Summary
holding that the merger price was the best indication of fair value for purposes of an appraisal because the merger "resulted from a competitive and fair auction, which followed a more-than-adequate sales process and involved broad dissemination of confidential information to a large number of prospective buyers"
Summary of this case from In re Appraisal of Dell Inc.Opinion
C.A. No. 19586.
Submitted: November 14, 2003.
Decided: December 19, 2003.
Thomas P. Preston, Esquire, BLANK ROME LLP, Wilmington, Delaware; Richard G. Menaker, Esquire and Samuel F. Abernethy, Esquire, MENAKER HERRMANN LLP, New York, New York, Attorneys for Petitioners.
Lewis H. Lazarus, Esquire, Michael A. Weidinger, Esquire, and Thomas E. Hanson, Jr., Esquire, MORRIS, JAMES, HITCHENS WILLIAMS LLP, Wilmington, Delaware, Attorneys for Respondent.
OPINION
This is an appraisal action in which shareholders who are all affiliated with the O'Brien family seek a determination of the fair value of their shares in Union Financial Group, Ltd. ("UFG"). The O'Brien family controlled approximately 38% of UFG's stock before UFG was acquired by an acquisition subsidiary of First Banks, Inc. on December 31, 2001. The patriarch of the O'Brien family, Albert O'Brien, founded UFG and more recently his son Denis O'Brien had served as the company's Chairman and Chief Executive Officer.
In 1999, Denis O'Brien was removed as CEO by the UFG board in a dispute over the wisdom of the business strategy he had committed UFG to pursue, and the financial prudence with which he had managed UFG. A disputatious period then ensued involving litigation between the O'Briens and UFG's board. The O'Briens failed to win a proxy fight at the 2000 annual meeting and Denis O'Brien was not re-elected to the board.
Meanwhile, the UFG board embarked on a strategy to ensure the solvency and future of UFG. That strategy involved exiting certain new businesses (e.g., the "pay-day loan business") that Denis O'Brien had caused UFG to enter and shoring up UFG's capital levels. This change in status left UFG as a bank holding company, owning two relatively small community banks operating in the suburban and rural portions of Illinois that are part of the St. Louis, Missouri metropolitan area. The impetus for this change in strategy was strong, as UFG was having problems meeting its payroll. Indeed, UFG was eventually forced to sign an agreement with the Federal Reserve after an examination of UFG and its subsidiaries revealed the company to be very poorly capitalized and to have inadequate loan-loss reserves. These and other serious problems caused the Federal Reserve to label UFG a "troubled financial institution."
Eventually, the O'Briens and UFG did agree on one thing: that UFG should be sold. This strategy was undertaken because UFG was unable to find equity financing given its condition — which included ongoing litigation with the O'Briens. Also, UFG was saddled with a large amount of debt, including a nearly $10 million debt that was owed in full by mid-2001. The holder of that debt had exhausted its patience with UFG's defaults and had signaled an unwillingness to refinance further. Other lenders were not enthusiastic about refinancing the debt.
Against this backdrop, UFG undertook a sales process. An investment banker working with the O'Briens cooperated with UFG's own banker in that process. After a diligent and effective search for buyers interested in buying UFG and its two banking subsidiaries, a final group of bidders emerged. From among these bidders, First Banks emerged as the winner of a fair auction and a "Merger Agreement" was signed.
The O'Briens, although unable to identify a higher bid, largely voted the shares they controlled against the "Merger" with First Banks in the vote held in September 2001. The other voters overwhelmingly supported the Merger, which was consummated on December 31, 2001.
At trial, the O'Briens sought an award of over $16 per share. This considerably exceeds the "Merger Price." That Price involved the payment of $9.40 per share up-front, with the possibility for UFG stockholders to receive two additional payments of 80 cents per share if the performance of UFG's loan portfolio did not fall below certain thresholds.
In this post-trial opinion, I conclude that the fair value of a UFG share as of the Merger date is the value of the Merger Price minus synergies. The appraisal award excludes synergies in accordance with the mandate of Delaware jurisprudence that the subject company in an appraisal proceeding be valued as a going concern. Because the sales process was an effective one that involved the provision of confidential information to numerous potential buyers and because there is no evidence that the UFG board or its investment banker sought to achieve anything other than the highest possible value, the Merger Price is the best evidence of fair value. Moreover, nothing occurred between the signing of the Merger Agreement and the effective date of the Merger that resulted in an increase in the value of UFG. As a check on the Merger Price, this opinion considers the parties' discounted cash flow ("DCF") estimates of UFG's worth, and concludes that the DCF value of UFG was likely lower than the Merger Price on the Merger date, if reasonable assumptions about UFG's cost of capital and future profit margins are used.
In this opinion, I also address two summary judgment motions filed by the parties disputing whether certain shares controlled by members of the O'Brien family are eligible for appraisal.
I. Factual Overview
As of the relevant valuation date in this appraisal action, December 31, 2001 (the "Valuation Date"), Union Financial Group was a bank holding company with two subsidiaries, Union Bank of Illinois ("Union Bank") and the State Bank of Jerseyville ("Jerseyville"). UFG was not a public company and therefore its shares were not listed for trading on a stock exchange.
Neither subsidiary was large, by banking standards. Union Bank had six branches that operated in the Illinois suburbs of St. Louis, Missouri, focusing on commercial and business lending and deposits. In the year preceding the Valuation Date, Union Bank's assets had generally been in the range of $176-$182 million.
Jerseyville was located in Jerseyville, Illinois, a rural community 35 miles from St. Louis. It was a full-service community bank, which also provided products of particular interest to the agriculture industry. Jerseyville had two branches and a drive-up facility. In the year preceding the Valuation Date, Jerseyville's assets had generally been in the range of $165-$168 million.
UFG — the holding company for Union Bank and Jerseyville — found itself in the position of being weaker than its subsidiaries. From 1997 to mid-1999, UFG had adopted an aggressive growth strategy, centering on going into the pay-day loan business and the "sub-prime" lending business, as well as opening a Missouri bank and acquiring other bank branches. This strategy was spearheaded by UFG's then-CEO, Denis O'Brien. The rationale for the strategy was that UFG's traditional banking operations would not generate large growth and that these new ventures would. Furthermore, the new pay-day and sub-prime loan businesses would yield high profit margins, if they panned out.
The problem is that these businesses were capital intensive and UFG did not have the funds to implement O'Brien's plans and meet its operating expenses. Although O'Brien tried to raise capital to implement his strategy, he fell far short of what was necessary. To help him with the company's finances, O'Brien brought in Thomas G. Barnett as chief financial officer in the late spring of 1999.
Within weeks of Barnett's arrival, the UFG board terminated O'Brien as CEO, after becoming convinced that his business plans had brought the company to the brink of insolvency. Barnett was named as O'Brien's replacement. At that time, the company was without the cash to meet its on-going expenses, and its primary lender, Firstar Bank, N.A., would not increase its credit facility. From among its own ranks, the board raised capital by subscribing for stock. O'Brien was excluded from this capital-raising effort.
As one might expect, O'Brien's termination led to litigation. He successfully challenged the terms of the stock offering, winning a very nominal victory, but failing to convince the court that the directors acted in anything other than good faith in infusing capital into the company on an emergency basis to help it meet its expenses. O'Brien and his family led a proxy contest at the spring 2000 annual meeting, which they lost, resulting in O'Brien's departure from the board. O'Brien and his family filed other derivative litigation, challenging, among other things, change-of-control severance agreements involving Barnett and other UFG managers.
On a very different record involving an earlier time period, this court's Master found that the fair market value of UFG's shares on June 30, 1999 was $17.99. Union Illinois v. Korte, 2001 WL 1526303, at *9 (Del.Ch. Nov. 28, 2001) (Master's Report). That decision was affirmed.
Meanwhile, Barnett and the board implemented a strategy to preserve the solvency of the company. They authorized the sale of the pay-day loan business and the sub-prime loan portfolio. They sought to trim costs. Payments to creditors were deferred when necessary to permit the company to meet payroll. At one point, the board even deferred payment of its own fees.
To meet its obligations, UFG was also drawing large payments of cash out of Union Bank and Jerseyville, in the form of dividends equal to 40% of their earnings. Because UFG had interest payments to make to Firstar that required it to obtain even more cash, UFG took a special dividend from Jerseyville in late 2000 ahead of schedule so that it could meet this obligation.
All in all, UFG was clearly under stress as an entity in the period 1999-2001. Its ability to survive as a going concern was doubtful and the probability that it could pay any dividends to equity holders within the coming decade was small. I now discuss two important factors that underscore this conclusion.
II. The Federal Reserve Bank Identifies UFG As A Troubled Financial Institution
In the autumn of 1999, the Federal Reserve Bank advised UFG of serious concerns it had about UFG's viability. In October 1999, the Federal Reserve's examiners rated UFG and its two subsidiaries. Jerseyville came out the best, with the Federal Reserve ranking it a 2 out of a 5-point scale, in which 1 was the best rating. The only concern noted was that Jerseyville was not at the "well-capitalized" level. Union Bank received a 3 or "marginal" rating, with serious concerns being registered about the adequacy of its loan-loss reserves and its low level of capitalization.
