The valuation gap.

Photographer: Matthew Busch/Bloomberg

Michael Dell Bought His Company Too Cheaply

Matt Levine is a Bloomberg View columnist. He was an editor of Dealbreaker, an investment banker at Goldman Sachs, a mergers and acquisitions lawyer at Wachtell, Lipton, Rosen & Katz and a clerk for the U.S. Court of Appeals for the Third Circuit.
Read More.
a | A

If you own a stock that you think is worth $10 a share, and I also own that stock and I think it's worth $20 a share, I can try to convince you that I'm right. I can make predictions about the company's future prospects and earnings, and build a discounted cash flow model to prove that the stock really is worth $20. And you can make different assumptions about the future, and build different models, and try to convince me that $10 is the right number. But this is all kind of dumb. I should just pay you $12 for your shares. Then you get paid more than you think the stock is worth, and I get stock that I think is worth more than I paid, and we never have to discuss it. This is called the efficient markets hypothesis.  

This is such a good and pleasant way of avoiding discussion that it is the principal way that people in finance make arguments. Every once in a while a famous investor will rent out a ballroom and give a three-hour presentation about why a stock is going to go up (or down). But these presentations are actually relatively rare. Mostly when investors think a stock is going to go up, they buy the stock. The buying is the argument. You don't even need a ballroom.

One argument that you hear a lot is that public stock markets are too focused on the short term, giving too much weight to quarterly results and not enough to long-term visions. Structurally, this argument tends to be expressed by the people with the long-term visions, who feel that they're being undervalued.  They make the argument in words, in shareholder letters and television appearances, but they also sometimes make it with money. Companies sometimes buy back their stock because their managers think the market undervalues it. And companies that aren't already public perhaps put off going public to avoid all the short-termism, or, if they do go public, do it in ways that insulate the managers from the markets' short-term focus. 

One person who thought his company's stock was undervalued by a short-term public market is Michael Dell, the founder and chief executive officer of Dell Inc., which in his long-term vision was moving from being a personal-computer manufacturer into an enterprise software company. He made this argument in boring verbal ways :

Mr. Dell lamented that the market just "didn't get" the Company. He thought that in spite of the Company‘s transformation, "Dell [was] still seen as a PC business." Mr. Dell conferred with his management team and hired consultants to devise strategies to help the market view the Company as "a sum of the parts." Mr. Dell regularly communicated his views to analysts.

Unsurprisingly this didn't work. So he made the argument in a more persuasive way: In 2012, with the market not getting Dell and valuing it at just $9.35 a share, he rounded up some financing sources and offered to pay about $12 or so for it.  

That argument worked, eventually. Michael Dell and his private-equity backers at Silver Lake ended up paying about $13.75 a share ($13.96 counting some dividends) in a $25 billion deal that they signed in February 2013 and closed in October of that year. There were twists and turns along the way; in particular, Carl Icahn held up the deal for a while by arguing that Dell was actually worth much more than $13.75 a share. But Icahn's arguments were not ultimately persuasive, because they consisted mostly of words and numbers and not money. (They were somewhat persuasive because they involved some money -- Icahn made a couple of leveraged recapitalization proposals for Dell that would pay some cash to shareholders that might have sort of looked like a higher price than $13.75 -- but that isn't quite the same as bidding $14 a share for all the shares in ready cash.) 

And then some shareholders sued in an appraisal lawsuit, arguing that the $13.75 was too low and that a Delaware court should award them more money. And then most of those shareholders were thrown out of court for hilarious legalistic reasons that don't concern us here, though they have concerned us previously. But on Tuesday Delaware Vice Chancellor Travis Laster ruled on the remaining appraisal claims and found, in a 114-page opinion, that Dell was really worth $17.62 a share, so everyone who successfully sued -- and managed to jump through the correct hoops -- is entitled to an extra $3.87 a share, with interest.

Appraisal is a weird bit of corporate law, because it undermines the usual nice clean process of deciding how much a stock is worth by just seeing what someone will pay for it. The market said Dell was worth $9.35, or whatever, and Silver Lake said it was worth $13.75 and paid that, and then some shareholders said it was worth even more. The normal form of financial-markets argumentation would be for them to offer more. But instead they get to sue, and make arguments to the court, and then Michael Dell and Silver Lake have to make arguments about why it's worth less. Their heart is unlikely to be in those arguments since, you know, they paid the $13.75, which implies that they think it's worth more than that. But their best argument will always be: If this company was worth more than $13.75 a share, why did no one offer more?

