In re Appraisal of Dell Inc., 2016 Del. Ch. LEXIS 81 (May 31, 2016)
The Delaware Court of Chancery’s favoring the value derived from a post-merger discounted cash flow analysis over the deal price in the Dell statutory appraisal action stunned the world of high finance. How was it possible that the merger price did not reflect fair value when the company’s board and Michael Dell, who led the management buyout that gave rise to the lawsuit, seemingly performed according to script? What about the Chancery’s recent decisions that expressly considered the merger price a better indicator of fair value than any results from the DCF and other valuation methods? In a 115-page decision that is heavy on appraisal jurisprudence but relatively light on valuation methodology, the Chancery explained its findings.
Michael Dell founded the company in 1983. In 1988, it went public. He left in 2004 but returned in 2007. Once back in charge, he decided to transform the company by reducing its reliance on PC sales and pursuing an expansion into enterprise solutions and services. Between 2010 and 2014, the company spent about $14 billion on acquiring new businesses. Michael Dell and the company’s management believed that over the long term the enterprise division would prove successful and the company would be able to grow its PC business by about 1% to 2% per year. In January 2011, management prepared a “sum-of-the-parts” analysis that valued the company at $22.49 per share (based on line of business) and $27.05 per share (based on business unit).
However, the company stock traded at $14 per share. Michael Dell recognized that the market still considered the company a PC business, and the outlook for that sector was increasingly grim. In a June 2012 meeting with analysts, Michael Dell and management tried to communicate their optimism, suggesting 12% annual growth in software sales and 22% growth in services revenue. Management also announced a $2 billion cost savings initiative. Market observers remained skeptical, and the company’s stock price declined to about $12 per share.
Meanwhile, Michael Dell began talking to two financial sponsors about a management buyout (MBO). He straightaway informed the company’s board of directors, saying he did not want to proceed without the directors’ approval. The board formed a special committee, which hired JPMorgan Chase & Co. as its financial advisor. At that time, the company’s earnings call for the second quarter of FY 2013 showed an 8% revenue decline and a 13% drop in earnings per share. Dell seemed undaunted: “We’re transforming our business, not for a quarter or a fiscal year, but to deliver differentiated customer value for the long term.” In contrast, analysts described the company as a “sinking ship” and its turnaround strategy as “fundamentally flawed.”
Focus LBO Analysis
In fall 2012, JPMorgan made a presentation to the committee about the feasibility of an MBO considering “the basic math by which financial sponsors determine what they are willing to pay in an acquisition transaction.” It noted that a sponsor typically expected to obtain a return on its equity investment within three to five years in an outright sale, public offering, or recapitalization. “Financial buyers evaluate investments with an internal rate of return (IRR) analysis, which measures return on equity,” JPMorgan explained. And “IRR will be used as the primary means to determine the appropriate purchase price by a sponsor.”
JPMorgan also said that it was not very likely that financial sponsors other than the two with whom Michael Dell had talked or any strategic bidders would be interested in buying the company. The upshot was that throughout the process JPMorgan and other financial advisors, including Goldman Sachs, prepared valuations based on a leverage buyout (LBO) analysis. By that calculation, JPMorgan suggested a financial sponsor wanting to achieve a range of IRRs of 20% to 25% would be willing to pay in the range of $11.75 per share to $13.00 per share or $13.25 per share to $14.25 per share, if it engaged in further recapitalizations of the company. Goldman Sachs, using the same company forecasts as JPMorgan did, said a sponsor could pay about $16.00 per share and still achieve a five-year IRR of 20%.
In late 2012, the market’s outlook for the PC industry and Dell, the company, became even more pessimistic and the committee, concerned that management projections were too optimistic for purposes of “valuation and sell-side estimates,” hired Boston Consulting Group (BCG) to prepare an independent set of forecasts. As a result of the negative outlook, one of the financial sponsors dropped out and the other—the eventual buyer—proposed to buy the company for $12.70 per share. The committee found the offer “inadequate.”