Union Bank was eventually subjected to a memorandum of understanding by the Federal Reserve. After UFG had taken steps to make Union Bank well capitalized, the memorandum of understanding was lifted in the summer of 2001.
But UFG itself came in for even harsher labeling. It was rated a 4, the second-worst score, and labeled a "troubled financial institution" and as having an "unsatisfactory condition." Absent management's efforts to divest the pay-day loan business and the sub-prime portfolio, the Federal Reserve suggested that the company might have received the worst-available score. Among the problems cited at UFG were: 1) inadequate cash flow; 2) excessive leverage; and 3) insufficient capital.
JX 59.
Id.
On February 16, 2000, the Federal Reserve made a presentation to the UFG board. In that presentation, the Federal Reserve repeated many of its earlier observations, noting serious deficiencies in UFG's previous management practices, the inadequacy of its capitalization levels, its questionable loan-loss reserves, its unsatisfactory earnings performance, and its excessive debt.
JX 85 at UFG0007367-7369.
Because UFG had changed strategic direction significantly and was undertaking responsible measures to bring its financial condition under control, the Federal Reserve chose not to place it under the mandatory supervision of a cease-and-desist order. Instead, the Federal Reserve proposed a memorandum of understanding ("MOU") to govern UFG's operations. The MOU required UFG to, among other things, "develop a capital improvement plan designed to increase and maintain capital ratios that are consistent with its financial condition and the risks facing the organization, and are above the `well-capitalized' regulatory standards." To that end, the MOU restricted UFG's ability to pay dividends or increase its borrowings without prior approval. To implement the MOU's goals, the MOU required UFG to develop short- and long-term plans to achieve financial health, which included features addressing the company's outstanding debt and loan-loss reserves. This plan was to be reported to the Federal Reserve and this reporting obligation was in addition to requirements for regular quarterly reporting to the Federal Reserve.
JX 90 at UFG0007230.
On March 29, 2000, UFG reported to the Federal Reserve in accordance with the MOU. Although its plan included using funds from the sale of its pay-day loan business to help Union Bank and Jerseyville achieve well-capitalized status, the plan had no time horizon for UFG itself to reach that status. The plan outlined cost strategies to help UFG manage its way through the situation, but admitted that there was little room for error in its implementation and that the company would have difficulty increasing its profit margins. Importantly, UFG had no firm plan to deal with its approximately $10 million debt to Firstar, which was to come due on July 31, 2000, but had hopes to refinance that debt. If it could not do so, UFG indicated it could "be forced to sell one or both of the banking subsidiaries."
JX 95 at UFG0007293.
The Federal Reserve generally accepted UFG's plan. But it was dissatisfied with the plan's failure to address when UFG would increase its capital to the well-capitalized level. The Federal Reserve therefore asked UFG to submit a plan getting itself to at least the "adequately capitalized" level by December 31, 2000 and to the well-capitalized level within a "reasonable period of time thereafter." In addition, the Federal Reserve suggested that UFG begin to look for alternatives to a refinancing with Firstar, given that the debt was coming due on July 31, 2000.
JX 103 at UFG00007233.
On May 18, 2000, UFG responded to the Federal Reserve. It explained that it would be unable to achieve even the adequately capitalized level in 2000 because it would need outside capital to achieve that level. In view of the pending litigation with the O'Brien family, UFG did not feel it could obtain such financing. Although UFG had taken steps to improve the capitalization of Union Bank and Jerseyville — such as substantial sales of loans and restrictions on operating costs — it could not commit to a timetable for becoming well capitalized itself. As to Firstar, UFG reported that it was making progress in seeking an extension of the debt but was also seeking out other sources of financing.
The Federal Reserve was again patient with UFG. It accepted the plan although it noted that UFG's plan did not "meet the capital targets contained in . . . the MOU." The Federal Reserve also asked to be updated on UFG's success in dealing with Firstar.
JX 116 at UFG0000661.
As a condition of the MOU, UFG's board was required to adopt the plan formally, which it did. As of the Valuation Date, UFG remained subject to the MOU, which required that UFG seek to increase its capital ratio to the well-capitalized level and which placed serious restrictions on UFG's right to pay dividends or receive them from its subsidiaries, or to increase its debt in order to pursue business growth.
III. UFG's Debt Problem
As has been adverted to, UFG's largest creditor during the period 1999-2001 was Firstar, which had provided UFG with a large credit facility. The credit extended by Firstar was secured by all of UFG's stock in Union Bank and Jerseyville.
From the middle of 1999 until the Firstar loan was paid off in July 2001 with funds loaned to UFG by First Banks in connection with the Merger, UFG was in default on its loan agreement with Firstar. In 1999 and 2000, Firstar extracted concessions from UFG in exchange for extending the maturity date of the loan. These concessions involved certain payments and stringent restrictions on UFG's, Union Bank's, and Jerseyville's use of cash as well as their agreements to meet certain performance targets (such as required returns on assets, specific loan-loss reserves, and capital ratios).
But, UFG and its subsidiaries were unable to meet these requirements. By the year 2001, UFG was facing an even more difficult problem, in that it was unable to make its March 2001 principal payment under the Firstar loan and had to seek an advance waiver of a default on its principal payment due in June 2001. These defaults were significant because the entire principal balance of $9.7 million was set to mature on July 31, 2001.
By 2001, Firstar had no appetite to further extend the loan and UFG's prospects for alternative financing were not promising. Barnett had tried to seek alternative debt financing without success. Moreover, given the litigation with the O'Brien family and UFG's balance sheet, outside equity investors were not interested in infusing capital into the bank.
As a result, in the absence of a strategic transaction, UFG faced the prospect of having Firstar force a sale of one or more of the subsidiaries or impose even more onerous restrictions on UFG's (and its subsidiaries') operations, as well as having Firstar exact a steeper financial toll for continuing to act as UFG's lender. Indeed, by June 2001, Firstar said it would not consider renewing its loan unless, among other things, UFG could demonstrate an ability to amortize the loan fully over a seven-year period and settle its litigation disputes with the O'Brien family on or before the end of July 2001.
The feasibility of meeting those goals was remote. In this connection, it bears mention that Barnett reported to UFG's board on June 20, 2001 that he expected the company to have cash on hand of less than $9,000 by the end of that month.
JX 233.
IV. The Sales Process
The preceding factors — the Federal Reserve MOU and UFG's debt crunch — were symptomatic of the company's overall condition in the timeframe of 1999 until the date of the Merger. These factors related in no small measure to UFG's ill-fated (and ultimately abandoned) expansion strategy, which stretched the company beyond its capacity for responsible growth, and left it on the brink of insolvency. But the expansion strategy was not UFG's only problem. The reality is that Denis O'Brien embarked upon that expansion strategy because he perceived Union Bank and Jerseyville as less-than-ideal vehicles to drive growth in shareholder value. As operating banks, both subsidiaries had profit margins — as illustrated by return on assets and net interest margin ("NIM") ratios — that were unimpressive when compared to their industry peers. Likewise, the quality of their loan portfolios was relatively poor.These confidence-shaking facts combined with another factor to make it difficult for UFG to raise capital: the ongoing feud between the O'Brien family and the company's incumbent board. Having historically controlled UFG and owning 38% of its stock, the O'Brien family did not accept having Denis O'Brien ousted from management with quiet resignation. Instead, it filed several lawsuits, most of which were active during this period. Thus, any investors who might be seeking less than full control were buying into an unsettled corporate governance structure. Although the O'Briens seek to downplay this factor because the lawsuits they filed were primarily derivative suits putatively brought to benefit UFG as a company, the reality is that investors rarely see such litigation as a positive factor in determining whether to put capital at risk in a company. When the company also faces a cash-flow crisis and is operating under Federal Reserve supervision, the existence of a disputatious atmosphere is likely to chill any lukewarm investor interest that might otherwise have existed.
In this case, that is exactly what happened, as all these factors made it impossible for the company to find equity investors. These factors also hampered the ability of the company to find an alternative source of debt financing or to obtain an extension from Firstar.
Ultimately, one thing the incumbent board and the O'Brien family could agree upon was that it would be advisable to sell UFG. For its part, UFG retained Stifel, Nicolaus Company ("Stifel") as its investment banker. Stifel was well-equipped for its duty, having carved out a leading niche in mergers and acquisitions involving small and medium-sized banks in the Midwest. Its team was led by a specialist in the banking field, Rick E. Maples. For their part, the O'Brien family hired Howe Barnes Investments, Inc. ("Howe Barnes") to help them sell their block or to generate interest in UFG as a whole.