And that is a pretty good argument! The best reply, especially in a conflicted management buyout where the buyer is also the CEO, is often along the lines of "because management deliberately undermined the sales process to prevent other bidders from seeing the company's true value, so they could take it for themselves on the cheap." But that's not really what happened here; the court found that, while there were some inherent conflicts of interest, Dell's independent directors basically did a bang-up job of running the sales process, and even Michael Dell himself -- despite being both the CEO and the prospective buyer of the company -- behaved like a prince. Actually, reading the opinion, you almost get the sense that Michael Dell didn't do this leveraged buyout for the money. It was just an intellectual engagement, a point of principle. He thought the stock was undervalued, and he wanted to argue his case, and the way you argue that is by buying all the stock.

But the court nonetheless disagreed with the price, and decided that the fair price was $17.62. To get there, first of all, Vice Chancellor Laster had to disregard the market price for the stock, which was $9.35 when the deal was first proposed, $13.42 when it was officially signed, and never closed above $14.51 afterward.  These lower prices are not too hard to explain away:

A second factor that undermined the persuasiveness of the Original Merger Consideration as evidence of fair value was the widespread and compelling evidence of a valuation gap between the market‘s perception and the Company‘s operative reality. The gap was driven by (i) analysts' focus on short-term, quarter-by-quarter results and (ii) the Company‘s nearly $14 billion investment in its transformation, which had not yet begun to generate the anticipated results. A transaction which eliminates stockholders may take advantage of a trough in a company‘s performance or excessive investor pessimism about the Company's prospects (a so-called anti-bubble). Indeed, the optimal time to take a company private is after it has made significant long-term investments, but before those investments have started to pay off and market participants have begun to incorporate those benefits into the price of the Company‘s stock.

That is: The market price didn't reflect fair value, not because the market didn't have the relevant information, but because it weighted it incorrectly. Analysts had a myopic "focus on short-term, quarter-by-quarter results," and couldn't understand the long-term value of the company's plan. This is true even though Dell's management explained its long-term plan and "tried to convince the market that the Company was worth more." The stock market, in Vice Chancellor Laster's view, was just incorrigibly short-termist, and couldn't take the long view and value Dell properly.

After all, that's why Michael Dell wanted to take the company private in the first place:

Mr. Dell identified the opportunity to take the Company private after the stock market failed to reflect the Company‘s going concern value over a prolonged period. He managed the Company for the long-term and understood that his strategic decisions would drive the stock price down in the short-term.

So he went to private-equity firms to fix his short-termism problem. This makes sense! If you have a public market that doesn't properly value a company because it is too short-term focused, you can ask a private-equity firm to buy the company and run it for the long term, without the pressure of quarterly earnings calls and microsecond stock-price moves. That was Michael Dell's thinking, and Silver Lake's, and it's a pretty standard move. And the next standard move would be to have an auction among private-equity firms to figure out how much they think the company is worth. That more or less happened here; the auction was imperfect, but ultimately three private-equity firms put a lot of effort into Dell, with KKR and Blackstone putting in bids at various points before dropping out of the running. The highest price that private equity was willing to pay for Dell was $13.75, give or take, significantly above the pre-announcement market price.

Meanwhile no strategic buyer -- no other company that might want to acquire Dell's business -- put in a bid.  So the best bid that anyone with a long-term perspective put in was $13.75.

So why wasn't that the right price?  The opinion on this is interesting, and it turns on the difference between an "LBO model" and a "DCF model." Here is the oversimplified difference :

  1. discounted cash flow model makes some assumptions about the company's future cash flows, assumes a reasonable capital structure, discounts those cash flows back to the present based on the company's cost of debt and a market-based cost of equity capital, and comes up with a value per share reflecting the present value of those cash flows.
  2. A leveraged buyout model does the same thing, except that the capital structure is more leveraged and, instead of a reasonable cost of equity capital, it assumes a 20 percent to 30 percent cost of equity capital.

