Summary
finding management projections unreliable in the context of an appraisal action because, among other reasons, management themselves did not regard them as reliable
Summary of this case from OptimisCorp v. WaiteOpinion
C.A. No. 19734.
Submitted: April 1, 2004.
Decided: May 20, 2004. Revised: May 21, 2004.
Norman M. Monhait, Esquire, ROSENTHAL, MONHAIT, GROSS GODDESS, P.A., Wilmington, Delaware; Thomas M. Skelton, Esquire, LOWEY DANNENBERG BEMPORAD SELINGER, P.C., White Plains, New York, Attorneys for the Petitioners.
Alan J. Stone, Esquire, James G. McMillan, III, Esquire, MORRIS, NICHOLS, ARSHT TUNNELL, Wilmington, Delaware, Attorneys for the Respondents.
MEMORANDUM OPINION
I.
This is an appraisal action, pursuant to 8 Del. C. § 262, filed as a result of a merger that cashed-out the petitioners' shares at a price of $28 per share. Both parties presented expert testimony to determine the fair value of the shares as of the merger date. For the reasons herein, the court concludes that the fair value of the shares as of the merger date is $32.76.
II.
A. Background
1. The Parties
Travelocity.com Inc. ("Travelocity"), a Delaware corporation, is the surviving entity of a merger between it and Travelocity Holdings Sub Inc. ("Holdings"), a wholly owned subsidiary of Sabre Holdings Corporation ("Sabre"). Because Sabre, through Holdings, owned more than 90% of the outstanding shares of common stock of Travelocity, the merger was authorized by Sabre's board of directors pursuant to 8 Del. C. § 253 and became effective on April 11, 2002 (the "Merger Date"). As a result of the merger, Travelocity is (again) a wholly owned subsidiary of Sabre.
Both Holdings and Sabre are Delaware corporations.
Travelocity's common stock began to trade on the public market in the first quarter of 2000 when Sabre, which owned 100% of Travelocity's business as a division, purchased Preview Travel, a publicly traded company. The combined entity was named Travelocity. Resp'ts Pre-Trial Br. at 5.
The petitioners owned 265,540 shares of Travelocity before the merger and were entitled to demand an appraisal of those shares pursuant to Section 253(d) of the DGCL. The parties stipulate that the petitioners have complied with the provisions of 8 Del. C. § 262, in timely filing their petition for appraisal and in perfecting their right to appraisal.
As of April 11, 2002, petitioner Cede Co. was the record holder of the 265,540 shares of Travelocity common stock, for the benefit of: Doft Co. Inc. (61,500 shares); First Trust Corp., as trustee FBO Alan Doft (3,800 shares); Elisabeth H. Doft (43,000 shares); Laurence Hoffman (300 shares); Maria Ivkovic (1,000 shares); Shirel Partners (8,000 shares); Blanche Romie Shapiro Charitable Remainder Unitrust dated 9/1/95 (4,000 shares); Edna R. Hoffman (600 shares); Blanche Shapiro 1999 Trust (8,000 shares); MJR Partners (124,340 shares); and, DB Securities, Inc. as custodian FBO Morton M. Maneker IRA dated 12/03/01 (11,000 shares).
2. The Online Travel Industry
Travelocity is in the business of providing online travel services. When Travelocity went public in 2000, the online travel industry was in nascent form and the future of the online travel industry was uncertain. By early 2001, the online travel industry was beginning to show profitability. By that time, Travelocity was the leading online travel agency.
The events of September 11, 2001, however, created great uncertainty in the online travel business. Even though the industry slowed in the period after September 11, analysts predicted that the negative effect would be temporary. Travelocity, however, also faced strong competition in the market at this time. Expedia, Travelocity's main competitor, surpassed Travelocity as the industry leader in early 2002 and Orbitz, a then brand new travel services provider, had become the third largest online travel agent in less than a year.
See Gompers Expert Report at ¶ 95.
Expedia quickly became more successful than Travelocity because of its early implementation of the "merchant model." The merchant model is a business plan in which travel agencies purchase the airline tickets, hotel rooms or car rentals at a negotiated rate from the suppliers and then resell them directly to consumers at a higher price. In the traditional agency model then used by Travelocity, the travel agent merely serves as a liaison between the supplier and the customer and receives a commission for the sale. The merchant model generates higher profit margins and much higher cash flows than the traditional agency model because the travel agent controls the price and works directly with both the supplier and the consumer.
In the fourth quarter of 2001, 42% of Expedia's revenues came from merchant model business while only 3.5% of Travelocity's revenue was from merchant model business, specifically merchant airline ticket sales.
Resp'ts Pre-Trial Br. at 6. Michelle Peluso, Travelocity's COO, described in great detail the benefits of the merchant model approach for both the stand-alone hotel business and for dynamic packaging and how "being a merchant is really critical to Travelocity's growth and profitability." Trial Tr. at 240.
Moreover, Travelocity was limited in its ability to develop the merchant model business because it was committed to working in partnerships with other entities to help build its merchant business. This partnership approach negatively affected Travelocity's ability to reap the full benefits of the merchant model business. For example, Travelocity had an exclusive contract running until 2005 with Hotel Reservations Network ("HRN"). HRN controlled Travelocity's relationships with hotels, the booking of hotel rooms and the price markup. Travelocity received only a commission on sales. Therefore, HRN and not Travelocity enjoyed the benefits of the merchant model plan.
Travelocity was also partnered with Contour, a small, start-up software company, to provide vacation package deals to its customers. Pursuant to its contract with Travelocity, Contour had control over the technology developed in setting up the vacation packaging business and Travelocity paid Contour a fee for changes in the software. Therefore, Travelocity was also sharing its profits with Contour. In addition, Travelocity was partnered with other entities to facilitate entry into the international market, and was dependent on third-party relationships with Internet portals, like AOL and Yahoo!, to direct consumers to its website.
Peluso testified that Travelocity should not continue outsourcing its merchant business and packaging business to Contour. She further testified that Travelocity's management had not as of the time of the merger decided whether it would continue to use Contour for the merchant model hotel business. Id. at 243.
