenneth R. French, A Five-Factor Asset Pricing Model, 116 J. Fin. Econ. 1 (2015) (JX 681) (finding evidence of size premium in asset pricing models), and Eugene F. Fama & Kenneth R. French, Common Risk Factors in the Returns on Stocks and Bonds, 33 J. Fin. Econ. 3 (1993) (JX 680) (finding evidence that stocks with smaller market capitalizations tended to have higher average returns), with Eugene F. Fama & Kenneth R. French, Size, Value, and Momentum in International Stock Returns, 105 J. Fin. Econ. 457 (2012) (JX 679) (finding no evidence of a size premium in any region based on analyses of international stock returns from November 1989 to March 2011).

26 Compare, e.g., Cliff Asness et al., Size Matters, If You Control Your Junk, 129 J.

Fin. Econ. 479, 479 (2018) (finding “[a] significant size premium . . . , which is stable through time, robust to the specification, more consistent across seasons and markets, not concentrated in microcaps, robust to non-price based measures of size, and not captured by an illiquidity premium” and arguing that challenges to the existence of the size premium “are dismantled when controlling for the quality, or the inverse ‘junk’, of a firm”), and Roger Grabowski, The Size Effect Continues To Be Relevant when Estimating the Cost of Capital, 37 Bus. Valuation Rev. 93 (2018) (responding to criticisms of Ang, infra), with Aswath Damodaran, The Small Cap Premium: Where is the Beef?, Musings on Markets (Apr. 11, 2015) (JX 682 at 1) (commenting that “the historical data, which has been used as the basis of the argument [for size premia], is yielding more ambiguous results and leading us to question the original judgment that there is a small cap premium” and that “forward-looking risk premiums, where we look at the market pricing of stocks to get a measure of what investors are demanding as expected returns, are yielding no premium for small cap stocks”), http://aswathdamodaran.blogspot.com/2015/04/the-small-cappremium -fact-fiction-and.html, and Clifford Ang, The Absence of a Size Effect Relevant to the Cost of Equity, 37 Bus. Valuation Rev. 87 (2018) (JX 732 at 3–4) (concluding from survey of empirical literature that either “(1) investors . . . do not believe a size effect exists and, therefore, do not demand compensation for it, or (2) investors . . . believe a size effect exists, but believe the adjustment for the size effect is not made in the cost of equity”).

137

Second, the experts debated the size of the equity risk premium. Zmijewski used a historic supply-side risk premium of 5.97% published by Duff & Phelps. See JX 837; JX 893. Duff & Phelps advised practitioners to deduct 1.08% from this measurement to account for “the WWII Interest Rate Bias.” JX 893 at 34. Zmijewski did not make the adjustment, explaining that it would not make sense to exclude the effect of interest rate controls during World War II, while failing to account for other periods of government control, such as the extreme phases of interest rate repression and quantitative easing that followed the 2008 financial crisis. Zmijewski Tr. 1042–43. Rosen used a forward-looking premium of 5.34%, derived from a model created by Aswath Damodaran. See JX 678. Zmijewski criticized the model, explaining that a user could generate approximately seventy different equity risk premiums by manipulating the inputs and objecting to some of Rosen’s selections. See JX 893 at 47; JX 894; Zmijewski Tr. 1053–54. This one dispute results in a valuation swing of $1.33 per share, accounting for approximately 15% of the difference between the two models.

Third, the experts disputed which set of commodity price forecasts to use to generate cash flows. Zmijewski relied on price forecasts prepared by another expert for Sibanye. JX 710 (Burrows Rep.). Rosen relied on price forecasts from Bloomberg. JX 654 ¶ 8.21 (Rosen Rep.). This one dispute results in a valuation swing of $0.82 per share, accounting for approximately 9% of the difference between the two models.

Zmijewski has acknowledged that “there is much weaker evidence of a size effect since the original [article finding the effect] was published.” JX 836 at 322.

138

Fourth, the experts diverged in their treatment of Stillwater’s exploration areas.

Sibanye argued that any valuation of these properties would be speculative and instructed Zmijewski not to try. Zmijewski Tr. 1074–75. Rosen estimated an “in-ground metal dollar value” for the properties, then relied on a report that examined PGM transactions in South Africa to estimate that exploration properties could be worth “between .5 percent and 2.5 percent of the estimated in situ dollar value of metal.” Rosen Tr. 277–78; see JX 765. The respondent’s mining expert identified many problems with Rosen’s method. See JX 768. The dispute over the exploration areas results in a valuation swing of more than $2.00, accounting for approximately 23% of the difference between the two models.

Four other disputes account for the remaining valuation swing of $3.00 per share.

Those disagreements concern how to account for the resources in mine-adjacent areas, the amount of excess cash, the value of inventory, and the value of Altar. As with the four major disputes, both sides have good reasons for their positions.

The legitimate debates over these inputs and the large swings in value they create undercut the reliability of the DCF model as a valuation indicator. If this were a case where a reliable market-based metric was not available, then the court might have to parse through the valuation inputs and hazard semi-informed guesses about which expert’s view was closer to the truth. In this case, there is a persuasive market-based metric: the deal price that resulted from a reliable sale process. Dell and DFC teach that a trial court should have greater confidence in market indicators and less confidence in divergent expert determinations. See Dell, 177 A.3d at 35–38; DFC, 172 A.3d at 368–70 & n.118. Compared to the deal-price metric, the DCF technique “is necessarily a second-best method

139

to derive value.” Union Illinois, 847 A.2d at 359. This decision therefore does not use it. See In re Appraisal of Solera Hldgs., Inc., 2018 WL 3625644, at *32 (Del. Ch. July 30, 2018).

III. CONCLUSION

The fair value of the Company’s common stock at the effective time of the Merger was $18.00 per share. The legal rate of interest, compounded quarterly, shall accrue on the appraised value from the effective date until the date of payment. The parties shall cooperate to prepare a form of final order. If there are additional issues that need to be resolved, then the parties shall submit a joint letter within fourteen days that identifies them and proposes a path to bring this matter to a conclusion at the trial level.

140

APPENDIX

Case Time Between Time from Total Time Termination Fee Other Deal Announcement Commencement for Protection of Deal and of Tender Offer Purposes of Measures Commencement to Closing Court of Tender Offer Decision Yanow v. Sci. 4 business days, 19 business 23 business Expense Window- Leasing, Inc., 1988 4 calendar days days, 28 days, 32 reimbursement shop, 16.6% WL 8772 (Del. Ch. calendar days calendar stock option Feb. 5, 1988) days lock-up In re Fort Howard 4 business days, 25 business 29 business $67.8 million; No-shop Corp. S’holders 4 calendar days days, 38 days, 42 1.9% of equity permitting Litig., 1988 WL calendar days calendar value target to 83147 (Del. Ch. days provide Aug. 8, 1988) information and

negotiate

(i.e., a

windowshop

).

In re KDI Corp. 4 business days, 24 business 28 business $8 million; 4.3% Window- S’holders Litig., 6 calendar days days, 35 days, 41 of equity value shop 1988 WL 116448 calendar days calendar (Del. Ch. Nov. 1, days 1988) In re Formica Corp. 3 business days, 30 business 33 business Graduated fee Strict no- S’holders Litig., 3 calendar days days, 43 days, 46 capped at 1.9% of shop 1989 WL 25812 calendar days calendar equity value (Del. Ch. Mar. 22, days 1989) Braunschweiger v. Single-step merger. No tender offer. 143 business 4.5% of equity None Am. Home Shield days, 205 calendar days, between announcement of value Corp., 1989 WL merger and stockholder vote approving deal. 128571 (Del. Ch. Oct. 26, 1989) Roberts v. Gen. 5 business days, 25 business 30 business $33 million; 2% of Window- Instr. Corp., 1990 7 calendar days days, 35 days, 42 equity value shop WL 118356 (Del. calendar days calendar Ch. Aug. 13, 1990) days McMillan v. Single-step merger. No tender offer. 102 business $3.1 million; 3.5% Window- Intercargo Corp., days, 148 calendar days between announcement of of equity value shop 768 A.2d 492 (Del. merger and stockholder vote approving deal. Ch. 2000) In re Pennaco 9 business days, 20 business 29 business $15 million; 3% of Window- Energy, Inc. 17 calendar days days, 28 days, 45 equity value shop S’holders Litig., 787 calendar days calendar A.2d 691 (Del. Ch. days 2001) In re Cysive, Inc. Single-step merger. No tender offer. 45 business Expenses up to Window- S’holders Litig., 836 days, 63 calendar days between announcement of $1.65 million; up shop with A.2d 531 (Del. Ch. merger and stockholder vote approving deal. to 1.7% of deal matching 2003) value rights

141

In re MONY Gp. Single-step merger. No tender offer. 82 business $50 million; 3.3% Window- Inc. S’holder Litig., days, 121 calendar days between announcement of of equity value; shop 852 A.2d 9 (Del. Ch. merger and stockholder vote approving deal. 2.4% of deal value 2004) In re Dollar Thrifty Single-step merger. No tender offer. 100 business $44.6 million with Window- S’holder Litig., 14 days, 144 calendar days between announcement of up to additional $5 shop with A.3d 573 (Del. Ch. merger and stockholder vote approving deal. million in matching 2010) expenses; 4.3% of rights deal value after

accounting for

options, RSUs and

performance units.

In re Smurfit–Stone Single-step merger. No tender offer. 89 business $120 million; Window- Container Corp. days, 123 calendar days between announcement of 3.4% of equity shop with S’holder Litig., 2011 merger and stockholder vote approving deal. value matching WL 2028076 (Del. rights Ch. May 20, 2011) In re El Paso Corp. Single-step merger. No tender offer. 51 business $650 million; Window- S’holder Litig., 41 days, 75 calendar days between announcement of 3.1% of equity shop with A.3d 432 (Del. Ch. merger and stockholder vote approving deal. value matching 2012) rights In re Plains Expl. & Single-step merger. No tender offer. 79 business $207 million; 3% Window- Prod. Co. S’holder days, 117 calendar days between announcement of of deal value shop with Litig., 2013 WL merger and stockholder vote approving deal. matching 1909124 (Del. Ch. rights May 9, 2013) C & J Energy Single-step merger. No tender offer. 130 business $65 million; Window- Servs., Inc. v. City of days, 189 calendar days between announcement of 2.27% of deal shop Miami Gen. Empls.’ merger and stockholder vote approving deal. value and Sanitation Empls.’ Ret. Tr., 107 A.3d 1049 (Del. 2014)