In re Appraisal of SWS Group, Inc.

Opinion

IN THE COURT OF CHANCERY OF THE STATE OF DELAWARE

IN RE APPRAISAL OF SWS GROUP, ) C.A. No. 10554-VCG INC. )

MEMORANDUM OPINION

Date Submitted: February 27, 2017 Date Decided: May 30, 2017

Marcus E. Montejo, Kevin H. Davenport, Eric J. Juray, Chaz L. Enerio, of PRICKETT, JONES & ELLIOTT, P.A., Wilmington, Delaware, Attorneys for Petitioners Merlin Partners, LP and AAMAF, LP.

Kurt M. Heyman, Patricia L. Enerio, Melissa N. Donimirski, of HEYMAN ENERIO GATTUSO & HIRZEL LLP, Wilmington, Delaware, Attorneys for Petitioners Lone Star Value Investors, LP and Lone Star Value Co-Invest II, LP.

Garrett B. Moritz, Eric D. Selden, Nicholas D. Mozal, of ROSS ARONSTAM & MORITZ LLP, Wilmington, Delaware; OF COUNSEL: William Savitt, Andrew J.H. Cheung, Noah B. Yavitz, of WACHTELL, LIPTON, ROSEN & KATZ, New York, New York, Attorneys for Respondents SWS Group, Inc. and Hilltop Securities Holdings LLC.

GLASSCOCK, Vice Chancellor

The Petitioners here are former stockholders of SWS Group Inc. (“SWS” or the “Company”), a Delaware corporation. They are seeking a statutory appraisal of their shares. The Company was exposed to the market in a sales process. As this Court has noted, most recently in In Re Appraisal of Petsmart, Inc.,1 a public sales process that develops market value is often the best evidence of statutory “fair value” as well. As noted below, however, the sale of SWS was undertaken in conditions that make the price thus derived unreliable as evidence of fair value, in my opinion. Methods of valuation derived from comparable companies are similarly unreliable here. I rely, therefore, on a discounted cash flow (“DCF”) analysis to determine the fair value of SWS, assisted by the learned but divergent opinions of the parties’ experts. My rationale for rejecting sale price, and my resolution of the disputed issues involved in the competing DCFs, follows.

This action arises from the Petitioners’ statutory right to receive a judicial determination of the fair value of their shares of SWS. On January 1, 2015, SWS merged into a wholly-owned subsidiary of Hilltop Holdings, Inc. (“Hilltop”), itself a substantial creditor of SWS. SWS shareholders received a mix of cash and stock worth $6.92. The Petitioners are a series of funds holding appraisal-eligible shares of SWS. The Petitioners bring this action challenging the merger consideration as

1 In Re Appraisal of Petsmart, Inc., 2017 WL 2303599 (Del. Ch. May 26, 2017).

2

unfair. It is my statutory duty to determine the fair value of the Petitioners’ shares as of the date of the merger.

This case presents two divergent narratives. The first is that the Company was on the brink of a turnaround before the sale, and had only been suffering due to unique and unprecedented market conditions. The second is that the Company had fundamental structural problems making it difficult to compete at its size. The reality is somewhere in the middle, in my view. The Company was a struggling bank which had a chance to modestly improve its outlook around the time of sale. It still faced a long climb, however.

Similarly, this case presents two divergent expert valuations. Neither party attempts to invoke the deal price, but for different reasons. The Petitioners argue that the sales process was so hopelessly flawed that the deal price is irrelevant. The Respondents argue that the deal price is improper here because it includes large synergies inappropriate to statutory fair value. Accordingly, neither party relies on price—though the Respondents argue any valuation should be reconciled or checked against the deal price. Each side instead relies on traditional valuation methods. Those traditional valuation methodologies result in almost mirror image valuations of 50% above and 50% below the deal price.

Upon review, I find the fair value of SWS as of the merger date to be $6.38 per share.

3

I. FACTS

The following are the facts as I find them after a four-day trial. I accord the evidence presented the weight and credibility I find it deserves. Because I do not find the merger price reliable on the unique facts here, I decline to focus extensively on the record as it relates to the sales process. In sum, as recited below, I find that Petitioners’ critiques of the sales process, and Hilltop’s influence on the process, are generally supported. However, Petitioners’ narrative that SWS was a company on the verge of a turnaround lacks credible factual support. Instead SWS consistently underperformed management projections and there is minimal record support that a turnaround was probable given its structural problems.

A. The Parties and Relevant Non-Parties There are several Petitioners in this action; each itself an entity. There is no dispute that the remaining Petitioners’ shares are eligible for appraisal. A collective 7,438,453 SWS common shares held by the Petitioners are at issue in this action.2 The share allocation of each remaining Petitioner is set out below:3

2 Pretrial Order and Stipulation at 5.

3 Id. at 5–6.

4

Entity Dissenting Shares Merlin Partners, LP 478,860 AAMAF, LP 429,803 Birchwald Partners, LP 1,425,423 Lone Star Value Investors, LP 1,400,000 Lone Star Value Co-Invest II, LP 2,850,000 Blueblade Capital Opportunities, LLC 696,578 Hay Harbor Capital Partners, LLC 157,789

SWS was a relatively small bank holding company. SWS entered a merger agreement with Hilltop on March 31, 2014 whereby SWS would merge into a subsidiary of Hilltop.4 That merger was consummated on January 1, 2015.5 Hilltop itself became a bank holding company following its acquisition of PlainsCapital in 2012.6 As discussed below, Hilltop, together with Oak Hill Capital Partners (“Oak Hill”), provided a substantial loan to SWS in 2011 that SWS needed to maintain proper capital and liquidity levels.7 Pursuant to the terms of the loan Hilltop’s Chairman, Gerald J. Ford (“Jerry Ford”), was appointed to SWS’s board in 2011 and remained a SWS director at all relevant times.8 Jerry Ford has approximately forty years of experience in the bank consolidation business,

4 Id. at 2. 5 Id. 6 See JX049 at 33. 7 See JX015. 8 See id. at 3; JX039 at 77.

5

including certain successful sales.9 Jerry Ford’s son, Jeremy Ford, is the President and co-CEO of Hilltop.10 In 2011 Jeremy Ford was named as Hilltop’s designated “observer” on SWS’s board, in connection with the loan, which permitted him to attend meetings, and review financial and operational reports “to oversee and protect Hilltop’s investment in SWS.”11 Oak Hill is a Texas based private equity firm which also participated in the 2011 loan to SWS.12 In connection with the loan, Oak Hill was also given a board seat and an “observer” on SWS’s board.13 B. The SWS Story 1. SWS’s Background

SWS was a Delaware corporation, incorporated in 1972, that traded on the New York Stock Exchange.14 SWS was a bank holding company with two general business segments: traditional banking (the “Bank”) and brokerage services (the “Broker-Dealer”).15 Under the brokerage services umbrella there were certain general sub-groups including retail brokerage, institutional brokerage, and

9 See JX015 at 3; JX039 at 77. 10 Trial Tr. 327:23–328:2 (Jeremy Ford). 11 Id. at 330:23–331:11 (Jeremy Ford); JX039 at 98. 12 See JX015 at 2. 13 JX008 at 2. 14 JX039 at 7. 15 See id.

6

clearing.16 The banking segment operated eight offices throughout the southwest.17 SWS had significantly more locations and resources dedicated to the brokerage business.18 In contrast to a traditional bank, SWS had minimal retail deposits— instead nearly 90% of SWS’s deposits were derived from overnight “sweep” accounts held by SWS’s Broker-Dealer clients.19 That is, SWS’s banking business lacked a “stand-alone deposit base.”20 On an employee, asset, and revenue basis the Bank was smaller than the Broker-Dealer.21 SWS’s CFO explained at trial that his view of the Company was that “really we were a broker-dealer with a bank attached.”22 2. SWS Faces Difficulty

SWS had a number of loans, backed by real estate in North Texas, that became impaired following the Great Recession.23 From 2007 to 2011 the Bank’s non- performing assets spiked from 2% of total assets to 6.6%.24 Federal regulators reacted to the impairment of the Bank’s assets. First, in July 2010 the Bank entered

16 See JX043 at 44. 17 See JX039 at 32. 18 See id. 19 Trial Tr. 116:18–117:12 (Miller). 20 Id. at 221:5–6 (Edge). 21 See JX038 at 19; JX759 at SWS_APP002395467. 22 See Trial Tr. 219:11–22 (Edge) (explaining that the banking line of business was acquired in 2000 and that the “roots” of the company were its broker-dealer operations). 23 See id. at 226:3–21 (Edge); JX011 at 12. 24 JX017 at 52.