Again, this is super oversimplified; please don't use this description to prepare for your investment banking interviews. But the basic idea is: Private-equity firms like to buy companies for less than they're worth, so that they can make 20-plus percent returns on their equity investments, so that their own investors will be happy with them and pay them their big fees. Public investors are satisfied with market-rate returns. Vice Chancellor Laster's DCF calculation used an 11.3 percent expected return for Dell's equity. If you expect an 11.3 percent return, you will be willing to pay more for a company than you will if you expect a 30 percent return. That is just math.

So you see the problem. Private-equity firms will only buy companies for cheap and lever them up, so they can get the 20-plus percent returns that their investors demand. So any price that a private-equity firm will pay for a public company is inherently suspect, since private-equity firms expect such high returns. So Vice Chancellor Laster ignored the price that Silver Lake actually paid, after a quasi-auction among private-equity firms. He chose his own DCF model, with lower expected returns, and calculated a higher price for Dell. And then he declared that that was the fair value.

I realize that this all sounds pretty naive when I put it like that, but it is kind of ... exactly what Vice Chancellor Laster says? This paragraph, near the end of the opinion, is deeply weird:

The fair value generated by the DCF methodology comports with the evidence regarding the outcome of the sale process. The sale process functioned imperfectly as a price discovery tool, both during the pre-signing and post-signing phases. Its structure and result are sufficiently credible to exclude an outlier valuation for the Company like the one the petitioners advanced, but sufficient pricing anomalies and dis-incentives to bid existed to create the possibility that the sale process permitted an undervaluation of several dollars per share. Financial sponsors using an LBO model could not have bid close to $18 per share because of their IRR requirements and the Company's inability to support the necessary levels of leverage. Assuming the $17.62 figure is right, then a strategic acquirer that perceived the Company‘s value could have gotten the Company for what was approximately a 25% discount. Given the massive integration risk inherent in such a deal, it is not entirely surprising that HP did not engage and that no one else came forward.

That is, Vice Chancellor Laster checked his work -- and his price of $17.62 -- by looking at the actual results of Dell's sales process, which got a price of $13.75. If Dell was worth $17.62, then no strategic buyer would be willing to bid more than $13.75, because at only a 25 percent discount to fair value "the massive integration risk" would make Dell unappetizing. And if Dell was worth $17.62, then no private-equity buyers would be willing to bid more than $13.75, "because of their IRR requirements and the Company‘s inability to support the necessary levels of leverage."  The proof that $17.62 was the fair price is that no one was willing to pay it:

  1. Public shareholders won't pay fair value for Dell, because they are obsessed with the short term and can't understand the long-term strategic vision.
  2. No strategic buyer would pay fair value to buy Dell, because that would be risky.
  3. No private-equity buyer would pay fair value to buy Dell, because private-equity firms only buy companies at a discount.  

So how could any merger deliver fair value to shareholders? This opinion creates its own weird valuation gap. Public-equity markets, let us assume, focus too much on the short term, and undervalue companies whose immediate results don't match their long-term prospects. The solution to that short-term-ism might be to go private so you can focus on the long term. But private-equity firms undervalue companies because, let us also assume, they demand an above-market rate of return and so will only buy companies at a discount. The only buyer willing to pay fair price for a company, apparently, would be a buyer with the relatively long-term focus of private-equity firms and the relatively low-return expectations of public-equity investors. Perhaps such a buyer exists. But it would be a little weird. After all, the higher returns are the compensation private equity gets for taking the longer view (and longer-term risks). And ample short-term liquidity is what allows public markets to afford the lower-return expectations. 

Why should there be mergers? Sometimes there are synergies: Company A and Company B will be better off together, because they can sell each others' products or cut back-office costs or whatever. Sometimes there are governance benefits: Company A's current managers don't know what they're doing, and an acquirer could install better managers with better ideas that make it worth more money. Either way, the merger creates value by making the business better. That's not what happened in Dell:

This was not a case in which the Buyout Group intended to make changes in the Company‘s business, either organically or through acquisitions. The Buyout Group intended to achieve its returns simply by executing the Company‘s existing business strategy and meeting its forecasted projections. Mr. Dell identified for the Committee the strategies that he would pursue once the Company was private, and the record establishes that all of them could have been accomplished in a public company setting. BCG recognized and advised the Committee that the only benefits Mr. Dell could realize by taking the Company private that were not otherwise available as a public company were (i) accessing offshore cash with less tax leakage (to pay down the acquisition debt) and (ii) arbitraging the value of the Company itself by buying low and selling high.