For example, Michael Gilliland, Travelocity's president and CEO as of May 2002, testified that about 15% of Travelocity's revenue came from its contract with Yahoo! and that Yahoo! Travel was in fact a private label for Travelocity. Id. at 188.
Additionally, airlines began reducing the traditional commissions paid to travel agencies for airline tickets in the mid-1990s. The airlines specifically targeted online travel agents in mid-2000 and began actively cutting commissions for online travel agents. In June 2001, in a further effort to reduce the commissions paid to online travel agencies, five major airlines created Orbitz to sell discounted airfares directly to online consumers. Orbitz had exclusive access to the discounted web fares offered by its owners and online travel agents were forced to renegotiate their relationships with major airlines in order to have access to web fares. Travelocity was hit harder than Expedia by Orbitz's formation and the resulting competition because it was still heavily dependent on airline ticket commissions while Expedia enjoyed substantial revenue from merchant model sales independent of those commissions.
For example, Northwest Airlines announced in early 2001 that it would no longer pay commissions to online travel agents. Id. at 162.
Gilliland testified to the extent that the formation of Orbitz and the general commission-cutting by airlines affected Travelocity because Travelocity, unlike Expedia, had not made progress in diversifying its revenue. Id. at 169.
Even though Travelocity was facing tough competition from Expedia in the fourth quarter of 2001, analysts expressed the belief that the gap in performance was temporary and that Travelocity would continue to be competitive. In fact, Travelocity's performance in early 2002 was ahead of the management forecast. Commenting on Travelocity's forecast for 2002, one analyst noted: "Although the addition of vacation packages comes two quarters after a similar move by Expedia, we believe that the leisure market remains large enough to support at least two dominant players, and we expect Travelocity to narrow the gap between Expedia over 2002."
See id. at 44.
Id. Ramesh Punwani, Travelocity's executive vice president and CFO until April 2002, however, testified that Travelocity had actively reduced operating expenses in the first quarter of 2002 in order to meet the proposed budget plan. Punwani's testimony on whether Travelocity would meet its projections for the year was pessimistic at best. Notably, Punwani and Brett Little, Travelocity's controller, were the top financial officers at Travelocity before the merger and were responsible for preparing Travelocity's projections and providing this forward-looking information to Sabre. Id. at 416.
Gompers Expert Report at ¶ 103, quoting CIBC World Markets, Travelocity, 4Q EPS As Expected; 2002 Growth Of, Jan. 17, 2002, p. 6.
In March 2002, Travelocity purchased Site59, a small online company with a limited working merchant model business. Site59 is in the business of providing last minute travel bookings at a substantial discount for customers who are flexible in their travel schedules. Travelocity's goal in acquiring Site59 was to build a merchant model hotel business and continue Site59's last-minute packaging business. Travelocity purchased Site59 with a loan from Sabre.
Trial Tr. at 197. Gilliland testified that the acquisition of Site59 did not help Travelocity in its competition with Expedia in the merchant model hotel business. He distinguished the merchant model hotel business from the last-minute packaging business. Id. Peluso, however, testified that Travelocity was working toward a merchant model hotel business. She testified that the process was extensive and that the goal was to acquire contracts with 4,000 hotels in an 18-month period. This number would put Travelocity at half the size of its competitors after 18 months. Peluso's testimony is indicative that in the purchase of Site59 Travelocity acquired "a team of people who had the ability at any given time with our merchant business to change the margins by hotel, by city, by chain, by region, depending what the market conditions are." Id. at 233. Therefore, even though Travelocity was handicapped by its contract with HRN and other partnership relationships, it was actively working toward regaining its competitive position, as evidenced by its purchase of Site59.
3. The Merger
In early 2001, Sabre began to consider buying back the public shares of Travelocity. Sabre launched "Project Tango" to examine the online travel business and to make recommendations on Travelocity's business approach. By September 2001, Project Tango was finished and resulted in the conclusion that "it was vital for Sabre to `own the customer,' to build strong relationships with suppliers and to grow its online presence."
Resp'ts Pre-Trial Br. at 9-10.
On February 16, 2002, William J. Hannigan, Sabre's Chairman and CEO, contacted Terrell B. Jones, Travelocity's President and CEO, and F. William Conner, a Travelocity director, and advised them that a Sabre board meeting was scheduled for February 18, 2002 to discuss a $23 per share cash tender offer to acquire all shares of Travelocity not already owned by Sabre. Hannigan also informed them that it was Sabre's intention, pending Sabre's board approval, to confirm the proposal in writing on February 18, and then publicly disclose the offer. On February 18, with the Sabre board's approval, Hannigan advised Travelocity's board members of its intention to start the $23 per share cash tender offer. That same evening, the Travelocity board met and established a special committee comprised of its two independent directors, Conner and Kathy Misunas (the "Special Committee"). The Special Committee retained Salomon Smith Barney, Inc. ("Salomon") as its financial advisor and Locke, Liddell Sapp LLP as its legal counsel.
On February 19, 2002, Sabre announced that it intended to make a tender offer for all of the outstanding publicly held Travelocity shares. The Special Committee then sent Sabre a letter inquiring if it would be interested in exploring alternatives to the going private proposal, such as the sale of some of Sabre's interest in Travelocity to a third party. Sabre responded the following day that it was not interested in selling any of its equity interest in Travelocity. On March 4, 2002, the Travelocity board met and the Special Committee delivered its initial report regarding Sabre's offer of $23 per share. At the meeting, Salomon representatives advised the board, orally and in writing, that the $23 per share offer was inadequate. The Special Committee then presented its report, concurring that Sabre's initial offer was inadequate.
Nonetheless, on March 5, 2002, Sabre began a $23 per share cash tender offer for all of Travelocity's publicly held common shares. On March 18, 2002, Sabre amended the offer by increasing the offering price to $28 cash per share. On March 18, 2002, the Special Committee and the Travelocity board voted to recommend that the Travelocity stockholders accept Sabre's amended offer and tender their shares. Sabre succeeded in acquiring approximately 95% of the outstanding shares of common stock by the close of the offer. Then, Sabre acted to effect the short-form merger under 8 Del. C. § 253. The merger became effective on April 11, 2002. Pursuant to the merger, the publicly held shares of Travelocity were converted into the right to receive $28 per share.
Both Special Committee directors voted in favor, one director voted against, and six directors abstained from the vote.
In accordance with 8 Del. C. § 262, the petitioners now seek a determination of, and payment for, the fair value of the Travelocity shares they held on the Merger Date.
B. The Experts
The petitioners' trial expert was William H. Purcell. Purcell has a B.A. in Economics from Princeton University and an M.B.A. from New York University. He has been an investment banker for more than 35 years, 24 years of which are with Dillon, Read Co. Inc. Over the span of his career, Purcell has worked on approximately 100 merger and acquisition related projects. He has performed numerous financial valuations of private and public companies in various industries. He also served as advisor to special committees of boards of directors in connection with corporate transactions. Purcell has testified many times as an expert regarding a wide range of investment banking matters, including a number of valuation issues. He has also testified as an expert before various regulatory agencies, including the Securities and Exchange Commission.
Purcell testified that the going concern value of Travelocity was at least $35 per share as of March 16, 2002. Purcell testified that he relied primarily on the most recent set of management projections in his valuation analysis. Purcell also looked to analyses performed by third parties to test the validity of his conclusions.
Purcell's valuation was as of March 16, 2002, the date that Salomon presented its fairness opinion to the Travelocity board regarding the merger. Therefore, Purcell's valuation, like Salomon's, does not factor in the Site59 acquisition. Notably, both experts treated the Site59 acquisition on a stand-alone basis because of its proximity to the merger date. Purcell did not add any incremental value to Travelocity as a result of the acquisition.
Specifically, Purcell looked at the Goldman Sachs presentations to Sabre and the Salomon presentations to the Special Committee. He also looked at analyses by various securities and industry analysts studying Travelocity in 2001 and 2002.
Travelocity's trial expert was Professor Paul A. Gompers of the Harvard Business School. Gompers has an A.B. in Biology from Harvard College, a M.Sc. in Economics from Oxford, and a Ph.D in Business Economics from Harvard University. He was an assistant professor of Finance and Business Policy at the Graduate School of Business at the University of Chicago for two years before joining the Harvard Business School faculty. He is also the Director of Research at the Harvard Business School and his research focuses on financial issues, valuation financing, and the markets related to young, growing technology companies. Although Gompers had never before testified as a trial expert, he had been retained 15 times as an expert in the area of finance and valuation of emerging technology companies in other legal matters.
Gompers reviewed various documents and materials on the online travel industry in general, as well as internal documents of Sabre and Travelocity. He also conducted interviews with some Sabre and Travelocity personnel. Gompers reached the conclusion that the going concern value of Travelocity as of the Merger Date was $20 per share.
Gompers did not interview Peluso or Gilliland in connection with his valuation analyses. Trial Tr. at 355.
Gompers did include the incremental value of the Site59 acquisition in his valuation. Gompers valued the acquisition to Travelocity stockholders at $36.2 million, approximately $0.72 per share in incremental value. Gompers Expert Report at ¶¶ 30, 203. Purcell stated that he would add at least $0.72 per share of incremental value if he accepted Gompers's criticism that the acquisition should be factored into the valuation. Purcell Rebuttal Report at ¶ 14(a).
C. The Valuation Methods Used
Both experts used essentially the same methods to value Travelocity's stock; i.e. a discounted cash flow analysis ("DCF") and a comparable company analysis. In performing their comparable company analyses, both Purcell and Gompers used Expedia as the single comparable company. Despite the similar approaches taken, the results arrived at by Gompers and Purcell vary widely. Gompers opines that, on a DCF basis, Travelocity common stock was worth between $11.38 and $21.29 per share. Using the same methodology, but using different inputs, Purcell opines that a share of Travelocity common stock was worth between $33.70 and $59.95 as of the Merger Date. The two experts' comparable company analyses also yield significantly divergent results because they disagree about the appropriate discount to apply to reflect Travelocity's competitive disadvantages.
III.
Pursuant to 8 Del. C. § 262, the petitioners are entitled to their pro rata share of the fair value of Travelocity's common stock as of the Merger Date. Fair value, as used in an appraisal setting, is defined as "the value of the Company to the stockholder as a going concern, rather than its value to a third party as an acquisition." Moreover, section 262(h) requires this court to determine fair value "exclusive of any element of value arising from the accomplishment or expectation of the merger." "In a statutory appraisal proceeding, both sides have the burden of proving their respective valuation positions by a preponderance of the evidence." The court may exercise independent judgment to assess the fair value of the shares if neither party meets its burden.
Gray v. Cytokine Pharmasciences, Inc., 2002 WL 853549, at *6 (Del.Ch. Apr. 25, 2002).
Id. (citation omitted).
Taylor v. Am. Specialty Retailing Group, Inc., 2003 WL 21753752, at *2 (Del.Ch. July 25, 2003), ( quoting M.G. Bancorporation, Inc. v. LeBeau, 737 A.2d 513, 520 (Del. 1999)).
Taylor, 2003 WL 21753752, at *2.
IV.
In determining the fair value of Travelocity's shares, the court may consider "proof of value by any techniques or methods which are generally considered acceptable in the financial community and otherwise admissible in court." Both parties used a DCF approach and a comparable company approach to value the shares. DCF involves projecting operating cash flows for a determined period, setting a terminal value at the end of the projected period, and then discounting those values at a set rate to determine the net present value of a company's shares. It is an exercise in appraising the present value at a set date of the expected future cash flows earned by the company. A DCF analysis is a useful tool for valuing shares and is frequently relied on by this court in appraisal actions.The utility of a DCF analysis, however, depends on the validity and reasonableness of the data relied upon. As this court has recognized, "methods of valuation, including a discounted cash flow analysis, are only as good as the inputs to the model." The problem in this case is that the most fundamental input used by the experts — the projections of future revenues, expenses and cash flows — were not shown to be reasonably reliable.
Neal v. Ala. By-Products Corp., 1990 WL 109243, at *9 (Del.Ch. Aug. 1, 1990) (citing S. Pratt, Valuing a Business: The Analysis and Appraisal of Closely Held Companies, 84 (2d. ed. 1989)), aff'd, 588 A.2d 255 (Del. 1991).