7

into a Memorandum of Understanding (“MOU”) with federal regulators.25 The MOU subjected the Bank to additional regulation limiting certain business and requiring higher capital ratios.26 Second, the MOU was followed by a formal Cease and Desist order in February 2011, similarly restricting the Bank’s activities and setting out heightened capital requirements.27 In light of this additional oversight and the need to improve the Bank’s capital position, SWS began seeking ways to prop up the Bank. Initially, SWS attempted to transfer capital from the Broker-Dealer to the Bank which included a “fire sale” of assets, however, this failed to solve the capital issue.28 In fact the transfer from the Broker-Dealer to the Bank caused the Broker-Dealer business to drop below threshold capital levels acceptable to counterparties and threatened to impair the Broker-Dealer business line.29 SWS had preliminary discussions with Hilltop in the “early fall of 2010 and entered into a non-disclosure agreement with Hilltop,” which began due diligence review of SWS.30 SWS, however, upon advice of counsel and advisors elected to pursue a public debt offering.31 In December 2010, SWS attempted to raise capital through a public offering of convertible unsecured debt,

25 See Trial Tr. 197:14–20 (Chereck). 26 Id. at 226:16–227:6 (Edge). 27 See JX009. 28 Trial Tr. 227:7–21 (Edge). 29 See id. at 252:14–253:15 (Edge). 30 JX011 at 13. 31 Id.

8

which failed due to lack of investor demand.32 Thereafter, SWS returned to the private market and finalized an arrangement with Oak Hill and Hilltop (the “Credit Agreement”).

a. The Credit Agreement

The terms of the Credit Agreement were finalized in March 2011,33 and later approved by stockholders, before the transaction closed on July 29, 2011.34 Pursuant to the Credit Agreement, Oak Hill and Hilltop made a $100 million senior unsecured loan to SWS at an interest rate of 8%.35 The Credit Agreement provided that SWS would issue a warrant to purchase 8,695,652 shares of SWS common stock to both Oak Hill and Hilltop exercisable at $5.75 a share.36 As a frame of reference, when SWS pulled its public offering in December 2010, SWS’s trading price dropped to slightly below $4.00 a share.37 Absent exercise of the warrants, which would eliminate the debt, or a permissible prepayment the loan would mature in five

32 Trial Tr. 227:22–228:21 (Edge). 33 See JX011 at 7. 34 JX015 at 2. 35 See JX011 at 1. 36 JX011 at 7, 17. The warrants covered the value of each of Oak Hill and Hilltop’s respective loan principal of $50 million. Id. 37 See Trial Tr. 228:12–17 (Edge).

9

years.38 Upon exercise of the warrants, Oak Hill and Hilltop would own substantial positions in the Company.39 The same day the Credit Agreement was finalized, SWS entered into an Investor Rights Agreement with Oak Hill and Hilltop that provided each company the right to appoint a board member and a board “observer” to SWS’s board.40 The Credit Agreement itself provided several protections to Oak Hill and Hilltop. This included, for example, certain anti-takeover clauses which would place the loan in default if the board ceased to consist of a majority of “Continuing Directors” or if any other stockholder acquired more than 24.9% of SWS stock.41 Importantly, a separate portion of the Credit Agreement included a “covenant prohibiting SWS from undergoing a ‘Fundamental Change’” which was defined to include the sale of SWS (the “Merger Covenant”).42 Hilltop was not willing to waive the Merger Covenant during SWS’s sales process.43 However, SWS was permitted to prepay the loan under certain conditions44—including if the stock price of SWS exceeded

38 See JX011 at 22. 39 When Hilltop’s original ownership of approximately 4% of SWS was combined with its later exercise of warrants for 8,695,652 shares, it eventually owned 10,171,039 shares or approximately 21% of the company. See JX042 at ix. 40 JX008 at 2. Oak Hill appointed J. Taylor Crandall to the SWS board and selected Scott Kauffman as its board observer. Id. 41 See JX016 at 38. 42 See JX042 at xii–xiii. 43 See id. 44 See JX016 at 10, 14–15.

10

$8.625 for twenty out of any thirty consecutive trading days.45 That is, if the stock price reached such a point an acquirer could essentially prepay the loan, and the Merger Covenant would fall away.

Around the time the Credit Agreement was being negotiated and finalized, Sterne Agee Group, Inc. (“Sterne Agee”) approached SWS about a potential acquisition.46 On March 26, 2011, Sterne Agee made an unsolicited conditional offer to acquire SWS at $6.25 a share, which the board rejected after attempts to “obtain further information about the offer, including the source of funding and ability to obtain bank regulatory approval . . . .”47 In rejecting the $6.25 proposal, the board framed the offer as “highly conditional” and concluded that it “substantially undervalues the future potential of SWS Group . . . .”48 SWS implemented defensive measures in response to the offer.49 Stern Agee followed up with a $7.50 per share cash offer on April 28, 2011.50 SWS rejected that follow-up offer on May 3, 2011 in favor of the Credit Agreement with Hilltop and Oak Hill.51 In rejecting the offer SWS’s board “unanimously determined that the Sterne Agee

45 See id.; JX042 at xiii. 46 See JX800. 47 Id. at 2. 48 Id. at Ex. 99.2. A corresponding press release by SWS indicated the offer would be reviewed, and disclosed that previous transaction proposals by Sterne Agee “were not in the best interests of SWS . . . .” Id. at Ex. 99.1. 49 JX801 at 2–3. 50 See JX014. 51 See JX013.

11

proposal is speculative, illusory, subject to numerous contingencies and uncertainties, and is clearly not in the best interests of SWS Group Stockholders.”52 The board cited numerous regulatory and financial barriers that Sterne Agee would face that created serious questions as to “Sterne Agee’s ability to complete a transaction on a timely basis.”53 Notably, Sterne Agee was not a bank holding company and would need to secure unlikely regulatory approval to facilitate an acquisition of SWS’s Bank.54 The SWS board found that the $7.50 bid would “deprive[] stockholders of the long term value of their shares” pointing out that the offer was at a substantial discount to SWS’s book value.55 Testimony at trial clarified that Sterne Agee was an unlikely acquirer and never made an “actionable” offer.56 b. SWS after the Credit Agreement Following the Credit Agreement, and the regulatory interventions SWS implemented a plan to turn the business around. The success of the “turnaround” was the subject of substantial litigation effort.

From 2011 through 2014 SWS management prepared annual budgets. The budgets were formulated by going to individual business sector heads, collecting

52 Id. at 1–2. 53 Id. at 2. 54 Id. See also Trial Tr. 10:18–12:6 (Sterling). 55 JX013 at 2. 56 Trial Tr. 10:18–12:6 (Sterling).

12

their projections, and then aggregating them.57 Frequently, management would ask the business heads for more “aspirational” goals or projections to get numbers they were comfortable taking to the full board.58 Single year projections were then extrapolated out into three year “strategic plans” which assumed each individual year’s budget would be met.59 SWS, however, never met its budget between 2011 and 2014.60 In that vein, management forecasts anticipated straight-line growth in revenue and profits, but SWS failed to hit the targets and continued to lose money on declining revenues.61 Robert Chereck became CEO of SWS in 2012, after being recruited by Jerry Ford.62 Chereck helped to implement changes at the Bank which ultimately led to the termination of the Cease and Desist order in 2013, presumably because the Bank had reached adequate capital levels and returned to prudent lending.63 SWS was able to reduce its volume of problem loans,64 but the Bank, overall, produced “very disappointing results.”65 The Broker-Dealer business line essentially remained stagnant.66 SWS was accruing “a deferred tax asset” in the form of a net operating

57 Id. at 256:4–257:2 (Edge). 58 Id. 59 Id. at 257:8–258:7 (Edge). 60 Id. at 258:8–10 (Edge). 61 See, e.g., JX028 at F-3; JX020 at F-3. 62 Trial Tr. 196:7–23 (Chereck). 63 See id. at 210:7–211:10 (Chereck). 64 See, e.g., JX043 at 47; JX017 at 35. 65 Trial Tr. 106:23–107:11 (Miller). 66 See JX028 at F-46.

13

loss.67 In June 2013, the Company made an accounting decision to write down, in the form of a valuation allowance, approximately $30 million of its net operating losses, because after several years of losses in a row, the Company did not believe it would be able to generate “enough income in the future to use up that operating loss in the requisite time frame.”68 This decision was made in the context of an audited accounting determination. I find that the decision—to provide for a valuation allowance because it was more likely than not that such losses could not be offset by income during the requisite period69—implies that managements’ straight-line growth and profitability projections were optimistic.70 The Respondents identify two “structural impediments” to growth which they assert were demonstrated by the trial record.71 First, the Respondents point to trial testimony regarding SWS’s size. For example, Tyree Miller of SWS’s board, testified that SWS “was subscale in every area” and such lack of scale impeded growth.72 Both regulatory requirements,73 and technology and back office costs,74

67 See Trial Tr. 237:2–21 (Edge); JX028 at 24. 68 Trial Tr. 237:2–21 (Edge). 69 See id. at 237:2–238:2 (Edge). 70 See id. I note that the company appears to have kept the tax deferred assets on the books, but placed a $30 million valuation allowance against it. See id.; JX028 at 24. See also JX503 at SWS_APP00094172–73. 71 Respondents’ Post-Trial Opening Br. 11. 72 Trial Tr. 108:23–24 (Miller); id. at 106:23–107:11 (Miller). 73 See id. at 198:15–22 (Chereck). 74 See id. at 300:3–302:14 (Roth). For example, compliance, online banking, and cyber security costs were spread over a much smaller number of clients than at larger banks. See id.