Stifel and Howe Barnes eventually worked together to market UFG under a joint cooperation agreement forged in September 2000. In October and November of 2000, thirty-six potential purchasers were contacted and asked if they were interested in possibly buying UFG, in whole or in part. Eventually, twenty-three companies signed a confidentiality agreement and received the package of confidential information that Stifel and Howe Barnes had prepared.
Four bids were then received, from Allegient Bancorp, Inc., First Banks, Manchester Partners, L.L.C., and Royal Bancshares. All of the bids were conditioned on the settlement of the pending litigation involving the O'Brien family. Of the four bids, the board decided that the First Bank, Manchester Partners, and Royal Bancshares bids warranted further inquiry, and those bidders were given the opportunity to conduct on-site due diligence and to submit revised proposals. After this due diligence was completed, the board decided to focus upon Royal Bancshares because it was the only bidder who, at that time, would negotiate before settlement of the O'Brien lawsuits.
Negotiations with Royal Bancshares continued until March but did not result in a deal. The next month a new bidder, Reliance Bancshares ("Reliance"), emerged as a potential buyer and commenced negotiations. Then, First Banks and Manchester expressed renewed interest.
By this time, the joint cooperation agreement with the O'Brien family had broken down to some extent. While the O'Briens still wished to sell UFG, they were seeking a substantial premium for their block of shares.
Therefore, when final and best offers were sought in July 2001 from Reliance, First Banks and Manchester, the O'Brien family favored Reliance's bid. That bid would have given $8.23 per share at closing to the non-O'Brien stockholders, with an additional $2.27 per share escrowed and payable if UFG's loan losses stayed below a certain threshold. Under Reliance's bid, the O'Brien family would have received $11.17 per share at closing, with $2.27 per share escrowed. This was a premium in two senses, because the up-front payment was higher and because the proportion of the payment to the O'Brien family that was escrowed was lower. Not only that, the Reliance bid was highly conditional, requiring that UFG agree to terminate its severance agreements with Barnett and Union Bank Chairman and President, Dennis Bielke. No evidence in this record supports a rational conclusion that those severance agreements were unreasonable.
First Banks submitted the best overall unconditional bid with the most up-front, guaranteed consideration. It offered to pay $9.20 per share at closing to all shareholders, with an additional $1.60 per share available if UFG met certain loan-loss targets. Manchester's bid involved the payment of $7.44 to $8.66 per share at closing, with the opportunity to receive up to a total of $10.50 per share eventually, again subject to the performance of UFG's loan portfolio.
Each of the bidders was afforded the opportunity to make an oral presentation to the UFG board, which was advised by outside legal counsel (including a firm that the O'Brien family helped select) as well as Stifel. The board then deliberated and concluded that the First Banks offer was the superior bid.
The board decided, however, to ask First Banks to increase its bid in two respects: 1) by offering 20 cents more per share at closing; and 2) by increasing the total consideration possible under the deal from $10.80 to over $11.00. First Banks agreed to the first demand, but not the second. Each of the other bidders was given the opportunity to bid again but each balked and none topped the First Banks bid.
The board approved the Merger with First Banks after receiving a fairness opinion and a thorough supporting presentation from Stifel. The Merger Agreement was prepared between UFG and First Banks. That Agreement called for the payment of $9.40 per share at closing. Additional payments of 80 cents per share were available on the first and second anniversaries of the Merger date, if certain contingencies were met.
Before the Merger vote, First Banks agreed to provide financing to UFG, which enabled UFG to pay off its debt to Firstar on July 31, 2001 and which provided it with $2 million in working capital to ease its cash crunch and pay Merger-related expenses. The financing that First Banks provided came after Firstar demanded a $25,000 payment in exchange for a 90-day extension after learning of the Merger Agreement. The $12 million First Banks credit facility was secured by a pledge of all the stock of Union Bank and Jerseyville. Loans under the credit facility became fully payable on the earlier of July 31, 2002 or six months after termination of the Merger Agreement.
The Merger Agreement was approved by the UFG stockholders at a September 24, 2001 meeting. Over 98% of the stockholders unaffiliated with the O'Brien family supported the Merger. The O'Brien family voted no with the shares over which they had direct voting power. After the Merger closed on December 31, 2001, the Federal Reserve terminated the MOU governing UFG's operations. Absent the Merger, the MOU would have continued in effect until the Federal Reserve was satisfied that its conditions had been met.
The Merger resulted from a professional and diligent sales process. As the O'Brien family's own advisor from Howe Barnes, testified: "My view is that we contacted all of the parties that would have been interested in acquiring [UFG]." Denis O'Brien himself described why Reliance did not bid again, when confronted with First Bank's ultimate offer: They were "frightened off" by that offer.
UFG Answering Post-Trial Br. App. R (Charles H. Barrow Dep.), at 36.
UFG Answering Post-Trial Br. App. S (Denis J. O'Brien Dep.), at 149.
In sum, the evidence supports the clear conclusion that UFG was marketed in an effective manner, with an active auction following the provision of full information to an array of logical bidders. The petitioners' allegation that the process was rushed and biased is unsupported by any substantial evidence. Rather, the evidence demonstrates that bidders were given an open opportunity to put their best offer on the table and that the winning bidder made the best unconditional bid.
The O'Briens contend that the highly conditional Reliance bid was better. The UFG board had abundant grounds to conclude it was not. These reasons include the fact that the Reliance bid was conditioned on termination of the management severance agreements and was only putatively higher in overall value because it contained higher escrowed payments, the value of which had to be highly discounted. Most important, the Reliance bid offered the non-O'Brien stockholders much less, in order to pay a "control premium" for a block of shares that, voting together, had recently failed to prevail in a proxy contest. Finally, it bears repeating that Reliance had the chance to top First Banks' final offer and did not.
This real-world market check is overridingly important evidence of value. This was not a situation involving a squeeze-out merger. Instead, this situation involved an effective process whereby third-party bidders were invited to buy UFG after receiving confidential data about the company's prospects. The O'Brien family's own advisor participated in stimulating the bidding process and preparing the information packet. Nothing suggests that the process was flawed in any material respect or that the timing of the process adversely affected the result.
V. The Management Projections
During the course of the sales process, Stifel and UFG management collaborated on developing financial projections for the company. These were designed to be best estimates of the results that UFG could achieve during the forecast period. The last set of projections was dated July 17, 2001 (the "Management Projections" or "Projections"). They were based on actual results as of June 30, 2001, and projected UFG's future financial results to the year 2005.
The management teams at Union Bank and Jerseyville helped formulate the Projections and they estimated the returns each bank would achieve, based on underlying assumptions they made regarding costs, loans, deposits, total assets, return on assets, and other relevant factors. The key factors in driving growth under the Projections are assets and earnings. UFG projected asset growth rates for the forecast period as follows:
2002 4.53%
2003 5.48%
2004 6.44%
2005 7.41%
The Projections also estimated a return on average assets of:
2002 0.47%
2003 0.47%
2004 0.48%
2005 0.48%
JX 249 at UFG0008181.
Id. at UFG00008180.
This return on average assets resulted — along with the effect of other assumptions — in an estimated net interest margin for the forecast period ranging from 3.61 to 3.67. Meanwhile, on the cost side, the Management Projections estimated that non-interest expense — such as employee costs — would grow at 5%, 6%, 7%, and 8% respectively in the years 2002-2005. As will be noted later, these cost projections were reasonable ones made by the management teams at Union Bank and Jerseyville.
Id. at UFG0008179; Trial Tr. at 357, 377-78.
The return — in thousands of dollars — to the bottom line — net income — that resulted from these assumptions was as follows:
2001 2002 2003 2004 2005 1,632 1,714 1,817 1,944 2,100
JX 249 at UFG0008179.
What is notable about these Projections is that they did not generate results that would permit UFG to achieve well-capitalized status by 2005. That is, even after four more years of operations, UFG was projected to be below the level of capitalization necessary to lift the MOU and to start generating returns (in the form of dividends) to the common equity holders.
In fact, UFG was not projected to even reach adequately capitalized status until 2003.
VI. The Contending Positions Of The Parties A. The O'Briens' Evidence
The O'Briens argue that the fair value of UFG as of the Valuation Date was over $16.00 per share. They base that aggressive assertion on the report of their expert, Michael G. Mayer, from Intecap, Inc. In his reports and testimony, Mayer relied heavily upon a DCF analysis.
Mayer's DCF started with the Management Projections as a jumping-off point. But he then deviated in significant ways that generated much higher free cash flow.
To justify his optimism, Mayer took the view that UFG's performance in the latter half of 2001 was greatly improving and was a harbinger of good things to come. He based this on some improvements in the company's NIM for the last six months of the year, resulting in a second-half 2001 average NIM of 3.92 as compared to the full-year NIM of 3.62. Mayer could not really explain the reason for this phenomenon, except to attribute it to lower interest rates in the wake of the tragic events of September 11, 2001. Mayer also noted that the Federal Reserve had lifted its MOU governing Union Bank, a sign of progress that also reflected UFG's efforts to increase Union Bank's capitalization by reducing assets and costs. Because of this supposed improvement, Mayer was optimistic that UFG would soon begin to perform in line with other comparable banks.
O'Brien Post-Trial Reply Br. Ex. D.
At trial, the most persuasive testimony on this score was that a decline in interest rates could improve UFG's NIM for a period of time, during which it could cut the amount it paid on time deposits before loans would reprice. Trial Tr. at 371-72, 791-94.
In keeping with his sunny viewpoint, Mayer did not adopt management's assumptions regarding return on average assets or NIM. Stated more accurately, Mayer rejected the NIM assumptions and projected that UFG would have a NIM of 4.29 during the projection period, a level of NIM he based on a ten-year average for all commercial banks operating in UFG's geographic market. This assumption was one of the largest drivers of value in his model and, stated simply, is based on nothing other than Mayer's assertions that a ten-year average is more indicative of the future than more recent averages for the same group of banks and that UFG could perform on par with this comparison group. The more recent five years produced an average NIM of 3.97 and the most recent year — i.e., 2000 — generated an average NIM of 3.68. As we shall see, Mayer's optimistic choice of a 4.29 NIM is one of the most important elements of his analysis. Without similar optimism, Mayer's model results in a much lower value.
Put more precisely, Mayer used data regarding all commercial banks in Illinois operating in the Federal Reserve District in which UFG operated. See O'Brien Post-Trial Reply Br. Ex. E. This included data for banks that were not holding companies. Mayer did not account for that factor or the relatively high degree of leverage at UFG.
The four-year average was an even lower 3.87. See PDX 6B.
Mayer also rejected management's projections of growth in non-interest expense. He reduced the projected growth rate of non-fixed expenses to 3.44% based on his view that management can control employee and other variable costs and that a rate of growth more in accord with inflation was to be expected. Mayer did not base this revision on discussions with management, historical growth in these cost areas, or depositions of management. He simply viewed it as possible and believed that the constriction on such costs would not impinge on growth.
In fairness, such expenses did drop sharply from the Management Projections during the second half of 2001. This drop was not explained and I have no rational basis to assume it was durable. It may have resulted to some degree from the departure of some key employees at its subsidiaries. To generate the expected growth, UFG would have had to replace these employees.
Mayer applied these revised assumptions to an adjusted balance sheet for UFG for year-end 2001. The balance sheet was adjusted to account for costs that Mayer perceived as being merger related. The effect of using a year-end balance sheet is modest in comparison with Mayer's other changes. Under Mayer's adjusted balance sheet, UFG's assets had fallen below the level expected as of December 31, 2001 in the Management Projections. This downward reduction in earning assets — which in large part resulted from UFG's concerted effort to improve its capitalization ratios, an effort that required it to restrict assets and therefore future growth — actually generates lower future returns, unless management's projected return on assets and NIMs for years 2002-2005 are raised considerably.
Mayer also valued UFG's preferred stock in a manner that was very favorable to his clients' position. Because I conclude that Mayer's adjusted balance sheet does not produce a value greater than the Merger Price minus synergies when reasonable NIM and cost-of-capital assumptions are used, I do not dilate on the proper valuation of the preferred in this opinion. Given what First Banks paid for the preferred in the Merger, the value used by the respondent's expert, Donald Clarke, strikes me as reasonable.
Mayer's DCF model accepted the Management Projections' (unrealistic) assumption that the outstanding Firstar debt — or the new and higher $12 million First Banks debt — could simply be carried at face value with no reduction in outstanding principal. Even with Mayer's higher returns premised on a much higher NIM than UFG achieved in 2001, UFG does not reach a well-capitalized level until sometime after 2005.
With his optimistic assumptions about cost reductions and improved profit margins, Mayer generated estimated returns for UFG that far exceeded those contained in the Management Projections. The difference is illustrated below:
--------------------------------------------------------------- | | | Management | | | | | Mayer Net | Projections | | | | Year | Income | Income | $ Difference | % Difference | |------|------------|-------------|--------------|--------------| | 2002 | $3,242,000 | $1,852,000 | $1,390,000 | 75% | | | | | | | | 2003 | $3,426,000 | $1,955,000 | $1,471,000 | 75% | | | | | | | | 2004 | $3,681,000 | $2,082,000 | $1,599,000 | 77% | | | | | | | | 2005 | $3,989,000 | $2,238,000 | $1,751,000 | 78% | ---------------------------------------------------------------
These differences depend on UFG achieving a return on average assets of 0.91% to 0.98%. This rate of return is about twice what management projected and far exceeds UFG's historic performance.
Finally, Mayer derived his cost of capital from Ibbotson's Cost of Capital 2001 Yearbook. Mayer used a standard industrial classification or "SIC" code for state commercial banks. He applied the median results of all banks in that category to derive a cost of capital of 10.43% using the capital asset pricing model ("CAPM") plus a size premium. Because UFG was not publicly traded, the beta was derived from comparisons to other companies. The beta Mayer used suggested that a rational investor would face less systematic risk by investing in UFG than in a large, publicly traded bank operating free of Federal Reserve supervision. That is, the beta suggested that an equity investor in UFG would require less than the average cost of capital for an equity investment in a banking institution.
Based on these various assumptions, Mayer came to a value for UFG of over $16 per share. In response to criticisms from the respondent, Mayer made certain adjustments that lowered his result somewhat but he basically stuck with a value in that range. After trial, Mayer modified his DCF model significantly, and provided an explanation of that modification only in the O'Briens' post-trial reply brief. This modification comes far too late in the day and will not be considered, as the evidence had long closed by then and Mayer had more than a fair opportunity to modify his model before trial.
In order to buttress his opinion in his reports, which was founded on his DCF analysis alone, Mayer also performed comparable-companies and comparable-transactions analyses that he believed reinforced the validity of his DCF model. Those analyses were, in my view, unrealistically generous to his clients' position.
B. The Respondent's Approach
The respondent's expert, Donald Clarke of the Griffing Group, took a different tack. In the end, Clarke believed that the Merger Price was the most reliable indicator of value. As a result, he opined that the fair value of a UFG share on the Valuation Date was $8.20, a value he derived largely by giving overriding weight to his estimate of the net present value of the expected Merger consideration (taking into account the possibility of receiving up to two additional 80-cent payments), net of synergies. That estimate — which was based on a reasonable synergy discount of 13% — generated an approximate net present value of $8.74.Clarke checked his view of value against the analysis he derived from other techniques, including a DCF analysis. In performing his DCF analysis, Clarke relied on the Management Projections without adjustment. In his view — which I conclude was generally reasonable — nothing material happened between July 17, 2001 (the date of the Management Projections) and December 31, 2001 that made it necessary to significantly alter those Projections. Even though Clarke concedes that UFG's results between July 17 and December 31, 2001 deviated from the Projections, he perceived nothing about those deviations that suggested a fundamental flaw in the Projections. Indeed, Clarke notes that the effect of adjusting for year-end results is to markedly reduce UFG's operating assets from the level contained in the Projections, a reduction that bodes ill for future returns.
Most importantly, Clarke hewed to the Management Projections' assumptions regarding costs, NIM, and return on assets. He did not revise costs downward and the return on assets or NIMs upward, believing there to be no basis to assume that UFG would improve its profitability markedly during the projection period. Using the Management Projections in this unadjusted manner, Clarke then discounted them using his estimate of the company's cost of equity under the CAPM, 16%, plus a company-specific discount of 2%. Clarke's estimate of a 16% CAPM rate was based on a comparison to large bank holding companies because Clarke believed that the calculation of beta for smaller, publicly traded state commercial holding banks was skewed by the infrequent trading in their shares, leading to an embedded understatement of systemic risk. Clarke also observed that UFG's high leverage and rocky history made it atypical of the smaller publicly traded bank holding companies, whose low leverage (he assumed) contributed to their low beta.
As for his advocacy of a 2% company-specific risk premium, Clarke acknowledged that it was not accepted by pure adherents to the CAPM but that it was an acceptable technique to those who had to value companies in the real world of commerce (and the surreal world of appraisal litigation). Because UFG, among other things, was highly leveraged, faced the need to pay off the $12 million First Banks debt imminently, operated under an MOU placing severe restrictions on use of cash and limiting growth and payments to equity holders, and was enmeshed in litigation involving the O'Brien family, Clarke believed that a prudent investor would demand a higher rate of return to invest in UFG than to invest in the average, publicly traded bank. In support of that conclusion, the respondent presented convincing evidence — in the form of Federal Reserve-prepared reports — that UFG's performance — when compared to a group of bank holding company peers tracked by the Federal Reserve Bank — was seriously deficient. Of particular note was how poorly capitalized and how highly leveraged UFG was in comparison to its peers. These factors accompanied indications that UFG's loan portfolio was of much lower quality than that of its peer group.
I do not choose to take this opportunity to enter into the debate about whether company-specific risk premiums can be added to come up with an accurate cost of capital for use in a discounted cash flow analysis. I understand that investors do consider company-specific risks in calculating the cost of capital they will use in investing money and that investment banks use company-specific risk premiums in advising clients. They particularly do so when a company's shares do not actively trade on a daily basis in a public market. Pure proponents of the CAPM argue that only systemic risk as measured by beta is relevant to the cost of capital and that company-specific risks should be addressed by appropriate revisions in cash-flow estimates.
Because I have awarded a value tied to the Merger Price, I need not heap on a higher risk premium. The effect of doing so would reduce DCF value even below the levels I identify as generous later in this opinion. I do note my view that it would have been reasonable for Clarke to monetize the risks he identifies by adjusting the Management Projections directly. For example, I think it would have been reasonable to assume that UFG would have had to reduce the outstanding debt to Firstar (later, First Banks) considerably during the Projections period and to have reduced cash flows accordingly. Trial Tr. at 339-40 (Firstar demanded, among other conditions, UFG to fully amortize loan over a seven-year schedule). This reduction would have been consistent with the substantial reductions in principal that Firstar demanded and received during the years 1999-2001.
See JX 359; RDX 2.
In his comparable-companies analysis, the O'Briens' expert identified a group of companies he believed to be comparable to UFG. Even Mayer conceded that UFG ranked in the lower quarter of this group of companies. My examination of his comparables leads me to conclude that UFG's performance placed it more accurately within the lowest decile.
Taken together, Clarke's assumptions generated a DCF value for UFG's shares of $5.06 per share. Notably, if Clarke only used a discount rate of 16%, rather than 18%, the value only rose to $5.77. Even using Mayer's discount rate of 10.43%, the value of a UFG share under the Management Projections was only $8.16.
Clarke also performed comparable-companies and transactional analyses that yielded values of $8.88 and $8.40 per share and that, in his view, generally confirmed his belief that the Merger Price was the best estimate of value.
VII. The Legal Framework
In an appraisal action, this court has broad discretion to determine the fair value of the shares of the petitioners. This is done in a jurisprudentially specific manner that is policy-based and that is different from that which would be undertaken to find the "fair market value" of the petitioners' shares. The Supreme Court has interpreted § 262 as requiring this court to determine fair value by awarding the petitioners their pro-rata portion of the company whose shares are subject to valuation, but with the proviso that the company be valued as a going concern.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. By its plain terms, § 262 only excludes from the amount awardable to the petitioners "value arising from the accomplishment or expectation of the merger" that gave rise to the petitioners' right to appraisal. The literal terms of § 262 do not preclude a court from considering, in using a comparable-companies analysis for example, that acquirers typically share a portion of synergies with sellers in sales transactions and that that portion is value that would be left wholly in the hands of the selling company's stockholders, as a price that the buyer was willing to pay to capture the selling company and the rest of the synergies. For that matter, the literal terms of § 262 do not preclude a court from using a comparable-transactions analysis that considers the price at which the subject company would likely sell in an auction. Again, such an approach would not award the petitioners value from the particular merger giving rise to the appraisal; it would simply give weight to the actual price at which the subject company could have been sold, including therein the portion of synergies that a synergistic buyer would leave with the subject company shareholders as a price for winning the deal.
8 Del. C. § 262(h) (emphasis added).
This observation should not be read as a statement that so-called "synergy" deals are actually done at higher premiums to pre-affected market price than other merger transactions. That is a highly contestable point as the finance literature, as I understand it, does not contain a reliable method for estimating the portion of a merger premium that results from expected synergy value. See, e.g., John C. Coates IV, "Fair Value" as an Avoidable Rule of Corporate Law: Minority Discounts in Conflict Transactions, 147 U. Pa. L. Rev. 1251, 1351-52 (1999) (discussing this issue); Agranoff v. Miller, 791 A.2d 880, 897-898 (Del.Ch. 2001) (same).
The exclusion of synergy value, rather, derives from the mandate that the subject company in an appraisal be valued as a going concern. Logically, if this mandate is to be faithfully followed, this court must endeavor to exclude from any appraisal award 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 a part of a larger enterprise, from which synergistic gains can be extracted.
The parties in this case accept, as do I, that this mandate binds me and that my attempt to value UFG must center on determining its value as a going concern. That requires me to exclude from any valuation analysis a consideration of even the portion of synergies that might be expected to be left with the UFG stockholders in a sale of the entity as a whole.
This understanding is an important one in this case. Because the definition of fair value used in a § 262 proceeding is not based on fair market value and involves policy considerations, such as the need to exclude synergies in order to value the entity as a going concern, the petitioners in an appraisal proceeding can be awarded a sum that deviates — upward or downward — from what an economist or investment banker or Warren Buffett would believe was the market value of the petitioners' shares. In this case, as we shall see, the fair value standard operates to leave the O'Briens, as petitioners, with less than they would have received had they accepted the Merger consideration.
My explanation of why that is so begins with my understanding of the discretion left to me under § 262. As I perceive it, I am free to consider all non-speculative elements of value, provided that I honor the fair value definition articulated by the Delaware Supreme Court. This does not mean that I must perform every conceivable valuation technique in the universe and then give some arithmetic weight to each of them. Rather, I am empowered to come up with a valuation, drawing on what I reasonably conclude is the most reliable evidence of value in the record.
Importantly, that evidence may include facts bearing on the market value of the subject company. This includes the transaction that gives rise to the right of appraisal, so long as the process leading to the transaction is a reliable indicator of value and merger-specific value is excluded. More generally, our case law recognizes that when there is an open opportunity to buy a company, the resulting market price is reliable evidence of fair value. Or, as then-Vice Chancellor, now Justice Jacobs, aptly put it: "The fact that a transaction price was forged in the crucible of objective market reality (as distinguished from the unavoidably subjective thought process of a valuation expert) is viewed as strong evidence that the price is fair." In an analogous context, Chief Justice Veasey recently noted that "in many circumstances a property interest is best valued by the amount a buyer will pay for it" and that the value placed on such an interest in an active bidding process (such as a liquid market) "provide[s] a measure of fair value superior to any estimate the court could impose." In this case, I apply that teaching in determining the fair value of UFG as an entity.
See Cooper v. Pabst Brewing Co., 1993 WL 208763, at *8-10 (Del.Ch. June 8, 1993).
See M.P.M. Enters., Inc. v. Gilbert, 731 A.2d 790, 797 (Del. 1999) ("A merger price resulting from arm's-length negotiations where there are no claims of collusion is a very strong indication of fair value."); Cooper, 1993 WL 208763, at *8-10 (basing valuation primarily on estimated actual market value of stock as determined by bidder's ultimately successful tender offer price); Van de Walle v. Unimation, Inc., 1991 WL 29303, at *17 (Del.Ch. Mar. 7, 1991) ("The most persuasive evidence of the fairness of the . . . merger price is that it was the result of arm's-length negotiations between two independent parties, where the seller . . . was motivated to seek the highest available price, and a diligent and extensive canvass of the market had confirmed that no better price was available.").
Van de Walle, 1991 WL 29303, at *17. Cf. Applebaum v. Avaya, Inc., 812 A.2d 880, 889-891 (Del. 2002) (when a stock is actively traded and the value of the stock is "tested . . . through the purchase and sale of the stock on the open market," the market price is fair value for the cash-out of fractional shares under 8 Del. C. § 155).
Applebaum, 812 A.2d at 889-890.
Therefore, I conclude that the price for UFG that resulted from the auction preceding execution of the Merger Agreement is the most reliable evidence of fair value. I peg my award to it, giving 100% weight to that factor. If I were to do otherwise, I would reduce the award because in my view, a rational DCF valuation of UFG would result in a lower, not higher, value than the Merger Price, even when the Merger Price is adjusted to subtract synergies.
I next explain that conclusion.
VIII. The Merger Price Is The Best Indicator Of Value
After a careful review of the evidence, I find no basis to fault the process used to sell UFG. As the O'Briens' own advisor from Howe Barnes admitted, the sale was undertaken in a sound way. It was not conducted in haste. There was no plausible motive on the part of the UFG board not to seek to obtain the highest price. There is every indication that the ultimate auction was conducted fairly and openly, and nothing convincing to the contrary.
Stated bluntly, the Merger in which First Banks acquired UFG was not a squeeze-out auction in which a parent company was the only available buyer. Rather, the Merger with First Banks resulted from a competitive and fair auction, which followed a more-than-adequate sales process and involved the broad dissemination of confidential information to a large number of prospective buyers.
The O'Briens' failure to sell their own block for an amount in excess of the Merger Price despite strenuous efforts to do so also supports the conclusion I reach in this case.
The O'Briens quibble that the Merger was not consummated until December 31, 2001 even though the Merger Agreement was approved in September 2001 and the Merger terms were set even earlier. This is not a forceful objection. The negotiation of merger terms always and necessarily precedes consummation.