I suppose that sounds derogatory, but there are those who think that "arbitraging the value of the company itself by buying low and selling high" is ... I don't want to say a noble purpose, but let's say, the most noble possible purpose. When Michael Dell proposed a buyout, the company's price was wrong. Everyone involved in this case agrees on that. Michael Dell thought -- and said -- that Dell was undervalued. The Dell board said that it was undervalued. Its financial advisers said it was undervalued. Silver Lake thought it was undervalued. When the deal was announced, Carl Icahn said it was undervalued. When the deal closed, the appraisal plaintiffs sued, saying it was undervalued. And now a judge has decided it was undervalued.

But Michael Dell and Silver Lake are the ones who did something about it. Everyone says Dell's market price was too low; they skipped the talk and just offered to pay more. This buyout didn't create value by changing Dell's business model; it created value by changing Dell's ownership -- by moving the shares from people who mostly didn't value them that highly (public markets) to people who did (private-equity buyers). It didn't create synergies or improve Dell's sales, sure, but it did correct an error. There are those who would say that's a worthwhile thing to do. The court, though, suggests that it isn't.

  1. I kid, I kid. It is called "economics."

  2. In principle there is no reason to think that "public markets are too focused on short-term results" would imply "public markets undervalue companies." Of course many companies are performing better now than they will in five years, and a short-term-oriented market will overvalue those companies. But you more usually hear arguments that imply that markets undervalue, like, research and development or whatever, than arguments that imply that markets overvalue, like, making money today, though there are exceptions.

  3. This is from Tuesday's appraisal opinion. Citations omitted, here and throughout.

  4. Here I simplify a long and complicated process. Michael Dell was first approached about a management buyout in June 2012, and first raised it with the board in August. He worked with KKR and Silver Lake on their proposals, which they first submitted in October 2012. From the appraisal opinion:

    Silver Lake proposed an all-cash transaction valued at between $11.22 and $12.16 a share, excluding shares held by Mr. Dell. KKR proposed an all-cash transaction valued at between $12.00 and $13.00 a share, excluding shares held by Mr. Dell and Southeastern, and based its illustrative analysis on a price of $12.50 per share. KKR‘s proposal also contemplated an additional $500 million investment by Mr. Dell. Dell‘s common stock closed at $9.35 that day.

    There were many further revisions. And it's not quite right to say that Michael Dell offered the $12:

    Mr. Dell had not been involved in determining the prices. He was content to participate at whatever pricing the financial sponsors obtained. 

  5. The merger consideration was $13.75 a share, plus a special dividend of $0.13 a share, plus a regular third-quarter dividend of $0.08 a share, regardless of when the deal closed. (See pages 36-37 of the appraisal opinion.) In the rest of this post I'm going to use $13.75 because I assume that that's the relevant comparison for appraisal purposes. That is, if you sought appraisal and turned down the merger consideration, you still got the 21 cents worth of dividends, so the fair comparison is between the appraisal price and the $13.75 deal price.

  6. The big loser was T. Rowe Price

    Tuesday’s ruling could have been far costlier for Mr. Dell and Silver Lake. Earlier this month, a judge disqualified about 30 million shares held by T. Rowe Price Group Inc. because the mutual fund giant, one of the deal’s most vocal opponents, mistakenly voted its shares in favor of the buyout. To be eligible for the court-ordered price bump, investors must have voted against the transaction.

    The mix-up, which T. Rowe has blamed on the complexities of the proxy voting system, means it will miss out on about $190 million, including interest.

    Oops!

  7. I mean, he decided they were entitled to $17.62 a share, with interest on the whole amount. It's a lot of interest really. In any case, though, there weren't that many of them:

    But the ruling likely won’t cost Mr. Dell and Silver Lake much because only a tiny fraction of shares are eligible for the bump, because of quirks in Delaware law. The largest holder is hedge fund Magnetar Capital, which has rights to about 3.8 million shares and stands to collect about $15 million. 

  8. Page 61 of the opinion:

    In this case, the Company‘s process easily would sail through if reviewed under enhanced scrutiny. The Committee and its advisors did many praiseworthy things, and it would burden an already long opinion to catalog them. In a liability proceeding, this court could not hold that the directors breached their fiduciary duties or that there could be any basis for liability.

  9. I really can't get over how he didn't have a say on the price in the private equity firms' initial bids, and "was content to participate at whatever pricing the financial sponsors obtained." He worked in parallel with both Silver Lake and KKR, and then later when Blackstone was kicking the tires he spent a lot of time with them too -- even though they mused publicly about getting rid of him if they bought the company. And to help Silver Lake get its bid up, he was willing to roll over his shares at a lower valuation. He just doesn't seem to have done any underhanded nickel-and-dime-y stuff; you get the sense that if someone was willing to pay more he would have tried to make it work. 

  10. The fact that it closed above the deal price does, of course, suggest that some people thought that it was worth more than that deal price. (And/or that Carl Icahn, or someone, would get them a better deal.)

  11. By the way, I am not sure this is exactly fair to the public analysts. Here, from page 33, is a description of the reaction to May 2013 earnings:

    "Most [analysts] were shocked at the profitability results." Revenues beat Street estimates by about 4%, but earnings per share came in approximately 29% below the consensus. According to BCG, the quarter "miss [was] HUGE." Analysts disagreed about the long-term implications for the Company. The Bernstein analyst focused on the enterprise solutions and services division, noting that it generated "essentially no profits," which was "a sober finding, given the company‘s strategic intent is to migrate the company to enterprise offerings." The analyst attributed the division‘s poor performance to "myriad factors – none of which have easy fixes."

    This isn't dumb short-termism. It's using presently available data -- the only available data -- to try to predict Dell's long-term prospects. What else were the analysts supposed to do?

  12. Amusingly, Vice Chancellor Laster writes that "KKR‘s investment committee had concluded that the Company‘s recent operating performance validated analysts‘ concerns." That is: KKR, a long-term private-equity investor, concluded that the public analysts might have been on to something. It wasn't alone: The special committee of independent directors also approached TPG, which "reported to the Committee that it believed the 'cash flows attached to the PC business were simply too uncertain, too unpredictable to establish an investment case.'"

  13. After the deal was signed, Dell's bankers at Evercore ran a "go-shop" process where they tried to find a higher bidder. They were hopeful for Hewlett-Packard, the obvious strategic buyer, but while HP signed a confidentiality agreement, it "never accessed the data room and did not submit an indication of interest." On the other hand:

    On March 21, 2013, GE Capital submitted a bid to acquire the Company‘s financial services unit for $3.6 billion in cash. GE Capital said it was happy to have the Committee consider its offer in connection with any other bid.

    That never seems to have gone anywhere. 

  14. The opinion has some interesting general statements about why courts shouldn't be too deferential to the market. Pages 47-48:

    Recent jurisprudence has emphasized Delaware courts‘ willingness to consider market price data generated not only by the market for individual shares but also by the market for the company as a whole. If the merger giving rise to appraisal rights "resulted from an arm‘s-length process between two independent parties, and if no structural impediments existed that might materially distort the 'crucible of objective market reality,'" then "a reviewing court should give substantial evidentiary weight to the merger price as an indicator of fair value."

    Here too, however, the Delaware Supreme Court has eschewed market fundamentalism by making clear that market price data is neither conclusively determinative of nor presumptively equivalent to fair value

    And pages 52-53:

    Writing as a Vice Chancellor, Chief Justice Strine observed that even for purposes of determining the value of individual shares, where the stock market is typically thick and liquid, the proponents of the efficient capital markets hypothesis no longer make the strong-form claim that the market price actually determines fundamental value; at most they make the semi-strong claim that market prices reflect all available information and are efficient at incorporating new information. The M&A market has fewer buyers and one seller, and the dissemination of critical, non-public due diligence information is limited to participants who sign confidentiality agreements. It is therefore erroneous to "conflate the stock market (which is generally highly efficient) with the deal market (which often is not)." It is perhaps more erroneous to claim that the thinner M&A market generates a price consistent with fundamental efficiency, when the same claim is no longer made for the thicker markets in individual shares.