14

burdened the Bank at its scale, as it had a smaller base to spread those costs across. Second, the Respondents point to testimony that SWS was a “people business,” and that its best assets were its people.75 This was particularly true of the Broker-Dealer business and SWS’s scale problems along with its publicized regulatory and capital problems made it difficult to retain client advisors. From 2009 to 2012 the Broker- Dealer lost approximately one third of its client advisors.76 The Bank business at SWS also struggled to retain and recruit loan officers in light of SWS’s well- publicized woes.77 The Petitioners narrative is that following termination of the Cease and Desist order and the changes implemented prompting the termination, SWS was on the brink of a turnaround. All parties agree that certain improvements were made to SWS’s problem assets78 and balance sheet. I find that the Company’s recent history and the record at trial supports the Respondents’ witnesses testimony that the Company would continue to face an uphill climb to compete at its size going forward.79 By August 2013, the board was becoming frustrated by the Company’s performance and directed SWS’s CEO to take action—specifically to cut costs by

75 Id. at 232:6–233:8 (Edge). 76 Id. at 232:20–233:15 (Edge). 77 Id. at 201:18–203:8 (Edge). 78 See, e.g., JX820. 79 See, e.g., Trial Tr. 36:6–7 (Sterling) (describing the company as “a melting ice cube for many years . . .”). See also id. at 244:5–10 (Edge).

15

10% within thirty days.80 The purpose of these cuts was not to stimulate growth, but rather to bring down the expense base in “an attempt to get margins up.”81 By the end of the year, nearly all of the cuts had been implemented. The savings expected were upwards of $18 million82—which included eliminating over 100 jobs, including thirty-two revenue-producing employees.83 Around this time federal bank regulators were conducting their annual review, which for the most part noted that SWS’s condition had improved, however, they raised a concern about SWS’s ability to repay the $100 million note.84 The board remained concerned about the Company’s condition and the ability of SWS to pay off its loan to Hilltop, and return to profitability and growth.85 3. The Sales Process

Prior to SWS launching a sales process there was noise by analysts in the market that SWS was an acquisition target,86 and that Hilltop, since it had recently become a bank holding company via its acquisition of PlainsCapital, was a likely fit for a synergies-driven transaction.87 SWS stock traded higher upon this speculation.

80 See id. at 244:14–245:1 (Edge). 81 Id. at 109:12–17 (Miller). 82 See JX089 at SWS_APP00002583. 83 See JX102 at SWS_APP00235079–82. 84 See Trial Tr. 248:4–15 (Edge). 85 See id. at 107:24–109:2 (Miller). 86 See, e.g., JX049 at 33. 87 Id. See Trial Tr. 429:16–432:10 (Eberwein). CEO of Petitioner Lone Star, Jeff Eberwein, invested on this thesis accumulating a position in SWS. See, e.g., id.; id. at 433:20–437:22 (Eberwein).

16

The analysts were correct—prior to SWS launching a sales process Hilltop was actively considering a purchase of SWS—and by October 2013 Jeremy Ford, Hilltop’s board observer, had drafted an analysis to present to Hilltop’s directors in support of an SWS acquisition.88 SWS was not aware of Hilltop’s interest at this time, however.89 In preparing his analysis Jeremy Ford had access to information via his position as a board observer that others in the market would not have had access to, including, for example, loan tapes,90 SWS board meeting materials,91 and access to SWS management. At no time did Jeremy Ford inform SWS of Hilltop’s interest, that it was analyzing SWS as a target, or that Hilltop was considering a tender offer.92 Hilltop’s internal projections reveal that following integration of PlainsCapital, an SWS acquisition would derive much of its benefits from cost- savings in reduction of overhead rather than SWS’s stand-alone performance.93 Thus, Hilltop’s acquisition thesis was synergies-driven.94 On January 9, 2014, Hilltop made an offer to acquire SWS for $7.00 per share, payable in 50% cash and 50% Hilltop stock.95 SWS’s trading price on January 9,

88 Id. at 365:14–366:2 (Jeremy Ford). 89 See id. 90 See, e.g., JX090; JX095. 91 See, e.g., JX089; JX091. 92 Trial Tr. 387:1–388:7 (Jeremy Ford). 93 See, e.g., JX906 at HTH00020915–17. 94 See id. at HTH00020921; Trial Tr. 340:4–341:11 (Jeremy Ford). See also JX002 at Ex. 15 (calculating Hilltop’s expected savings per share). 95 JX153.

17

2014—with some merger speculation in the market but prior to the announcement of the offer—was $6.06, and the one-year average of SWS in the previous year was $5.92. SWS responded by creating a Special Committee to consider the offer on January 15, 2014.96 The Special Committee “knew there were very, very strong synergy values already partly reflected . . .” in the initial offer but wanted to “convince Hilltop” to share more of the synergies with SWS shareholders.97 The process the Special Committee ran, and whether it was independent or “straightjacketed,” was also the subject of substantial litigation effort. As I do not rely on the deal price, I need only briefly address the matter here. The Committee was represented by legal and financial advisors.98 The financial advisor retained by the Special Committee asked management to update its most recent three year projections, which at the time ended in June 2016, to run through the end of calendar year 2017.99 While management dialed back some of the growth assumptions, due to the failure to meet prior-period projections,100 management projections were still “optimistic” and projected growth and “net additions to the business.”101 That is, the

96 See JX177. 97 Trial Tr. 114:15–115:10 (Miller). 98 JX187. I note, however, the Petitioners attack the selection of the financial advisor and suggest that the advisor was conflicted. See, e.g., Petitioners’ Post-Trial Opening Br. 9–10. 99 See Trial Tr. 15:6–16:14 (Sterling). 100 See id. at 258:11–259:9 (Edge).

101 Id. at 259:10–20 (Edge).

18

revised management projections still relied on a number of favorable assumptions.102 A visual representation of those projections are set out in Figure 1 below.

Figure 1103

Following Hilltop’s bid, the Special Committee’s financial advisor contacted seventeen potential merger partners for SWS in early February 2014.104 Besides Hilltop, two other entities expressed interest, as discussed below.

a. Esposito

Esposito is a small Dallas, Texas broker-dealer.105 Esposito had approximately $10 million in capital.106 Esposito made an expression of interest in

102 See id. at 20:6–15 (Sterling); id. at 205:9–16 (Chereck); id. at 116:3–13 (Miller).

103 This demonstrative is for ease of explanation and condenses a number of factual sources from the record. It can be found in the Respondents’ expert report. See JX002 at 8. 104 JX042 at 240.

105 See Trial Tr. 118:6–119:2 (Miller).

106 Id.

19

SWS at $8.00 per share on February 12, 2014, subject to a slew of conditions, including securing financing.107 Shortly thereafter Esposito released a press release publicizing its $8.00 expression of interest.108 Esposito was unknown to the entire Special Committee despite their decades of experience in the area.109 Nonetheless the Special Committee engaged with Esposito to try to obtain additional information regarding its plans to finance the transaction and secure regulatory approval.110 This revealed that Esposito would need the assistance of another small regional bank— Triumph Bancorp, who together with Esposito, would seek out $300 million from three private equity firms to finance the deal.111 Certain communications indicate that SWS “stiff-armed” Esposito.112 Stiff-armed, or otherwise, Esposito was not able to pull together the requisite financing and secure a path towards regulatory approval; thus, neither Esposito nor Triumph made a formal offer.113

107 JX222. I note this indication of interest appears to have been made, at least in part, at the suggestion of the CEO of one of the Petitioners here—Lone Star. See JX212; JX195. 108 JX236.

109 Trial Tr. 118:6–16 (Miller).

110 See JX261; Trial Tr. 118:10–119:24 (Miller).

111 See Trial Tr. 118:10–119:24 (Miller); JX292. As of March 15, 2014 Triumph’s CEO still had “no idea whether this deal makes sense at $8.00 per share (or any price for that matter).” JX335. 112 JX232. Specifically, the Special Committee’s financial advisor indicated on February 14, 2014 that he was going “to stiff arm [Esposito] shortly.” Id. Esposito also felt “stiff-armed.” See JX212 (indicating that Esposito received “a clear stiff arm” from the financial advisor). 113 See Trial Tr. 120:1–3 (Miller).