Nothing in the record persuades me that UFG was more valuable by December 31, 2001 than it was when the Merger terms were set. The O'Briens have not been 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. Although UFG experienced a modest upward adjustment in its NIM in the second half of 2001, there is no evidence that this increase was sustainable or that it portended an end to UFG's problems. Rather, as of December 31, 2001, UFG actually had lower earning assets than management had estimated, a result of its on-going effort to achieve adequate capitalization levels. Furthermore, although UFG had improved the capitalization of Union Bank and Jerseyville by the end of 2001, its own capitalization remained well below (and over a half-decade away) from the level required to be achieved in the MOU. The O'Briens also do not confront the reality that UFG remained cash strapped and as recently as mid-summer 2001 could not service — i.e., had defaulted on its obligation to repay — its debt and measured its month-end cash flow in thousands of dollars.
First Banks provided a loan that gave UFG a brief reprieve, as it allowed UFG to pay off the Firstar loan and gave it access to additional cash. But absent consummation of the Merger, UFG faced the need to pay off that $12 million loan in full in 2002 or refinance it. UFG could not do the former and its ability to do the latter was highly questionable, and would, in any event, come at a significant price that would eat into the company's estimated cash flow. Furthermore, as of December 31, 2001, UFG remained enmeshed in ongoing conflict between its incumbent board and the O'Brien family, a conflict that deterred capital investors who could not obtain complete control of the company. Considered 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.
I have not burdened the reader with a painstaking recitation of all the evidence that supports my conclusion. Among the other factors is evidence that UFG's loan-loss reserves were less than adequate, that its loan portfolio was deteriorating in the autumn of 2001, and that UFG had not adjusted its loss reserves properly in late 2001 because of the upcoming Merger; Denis O'Brien's own view, documented in record evidence, that Union Bank and Jerseyville were not businesses of the kind that would generate rapid growth that would result in an attractive equity price for UFG; the need to replace key employees at the subsidiaries; the September 30, 2001 Federal Reserve bank holding company report that indicated that UFG was lagging well behind its industry peers in critical areas bearing on the quality of its loan portfolio and its financial stability; and UFG's inability to repay its debt to First Banks in the event the Merger was not consummated and its poor refinancing prospects. Putting a point on this conclusion is the fact that all of the bidders in the auction insisted on putting a substantial amount of the total consideration they would potentially pay for UFG in escrow, subject to UFG's loan portfolio meeting certain performance conditions. Thus, in the real world, the bidders for UFG perceived the risk that its assets (i.e., its loan portfolio) would not generate the returns UFG promised.
IX. The DCF Valuation Technique Is An Inferior Estimate Of Value When Compared To The Merger Price
In view of the market's opportunity to price UFG directly as an entity, the use of alternative valuation techniques like a DCF analysis is necessarily a second-best method to derive value. A DCF analysis depends heavily on an assumption about the cost of capital that rational investors would use in investing in UFG, and assumptions about the accuracy of UFG's cash-flow projections. The benefit of the active market for UFG as an entity that the sales process generated is that several buyers with a profit motive were able to assess these factors for themselves and to use those assessments to make bids with actual money behind them.For me (as a law-trained judge) to second-guess the price that resulted from that process involves an exercise in hubris and, at best, reasoned guess-work. For the sake of completeness, however, I will make some observations about the DCF analyses in the record and indicate why I agree with the respondent's expert, Donald Clarke, that a proper DCF analysis generates a value lower than the Merger Price, adjusted for synergies.
Cf. Richard A. Brealy Stewart C. Meyers, Principles of Corporate Finance 70 (7th ed. 2002) (noting, with their tongues only partly in cheek, that if the DCF analysis you perform of a stock does not match the market price, you have probably used poor forecasts).
First, I believe that there is no rational basis to conclude that UFG's performance was improving at the rate that Mayer assumed. Mayer projected that UFG would have a NIM of 4.29 during the projection period. That NIM exceeded any NIM that UFG achieved in any month of 2001. The improvement in NIM in the last part of 2001 that Mayer emphasizes so heavily is, in my view, based on only a few months and Mayer did not persuasively explain why that brief uptick signaled a sustainable trend of the strength necessary to increase UFG's NIM to 4.29. I acknowledge that UFG performed better in 2001 than expected, but I note that, overall, UFG had a NIM of only 3.62 for 2001. Even if one assumes that UFG would achieve a NIM of 4.0 throughout the forecast period, and even if one uses Mayer's own adjusted forecasts and his 10.43% cost of capital, his own model produces a per-share value of only $2.24. This demonstrates that Mayer's estimate of value derives largely from his optimistic assumption that UFG's NIM (and accordingly its return on average assets) would exceed the Management Projections' estimates in a huge way. But this optimism was unaccompanied by any explanation of why UFG's performance — which had long lagged that of its competitors — would surge in this way.
Despite some improvements in performance, UFG remained well below the adequately and well-capitalized levels at the Valuation Date. E.g., JX 299 at UFG0005017. Lacking access to outside capital, UFG's ability to grow was therefore restricted because it was forced to retain earnings and cut assets to improve its capital ratios.
In his revised model that was explained only in the O'Briens' post-trial reply brief, these assumptions generate a value of $11.10, a shockingly large disparity that illustrates the unfairness of allowing the O'Briens to change their model so enormously after trial.
See JX 359 at 1 (UFG ranks in 13th percentile of peers for NIM, the low percentile reflecting poor performance).
UFG's failure to increase its loan-loss reserves in the face of adverse developments in its loan portfolio, see, e.g., Trial Tr. at 742-43, results in an overstatement of its net income in Mayer's adjusted 2001 balance sheet.
Given the problems that UFG objectively faced and its historic performance, I see no basis to believe that its return on average assets or NIM would significantly exceed that estimated in the Management Projections. Although it might be reasonable to use somewhat more optimistic assumptions, the assumption that UFG would achieve a NIM of 4.29 — a figure based on the unproven assertion that a ten-year average of NIMs for all commercial banks (large and small) in UFG's region is a good forward-looking projection — is unjustified. The use of such a figure has the effect of markedly increasing UFG's return on average assets and its net profitability. So too did Mayer's decision to reduce management's projected variable costs, a decision for which I conclude there to be no rational basis. Management was in the best position to estimate such costs and Mayer did not perform the detailed look at UFG's operations necessary to justify a deviation from their assumptions.
I note that Mayer criticized Clarke for relying upon large bank holding companies to derive his cost of capital for UFG. Mayer, however, derived his NIM estimate from data that includes banks much larger than UFG. He also derived his NIM from a comparison group that was not comprised solely of bank holding companies, a group which Clarke testified tends to have lower NIMs. Trial Tr. at 786-89.
UFG's CEO Barnett testified that he did not expect a leap to this level given UFG's historical performance and that something "extraordinary" would have been necessary to generate such an increase. Trial Tr. at 370. Mayer failed to identify why a jump of this kind was to be expected and, as important, likely to be sustained. Furthermore, Mayer's testimony that UFG usually performed at NIM levels equal to other banks in his comparison group does not hold up. Excluding years when UFG's participation in the pay-day and sub-prime loan businesses pumped up its NIM, UFG generally lagged Mayer's comparables. As important, Mayer's use of a ten-year average minimizes — without credible explanation — the declining NIM trends displayed in his data in more recent years. Trial Tr. at 108-09, 115-116.
All in all, I find both Mayer's NIM assumptions and his cost assumptions untenable.
Likewise, Mayer's cost of capital of 10.43% is unrealistically low and is based on an implied beta of only 0.3. As noted, UFG was operating under Federal Reserve supervision. It was a troubled financial institution with high leverage. It had defaulted on its loan obligations. It did not have cash to pay off the First Banks debt if the Merger did not go through and it had no readily available source of debt financing. Its efforts to raise equity capital in the recent past had yielded no takers. The idea that a rational, diversified investor would have demanded a rate of return of only 10.43% is hard to accept, given the real risk that UFG would have had to divest one or more of its subsidiaries to address the $12 million First Banks loan. This is especially the case given that UFG would not have been able to provide returns to equity holders until it became well capitalized, an eventuality that does not occur under Mayer's model under after 2005 at the earliest.
Mayer did not adjust the beta from the Ibbotson data despite Ibbotson's own advocacy of an adjustment. Trial Tr. at 165-69.
Indeed, Mayer acknowledges that 97% — I repeat, 97% — of the value incorporated in his DCF model at trial was derived from the terminal value. This back-loading highlights the very real risks that would have been involved in investing in UFG, and undermines the reliability of applying the DCF technique to UFG. Notably, an investor in UFG would not have undertaken this risk in the context of investing in a technology-intensive start-up from which current returns might not be expected but from which very rapid future growth might be achieved. Rather, an investor in UFG would have undertaken the risk of investing in a small, community bank facing growth constraints because it was inadequately capitalized and therefore unable to pay any returns to equity holders for a half-decade or more into the future. To put a point on that, it must be remembered that Denis O'Brien did not regard Union Bank or Jerseyville as vehicles for growth; he had UFG enter riskier fields because he believed the opposite.
