SGI, Rackable, and the stalking horse

Print Friendly, PDF & Email

SGI sent out a letter to its support customers, which you can read here, that supplies a little more detail and a magic phrase that led to some interesting insight into their process with this bankruptcy. Here is the relevant section

SGI logoThe way that we’ll accomplish the merger will be a little unique, but given today’s economic environment, this approach will give us the desired result in the shortest amount of time. We’ll put SGI through a “Section 363” process under the U.S. Bankruptcy code that will enable Rackable to acquire the company without assuming the debt.  During a relatively short 25 day period, we’ll continue to operate as independent companies, and you should expect and receive the same levels of support as before.

Investopedia describes a stalking horse bid

An initial bid on a bankrupt company’s assets from an interested buyer chosen by the bankrupt company. From a pool of bidders, the bankrupt company chooses the stalking horse to make the first bid.

In this case Rackable is the “stalking horse” in SGI’s section 363 sale.

Having a stalking horse is apparently not all that uncommon in a section 363 sale these days. The horse is put in in many cases to establish a valuation floor before the sale of assets in the bankruptcy — in other words, to ensure that a minimum bid is achieved.

This paragraph from an article at Focus Management Group’s web site precisely describes the Rackable/SGI deal (note, according to their web site the Focus Management Group “provides turn-key support to underperforming companies and their stakeholders, including secured lenders and equity sponsors”):

In some cases, stalking horse identification and selection is the end result of a disciplined auction process seeking the highest value for a debtor’s assets. In such cases, the pre-filing process usually determines the debtor’s business unviable on a stand-alone basis. In this circumstance, the company must be sold to a strategic competitor with the strength to maximize the value of the debtor’s assets; the debtor’s business, which requires significant equity capital and perhaps new management to reorganise, can only be provided by a deep-pocket strategic or financial buyer. After the auction process, the 363 sale gives the buyer the legal protections provided by the Code and allows for an orderly distribution of funds to secured and unsecured creditors in accordance with the priorities set forth in the Code. In essence, the stalking horse is the successful bidder as a result of a fair and efficient auction process, and the 363 sale confirms that fact while still leaving the door open for a higher and better offer.

Note that what happens in the “short 25 day period” that SGI references in its customer letter is that the floor is open for bids from companies other than Rackable who might want to acquire SGI for more than Rackable has offered. But the word “short” is part of the problem from the perspective of those other bidders. Rackable has (likely) had a much longer amount of time for due diligence on the sale and, according to the Focus article probably has stacked up some other benefits as well

Other stalking horse benefits incorporated in orders approving 363 bidding procedures include expense reimbursement, bidding procedure influence leading to a favorable advantage and stipulations in the bidding procedures that call for: (i) the stalking horse to receive a breakup fee if another bidder offers more, (ii) competing overbids higher by at least a specified minimum amount, and (iii) short time periods for other bidders to conduct due diligence and value the business. With these tactical advantages, it’s not surprising that, in most 363 sales, the stalking horse bidder is ultimately the successful bidder.

All of which adds up to Rackable being the likely, but not necessarily the guaranteed, winner.

Companies considering a 363 sale do not always need to go in with a stalking horse. This is called “going naked,” and it may be the right decision if there is expected to be a high degree of interest in the failing company’s assets that would result in a competitive bidding process and a higher price. Given the dynamics of the HPC market and SGI’s very large load of secured debt, I think that they were probably unlikely to find another suitor willing to offer more, so having a stalking horse probably makes sense in this case.

If no bids are are received during the 25 day waiting period that will establish that the Rackable bid is at least fair  (might be well over fair value, but that is not revealed by this proces), and that satisfies the market’s demands for a disposition at fair value and makes sure that Rackable isn’t taking advantage of the situation. Still, have a stalking horse doesn’t always lead to the best outcome for the company’s owners

On the other hand, it appears that the tactical leverage, provided the stalking horse has, in many cases, dissuades other qualified bidders who might have bid more from even entering the fray. There have been cases where other potential bidders choose not to participate rather than spend more time and effort trying to overcome the bidding advantages offered to the stalking horse.

For example, in a recent case {PSA Quality Systems, Case No. 04-13030(MFW)}, the bidding procedure and tight timeframe favouring the stalking horse dissuaded several potential buyers expressing an interest in the debtor’s assets from even attending the auction.

Does Rackable have to consummate the sale? Probably not, though I don’t know whethere there are specific covenants in this deal that would force specific performance. Again, from the Focus article

To guard against the stalking horse bidder not closing the transaction, a sizable deposit should be received which is only refundable if: (i) the stalking horse is outbid, and (ii) the stalking horse does not bid after its stalking horse bid is exceeded by another bidder.