BCG said the low valuation did “not match apparent company strengths.” Instead, it reflected “investor concerns.” JPMorgan agreed that investors were focused on short-term results at the expense of the company’s long-term value. Michael Dell, himself, made a presentation to the board explaining the strategic initiatives he planned to pursue once the company was a private company and noting an MBO would be in the best interest of the company’s shareholders “because they would receive a portion of the potential upside from these initiatives without bearing the risk.”
In early 2013, BCG prepared a series of projections. The most consequential ones in terms of the later trial expert valuations were two that turned on the company’s likelihood of achieving $3.3 billion in cost savings. Publicly, management had only announced $2 billion in savings to give itself a “cushion.” Under one scenario, the “BCG 25% Case,” BCG assumed the company would be able to realize 25% of the savings. BCG believed that goal was achievable and resulted in the most reasonable set of projections. The “BCG 75% Case” assumed the company would realize 75% of the savings. The committee was doubtful about this scenario, which implied margins for FY 2015 that were higher than the company or its competition had ever achieved.
Another set of projections, the “Bank Case,” related to a presentation the buyout group and Michael Dell made in fall 2013 to banks that would finance the merger. The forecasts assumed lower gross profit and EBITDA relative to an earlier set of projections submitted to banks but higher revenues. The presentation modeled $3.6 billion in cost savings, with $2.6 billion appearing in the forecasts and $1 billion as a separate line item.
Also, in early 2013, the committee hired a second financial advisor, Evercore, to conduct a potential go-shop period. In pitching itself for the job, Evercore prepared a DCF analysis that valued the company at between $14.27 per share and $18.40 per share. Its LBO analysis suggested a range of between $12.36 per share and $16.08 per share assuming an IRR of 15% to 25%.
Negotiations between the buyout group and the committee kept going. The committee wanted $13.75 per share, and the buyout group offered $13.50 per share. Under threat of a stalemate, Michael Dell himself offered to roll over his shares at a lower price than what public stockholders would receive. Ultimately, the parties agreed to a $13.65-per-share price, which both financial advisors considered fair to unaffiliated stockholders from a financial point of view.
The merger agreement provided for a 45-day go-shop period. During that period, there were two offers of promise, one by Blackstone Management Partners LLC and the other by Carl Icahn. Blackstone suggested the company’s existing shareholders could elect to receive a cash payment of at least $14.25 per share or a package of stock in a new entity worth $14.25 per share and subject to a cap on the total amount of equity that the new entity would issue. Icahn proposed a similar transaction offering stockholders a combination of cash and stock subject to a cap on the amount of cash. Evercore found the proposal amounted to a value of between $13.37 per share and $14.42 per share.
Ultimately, neither bid went anywhere, but both had the effect of bumping up the original merger price. In a final bid, the buyout group agreed to pay an additional $0.10 per share, increasing the total merger consideration to $13.75 per share. It also agreed to a special cash dividend of $0.13 per share. A majority of the holders of outstanding shares approved the deal.
Fair Value Determination a Different Animal
However, a number of shareholders filed a petition under Delaware’s appraisal status asking the Chancery for a fair value determination. The court, performing a discounted cash flow analysis that drew on analysis from both sides’ experts, achieved a value of $17.62 per share. In other words, Michael Dell sold the company for about $7 billion too little.
The crux of the Chancery’s decision is that a statutory appraisal determination is not an inquiry into a breach of fiduciary duty claim. “In a fiduciary duty class action, the court is faced with the question of holding individual directors personally liable for having breached their duties to the stockholders,” the court explained. “An appraisal action asks a substantially more modest question: did the stockholders get fair value for their shares in the merger?” According to the court, “a sale process might pass muster for purposes of a breach of fiduciary claim and yet still generate a sub-optimal process for purposes of an appraisal.” This was the case here. No one breached a duty to the shareholders. In fact, the board, the committee, and its advisors “did many praiseworthy things,” the court noted. Regardless, factors related to the process “undercut the relationship between the Final Merger Consideration and fair value,” the court found.
Pre-signing phase problems. The problems started with the pre-signing phase, which produced an original merger price ($13.65 per share) that was below fair value, the Chancery said.
The paramount problem was that the players all were financial sponsors—there was no outreach to strategic bidders—and all the valuations driving the merger were premised on LBO models, which calculate what a financial investor would be willing to pay to achieve a certain internal rate of return.
In contrast, “fair value” under the applicable statute is “the value to a stockholder of the firm as a going concern as opposed to the firm’s value in the context of an acquisition or other transaction.” The court noted that the company conceded that an LBO model was not “oriented toward solving for enterprise value.” Here, JPMorgan performed a going-concern analysis using the DCF method and showed a value of between $20 per share and $27 per share. But, said the court with emphasis, “using the same projected cash flows in an LBO model,” JPMorgan found a financial buyer would pay a maximum of $14.13 per share—if the price were any higher, the sponsor could not achieve a minimum 20% IRR over five years. Assuming the buyer wanted to achieve IRRs in the range of 20% to 25%, the likely range of prices was even lower, JPMorgan determined.
The financial sponsors behaved as the committee’s financial advisors predicted they would, and they used similar LBO models. Competition between or among financial sponsors depends on their willingness to sacrifice IRR, but it “does not lead to intrinsic value,” the court observed. Michael Dell and the committee also did not focus on fair value, the court added. In fact, in its proxy to the shareholders, the committee stated it did not pursue a premerger going-concern valuation but used other metrics, such as what the LBO models indicated a financial sponsor would pay. “As a practical matter, the Committee negotiated without determining the value of its best alternative to a negotiated acquisition,” the court said.
The court also noted a “valuation disconnect” between the market’s perception and the company’s operative reality. The market focused on short-term results and on the company’s $14 billion investment to transform itself, but this narrow view might cause it to “excessively discount the value of long-term investments,” the court said. It also noted that proposing an MBO when the stock market is low meant using the low stock price to anchor price negotiations. Empirical evidence shows that the ensuing sales process is more likely to generate an undervalued bid, the court said. Here, all the participants to the transaction recognized the value disconnect, the court observed. In fact, Michael Dell and management tried to convince analysts that the company was worth more. Even though, during the sales process, the committee and its advisors used the market price as a key input, the court observed. “In fact, the trading price was the only quantitative metric the Committee cited in the Company’s proxy statements when explaining its recommendation that stockholders approve the merger,” the Chancery pointed out.
Post-signing issues. The post-signing go-shop period did not cure the initial undervaluation, the court found. True, the Blackstone and Icahn bids that came in during the go-shop period led to a 2% increase in the final merger consideration, but they ultimately went nowhere. If anything, the court said, the two bids showed that the original merger price was too low. Worse, in terms of the respondents’ case, the fact that the two bids exceeded the final merger price “undercut the notion that the Final Merger Consideration provided fair value,” the court observed.
DCF Analysis Tops Merger Price
To determine fair value, the court turned to the DCF analysis—the method both parties’ experts used, albeit to startlingly different effect. The petitioners’ expert said the company had a fair value of $28.61 per share on the merger closing date; the respondents’ expert said the value was $12.68 per share. “Two highly distinguished scholars of valuation science, applying similar valuation principles, thus generated opinions that differed by 126%, or approximately $28 billion. This is a recurring problem,” the Chancery observed.
Neither expert analysis was entirely credible, the court found. The petitioner expert’s result suggested that the merger undervalued the company by $23 billion. “Had a value disparity of that magnitude existed, HP or another technology firm would have emerged to acquire the Company on the cheap,” the court noted. On the other hand, the respondent expert’s result was below the merger price. If, as the court found, the deal price undervalued the company, the respondent expert’s valuation did even more so.
The court declined to “exhaustively describe the DCF methodology” but instead discussed areas of major disagreement between the experts and its preferences.