See Gray v. Cytokine Pharmasciences, Inc., 2002 WL 853549, at *9 (Del.Ch. April 25, 2002) (noting that the results of a DCF valuation must be regarded with great suspicion and given little weight when the terminal value accounts for over 75% of a DCF analysis).
Under the Management Projections, UFG did not reach even adequately capitalized status until 2003 and remained below the MOU-required well-capitalized level beyond the Projections period.
In my view, Clarke's approach of basing UFG's cost of capital on that of larger banks with active trading in their shares is an acceptable one in view of the differences that exist between UFG and publicly traded small commercial state banks. The proposition that UFG's beta was 70% less than "I" and that its cost of capital was 70% cheaper than the average equity investment defies the known facts, including UFG's inability to find debt or equity financing in the timeframe relevant to the Valuation Date. One reason that smaller banks tend to have lower betas is (I infer) that they tend to be more stable, less-leveraged institutions — characteristics that UFG did not share. Moreover, at trial, Rick Maples, the lead banker from Stifel, testified persuasively that no one would have invested in UFG in 2001 at an expected return of the level used by Mayer, given UFG's prospects and the availability of other investment opportunities in the banking field. Using Clarke's CAPM rate of 16% and applying it to Mayer's other assumptions (including his large inflation of NIM and his sharp reduction in management's estimate of costs) in his trial model results in a per-share value of only $3.54.
Other reasons might include thin trading in their stock and greater sensitivity to regional, rather than national, economic trends.
Although the parties provided me little help in explaining this phenomenon, I note that Mayer's Ibbotson data shows that the smaller banks in the composite had far less leverage than the larger banks in the composite. JX 375. By contrast, UFG's leverage was high even by large-bank standards.
By any objective measure, UFG was highly leveraged. For example, the Federal Reserve's September 30, 2001 Bank Holding Company Report ranked UFG in the 98th percentile compared to its peers for total debt to equity capital and in the 99th percentile for short-term debt to equity capital. JX 359 at 18. On that test, 98% and 99% are the opposite of "A" scores.
See Trial Tr. at 634-44. Stifel's own DCF model confirms the conclusion that UFG's value as a going concern did not exceed the Merger Price, net of synergies. Its model added in synergies to come up with values that, using the most reasonable assumptions about cost of capital, were below the value that I award. See JX 255 at ST-00186, ST-00187.
Mayer's use of the Ibbotson book also failed to take into account the much higher leverage of UFG compared to the banks whose betas generated Mayer's discount rate. According to Brealy and Meyers, the CAPM theory assumes that the equity beta increases with leverage. See Brealy Meyers, supra note 43, at 229 ("Financial leverage does not affect the risk or the expected return on the firm's assets, but it does push up the risk of the common stock. Shareholders demand a correspondingly higher return because of this financial risk." (emphasis deleted)).
In order to ensure that I was being fair to the O'Briens as petitioners, I constructed a DCF analysis of my own using the DCF model Mayer presented at trial. In that model, I used Mayer's adjusted balance sheet, including his (in my view, unjustifiably) reduced variable-cost assumption. I also used a more optimistic NIM than was contained in the Management Projections, increasing the NIM to 4.10 for each year. This increase is not, in my view, justifiable as it was more likely that UFG would perform in accordance with the Management Projections' estimate of NIM. Moreover, the 4.10 NIM assumption represents a higher figure than the five-year trailing average of 3.97 for the same comparison group of banks from which Mayer generated his 4.29 estimate. But the 4.10 estimate gives more weight to recent experience given the lack of evidence that there is anything reliable about a ten-year NIM average.
I note that using Mayer's year-end 2001 adjusted balance sheet but applying the rest of the assumptions in the Management Projections yields a DCF value less than that yielded by Clarke's model. See Trial Tr. at 803-07.
Use of the four- or five-year NIM averages from Mayer's comparison group generated no value under Mayer's trial model even using his 10.43% cost of capital. Trial Tr. at 196.
Finally, although I believe Clarke's use of a 16% CAPM cost of capital was reasonable and is closer to the real cost of capital that an investor in UFG would expect as a price for investing, I used a lower figure based on the same comparison group of state commercial banks from which Mayer derived his 10.43% estimate of UFG's cost of capital. In contrast to Mayer, however, I used the three-factor Fama and French CAPM cost of capital of 13.53%. The advantage of using that formula is that it attempts to better account for certain factors that explain equity return than does the original CAPM. These factors include the relationship of market returns to underlying book value, which is a proxy that, among other things, helps capture the risk associated with possible insolvency and other problems in highly leveraged firms. Although the Fama-French three-factor CAPM model is not wholly accepted, neither is the original CAPM itself. By better factoring in the real risks of leverage, the Fama-French model captures useful data that contributes to a more reliable and real-world cost of capital. In the real world, a diversified investor — even one interested in small, regional banks — would seek a higher return if asked to invest in a highly leveraged small bank under Federal Reserve supervision. Why? Because she could easily avoid the extra risks associated with that bank by investing in other similar firms with much lower leverage.
JX 375.
Id. I note in this regard that the O'Briens did something rather nifty in presenting me with a sensitivity analysis using the Fama-French model. Rather than hew to their original use of a comparison group comprised of all state commercial banks — the group from which Mayer derived his 10.43% cost of capital — the O'Briens presented me with a Fama-French cost of capital based on a group of the smallest state commercial banks. The logic behind this seems simple to comprehend, if not attractive for me to adopt as my own. When presenting its cost of capital at trial, the O'Briens were able to use a lower number by using the median of all state commercial banks (CAPM plus size premium equals 10.43%) than of the small-bank composite (CAPM plus size premium equals 11.38%). The O'Briens justified that decision by noting that UFG was larger than the banks in the small-bank composite. When using the Fama-French model, however, use of the median of all state commercial banks would have resulted in a cost of capital of 13.53%. By switching to the composite of small banks, the O'Briens were able to argue for a cost of capital of only 11.03%. That is, whichever comparison group generated the lowest cost of capital is the one that the O'Briens used.
The risk of an economic downturn to a bank like UFG that has severe cash constraints, a relatively weak loan portfolio, and an inability to amortize its debt is higher than to a more stable, better-capitalized small bank.
See generally Eugene F. Fama Kenneth R. French, The Cross-Section of Expected Stock Returns, 47 J. Fin. 427 (1992); Eugene F. Fama Kenneth R. French, Multifactor Explanations of Asset Pricing Anomalies, 51 J. Fin. 55 (1996). See also Richard A. Brealey Stewart C. Myers, Principles of Corporate Finance 209-212 (6th ed. 2000) (explaining Fama-French three-factor model and the fact that it is possibly a better estimate of the cost of capital because it considers two other factors related to market returns); Tom Copeland et al., Valuation: Measuring and Managing the Value of Companies 265-66 (2d ed. 1995) (noting the problems with the original CAPM in explaining market returns, and citing the advantages of using a multifactor cost-of-capital equation, such as the Fama-French model or the Arbitrage Pricing Model, that takes into account firm size and the ratio of the firm's book value to market price); Ronald J. Gilson Bernard S. Black, The Law and Finance of Corporate Acquisitions 117-33 (2d ed. 1995) (discussing the debate over whether the original CAPM or multifactor variations such as the Fama-French model more accurately measure the cost of capital).
As an additional matter, the use of the Fama-French model seems particularly useful and pertinent in industries — such as banking — for which estimates of book value are considered to be important indicators of value. Eg., UFG Answering Post-Trial Br. App. S (Denis J. O'Brien Dep.), at 67-69, 96-97 (O'Brien admitting that book value is an important value metric for banks); JX 3 at P00248; JX 7 at UFG0005635; JX 13 at UFG0000305. See also Trial Tr. at 590-92 (Maples saying same); JX 121 at UFG0000513. In the banking industry, I note, the mergers and acquisitions discussions center on book value and the price at which institutions sell in comparison to book. Moreover, banks are heavily regulated and their calculation of book value is accordingly well understood, a meaningful insight into value, and reliable. The Fama-French calculus takes the relationship of market price to book value into account.
When using all these factors — i.e., Mayer's balance sheet, a 4.10 constant NIM, and a 13.53% cost of capital — which, I repeat, involve assumptions more favorable to the O'Briens than I believe warranted by the evidence — Mayer's DCF model generates a per-share value for UFG of only $5.44. If a more reasonable NIM of 4.0 is used — a figure in line with the five-year average for banks in UFG's region, a figure significantly higher than UFG achieved in 2001, and a figure higher than that contained in the Management Projections — the value drops to $1.35 per share.
Even using the inadmissible model presented for the first time after trial, these assumptions generate a value of only $7.76 per share. This is lower than the value I award based on the Merger Price.
The inadmissible post-trial Mayer DCF model results in a per share value of $6.52 under these assumptions.