"Inputs in a discounted cash flow are predictions which are necessarily speculative in nature. The quality of these predictions is therefore central to the reliability of the underlying methodology." Harris v. Rapid Am. Corp., 1990 WL 146488, at *6 (Del.Ch. Oct. 2, 1990), aff'd in relevant part, and rev'd on other grounds, 603 A.2d 796 (Del. 1992).
Delaware law clearly prefers valuations based on contemporaneously prepared management projections because management ordinarily has the best first-hand knowledge of a company's operations. Here, management prepared the 5-year projections for the period 2002-2005 and gave them to Sabre for use in its routine planning processes. Often, projections of this sort are shown to be reasonably reliable and are useful in later performing a DCF analysis. In this case, however, the court is persuaded from a review of all the evidence that the Travelocity 5-year plan does not provide a reliable basis for forecasting future cash flows.
See Gilbert v. MPM Enters., Inc., 709 A.2d 663, 669 (Del.Ch. 1997), aff'd, 731 A.2d 790 (Del. 1999) (concluding that management was in the best position to forecast the company's future before the merger); Gray, 2002 WL 853549, at *8 (rejecting valuation that inexplicably ignored management projections).
To begin with, Travelocity's management held the strong view that these projections should not be relied upon because the industry was so new and volatile that reliable projections were impossible. At trial, Punwani, Travelocity's CFO, characterized the 5-year projections as "simulations" and "thought studies" and said that they were never reviewed by any of the operating departments at Travelocity. Punwani further testified that because of the limited financial history of Travelocity, together with a rapidly evolving marketplace, it was difficult "to forecast the next quarter, let alone five years out." He also confirmed that the events of September 11 led to more doubt about the future of the industry and Travelocity's positioning in the market.
In Gray, the court relied on management's determination that their projections were reliable even though prepared in an industry with a high degree of speculation due to the facts that the company's product required regulatory approval and that there was an unknown market share for drug delivery products. In Gray, management submitted their projections to Merrill Lynch to use in its independent valuation of the company's shares in connection with a merger. Gray, 2002 WL 853549, at *8. To reflect the inherent risks involved in achieving those projections, Merrill Lynch applied a discount rate as high as 50%. Id. at *11. Those facts are easily distinguishable from the case at hand Here, Punwani clearly testified that the projections that he prepared were merely speculative and too unreliable to give to Salomon in their independent valuation of Travelocity.
Trial Tr. at 381.
Id. "We were really not in a position to be able to put any credence on the numbers, both on the revenue and on the cost side. And the only way to get credibility in our numbers would have been to take those models and put them through reasonability checks . . . [that] were never done because, when we built these frameworks, I'll call them, in the year 2000, we were in a period of explosive growth. We were growing at 150 percent per year. . . . No one really knew what the right number was." Id. at 381-82.
Id. at 383. "It was bad enough before when we did the data, and we had this new variable that got thrown into our lap, which totally destroyed our ability to have any confidence in projections beyond one quarter out." Id.
Although it was aware of the 5-year forecasts, Salomon did not conduct a DCF analysis of Travelocity as part of its work in connection with the merger. The testimony of Anwar Zakkour, Salomon's managing director, is especially relevant on this issue:
Purcell notes that "it is very unusual for an investment banking firm not to employ a DCF analysis in a valuation study, appraisal study or in a fairness opinion." Purcell Rebuttal Report at ¶ 5.
Goldman Sachs did do a DCF analysis of Travelocity for Sabre. The Goldman Sachs report, however, is not helpful for this court's inquiry into the fair value of Travelocity as of the Merger Date. First, it was prepared nine months before the merger and before September 11. Second, it was prepared for Sabre, not Travelocity. Third, there is nothing in the record that indicates that Goldman Sachs used Travelocity's management projections in its analysis. In fact, the record shows the opposite. Punwani testified that Sabre did not have direct access to Travelocity's financial data and that the projections he did give to Sabre were highly qualified as to their reliability. Trial Tr. at 372-73. Punwani also testified that he did not have any discussions with Sabre regarding the use of his numbers in Goldman Sach's DCF analysis. Id. at 415-16. Furthermore, Gilliland testified that these management projections "are about as good as the weight of the paper they're written on." Id. at 158.
Q. Did Salomon Smith Barney prepare a discounted cash flow analysis of Travelocity in connection with this transaction?
A. Absolutely not.
* * *
Q. Why was no discounted cash flow analysis prepared in connection with this transaction?
A. Because this was an industry that was in flux. And the management team itself, which should have been the team that was most able to put together a set of projections, would have told you it was virtually impossible to predict the performance of this company into any sort of reasonable future term. And they in fact had very little confidence with even their 2002 forecast numbers because of that.
September 11th didn't help the pace of migration from off-line to online. It didn't help. The airlines being very focused on cutting their distribution costs didn't help. These were all things that were happening real time. Travelocity going from being the number one player to being very unfavorably compared to Expedia and certainly losing its number one position to them in a very short time didn't help. These are all things that support that. And other than maybe God himself, I suspect nobody could really predict what this business is going to do in the next five years.
Q. Is a discounted cash flow methodology a methodology that is commonly used by Salomon Smith Barney in valuing companies?
A. Valuing mature companies, yes.
Zakkour Dep. at 35-37.
Purcell's DCF relies more or less uncritically on the Travelocity 5-year plan. Purcell justifies his reliance on these projections because they were provided to Sabre for its 5-year planning and later used by Goldman Sachs in its presentations to Sabre. Punwani, however, explained at trial that these numbers were given to Sabre as a routine requirement for Sabre's internal planning process and with express caveats as to their reliability, and that he personally told both Sabre's CFO and controller that the numbers were only simulations. Moreover, Punwani was presented on cross-examination with several Sabre documents showing projections for Travelocity, and testified credibly that he had never seen the documents before nor was he familiar with how Sabre used Travelocity's projections in its business planning. Despite the normal preference for management projections, the court concludes that the petitioners failed to prove that Purcell's reliance on these projections was justified. Thus, the court must disregard Purcell's DCF analysis.
Purcell's DCF is flawed for other reasons that the court will not describe in detail. Generally, Purcell makes certain assumptions and observations that are unsubstantiated in his report or in his testimony. Moreover, his report only provides the court with only the most skeletal mathematical calculations to back up his analysis.
Trial Tr. at 383-84, 410.
Id. at 412-16.
Gompers takes a different approach, after concluding that the 5-year projections were "merely meant as a rough plan and were considered to be optimistic targets" and not a reliable basis for a DCF analysis. Instead of eschewing a DCF analysis, however, Gompers sets about to create a new set of projections, covering periods of 10 and 15 years into the future, based on his expert analysis of Travelocity and post-merger discussions with certain members of its management. As a preliminary matter, this court is inherently suspicious of post-merger, litigation-driven forecasts because "[t]he possibility of hindsight bias and other cognitive distortions seems untenably high." As important, in this case, Gompers's exercise is strikingly at odds with the views of Travelocity management and Salomon that no one could reliably predict Travelocity's future cash flows.
Gompers Expert Report at ¶ 27.
Agranoff v. Miller, 791 A.2d 880, 892 (Del.Ch. 2001).
The reliability of Gompers's projections is further undermined by the fact that he selectively picks and chooses variables from management's 5-year forecast that conveniently fit into his exercise in creating less "optimistic" projections. Although Gompers's valuation is facially more credible than Purcell's, in that he provides both the numerical calculations and the academic theories for his assumptions, his selective reliance on aspects of management's projections is suspect. Gompers starts reasonably by using Travelocity's 2002 revenue projection, adjusted for Travelocity's actual performance in the first three months of 2002. He then generates 10-year and 15-year revenue projections by assuming that the revenue growth rate will (i) decrease in a linear fashion to 17.2%, the 2005 revenue growth rate found in the 5-year forecast, and then (ii) will continue to slow in a linear fashion until it reaches the "steady state of growth" in 2011 or 2016. Gompers does not explain why only the 2005 growth rate from the Travelocity 5-year plan is reliable and ignores that the Travelocity 5-year plan predicted much higher intervening growth rates. Gompers then uses the operating margins found in the Travelocity 5-year plan through 2005 and uses the 2005 operating margin in perpetuity to derive his projections for operating income.
Trial Tr. at 342.
Gompers Expert Report at ¶ 155. Gompers testified that it is standard practice in both his teaching and his valuation exercises to project out for a longer period of time when valuing start-up companies or venture capital firms with the goal of the young industry reaching a steady state of growth: "most of the time when I ask my students to project out, or if I do it as a board member, or do it as an advisor to people raising capital, I tell them to project out ten years or fifteen years, to get out to the point where the industry is more mature and their prospects look as though they grow in line with the overall economy." Trial Tr. at 277-78.
See In re Emerging Communications, Inc. S'holder Litig., 2004 WL 1043794, at * 15 (Del.Ch. May 3, 2004) (dismissing an expert's unsubstantiated adjustments to management projections because the "adjustment amounts to [the expert] substituting his personal judgment of what [the input] should be for the non-litigation business judgment of [the company's] management.").
"None of the long-term forecasts were provided to or approved by senior management or the board o[f] directors, much less the public. Projections beyond 2003 were merely meant as a rough plan and were considered to be optimistic targets, i.e., they were not the expected cash flows." Gompers Expert Report at ¶ 27.
The respondents argue that this selective use of management projections is acceptable because "they are reasonable or somewhat optimistic" and that since the petitioner's valuation wholly relies on the Travelocity 5-year plan that it is somehow estopped from arguing that Gompers selective use is unacceptable. Neither of these arguments is persuasive. The only reasonable conclusion the court can draw from the record evidence is that no one, including Professor Gompers, is able to produce a reliable set of long-range projections for Travelocity, as of the Merger Date. This conclusion is substantially reinforced by the observation that Gompers's DCF produced values ranging from $11.38 to $21.29 relative to a squeeze-out merger in which Travelocity's 70% parent agreed to pay $28 per share to acquire the minority interest.
For these reasons, the court reluctantly concludes that it cannot properly rely on either party's DCF valuation. The goal of the DCF method of valuation is to value future cash flows. Here, the record clearly shows that, in the absence of reasonably reliable contemporaneous projections, the degree of speculation and uncertainty characterizing the future prospects of Travelocity and the industry in which it operates make a DCF analysis of marginal utility as a valuation technique in this case. If no other method of analysis were available, the court would, reluctantly, undertake a DCF analysis and subject the outcome to an appropriately high level of skepticism. The court, however, now turns to the other method of valuation offered by the parties.
D. The Comparable Company Approach
A comparable company analysis is often used in connection with a DCF analysis. The court, however, may use a comparable company valuation on a stand-alone basis in an appraisal action when it is the only reliable method of valuation offered by the parties. In Borruso v. Communications Telesystems Int'l, the court relied on a comparable company analysis because neither expert was comfortable using a DCF analysis to value the company's shares due to the limited financial data of the company available as of the merger date. 753 A.2d 451, 455 n. 5 (Del.Ch. 1999).
The comparable company approach entails the review of publicly traded competitors in the same industry, then the generation of relevant multiples from public pricing data of the comparable companies and finally the application of those multiples to the subject company to arrive at a value. The true utility of a comparable company approach is dependent on "the similarity between the company the court is valuing and the companies used for comparison." Both experts and Salomon use Expedia as the single comparable company in their analyses, but disagree on the appropriate discount to be applied to the multiples derived from their analyses of Expedia. The court agrees that Expedia is clearly comparable to Travelocity.
See Taylor, 2003 WL 21753752, at *7.
Gray, 2002 WL 853549, at *9 (quoting In re Radiology Assoc., Inc. Lit., 611 A.2d 485, 490 (Del.Ch. 1991)).
Gompers does not challenge Salomon's valuation, but he dismisses Purcell's valuation because "it is applied in an ad hoc manner with little understanding of the proper measure of comparison and the factors that affect comparable multiples." Gompers states that the discount to Expedia should be at least 40% and concludes that Travelocity's valuation as of the merger date is $22.08.
Gompers Expert Report at ¶ 172.
Gompers testified that even a 60% discount could be justified. Trial Tr. at 332.