20

b. Stifel

In February 2014, Stifel emerged as a second interested acquirer. The parties heavily dispute whether Stifel was truly interested and capable of consummating a transaction with SWS. The Petitioners argue that Stifel was improperly shut out of the sales process despite having the means and the interest to submit a topping bid to Hilltop’s proposal. The Respondents’ narrative is that Stifel had a “reputation” and “history” of pursuing sales processes, backing out, and poaching key employees.114 Nonetheless the Special Committee instructed its financial advisor to solicit interest from Stifel,115 and Stifel expressed interest at $8.15 a share. The Respondents assert that Stifel was then “difficult” in carrying out due-diligence, arguing that Stifel insisted on “unusually personalized diligence.”116 SWS and Stifel engaged in robust negotiation over a non-disclosure agreement (“NDA”).117 The process of consummating a NDA was protracted; Stifel finally signed it on March 18, 2014.118 The Special Committee, apparently dragging its feet, did not countersign the NDA immediately,119 and by March 21, Stifel had withdrawn its signature.120

114 See, e.g., id. at 38:16–40:11 (Sterling).

115 Id. at 38:4–7 (Sterling).

116 Respondents’ Post-Trial Opening Br. 25.

117 See, e.g., Trial Tr. 121:18–122:16 (Miller).

118 JX355; Trial Tr. 70:8–10 (Sterling).

119 See e.g., JX368; Trial Tr. 74:15–79:16 (Sterling).

120 See JX380.

21

As discussed below, an initial handshake deal was reached between Hilltop and SWS on approximately March 20, 2014. Stifel, unaware of this, continued its expression of interest, at a price above Hilltop’s offer.121 This information was taken to the Special Committee at a March 24, 2014 meeting, which initially favored signing a NDA.122 However, when this information was relayed to Jerry Ford he “blew his top,” and demanded that the deal be signed with Hilltop by March 31, 2014 or he was withdrawing his offer and resigning from the board.123 Further Jerry Ford indicated that Hilltop would not waive the Merger Covenant.124 A NDA was eventually executed with Stifel, and by March 27, 2014 Stifel made a proposal at $8.65 a share.125 According to Stifel’s March 27 letter to SWS, the proposal was non-binding and subject to due diligence, and Stifel stated that it believed its proposal “would not be subject to blocking” by the Merger Covenant.126 Stifel proposed to finish diligence by March 31,127 and internal Stifel documents demonstrate that its price was driven significantly by synergies.128 Stifel’s access to SWS’s building and the diligence data room in the days leading up to the March 31

121 Trial Tr. 79:17–21 (Sterling).

122 JX388 at 2.

123 See Trial Tr. 80:11–81:4 (Sterling); JX388 at 2.

124 See id.

125 See Trial Tr. 81:9–82:5 (Sterling).

126 JX421.

127 See JX426.

128 See JX482 at STIFEL0000082 (indicating Stifel expected to save or cut costs by approximately 35%).

22

deadline is in dispute. The same is true for whether SWS and the committee were adequately cooperating with Stifel, and whether Stifel’s interest at its announced price-point was genuine. Shortly before the deadline the Special Committee asked Stifel if it would raise its offer to $9.00 per share.129 Stifel was not able to complete its diligence to its satisfaction and asked for an extension via letter of March 31, 2014.130 The extension request also suggested that the Merger Covenant now presented a problem for Stifel.131 No extension was granted.

c. Hilltop and the Committee Recommendation Hilltop’s initial $7.00 per share offer was rejected by the committee as “inadequate” and “undervalued” SWS per the Special Committee’s meeting minutes.132 As mentioned above, however, at trial members of the Special Committee testified to their belief that the initial offer significantly shared synergies, and that going forward the object of bargaining would be to extract additional synergy value for SWS shareholders.133 On March 19, 2014 Hilltop raised its offer to $7.50 a share with a ratio of 25% cash and 75% Hilltop stock.134 The Special Committee countered at $8.00.135 On March 20, while Stifel’s NDA was still

129 Trial Tr. 163:14–22 (Miller); JX486.

130 See JX509.

131 See id.

132 JX269.

133 See Trial Tr. 114:15–115:10 (Miller).

134 JX367.

135 Trial Tr. 343:1–344:17 (Jeremy Ford).

23

pending the Special Committee met and instructed the financial advisor to ask Hilltop to increase its offer to $7.75.136 Hilltop believed it had a “handshake” deal at $7.75.137 As discussed above Hilltop become upset at the prospect of another bidder entering the picture, which it viewed as a “retrade” or suspected negotiation tactic, and made clear that $7.75 was best and final.138 Thus, Hilltop set the March 31, 2014 deadline to accept or reject its offer.139 The Special Committee met on March 31, 2014 to consider Hilltop’s offer and review the sales process.140 The Committee’s financial advisor provided a fairness opinion which opined the proposed transaction was fair to SWS’s stockholders.141 The financial advisor did, however, recognize that the Company informed it that the Credit Agreement may place “significant constraints on the Company’s ability to sell itself . . . .”142 As of the self-imposed March 31 deadline Hilltop was the only acquirer that had made a firm offer.143 The Committee viewed the offer as “a very solid offer” that they knew could actually close and determined that accepting it was the appropriate course of action in light of the Company’s “precarious financial

136 See id. at 76:12–78:10 (Sterling).

137 Id. at 343:1–344:10 (Jeremy Ford).

138 Id. at 343:20–344:17 (Jeremy Ford).

139 See id.

140 JX516.

141 See id.; JX500.

142 JX530 at SANDLER00014168.

143 See Trial Tr. 140:21–141:21 (Miller).

24

position.”144 Further, in light of the financial advisor’s opinion that the offer was fair, the committee recommended it to the full board.145 The SWS board approved the merger later that day on the terms described above: $7.75 a share with 75% Hilltop stock and 25% cash.146 4. Post-Deal Developments Shortly after the deal was announced, certain Petitioners started accumulating shares for appraisal investment funds. The world of appraisal arbitrage does not lack for irony: Included in these Petitioners’ solicitations of investments was the disclosure that a prime investment risk to their business strategy of dissent from the merger was that a majority of stockholders would do the same. 147 In that case, the deal would not close and they would remain investors in SWS as a going concern.148 Prior to the record date for the merger, Oak Hill exercised the majority of its warrants on September 26, 2014, acquiring 6.5 million SWS shares thereby eliminating $37.5 million in debt.149 On October 2, 2014, Hilltop exercised its warrants in full and received approximately 8.7 million SWS shares, and as a result $50 million in SWS debt was eliminated.150 A proxy advisory service noted that SWS’s viability as a

144 See id.; JX516.

145 See Trial Tr. 140:21–141:21 (Miller); JX516.

146 See JX524.

147 See, e.g, JX600 at LSV0002117, LSV0002121.

148 See id. This same investment group also threatened a proxy contest in 2014 to replace certain directors. See JX616. 149 JX656.

150 JX670.

25

stand-alone entity was harmed by both market conditions and its poor performance over the past five years.151 However, this same proxy advisor, although it supported the merger, also indicated that the “merger consideration is clearly not the optimal outcome of the 2014 sales process, but it may, cumulatively, be an acceptable outcome when considering the entire 2011-2014 process.”152 SWS continued to struggle to turn a profit. Financial results for fiscal year 2014, released on September 26, 2014 revealed a decline in net revenue from $271 million to $266 million.153 While some sectors of SWS’s business improved, management forecasts were not met and SWS recorded a net loss of $15.6 million.154 The merger was approved by a special stockholder meeting on November 21, 2014 and closed on January 1, 2015. In the several months between the announcement of the merger agreement and the stockholder vote, no other bidder emerged. Due to fluctuation in Hilltop’s stock, the value of the merger consideration had decreased to $6.92 per share.

C. The Experts As is typical in these proceedings, the experts present vastly divergent valuations. In sum, neither expert attempts to invoke the deal price in light of the unique relationship between the buyer and seller and the sales process outlined

151 See JX705 at SWS_APP00193843.

152 Id. at SWS_APP00193835.

153 JX039 at 36.

154 JX759 at SWS_APP00239432–33.

26

above. The Petitioners’ expert, David Clarke, is a well-seasoned valuation expert with over thirty-five years of providing valuation opinions and expert testimony in various types of valuation litigation.155 Clarke employed a valuation which places 80% weight on his DCF analysis and 20% on a comparable companies analysis. Clarke arrives at a fair value of $9.61 per share, for a total value of SWS of $483.4 million.156 The Petitioners offer several purported explanations for the divergence from the deal price, including flaws in the sales process, and the failure to account for SWS being on the verge of a turnaround. The Respondents’ expert, Richard Ruback, a Corporate Finance Professor with substantial experience in expert testimony, places 100% weight on his DCF analysis.157 His analysis results in a $5.17 per share valuation. The Respondents’ explanation for its expert’s valuation falling below the merger price is that certain “shared synergies” are included in the merger price, but not properly considered fair value in an appraisal action. The experts’ positions are discussed in more detail in the analysis portion of this Memorandum Opinion.