Under Delaware law, it would be appropriate for me to give heavy weight to the value of UFG as implied by a DCF analysis. For example, I could use the generous assumptions I used to test the Merger Price and award the O'Briens $5.44 per share. Or, I could give that value equal weight to the Merger Price.
Rather than do that, I intend to put full weight on the Merger Price, as that Price is the best estimate of value.
X. The Award
For the reasons I have explained, I will award the value of the Merger Price net of synergies, which I estimate to be $8.74. In coming to this figure, I assume that as of the Merger date, it was equally likely that none or all of the two escrowed payments of 80 cents per share would be paid. Pre- and post-judgment interest will be awarded at the legal rate, compounded monthly. Although the respondent argues that I should not compound the pre-trial interest because the O'Briens sought and obtained a postponement of the original trial date, I do not believe that such a sanction is justified. Although the postponement came very late and was not convenient for the respondent or the court, this action did not proceed with the kind of inexcusable torpor that would justify the imposition of simple interest, especially because the respondent has had use of the portion of the Merger consideration attributable to the O'Brien shares during the entire time since the Merger and because the financial market standard is now based on compound, not simple, interest.
As a matter of fact, none of the escrowed funds were ultimately paid as UFG did not satisfy the financial performance conditions in the Merger Agreement necessary for the UFG stockholders to receive payment of those funds.
XI. The Summary Judgment Motions
Before concluding this opinion, I must deal with the summary judgment motions the parties have filed regarding two discrete issues.
The first involves the question of whether 3,699.3154 shares beneficially owned by Denis O'Brien through the UFG Employee Stock Option Plan should be included with the shares eligible for appraisal. O'Brien alleges that the ESOP trustee, the Jerseyville bank subsidiary, breached a duty owed to O'Brien when it failed to seek appraisal for his shares. The problems for O'Brien's position are several. First, he never demanded that Jerseyville seek appraisal for these shares, even though Jerseyville, as trustee, informed him that a mere instruction to vote against the merger would not be regarded as a demand for the trustee to seek appraisal. Thus, O'Brien concedes that no appraisal demand was made for his ESOP shares. Second, to the extent that O'Brien argues that Jerseyville breached its fiduciary duties as trustee by failing to provide him with clear instructions about how to seek appraisal, as well the option of doing so, he seeks relief that is beyond the proper scope of this action. His dispute is essentially between himself as beneficial holder and Jerseyville as record holder, in a circumstance when it is undisputed that neither he nor Jerseyville demanded appraisal as to his ESOP shares. Appraisal actions must be confined to the issues contemplated by § 262. The respondent's motion for summary judgment that the ESOP shares are not part of the appraisal class is granted.
See Cede Co. v. Technicolor, Inc., 542 A.2d 1182, 1187 (Del. 1988); Andaloro v. PFPC Worldwide, Inc., 830 A.2d 1232, 1234 (Del.Ch. 2003).
The second issue involves the attempt by the Douglas Alan O'Brien Family Irrevocable Trust (the "Douglas Trust") to withdraw part of its holdings from the demand for appraisal, while continuing to seek appraisal as to the remainder of its UFG shares. The parties agree that the Douglas Trust voted all of its 37,691.67 shares against the merger. The Douglas Trust then demanded appraisal as to all of its shares but later purported to withdraw that demand as to 21,875 of its total shares within the time permitted by § 262(e). The question is whether it could do so or whether its attempt to withdraw a portion of its total shares constituted a withdrawal of all of its shares.
This is an open issue of Delaware law. The statute's terms appear to treat each stockholder as voting and demanding appraisal as to its shares as an aggregate. Moreover, § 262(e) gives "any stockholder" the "right to withdraw such stockholder's demand for appraisal." This language tends to support the respondent's argument that a partial withdrawal is not contemplated or permitted by the statute. What the Douglas Trust possessed was the right to "withdraw" its "demand for appraisal." Once it did so, it had to withdraw all of its shares as its singular demand was no longer in existence.
See 2 David A. Drexler et al., Delaware Corporation Law and Practice § 36.04, at 36-09 to 36-10 (2002); 2 Rodman Ward, Jr. et al., Folk On The Delaware General Corporation Law § 262.4.2, at GCL-IX-198 to GCL-IX-199 (4th ed. 2003).
See 8 Del. C. § 262(a) (stockholder who has not voted in favor of a merger is entitled to an appraisal in certain circumstances to determine the "fair value of the stockholder's shares of stock"); id. § 262(d) (stockholder "electing to demand appraisal" shall deliver a written demand for "such stockholder's shares").
Complicating this reading, however, is venerable case law. It is understood by now that an entity like Cede Co. that is record holder (but not beneficial holder) of a company's shares can vote certain of those shares against a merger, and others in favor, and seek appraisal as to the dissenting shares. This makes sense, of course, because Cede often represents a large number of beneficial holders in the same company, holders whose views on the advisability of a merger might diverge. Still, because § 262 deals with record holders, tolerance of this practice cuts against a hyper-literal reading. Cutting even more against this is the teaching of the Supreme Court in Olivetti Underwood Corp. v. Jacques Coe Co. Therein, the Supreme Court stated — in response to an argument as to why the respondent corporation had a right to inquire whether a record holder was acting in accord with the wishes of beneficial holders in seeking appraisal — that it would permit a stockholder to accept the consideration from a § 253 merger for some of her shares and seek appraisal as to the remainder. In so ruling, the Court stated: "Insofar as that choice [between accepting the merger consideration and demanding appraisal] is concerned, we find no valid reason in the letter or spirit of the Statute, in the cases, or in considerations of fairness, which would bar a stockholder from `hedging' his position by electing to accept the offered price as to some of his stock and demanding appraisal as to the rest. We think that stockholders should have such flexibility and freedom of choice; and we so hold."
See, e.g., Reynolds Metal Co. v. Colonial Realty Corp., 190 A.2d 752 (Del. 1963) (stockbroker holding shares in its name for customers could seek appraisal as to shares it voted against merger even though it voted other shares it held in its name for other customers in favor of the merger).
217 A.2d 683 (Del. 1966).
Id. at 687 (emphasis added).
Here, I face similar situation. The Douglas Trust did what it could to stop the Merger. It voted all of its shares against it; thus the Douglas Trust did not hedge its bets on the merger vote itself. There was no option after the merger vote for the Douglas Trust not to be cashed out. The question is then like the one the Supreme Court confronted in Olivetti: Can the Douglas Trust hedge its bets on the appraisal award? Or does a different and more stringent rule apply for stockholders who are both record holders and beneficial holders than applies to entities like Cede? Given the teaching of Olivetti, I am hard-pressed to read § 262 as precluding a partial withdrawal of shares by a beneficial holder who did not vote any shares in favor of the merger. This is particularly so when there are serious collective-action and cost constraints that make appraisal a sometimes-difficult remedy for stockholders to pursue. During the pendency of an appraisal, the petitioners' shares are held and the petitioners receive no part of the merger consideration. Hedging, as the Douglas Trust has done, is one way a dissenter might choose to help defray the cost of proceeding with the appraisal proceeding. Olivetti read § 262 as not precluding stockholders from exercising tactical options of this kind and the petitioners contend that I should leave it to the Supreme Court to conclude that its own precedent should be altered, or interpreted as not applicable to this situation. I believe that the petitioners are correct and grant their motion for summary judgment. For obvious reasons, I also reject the respondent's argument that hedging of this kind constitutes acquiescence; Olivetti obviously teaches otherwise.
See, e.g., Andra v. Blount, 772 A.2d 183, 194-96 (Del.Ch. 2000) (discussing "Litigation-Cost Benefits" of unfair dealing action as compared with appraisal remedy).
It is, I suppose, technically possible to distinguish Olivetti on narrow statutory grounds. By its plain terms, Olivetti contemplated a situation in which a stockholder would make a demand for appraisal as to only some of its shares, with the remainder never being part of the appraisal demand Here, I deal with a situation in which the Douglas Trust made a single demand for appraisal as to all of its shares. By withdrawing under § 262(e), the Douglas Trust withdrew its (by statutory language, singular) "demand," which pertained to all of its shares. But this same kind of reasoning would preclude Cede and other record holders from engaging in long-accepted position-splitting and would also preclude the very conduct that Olivetti expressly authorized. I note that the respondent argues for a policy interpretation of § 262 that has literalism go out the window as to record holders like Cede but not as to record holders who are also beneficial holders. This policy-based distinction might well be sensible but it does not manifest itself in § 262. This contrasts with the Model Business Corporation Act, which explicitly permits record holders, but not beneficial holders, to split their shares. See Model Bus. Corp. Act Annotated § 13.03, at 13-44 to 13-46 (3rd ed. 2000-02 Supp.). The policy choice made by the Model Business Corporation Act might commend itself to the General Assembly, which is free to disagree with Olivetti and to clarify this area of our law through a statutory amendment. I, however, am not and I find no credible basis to conclude that the teaching of Olivetti does not logically apply to the circumstances of this case.
XII. Conclusion
The parties shall present an implementing final order within ten days.