Purcell critiques Gompers's valuation in that it is significantly lower than any valuation done of Travelocity and, more importantly, inexplicably less than the $28 paid by Sabre in the merger. Purcell also criticizes Gompers's comparable company analysis in that it is "wildly divergent" from his DCF calculation when Gompers states that his comparable company valuation serves as a check on his DCF. Purcell states that a 10% discount to Expedia is appropriate and concludes that the value should be no less than $35 a share.
Gompers testified that he did not inquire as to why Sabre was willing to pay $28 per Travelocity share as a check against his significantly lower valuation. Id. at 359.
Additionally, the petitioners argue that Gompers's valuation is fatally flawed because he relied on post-merger information in his valuation. The petitioners rely on Cavalier Oil Corp. v. Harnett, 1988 WL 15816, at *14 (Del.Ch. Feb. 22, 1988), where the court held that the expert's DCF analysis was flawed because it relied on actual earning and expense data from a period after the merger. Gompers use of Expedia's first quarter 2002 results publicly announced on April 23, 2002, twelve days after the merger, does not implicate the credibility of Gompers's valuation. The court in Cavalier held that the post-merger data was suspect because it was not available until after the merger and it "could not have been known or susceptible of proof" at the time of the merger. Id. (quoting Weinberger, 457 A.2d at 713). Here, Expedia's first quarter performance "could have been known or susceptible of proof" before the actual data was released after the merger date. In Kleinwort Benson Ltd. v. Silgan Corp., the court allowed the expert's reliance on a balance sheet released after the merger date and noted that data released on a balance sheet pertains to events that happen before the balance sheet is released. 1995 WL 376911, at *7 (Del.Ch. June 15, 1995). Expedia first quarter 2002 information was clearly knowable or "susceptible of proof" before the actual balance sheet was released only twelve days after the merger. Moreover, trial testimony clearly shows that Travelocity's management had general knowledge of Expedia's first quarter performance. Trial Tr. at 401-02. Therefore, the statutory requirement that the valuation must exclude elements of value "arising from the accomplishment or expectation of the merger" is clearly not implicated by Gompers's valuation. See 8 Del. C. § 262(h).
Salomon applies a 20%-30% discount range to Expedia and concludes that the appropriate value is between $24 and $32 a share. The independent valuation performed by Salomon provides the court with a neutral framework from which to analyze Purcell and Gompers's divergent values.
Salomon also uses two other valuation methodologies in connection with its independent valuation of Travelocity. It looks at precedent squeeze-out transactions and a Sabre "ability to pay" analysis. The precedent squeeze-out transaction premiums comparison involves taking a list of 40-plus companies and looking at the premiums paid and then applying them to Travelocity's various stock prices. This valuation supports a price range of approximately $30 to $35 per share. The "ability to pay" analysis based on 2002 and 2003 EPS (assuming no multiple expansion) supports a price well in excess of $35 per share. The "ability to pay" analysis factors merger synergies and is therefore not relevant to the court's analysis. See Zakkour Dep. at 50.
Zakkour testified in his deposition about Salomon's approach to the valuation and discusses the metrics of the valuation that were emphasized and why. The court adopts Salomon's valuation as a framework, and isolates the valuation metrics that should be of greater or lesser importance in determining the appropriate value for Travelocity's shares. Notably, Zakkour's extensive and detailed testimony in his deposition about Travelocity's "lost momentum" to Expedia evidences Salomon's awareness of Travelocity's positioning in the market vis-à-vis Expedia. See id. at 51-54.
1. The Appropriate Discount
The experts disagree on the appropriate discount that should be applied to Expedia as a comparable company. Purcell adopts Salomon's initial discount to Expedia of 10% and Gompers uses a minimum 40% discount. Salomon derives its discount range of 20%to 30% comparing the historical discounts of Travelocity's multiples of firm value to EBITDA and share price to estimated 2002 earnings per share relative to corresponding multiples for Expedia. The court finds Gompers's detailed analysis of Travelocity's risk and expected future growth rates reasonable. Furthermore, when asked why Salomon adjusted its initial discount rate, Zakkour testified at length about discussions with Travelocity's management as to the difficulties it faced in catching up to Expedia and successfully implementing a merchant model business. Gompers, like Zakkour, discusses the difference in the business models of the companies and the significance of this difference in the comparable company valuation.
Id. at 77-81.
Gompers testified that he discussed the business models of the companies and the respective cash flows of each model with Punwani who verified that this difference must be incorporated in discounting the cash flow multiples in the comparable company valuation. Trial Tr. at 328-29.
Purcell relies on the early 2002 positive analyst research reports as proof that Travelocity should only be at a "moderate," if any, discount to Expedia. Purcell gives great weight to James Hornthal's testimony about Travelocity and its potential. Hornthal characterized the Expedia-Travelocity competition as a "cat-and-mouse game" where the two companies were "jockeying back and forth" in the market. Hornthal relies on the Site59 acquisition as a beacon of light for Travelocity in its ability to catch up to Expedia after Expedia had pulled ahead in the fourth quarter of 2001. Peluso's testimony on Site59's ability to "transform" Travelocity's business model is persuasive: the acquisition of Site59 while being a step in the right direction did not equal a fully operational merchant model business. Hornthal's optimistic view of Travelocity's ease in catching up to Expedia, on which Purcell relies, is too speculative when compared to the clear evidence in the record that Travelocity still faced significant challenges in the development of its merchant model business. Purcell also places great importance on the fact that Travelocity was going to meet or exceed its 2002 expectations, but Punwani testified that it was only going to meet its projections through strategic cost-cutting that could not be sustained long-term. Moreover, Salomon adjusted its initial 10% discount (on which Purcell relies) to a 20% to 30% range after discussing Travelocity's strengths and weaknesses with management. Therefore, the record shows that Purcell's assumptions vis-à-vis the appropriate discount to be applied in comparing the companies are unduly optimistic.
"[T]he research analysts discussed Travelocity in January and February of 2002 in such positive terms as a company with a strong business model that can make money with gross margins of 63% ( i.e., Bear Stearns); a company whose sales are back on track, with a healthy outlook for 2002, and expected solid earnings growth with a possible multiple (price earning ratio) expansion ( i.e., Weisel); a company on target with revenue growth between 20% to 30% in 2002, with expectations of narrowing the gap with Expedia ( i.e., CIBC); and a company with travel-bookings now running close to pre-September levels ( i.e., Untenberg)." Purcell Expert Report at ¶ 41.