    On the other hand, there are limits. Pages 83-84:

    Based on this evidence, the Company makes a straightforward argument: Capitalists want to make money, and America is full of capitalists, so it is counterintuitive and illogical—to the point of being incredible—to think that another party would not have topped Mr. Dell and Silver Lake if the Company was actually worth more. In my view, this argument has force for large valuation gaps, and is sufficiently persuasive to negate the valuation of $28.61 per share that the petitioners advanced. If the Company was really worth more than double what the Buyout Group was paying, then a strategic bidder like HP would have recognized a compelling opportunity and intervened.

  15. Here is the footnote acknowledging that there are other differences. Also the LBO model is usually expressed as solving for a 20-30 percent internal rate of return hurdle rather than using a 20-30 percent cost of equity capital and solving for a price. Here's how the opinion puts it (page 63):

    When proposing an MBO, a financial sponsor determines whether and how much to bid by using an LBO model, which solves for the range of prices that a financial sponsor can pay while still achieving particular IRRs. What the sponsor is willing to pay diverges from fair value because of (i) the financial sponsor‘s need to achieve IRRs of 20% or more to satisfy its own investors and (ii) limits on the amount of leverage that the company can support and the sponsor can use to finance the deal. Although a DCF methodology and an LBO model use similar inputs, they solve for different variables: "[T]he DCF analysis solves for the present value of the firm, while the LBO model solves for the internal rate of return."

  16. See pages 107-109 of the opinion, calculating the weighted average cost of capital:

    • He assumes a capital structure of 75 percent equity, 25 percent debt.
    • The cost of debt is 4.95 percent, with a tax rate of 21 percent (see pages 105-107), for an after-tax cost of 3.91 percent.
    • The risk-free rate is 3.31 percent, the beta is 1.31, and the market equity risk premium is 6.11 percent, for a cost of equity of 11.31 percent.
    • The weighted average cost of capital is thus 9.46 percent. 
  17. I mean, it cheats a bit on the math, in that the LBO model uses a lot more leverage than Vice Chancellor Laster's 75-percent-equity DCF model, and with that much leverage the capital-asset-pricing-model cost of equity would be much higher.

  18. By the way, DCF models aren't perfect either. Page 90:

    The petitioners‘ expert, Professor Bradford Cornell, used a DCF analysis to opine that the Company had a fair value of $28.61 per share on the closing date. The respondent‘s expert, Professor Glenn Hubbard, used a DCF analysis to opine that the Company had a fair value of $12.68 per share on the closing date. Two highly distinguished scholars of valuation science, applying similar valuation principles, thus generated opinions that differed by 126%, or approximately $28 billion

    Meh, whatever. 

  19. The opinion is full of stuff like this. Page 86: The fact that Blackstone and Icahn emerged during the go-shop period "indicates that the Original Merger Consideration was low not only when judged against a fair value metric, but also when judged by an LBO model." That is: An LBO model intrinsically undervalues a company.

    To be fair Vice Chancellor Laster has some support for this view in this case. Page 65:

    Using the same set of projections for both a DCF analysis and an LBO analysis, JPMorgan showed why a financial sponsor would not be willing to pay an amount approaching the Company‘s going-concern value. Using the September Case and a DCF analysis, JPMorgan valued the Company as a going concern at between $20 and $27 per share. But using the same projected cash flows in an LBO model, JPMorgan projected that a financial buyer‘s willingness to pay would max out at approximately $14.13 per share, because at higher prices, the sponsor could not achieve a minimum return hurdle of a 20% IRR over five years. That $14.13 figure also assumed the sponsor engaged in further recapitalizations of the Company. Assuming a financial sponsor wanted to achieve IRRs in the range of 20% to 25%, JPMorgan placed the likely range of prices at $11.75 to $13.00, or $13.25 to $14.25 if the sponsor engaged in further recapitalizations of the Company.

  20. Because they have high IRR hurdles. Because buying companies in leveraged buyouts is risky!

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

To contact the author of this story:
Matt Levine at mlevine51@bloomberg.net

To contact the editor responsible for this story:
James Greiff at jgreiff@bloomberg.net