155 See JX001 at Appendix B.

156 See id. at 53; JX004 at 34 (correcting initial per share valuation for the proper number of shares outstanding). 157 See JX002 at 27, 29, 38–40.

27

D. Procedural History Several separate appraisal petitions were initially filed in January 2015, and the petitions were later consolidated. A four-day trial was held in September 2016, followed by extensive post-trial briefing. After the conclusion of post-trial briefing, closing argument was held on December 14, 2016. At the conclusion of closing argument I requested that the parties submit essentially a stipulated list of issues arising from the evidence of value.158 That exercise proved helpful in highlighting the differences between the parties. However, it failed to result in a stipulated list of issues, and led to further motion practice.159 What follows is my decision on the fair value of SWS.

II. ANALYSIS

This is a statutory proceeding pursuant to 8 Del. C. § 262 (the “Appraisal Statute”). Once the procedural strictures are met and entitlement to appraisal is perfected, the Appraisal Statute provides shareholders who did not vote in favor of certain transactions a statutory right to have this Court value their shares.160 The only issue before me here is the value of the Petitioners’ shares.

The Appraisal Statute provides that “the Court shall determine the fair value of the shares exclusive of any element of value arising from the accomplishment or

158 See Dec. 14, 2016 Oral Argument Tr. 122, 124.

159 See Dkt. No. 222.

160 See 8 Del. C. § 262.

28

expectation of the merger . . . .”161 Unlike traditional adversarial legal proceedings, the burden of proof is not specifically allocated to a party—rather the Court, via statute, has the duty to determine the fair value of the shares.162 Therefore, “[u]ltimately, both parties bear the burden of establishing fair value by a preponderance of the evidence.”163 The corporation is to be valued as a going concern,164 taking “into account all relevant factors,”165 including the “‘operative reality’ of the company as of the time of the merger.”166 That is, the fair value calculation focuses on “the value of the company as a going concern, rather than its value to a third party as an acquisition.”167 Despite the burden of articulating fair value ultimately falling on the Court, I am, as a practical matter, generally guided in my valuation by the adversarial presentations of the parties. After evaluating those presentations and the trial record, the Court may “select one of the parties' valuation models as its general framework, or fashion its own, to determine fair value in [an] appraisal proceeding.”168 The

161 8 Del. C. § 262(h) (emphasis added).

162 See, e.g., Laidler v. Hesco Bastion Envtl., 2014 WL 1877536, at *6 (Del. Ch. May 12, 2014)

(explaining that “[i]n an appraisal proceeding, the burden to establish fair value by a preponderance of the evidence rests on both the petitioner and the respondent”) (citation omitted). 163 Merion Capital LP v. BMC Software, Inc., 2015 WL 6164771, at *11 (Del. Ch. Oct. 21, 2015).

164 See M.G. Bancorporation, Inc. v. Le Beau, 737 A.2d 513, 525 (Del. 1999) (citation omitted).

165 8 Del. C. § 262(h).

166 Le Beau, 737 A.2d at 525 (citation omitted).

167 In Re Appraisal of Petsmart, Inc., 2017 WL 2303599, at *27 (quoting M.P.M. Enters., Inc. v.

Gilbert, 731 A.2d 790, 795 (Del. 1999)). 168 Cede & Co. v. Technicolor, Inc., 684 A.2d 289, 299 (Del. 1996) (citation omitted).

29

Court has “significant discretion to use the valuation methods it deems appropriate . . . .”169 That is, “appraisal is, by design, a flexible process” and “vests the [Court] with significant discretion to consider ‘all relevant factors' and determine the going concern value of the underlying company.”170 A. The Appropriate Valuation Methodology Here A line of decisions in this Court have invoked the merger price as the best indication of fair value.171 Certain common threads run through these decisions making the merger price, in those circumstances, the best indicator available— including a sales process which exposed the company sufficiently to the market such that if the market valued the asset at a higher price, it is likely that a bidder would have emerged.172 Similarly, cases invoking the merger price generally involve a relatively clean sales process. However, when the merger price represents a transfer to the sellers of value arising solely from a merger, these additions to deal price are properly removed from the calculation of fair value.173

169 See In re Appraisal of DFC Glob. Corp., 2016 WL 3753123, at *5 (Del. Ch. July 8, 2016)

(citation omitted). 170 Golden Telecom, Inc. v. Glob. GT LP, 11 A.3d 214, 217–18 (Del. 2010).

171 See Merion Capital L.P. v. Lender Processing Servs., Inc , 2016 WL 7324170, at *30–31 (Del.

Ch. Dec. 16, 2016) (collecting recent cases relying on the deal price). 172 See, e.g., LongPath Capital, LLC v. Ramtron Int'l Corp., 2015 WL 4540443, at *24 (Del. Ch.

June 30, 2015) (relying on the deal price and concluding that “[t]his lengthy, publicized [sales] process was thorough and gives me confidence that, if Ramtron could have commanded a higher value, it would have”). See also In Re Appraisal of Petsmart, Inc., 2017 WL 2303599, at *2 (adopting the deal price where “the evidence does not reveal any confounding factors that would have caused the massive market failure, to the tune of $4.5 billion . . .”). 173 Ramtron Int'l Corp., 2015 WL 4540443, at *25–26 (relying on the deal price and excluding proven synergies arising from the specific transaction).

30

In this case, in light of the facts recounted in the background section of this Memorandum Opinion, certain structural limitations unique to SWS make the application of the merger price not the most reliable indicia of fair value. Neither party relied on deal price to demonstrate fair value. Here, because of the problematic process, including the probable effect on deal price of the existence of the Credit Agreement under which the acquirer exercised a partial veto power over competing offers, I find it inappropriate to rely on deal price and instead perform my statutory duty by employing traditional valuation methodologies.

The parties have presented two valuation methodologies: a comparable companies valuation by the Petitioners, and dueling DCF analyses by both the Petitioners and the Respondents. The selection of valuation methodologies is fact specific and necessarily dependent on the support in the trial record. A comparable companies analysis is appropriate only where the companies selected are truly comparable.174 The burden of establishing that companies used in the analysis are actually comparable rests upon the party seeking to employ the comparables method.175 The selected companies need not be a perfect match; however, to be useful the methodology must employ “a good sample of actual comparables.”176

174 See, e.g., Laidler, 2014 WL 1877536, at *8 (rejecting a comparable companies analysis where the proponent failed to demonstrate the companies were “truly comparable”). 175 See ONTI, Inc. v. Integra Bank, 751 A.2d 904, 916 (Del. Ch. 1999) (“The burden of proof on the question whether the comparables are truly comparable lies with the party making that assertion . . . .”). 176 In re Orchard Enterprises, Inc., 2012 WL 2923305, at *10 (Del. Ch. July 18, 2012).

31

Here the companies selected by Clarke in his comparable company analysis diverge in significant ways from SWS in terms of size, business lines, and performance. The record reflects that SWS, because of its unique structure, size and business model had few, if any, peers. Thus, finding comparables is difficult. Clarke compounded this challenge by selecting companies in both the Banking and Broker- Dealer lines of business that were dissimilar in size to SWS,177 some of which also had other characteristics making them not truly comparable.178 On the facts of this case, I do not find Clarke’s comparable-companies analysis sufficiently supported by the record to be reliable; thus, I employ the DCF methodology exclusively here.

B. The Court’s DCF Below I review the experts’ positions on contested inputs to the DCF valuation, and then decide the appropriate value of each input in light of the record established at trial and the law of this State. The DCF valuation, although complex in practice, is rooted around a simple principle: the value of the company at the time of the merger is simply the sum of its future cash flows discounted back to present value. The calculation, however, is only as reliable as the inputs relied upon and the

177 See JX005 Exs. 9.2, 10. 178 See, e.g., id. at 21–22 (explaining that a banking comparable used by Clarke, Green Bancorp, was a new public company pursuing a high growth rate via strategic acquisitions); id. at 20–21 (explaining how other comparables had undergone mergers during the relevant time).

32

assumptions underlying those inputs. Below, I select the inputs I find best supported by the factual record.

1. The Appropriate Cash Flow Projections This Court has long expressed its strong preference for management projections. Naturally, prior appraisal decisions have recognized that it is proper to be skeptical of “post hoc, litigation-driven forecasts” by experts.179 Similarly, the cash flow projections have been described by this Court as the “most important input” in performing a DCF, and that absent reliable projections “a DCF analysis is simply a guess.”180 Reliable management projections of cash flows in advance of the merger are favored over litigation-facing expert derived projections.181 As described earlier, management routinely prepared three-year projections which, in connection with the sales process, management extended at the request of the Company’s financial advisor to run through December 2017.182 All parties rely on these projections,183 with reservations. The Petitioners refer to the management projections as “Downside” projections because they had been adjusted downward

179 See Owen v. Cannon, 2015 WL 3819204, at *21–22 (Del. Ch. June 17, 2015).

180 Delaware Open MRI Radiology Assocs., P.A. v. Kessler, 898 A.2d 290, 332 (Del. Ch. 2006).

With reliable inputs, a DCF valuation may be considered an educated guess. 181 See id. See also Cede & Co. v. JRC Acquisition Corp., 2004 WL 286963, at *2 (Del. Ch. Feb.