Hornthal was the founder and chairman of Preview Travel, a travel agency that started with a television platform in the mid-1980s and later moved online, which went public and then merged with Travelocity in 2000. He became vice chairman of the combined companies after the merger. See Hornthal Dep. at 10-16.
Id. at 91-93.
As already discussed, Peluso testified extensively on how Travelocity needed to develop its merchant model business and the obstacles it faced in doing so. See supra notes 6, 7, 14 and accompanying text.
Trial Tr. at 397-401.
Salomon used a 10% discount rate in its initial presentations to the Special Committee and the Travelocity board. JX 15 ("Project Roundtrip" — Salomon's February 27, 2002 Presentation on Travelocity). Zakkour explained that this initial number was more of a preliminary guess by Salomon before it had spent any time with Travelocity's management to "really understand how Travelocity and Expedia compared." Zakkour Dep. at 77.
Gompers concludes that the discount to Expedia should be at least 40% because Travelocity had a higher cost of capital, a lower growth rate, and a lesser ability to generate cash. He states that at the time of the merger, "Travelocity had lost momentum and was facing new competition that made its prospects potentially tenuous." The record is clear that even though Travelocity was actively working to remedy its outdated model, it still faced significant challenges at the time of the merger. The court notes that there was no evidence presented at trial or in the record to quantify the actual cost of building a merchant model or any necessary technological upgrades. With all of these factors in mind, the court concludes that it should apply a 35% discount to the valuation multiples derived from the analysis of Expedia, to reflect that competitive obstacles Travelocity confronted as of the Merger Date. This decision reflects the court's view that Gompers is substantially correct, albeit unduly pessimistic, in his critical comparison of Travelocity to Expedia. Instead of relying on Gompers's assessment that a discount of at least 40% is warranted, the court adopts, instead, the mid-point of Gompers's 40% and the high end of Salomon's 20%-30% range.
Gompers Expert Report at ¶¶ 194, 195.
Id. at ¶ 136.
See Trial Tr. at 246.
2. Valuation Multiples
Gompers and Purcell agree that firm value to EBITDA is the most important valuation metric. Purcell isolates firm value/EBITDA as " by far the most relevant and important statistic for comparison purposes." Purcell argues that this is the most important statistic because Travelocity has a great deal of noncash expenses, including depreciation, amortization, and the amortization of intangibles such as goodwill. Gompers agrees with Purcell that the EBITDA multiples are the "preferred multiple to examine" because they "are closest to cash flow and are a better proxy for the firm's on-going concern value."
EBITDA is earnings before interest, taxes, depreciation and amortization.
Purcell Expert Report at ¶ 47.
Id. at ¶ 48.
Gompers Expert Report at ¶ 173.
Zakkour testified in his deposition that even though a range of valuation metrics were used in Salomon's report, the most important valuation metric for comparing the companies was the price to earnings multiple because Travelocity was less profitable than Expedia. Zakkour further testified that Travelocity had a lot of work to do to catch up to Expedia, not only because Expedia was growing faster than Travelocity, but also because Travelocity had to basically transform its business model to remain competitive.
In calculating its reference range for comparison of the two companies, Salomon sets up a table of four statistical parameters: firm value/estimated 2002 EBITDA; firm value/estimated 2002 EBIT (earnings before interest and taxes); share price/estimated 2002 EPS; and share price/estimated 2003 EPS. Salomon derived a reference range of $24 to $32 per Travelocity share by applying 20%to 30% discount on the Expedia multiples. See JX 8 (Letter to shareholders and SEC Schedule 14D-9 for Travelocity.com).
"In this case, which is what I'll comment on, because every situation is unique, in this case, there is no doubt that PE multiples is by far the most important metric." Zakkour Dep. at 69. Furthermore, Zakkour testified that Goldman Sachs in its valuation of Travelocity also considered the PE multiples as the most important valuation metric. Id. Salomon defines the price/earnings (PE) multiple as earnings per share before noncash expenses. See JX 14 at 36 (Salomon February 25, 2002 Project Roundtrip Presentation).
Based on the expert reports and Zakkour's testimony, the court isolates the 2002 EBITDA multiple and the price-to-earnings multiple as the most important multiples in calculating Travelocity's firm value. Since Purcell does not present any calculations to back up his comparable company valuation, the court looks to Gompers's analysis in deriving the correct multiples. Gompers provides detailed and reasonable calculations for both Travelocity and Expedia's financial multiples, and the court agrees that these multiples are appropriate in comparing the companies.
See Gompers Expert Report at ¶ 197 and Exhibits C25 and C26 thereto.
Discounting Expedia's EBITDA multiple (34.8 x) by 35% produces an EBITDA multiple of 22.62 x. Applying this multiple to Travelocity's expected 2002 EBITDA of $47.80 million yields a value of $1,081,236,000. Discounting Expedia's EPS multiple (50.77 x) by 35% produces an EPS multiple of 33.00 x. Applying this multiple to Travelocity's expected 2002 net earnings of $39.45 million yields a value of $1,301,850,000. The court gives 2/3 weight to the EBITDA calculation and 1/3 weight to the PE calculation, yielding an enterprise value of $1,154,774,000. To determine the equity value, Gompers adds back the cash of $114 million and subtracts out the debt of $4.03 million. This leads to an equity valuation of $1,264,744,000, or $25.20 per share.
There were approximately 50.19 million shares outstanding.
E. Application Of A Control Premium
Delaware law recognizes that there is an inherent minority trading discount in a comparable company analysis because "the [valuation] method depends on comparisons to market multiples derived from trading information for minority blocks of the comparable companies." The equity valuation produced in a comparable company analysis does not accurately reflect the intrinsic worth of a corporation on a going concern basis. Therefore, the court, in appraising the fair value of the equity, "must correct this minority trading discount by adding back a premium designed to correct it."
Agranoff, 791 A.2d at 892.
Id. at 893.