Forecasts. The projected cash flows underlying the experts’ analysis accounted for much of the difference in value, the court said. The petitioners’ expert used the BCG 25% Case and BCG 75% Case and weighted them equally. He also considered the Bank Case. The respondents’ expert used the BCG 25% Case but adjusted for weaknesses, including the risk that these forecasts were stale by the time of the merger. They were prepared in January 2013 and never updated. The adjustments included updating revenue projections based on PC industry data and updating revenues flowing from the sale of secondary products. Moreover, the expert adjusted for stock-based compensation and created a five-year transition period in the projections to better capture the company’s operative reality and the likely schedule for the transformation plan to show results.
He also found the Bank Case reasonable but adjusted to account for nonrecurring restructuring expenses and stock-based compensation.
The Chancery normally is skeptical about litigation-driven adjustments, the court noted. It prefers valuations “based on contemporaneously prepared management projections.” In this instance, however, the respondents’ expert “persuasively justified his changes.” The court agreed that, even though the BCG projections were “impressively thorough, with over 1,100 assumptions,” they risked being out of date by the time the merger closed (the valuation date). It found the Bank Case was perhaps too optimistic, but ultimately it approved of both sets of forecasts, as adjusted by the respondents’ expert.
Tax considerations. Taxes also figured prominently in the valuations. The petitioners’ expert used a 21% tax rate throughout his forecast period based on rates in the valuation’s models the company’s financial advisors prepared. The respondents’ expert used a 17.8% rate during the projection and transition periods but a 35.8% marginal tax rate for the terminal period. He justified the latter by citing to academic literature.
The court adopted the petitioner expert’s 21% rate. It pointed out the company had not paid taxes at the marginal rate since at least 2000. In the five years leading up to the merger, it paid effective rates of between 16.5% and 29.2%. Its cash tax rates ranged from 9.6% to 24.1%. The low effective rate stemmed from the company’s “indefinite reinvestment election,” meaning the company represented to auditors that it planned to defer indefinitely paying U.S. taxes on overseas profits. To suggest, as the respondent expert’s model did, that the company would in the near future pay a marginal tax rate of 35.8%, not to mention perpetually, contradicted historical practice, the court found.
The respondents’ expert also deducted $2.24 billion for deferred taxes attributable to the company’s foreign earnings and profits. However, the evidence showed the company’s effective tax rate accounted for the deferred taxes and the company did not plan to repatriate funds. Consequently, “a proper valuation would have to back out any deferred taxes on foreign earnings from its effective tax rate,” the Chancery found.
Finally, the respondents’ expert deducted a contingent liability of $3.01 billion from the enterprise value for unrecognized tax benefits. This liability equaled 100% of the reserve the company was required to have on its balance sheet to cover unpaid taxes in the event a tax position it had taken earlier proved incorrect (FASB Interpretation No. 48 (FIN 48)).
To subtract the full FIN 48 liability was excessive, the court found. However, there was evidence that the company would pay out $650 million from the FIN 48 reserve in connection with a dispute resolution. The court said it was reasonable to subtract this amount as a non-operating liability.
Equity risk premium. The experts disagreed over every input, except the risk-free rate, in computing the company’s weighted average cost of capital. Notably, the petitioners’ expert used a forward-looking equity risk premium of 5.50%, whereas the respondents’ expert used a blended historical and supply-side ERP of 6.41%. The court adopted neither approach but opted for a supply-side ERP of 6.11%.
Beta. The petitioners’ expert arrived at a beta of 1.35 by analyzing peer companies. The respondents’ expert generated a beta of 1.31 by analyzing weekly observations over a two-year period. The court favored the respondent expert’s approach, noting a “beta specific to the Company is more targeted than a blended beta calculated from peer companies, particularly when both experts opined that the Company had few peers.”
In the final analysis, the court arrived at a WACC of 9.46%.
The court concluded that in this case the sales process “functioned imperfectly as a price discovery tool.” At the same time, it was impossible to quantify the “sales process mispricing.” Accordingly, the Chancery decided not to give any weight to the final merger price but rely exclusively on the result of its DCF analysis as the indicator of fair value.