10, 2004) (providing that this “Court prefers valuations based on management projections available as of the date of the merger and holds a healthy skepticism for post-merger adjustments to management projections or the creation of new projections entirely”). 182 See supra notes 99–103 and accompanying text.

183 See Petitioners’ Post-Trial Opening Br. 41 (stating “both experts relied on Management Projections”).

33

from previous projections.184 The Respondents characterize the projections as overly optimistic, as SWS’s actual performance “never came close to Management Projections.”185 The Respondents’ expert, Ruback, takes the management projections as they are, without adjustment.186 The Petitioners’ expert, Clarke, made several major adjustments to the management’s projections of cash flows. Clarke also chose to extend the projections by two years.187 As do the parties, I adopt the management projections as my starting point. I review each proposed alteration in light of the record.

a. The 2018 and 2019 Extension The first major alteration advanced by the Petitioners is Clarke’s extension of management projections for two additional years. The Petitioners frame this issue as whether SWS reached a “steady state” by the end of the management projections. They assert that a second-stage period of two years, covering calendar years 2018 and 2019, is necessary to “normalize SWS’s financial performance before calculating a terminal value.”188 The Petitioners’ primary contention is that as a matter of valuation methodology the Company had not reached a “steady state” by the end of management projections, thus it is necessary to extend the projection until

184 Id. at 39.

185 Respondents’ Post-Trial Answering Br. 23.

186 See Dkt. No. 221, Ex. at 1.

187 See Petitioners’ Post-Trial Opening Br. 41.

188 Dkt. No. 230 at 2 (Petitioners’ List of DCF Disputes).

34

they reached such a state before performing the terminal value calculation.189 The basis for the Petitioners’ conclusion that a steady state was not reached is that SWS’s profit margin at the end of management projections “was well below projected comparable company margins . . .” and that ROAA (return on average assets) was not in line with peers.190 There are a number of subsidiary assumptions necessary to allow the Petitioners’ premises to stand, and the extensions to be factually supported. Those include that the so-called peer firms are actually comparable,191 and that SWS, in light of its scale problems, could ever have performance similar to or greater than larger entities.192 Further, adopting Clarke’s specific projection extensions would require me to find that SWS would continue an additional two years of unprecedented straight-line growth, reaching a profit margin far exceeding any management projections, despite the Company’s structural issues and performance

189 Petitioners’ Post-Trial Opening Br. 41–42.

190 See id. at 42, 48–49.

191 Most of the “comparables” were significantly larger—and therefore less likely to face SWS’s persistent scale problems. See, e.g., JX005 at Ex. 10. 192 For example, the Petitioners argue ROAA needed to reach 1% before a steady state was reached.

See Petitioners’ Post-Trial Opening Br. 48–49. However, SWS had averaged a ROAA of 0.22% since the year 2000. See JX005 at Ex. 2. Further, the Broker-Dealer operation provided lower ROAA than the Bank, thus the Bank would have to significantly exceed 1% ROAA in order for SWS to have an overall ROAA of 1%. See Trial Tr. 223:6–224:12 (Edge); Trial Tr. 670:12–23 (Clarke).

35

problems.193 I note that the Respondents’ expert concluded that SWS had reached a steady state, and did so based on SWS’s ability to perform against similar firms. 194 I find the premises underlying the rationale for the extension unsupported,195 and that Clarke’s post hoc extensions to management’s projections are not proper here. On the eve of the merger SWS was continuing to lose money on declining revenues.196 Similarly, the record, on balance, supports a finding that at the end of three years the Company would reach a steady state.197 On the record before me, there is inadequate evidence to support the extension of straight-line unprecedented growth and I employ the three-year management projections as the starting point.198 Ruback’s DCF model uses management’s three year projections, as I have found supported here. Therefore, I begin with Ruback’s general model subject to

193 See, e.g., Trial Tr. 264:2–13 (Edge).

194 See JX005 at 5–6.

195 See Trial Tr. 264:2–13 (Edge) (testifying that management would not have signed off on Clarke’s extensions as the profit margin Clarke argued SWS needed to reach a steady state “would not be reasonable”). 196 See, e.g., JX036 at 36.

197 See JX005 at 6. See also Trial Tr. 261:8–12 (Edge) (“And then we thought it was appropriate to have obviously one full year of kind of steady state, stand-alone, didn’t have the noise of the transaction or anything. And that’s how we settled it going through the end of 2017.”). 198 See also Trial Tr. 708:1–710:5 (Ruback) (testifying that the appropriate measure for a steady state here is when SWS was “as good relative to [its] peers as [it] c[ould] be”). Ruback concluded that 2017 was a reasonable time at which to stop the projections as SWS’s turnaround would have slowed or been complete. See id. at 713:19–22 (Ruback). I find that conclusion reasonable on the facts here. See also id. at 15:18–16:4 (Sterling) (explaining that management thought extending projections beyond 2017 presented “too much uncertainty”).

36

the adjustments set out below.199 That is, management’s projections of net income for calendar years 2015 through 2017 of $37,075,000, $35,465,000 and $28,283,000, respectively serve as the starting point for my calculation.200 b. The 2014 warrant exercise and SWS’s Capital Level The next major adjustment advocated by the Petitioners intertwines two issues: should the warrant exercise be considered in valuing SWS, 201 and what, if any, excess regulatory capital SWS held should be distributed in the valuation model. That is, if the warrant exercise is considered part of the Company’s operative reality as of the merger date, in the Petitioners’ view the Company will have less debt and thus greater excess regulatory capital. The parties present me with binary and divergent positions. They differ as to whether the warrant exercise should be part of the operative reality of the company as of the merger date. Partly as a result, the Respondents and the Petitioners advocate that fair value should include $0 and $117.5 million, respectively, as the amount of excess regulatory capital distributable. I consider their positions, below.

199 That is, my calculations below are made using the described adjustments to Ruback’s model supplied to the Court. That framework is located in Ruback’s Expert Report. See JX002 at Exs. 6, 7. 200 See JX001 at 25; JX002 at 8.

201 Neither party disputes that the warrant exercise caused an increase in regulatory capital; the Respondents argue, however that this increase should not be considered here, because it arose from the merger, and it introduced no additional cash to SWS but instead simply canceled SWS debt. See Respondents’ Post-Trial Opening Br. 60 n.245.

37

i. The Warrant Exercise was Part of SWS’s Operative Reality

In an appraisal proceeding the Court is to exclude speculative elements of value that arise from the “accomplishment or expectation” of a merger.202 However, the “accomplishment or expectation” of the merger exception is “narrow” and is designed to eliminate speculative projections relating to the completion of a merger.203 Further, the “narrow exclusion does not encompass known elements of value, including those which exist on the date of the merger . . . .”204 Here, it is undisputed that the warrant exercises were known well in advance of the merger closing: in fact, the record indicates that the warrants were exercised to enable the holders “to vote for the merger.”205 The shares issued in the warrant exercise, totaling approximately 15,217,391, were all voted in favor of the merger. The Respondents argue the warrant exercise should be excluded and the changes it worked to SWS’s capital structure should not be considered.206 They essentially advance a “but for” test; but for the merger these warrants would not have been exercised when they were, and therefore they are an element of value arising solely

202 8 Del C. § 262(h); Weinberger v. UOP, Inc., 457 A.2d 701, 713 (Del. 1983).

203 See Cede & Co. v. Technicolor, Inc., 684 A.2d 289, 299 (Del. 1996).

204 Id. (emphasis added) (citation omitted).

205 See Respondents’ Post-Trial Opening Br. 66. Similarly, and unlike the facts in certain cases relied on by the Respondent, here the warrant exercise was not conditioned in any way on the merger: here those exercising the warrants simply made the independent decision to exercise in- the-money warrants before the record date to vote for the merger. 206 See Respondents’ Post-Trial Answering Br. 38.

38

out of the merger. Thus, they assert that I should use “the expected capital structure of the target company as a going concern.”207 The Petitioners point out that the warrants had, in fact, been exercised prior to the date of the merger; the exercise was not contingent or directly tethered to the merger itself, and the resulting shares were voted in favor of the merger. Logic, equity, and precedent, they argue, require the exercise of the warrants to be considered part of the operative reality of SWS.

The exclusion of changes in value resulting from the “accomplishment or expectation” of the merger is applied narrowly. It is applied properly where the change in the company is directly tied to merger.208 Here, two creditors made the economic decision to exercise warrants in advance of the merger, and prior to the record date, in order to vote those shares in favor of the merger. That is, this case is unlike certain other decisions of this Court which look to actions taken by the subject company, with an eye towards the merger, that changed the company’s balance sheet.209 Here, I note, the warrant shares are included in both parties’ calculations of the total number of shares outstanding over which to divide SWS’s total value in

207 Id. at 40.

208 See, e.g., Cede & Co. v. JRC Acq. Corp., 2004 WL 286963, at *7–8. (Del. Ch. Feb. 10, 2004)

(excluding debt incurred to finance a merger, and distinguishing a case that included transactions with some relation to a merger as part of the “operative reality” where those transactions were in place at the time of the merger). 209 See BMC Software, Inc., 2015 WL 6164771, at *13 (excluding excess cash the company conserved in contemplation of the merger); Gearreald v. Just Care, Inc., 2012 WL 1569818, at *8 (Del. Ch. Apr. 30, 2012) (employing the theoretical capital structure the company would have maintained as a going concern where the company paid off all of its debt only “as a condition of the Merger Agreement”).

39

the per-share value calculation.210 I find the operative reality as of the date of the merger was that the warrants were exercised three months prior to close, by third parties acting in their own self-interest, and that the exercise was part of the Company’s operative reality as of the merger date.

ii. Excess Regulatory Capital The Petitioners argue that “excess capital must be valued separately as a matter of law” and accounted for in a valuation.211 It is true that excess cash not being redeployed into the business must be added to the result of the DCF valuation.212 The Petitioners argue the same is true for excess regulatory capital in the context of a bank holding company.213 The Respondents counter that the Petitioners are improperly conflating regulatory capital with freely distributable cash, and improperly assuming that a massive distribution would have no effect on the company meeting management projections, which do not envisage any such bulk distributions.214 Here, the warrant exercise created some additional excess regulatory capital.

By regulatory capital I mean generally the ratio which federal regulators require

210 See JX002 at Ex. 8; JX004 at 34.

211 Petitioners’ Post-Trial Answering Br. 28 (relying, in part, on Gholl v. Emachines, Inc., 2004 WL 2847865, at *13 (Del. Ch. Nov. 24, 2004)). 212 See Gholl, 2004 WL 2847865, at *13 (observing that “in determining the fair value of a corporation, excess cash must be added to the result of the DCF valuation”). 213 See Petitioners’ Post-Trial Answering Br. 28 (arguing that “in the context of a bank holding company, Delaware law treats excess capital the same way” as excess cash). 214 See Respondents’ Post-Trial Answering Br. 31–37.

40

banks and bank holding companies to maintain between their capital and their assets.215 Capital in this context is roughly equivalent to stockholder’s equity.216 The exercise of the warrants did not directly put a single cent into the company— that money had already been received and deployed by the Company upon execution of the Credit Agreement in 2011. Rather, exercise of the warrants worked a capitalization change, cancelling $87.5 million in debt owed in exchange for issuing over 15 million shares in consideration for cancelling the debt. That change increased regulatory capital. It did not, necessarily, create excess capital in the sense of “excess cash” or marketable securities beyond what was needed to run the business to meet management projections.217 Clarke alters management projections by distributing to shareholders $87.5 million in year one of his projections (the year of the warrant conversion), and then $30 million more in year three.218 Clarke’s valuation model, which distributes over $117 million in three years, while assuming no impact on SWS’s ability to generate

215 See, e.g., JX005 at 13 (explaining that “[r]egulatory capital is a book-value-based measurement that is specified by government regulators” and that it “is not the same as excess cash readily available for distribution”). 216 See Trial Tr. 736:11–737:11 (Ruback) (explaining that what “excess capital means is that you have more equity than required by regulators”); id. at 408:20–409:13 (Jeremy Ford) (explaining that “excess capital is really the equity component, and it relates for these regulated businesses . . .”). 217 See, e.g., id. at 205:17–206:19 (Chereck) (testifying to the impracticability of a dividend in 2014, and that the Company “needed that capital to support the growth that we were projecting . . .”). 218 JX001 at Schedule 2-A.

41

cash flow, is hard to accept on its face: it assumes that SWS would distribute to shareholders over half of its pre-merger market capitalization of $198 million with no effect on the Company or its income. I also find Ruback’s approach, making no alterations to distribute excess regulatory capital in light of the structural changes resulting from exercise of the warrants, somewhat problematic. However, on the record here, I am persuaded that his approach is correct given the treatment of cash flows in the management projections. Importantly, management assumed a warrant exercise in 2016, but they do not project excess cash distributable as a result.219 I have no way to judge, on the record, how much capital, if any, would actually be distributable as of the merger date, January 1, 2015, without altering downward management’s projections of cash flow as a result.220 Clarke’s $87.5 million immediate distribution is linked to the warrant exercise.221 Management projections were made on an assumption of a warrant exercise in July 2016.222 Thus management’s projections included that transaction, yet declined to assume a bulk distribution in projecting the Company’s cash flows. The record does not reflect any

219 See Trial Tr. 261:20–262:11 (Edge); id. at 16:19–17:9 (Sterling). See also Respondents’ Post-

Trial Opening Br. 20 (arguing that management projections rested on the “favorable assumption” that “Oak Hill and Hilltop would exercise their warrants in July 2016”). 220 See, e.g., Tr. 252:14–253:15 (Edge) (testifying to the de-facto requirement in the Broker-Dealer business of having $100 million in excess capital for counterparties to transact business with SWS, and that counterparties would cut SWS off when they dropped below $100 million in excess capital). 221 JX001 at 30.

222 See supra note 219.

42

persuasive reason to second-guess management’s implied judgment. Further, I find it facially unreasonable to assume, as does Clarke, that such a distribution could be made without effect on the Company’s ability to generate cash flow consistent with the projections. In addition, the record makes me doubtful, in light of SWS’s recent emergence from major regulatory intervention, and its continuing business line in a highly regulated industry, that such a massive distribution would be possible from a regulatory prospective.223 It is true as a matter of valuation methodology that non-operating assets—

including cash in excess of that needed to fund the operations of the entity—are to be added to a DCF analysis.224 The Petitioners seem to conflate distributable cash or assets with a balance sheet increase in regulatory capital as the result of the conversion of debt to equity in the form of Hilltop and Oak Hill’s new shares. The Petitioners rely on In re PNB Holding Co. Shareholders Litigation225 for the proposition that excess regulatory capital must be accounted for in valuing a bank holding company. I note that PNB rejected a lump-sum distribution as proposed by Clarke’s valuation, however.226 Rather, the Court explained that there was “no basis

223 See, e.g., Trial Tr. 205:17–206:11 (Chereck) (testifying that in 2014 it would have been “very difficult” to get permission from federal regulators to dividend bank capital up to the holding company level). 224 See Gholl, 2004 WL 2847865, at *13 (explaining that non-operating assets should be added to the valuation and that “excess cash must be added to the result of the DCF valuation”). 225 In re PNB Holding Co. Shareholders Litig., 2006 WL 2403999 (Del. Ch. Aug. 18, 2006).

226 Id. at *26–28 (Del. Ch. Aug. 18, 2006) (“Despite its high Tier-1 Ratio as of the Merger date, though, there is no basis in equity to assume that [the bank] was required to premise the Merger

43

in equity” to add to the DCF calculation a one-time dividend of excess regulatory capital.227 For the reasons above, I defer to management projections, which assume a warrant exercise in July 2016. In light of the fact that the operative reality here is that the warrants were exercised earlier than implied in those projections, however, other adjustments are proper, as discussed directly below.

c. Interest Expense Adjustments Because the warrant exercise occurred earlier than management expected in its projections, I do find it appropriate to reduce the interest expense accordingly to reflect the Company’s operative reality. That is, management projections assumed a warrant exercise in July 2016, implying interest payable through that date. Interest expense for the gap between actual and projected exercise must be backed out accordingly.

price on a reduction of its starting Tier-1 Ratio.”). The other case relied upon by the Petitioners, in support of the major lump sum distribution advanced here, involved a discounted net income analysis of a small community bank where both experts agreed it was proper to distribute certain excess capital, and only disagreed as to the amount. See Petitioners’ Post-Trial Answering Br. 28 n.133 (citing Dunmire v. Farmers & Merchants Bancorp of W. Pa., Inc., 2016 WL 6651411, at *16 (Del. Ch. Nov. 10, 2016)). 227 In re PNB, 2006 WL 2403999, at *26–27. I note the PNB Court observed, in rejecting a large lump-sum distribution, that “it also is inappropriate to assume that PNB would retain cash simply to remain well above the well-capitalized threshold.” Id. The PNB Court handled the excess regulatory capital issue by distributing income in the future, and only retaining the amount required to remain at what the Court set as a reasonable capitalization level. Id. The evidence on which to perform a similar calculation here is lacking on this record.

44

The warrant exercise removed $87.5 million in debt which was owed at an 8% interest rate. This adjustment results in the removal of $7 million in interest expense for 2015, and $4.027 million for 2016.228 Given the assumed tax rate of 35%,229 this reduction in interest expense has the effect of increasing net income by $4.6 million in 2015 and $2.6 million in 2016.230 Accordingly, I add these to the management projections of net income in those two years.231 2. The Terminal Value Growth Rate Clarke employed a 3.00% terminal growth rate after performing his recommended adjustments to management projections. Ruback set his terminal growth rate slightly higher, at 3.35%, which he derived from the midpoint of the long term-expected inflation rate of 2.3% and the long-term expected economic growth rate of the economy at large of 4.4%.232 Ruback’s rate was set without the major adjustments to Company cash flows performed by Clarke. In his rebuttal

228 See JX001 at Schedule 2-D.

229 Id. See also JX005 at 16 n.51.

230 JX005 at 16 n.51. See also JX004 at 20 (indicating alterations to the interest expense result in an additional few cents per share). I note that there is some apparent confusion or disagreement as to the proper tax treatment of this reduction in interest expense. See JX004 at 9-C (adjusting net income by $7 million in 2015, but only $2.618 million in 2016, and including a $6.791 million tax expense in 2016). I find the approach I employed above the line reasonable here, and adopt it. 231 That is, I add 4.6 million and 2.6 million into cells A1 and B1, respectively, of Ruback’s model.

JX002 at Ex. 7. 232 Id. at 12.

45

report Clarke accepts Ruback’s growth rate as reasonable.233 On the facts here, I adopt 3.35% as the proper terminal growth rate.

3. The Proper Discount Rate Both parties and their experts rely on the Capital Asset Pricing Model (“CAPM”) to calculate the cost of equity. The basic CAPM formula employed here is the risk free rate, plus the product of beta times the equity risk premium, plus the size premium.234 The parties and their experts agree that the risk free rate of return is 2.47%, but disagree as to the three other inputs: the equity risk premium (“ERP”), equity beta, and size premium.

a. Equity Risk Premium

The skirmish over this input is whether historical ERP or supply-side ERP is the proper method for calculating ERP. The Respondents concede that recent decisions of this Court have adopted supply-side ERP, but observe that ERP must be decided on the facts of each case.235 Here, Ruback used an ERP of 7.0% which represents the applicable historical ERP. Clarke, in contrast used the supply-side ERP of 6.21%. While there was vigorous debate on this issue, I find that the supply- side ERP provided by Clarke is proper here.236 While it is true that a case-by-case

233 JX004 at 23.

234 That is: Risk Free Rate + (Beta * Equity Risk Premium) + Size Premium = Cost of Equity.

235 See Respondents’ Post-Trial Answering Br. 41.

236 See Glob. GT LP v. Golden Telecom, Inc., 993 A.2d 497, 517–18 (Del. Ch. 2010). See also In re Orchard Enterprises, Inc., 2012 WL 2923305, at *19 (citing Golden Telecom and finding that

46

determination of ERP remains appropriate, here there is no basis in the factual record to deviate from what this Court has recently recognized as essentially the default method in these actions.237 Therefore the proper ERP here is 6.21%.

b. Beta

The experts also disagree as to the appropriate beta. Clarke employs a beta of 1.10, whereas Ruback uses a beta of 1.18.

Ruback derived his beta from SWS’s performance rather than peer returns, which Clarke employed. The Respondents argue that the “peers” are not actually peers.238 Thus, the Respondents argue that a more targeted, company-specific beta, as employed by Ruback, is appropriate.239 Ruback used two years of SWS weekly stock returns ending on January 3, 2014, that is, data from the two years preceding the announcement of Hilltop’s initial offer.240 I cannot accept Ruback’s beta on this record. Ruback’s measurement period covered times where a “merger froth” and corresponding volatility were likely reflected in SWS’s trading and price.241 Conveniently for the Respondents, Ruback’s weekly two-year lookback period

the party advancing a historical risk-premium did “not provide[] me with a persuasive reason to revisit the supply-side versus historical equity risk premium debate”). 237 See id. See also Gearreald, 2012 WL 1569818, at *10.

238 Respondents’ Post-Trial Answering Br. 44.

239 Id.

240 JX002 at 16–17.

241 See, e.g., Respondents’ Post-Trial Answering Br. 60 (arguing that “Petitioners are therefore wrong to say that the merger froth in SWS’s stock is speculative, because the uncontroverted evidence demonstrates that the market anticipated a synergies-driven deal for SWS, and likely one involving Hilltop”).

47

reflects this; it yields a beta of 1.18, which is higher than the five-year monthly lookback of 0.81 and the five-year weekly lookback of 1.09.242 The Respondents argue that Clarke “supplied no explanation for his beta.”243 Clarke, however, used multiple data points:244 he surveyed possible betas and concluded a blended median was proper.245 Clarke’s beta was derived in part, however, with reference to companies that were not closely comparable.246 Clarke’s beta has drawbacks, then, including the extent of comparability to SWS of the entities from which he derived it. Nonetheless, under the facts here I find it best comports with the record. Therefore, I adopt Clarke’s beta of 1.10.

c. Size Premium

The experts agree that a size premium is appropriate here and that Duff & Phelps is the appropriate source to employ to estimate the size premium. However, they disagree as to which size premium should be used. Clarke uses a size premium of 2.69%, whereas Ruback uses a size premium of 4.22%.

242 JX001 at Schedule 3-B. See also In re Appraisal of DFC Glob. Corp., 2016 WL 3753123, at *10 (observing that “[a] five-year period is the most common for measuring beta and generally results in a more accurate measurement, although two-year periods are used in certain circumstances”). 243 Respondents’ Post-Trial Answering Br. 46.

244 See JX001 at 33.

245 Trial Tr. 541:21–543:22 (Clarke) (testifying to how he derived his beta and explaining that “I think it's appropriate, when looking at beta, to get as many measurements as you can, to try to triangulate something that is supportable both by the company itself and by peers”). See also JX001 at Schedule 3-B. 246 See JX001 at Schedule 3-B.

48

The divergence arises from the overall valuation of the company. Each expert took a different approach to derive the appropriate “decile” which thereby provides the size premium. Ruback selected the size premium based on the market capitalization of SWS prior to Hilltop’s offer, which was approximately $198.5 million.247 Clarke performed calculations to arrive at a preliminary valuation based on his DCF and other metrics, and used that value of $464 million to select the size premium for the decile in that range.248 Ruback’s approach places SWS in a decile that runs from approximately $190 million to $301 million, 249 whereas Clarke’s approach places SWS in a decile that runs from $301 million to $549 million.250 The Respondents point out that Clarke’s approach is “circular,” and that his approach is only “occasionally used” for computing size premiums for private companies where market capitalization is not easily derived or reliable.251 Recent cases in this Court, I note, are consistent with the criticism of Clarke’s approach in selecting a size premium in valuing this public company.252 The Petitioners counter

247 JX002 at 17; JX005 at 19. I note this market capitalization figure excludes the warrant exercise which I have found was part of the Company’s operative reality. 248 See JX001 at 34.

249 JX002 at 17.

250 JX001 at 34.

251 Respondents’ Post-Trial Answering Br. 47–48.

252 See, e.g., Merion Capital, L.P. v. 3M Cogent, Inc., 2013 WL 3793896, at *19 (Del. Ch. July 8, 2013) (observing that the “Court of Chancery consistently has used market capitalization as the benchmark for selecting the equity size premium”). See also In re Appraisal of DFC Glob. Corp., 2016 WL 3753123, at *14 (observing that “the size premium itself is calculated using market value, when available, as it is here”).

49

that while using market capitalization is generally appropriate for public companies, the “capital structure” here (including the large amount of outstanding warrants— 17,391,304—where the total shares outstanding were only 32,747,990) makes the market capitalization approach imperfect and inappropriate.253 They contend that SWS has enough in common with a private company for an iterative calculation to be appropriate.254 Both sides have presented some support for their respective size premiums that I find persuasive. SWS was a public company thus making it generally susceptible to Ruback’s market capitalization approach. However, it had a substantial amount of in-the-money warrants and significant influence by certain major creditors—making it in some ways more analogous to a private company. I find it appropriate in these circumstances to use the midpoint of these approaches, and I find the applicable size premium is 3.46%.

III. CONCLUSION

For the reasons stated above, and using the valuation inputs I have described, I find the “fair value” of the Petitioners shares of SWS as of the date of the merger was $6.38. The Petitioners are entitled to the fair value of their shares together with interest at the statutory rate. I note that the fact that my DCF analysis resulted in a value below the merger price is not surprising: the record suggests that this was a

253 Petitioners’ Post-Trial Answering Br. 50.

254 Id. at 50–51.

50

synergies-driven transaction whereby the acquirer shared value arising from the merger with SWS.

The parties should confer and provide a form of order consistent with this Memorandum Opinion.