The parties are silent on the proper application of a control premium. Purcell states summarily that if the court is to accept the theory that "some minority discount from going concern value" is appropriate in a comparable company analysis, then the correct valuation would be above his stated value. Salomon conducted a review of precedent minority squeeze-out transactions and found that the average premium paid for a control block when compared to the stock price was approximately 50%. Travelocity, however, is not directly comparable to the companies in Salomon's data survey. In fact, the online travel industry, as already discussed in great detail, is unique when compared generally to publicly traded companies. Moreover, the recent appraisal cases that correct the valuation for a minority discount by adding back a premium "that spreads the value of control over all shares equally" consistently use a 30% adjustment.
Purcell Expert Report at ¶ 52.
Salomon looked at both negotiated and unilateral squeeze-out transactions to determine whether Sabre's initial offer was adequate. Salomon determined that Sabre's offer was inadequate when compared to other squeeze-out transactions by acquirers with greater than 50% ownership and transaction values greater than $50 million on completed transactions announced from January 1999 to February 2002. See JX 16 (Salomon Presentation to the Travelocity Board of Directors, March 4, 2002); JX 14 ("Project Roundtrip" — Salomon's February 25, 2002 Presentation on Travelocity).
Notably, Salomon's final presentation to the board looked only at the unilateral precedent squeeze-out transactions. The actual numbers presented by Salomon on March 4, 2002 are: a 52.4% premium over 1 day prior; a 54.3% premium over 30 days average; and a 51.1% premium over 60 days average. JX 16 (Salomon Presentation to the Travelocity Board of Directors, March 4, 2002).
See Agranoff, 791 A.2d at 887; Borruso, 753 A.2d at 459; Bomarko v. Int'l Telecharge, Inc., 794 A.2d 1161, 1186 n. 11 (Del.Ch. 1999), aff'd, 766 A.2d 437 (Del. 2000).
Relying on recent precedents, the court will adjust the $25.20 per share value by adding a 30% control premium. This results in a per share value of $32.76.
See Borruso, 753 A.2d at 458-59 n. 10.
The court will not adjust this figure to reflect any incremental value inherent in the acquisition of Site59. While that acquisition held significant future promise to allow Travelocity to develop a merchant model for its business, there is no reason to believe that it was immediately additive to value. Notably, Salomon did not factor the Site59 acquisition in its valuation for several reasons. First, at the time of the fairness opinion, the deal was still in negotiations. Second, the deal was relatively small so that the effect on Travelocity stock would be at least in the short term, value-neutral. Third, even though it was a step in the direction of a merchant model approach, the acquisition of Site59 was only an initial step in a long process of transforming Travelocity's business model. Fourth, Zakkour stated in his deposition that the decision not to include the acquisition in the valuation of Travelocity was a "consensus view" by Salomon, the Special Committee, and management. See Zakkour Dep. at 22-24.
F. Interest
The petitioners are entitled to interest on the fair value of their shares pursuant to 8 Del. C. § 262(h). Moreover, section 262(i) states that "[t]he Court shall direct the payment of fair value of the shares, together with interest, if any, by the surviving or resulting corporation to the stockholders entitled thereto. Interest may be simple or compound as the Court may direct." This court has consistently awarded compound interest in appraisal proceedings.
"After determining the stockholders entitled to an appraisal, the Court shall appraise the shares, determining their fair value . . . together with a fair rate of interest, if any, to be paid upon the amount determined to be the fair value." 8 Del. C. § 262(h) (emphasis added).
8 Del. C. § 262(i) (emphasis added).
See Onti, Inc. v. Integra Bank, 751 A.2d 904, 926 (Del.Ch. 1999) (holding that a compound interest award in an appraisal proceeding is consistent with "fundamental economic reality").
There is no precise formula the court must use in determining the appropriate rate of interest, and "[e]ach party bears the burden of proving the appropriate rate under the circumstances." The petitioners argue that the appropriate rate of interest to be applied is 9.53%. Purcell reached this conclusion by averaging the petitioners' lost opportunity costs at a prudent investor rate (10.95%), and the respondents' borrowing costs (8.1%).
Grimes v. Vitalink Communications Corp., 1997 WL 538676, at *9 (Del.Ch. Aug. 28, 1997), aff'd, 708 A.2d 630 (Del. 1998).
Purcell Expert Report at ¶ 78.
Although it is reasonable to base the appropriate rate of interest on the average of prudent investor rate and a company's cost of borrowing, the court does not accept Purcell's calculation of pre-judgment interest. First, Purcell states that "a prudent investor would likely invest in a combination of long- or medium-term and short-term investment vehicles that would generate the highest return available, such as a mix of treasury and corporate bonds." He then assumes based on his "experience regarding portfolio mix allocations" that a prudent investor would invest 50% in three-year treasury bonds and 50% in Baa-rated corporate bonds. He offers no explanation, however, why a "prudent" investor, such as any of the plaintiffs, would not invest a portion of available funds in the equity market. Since Purcell does not provide the court with the necessary details to support his opinion, the court rejects his calculation of the prudent investor rate.
Id. at ¶ 79.
Id.
Moreover, Purcell inexplicably relies on a 2002 KPMG Consulting valuation of Travelocity's tangible and intangible assets to determine Sabre's cost of borrowing. Purcell relies on the cost of debt used in the KPMG report without offering any evidence as to why the court should adopt this calculation nor does he address Sabre's actual cost of borrowing.
The KPMG report assumes that Sabre's cost of borrowing is equal to "Moody's Baa Industrial Yield Average Bond Rate." Id. at ¶ 83.
Since the petitioners have failed to develop a credible record on the issue, the court looks to the legal rate of interest. The legal rate of interest, as defined by 6 Del. C. § 2301, is 5% over the Federal Reserve discount rate. Because the court will award the legal rate of interest, the appropriate compounding rate is quarterly.
Chang's Holdings, S.A. v. Universal Chems. Coatings, Inc., 1994 WL 681091, at *3 (Del.Ch. Nov. 22, 1994) (stating that the legal rate of interest is "a useful default rate when the parties have inadequately developed the record on the issue").
See Taylor, 2003 WL 21753752, at *13 (holding that the appropriate compounding rate for the legal rate of interest is quarterly because "the legal rate of interest most nearly resembles a return on a bond, which typically compounds quarterly").
The petitioners shall submit a form of final order, on notice, within 10 days.