So they could walk away, but if SGI was smart it will cost Rackable a bundle.


  1. […] SGI, Rackable, and the stalking horse | Share and Enjoy: […]

  2. […] I should point to John West’s very intelligent story on the 363 process at InsideHPC. Here he looks at how this could play out. I don’t think its going to be clean. At $0.03 on […]


  1. Rackable has posted five consecutive quarterly losses. I believe the plan is this:

    Phase 1: One money losing company buys another money losing company.

    Phase 2: ????

    Phase 3: Make a profit.

    See South Park, episode 217 if you don’t get the joke.

  2. John West says

    Paul – the same plan that SGI had when they bought LNXI, right?

  3. Christine C says

    How much will my shiny purple SGI mug fetch on ebay now?

  4. John Leidel says

    I have a yellow one 🙂

  5. apart from the parallel one has to draw with all the Wall Street antics coming to light… Loss + Loss = ?? ..this is a very informative and educational article. I can’t say I understand why they chose to use “stalking horse” – it just doesn’t seem like a creature that would “stalk” does it? And, by description, it almost sounds more like a Trojan Horse – as in it’s not really what it seems to be. Oh well, let the soap opera continue.

  6. Bob Ewald has developed a pretty amazing resume. His bio on the SGI web site says he has had “a successful track record as a CEO”. But this consists of the following CEO positions:
    * E-Stamp (out of business)
    * Scale Eight (out of business)
    * Linux Networx (out of business)
    * SGI (soon to be out of business)

    Bob’s a nice guy, and I know put a lot of time and effort into trying to help these companies succeed. But if he lands another job as CEO, short the stock!

  7. I see the benefits for SGI outweigh those of Rackable here. It would give SGI quite a bit of money very quickly, and they would lose most of thier debt. I can not believe a 250 (+/-) person company would get rid of SGI people with thier knowledge on HPC. Most people are writting in thier blogs as if this decision was made on March 31st and filed on April 1st. I would think that this process started last year, and have been caculated through many times to ensure both companies would come out ahead. It does seem that the caculation will not work out (loss+loss=success) however, how many people here were invested in winning companies that have lost everything over the past few months?

    I think we will see a stronger SGI/Rackable come out of this one, and perhpas SGI will be forced to lower their overpriced products to make the competitive again.

    • John West says

      Joe – I certainly hope you are right, but there are significant problems and a very cloudy future. Yes, SGI has great people and great technology, but much of their most successful product line competes directly with Rackable’s products (I’m thinking ICE specifically here). I certainly feel that the shared memory technology, which I’m most keen to see stay around, is at risk, and I’m concerned about development efforts as well. It all really comes down to what Rackable’s broader intentions are with respect to HPC and the federal space, which is something we won’t know until we are further into the process. As you alluded to, I think there is reason to believe that they wouldn’t have bothered with the purchase if they didn’t intend to grow here. But there are at least as many reasons to be concerned as there are reasons to be optimistic.

      One of the primary reasons to be concerned is the very reason you point out: the deal appears from the outside to be lopsided, with SGI getting more. In very very broad terms, SGI is smart and Rackable is rich (yes, I know that probably goes so far as to be incendiary, but you get the point). This means that the cultural problems, which people seem to want to pooh pooh at the start of a deal and then always point to and nod sagely when the thing falls apart, are significant. Not insurmountable. But significant. I do think that if anyone can get past these types of problems it’s Bo.

      In terms of the wisdom generated by many people calculating through things many times before a decision is made to ensure that both companies come out ahead, I’m sorry I don’t share your enthusiasm. I’d point to deals that are much much larger with way more smart people calculating: Daimler and Chrysler (devolving), HP and Compaq (thousands laid off, Dell gained market share, took four years to even out), AOL and Time Warner (B school case study on how not to do mergers), and the between 50% and 70% of all mergers and acquisitions that are judged to have failed (Google it). Oh, and SGI/LNXI. Smart people with the calculators don’t have numbers for all the factors that contribute to success or failure. All we know from them at the start of a deal is that the numbers don’t guarantee failure. We don’t know anything about the potential of success.

  8. I guess we will find out today Thursday 4/30/2009- the final outcome. They should announce it at 5PM EST or at least by Friday morning the following day……..

  9. You Made Some Good Points There. I Did a Search on the Topic and Found Most People Will Agree with Your Words. Thank You!

    P.S: Please Take a Minute to Visit My Website as Well: