Eddie Lampert, pleased with himself.

Sears Has a Deal to Offer Its Shareholders

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.
( Updated
)
a | A

When you're a public company and you need money, you can raise it in a bunch of different ways, each of which conveys a different amount of desperation. Right near the top of the desperation pile is the rights offering, in which you go to all your shareholders and threaten and cajole them into buying more shares by telling them that if they don't they will be diluted and the value of their stock will drop. In the U.S. this sort of thing is perceived as a bit shameful, so it tends to be a last resort, though it's much more common in Europe.

Here, on the other hand, is a much stranger rights offering from Sears Holdings, which is a U.S. company, albeit also sort of a Canadian one. Sears Holdings' innovation is that its rights offering is not an offering of stock but of bonds and warrants. For every 85.1872 shares of common stock that you own, you get a right to:

  1. Give Sears Holdings $500.
  2. Get a $500 bond (five-year maturity, 8 percent coupon, senior unsecured at Sears Holdings).
  3. Also get 17.5994 warrants to buy Sears Holdings common stock (five-year expiry, strike price of $28.41 per share).

The total issuance would be $625 million of bonds and warrants to buy 22 million new shares, or about 20.7 percent of the shares currently outstanding.

That's weird! A typical rights offering involves cajoling your shareholders to buy more shares, because if there's one thing you know about your shareholders it's that they hold your shares. That doesn't necessarily mean they want to buy more of them, but it's at least a defensible inference. It's less clear that they'd want to buy bonds and warrants. Lots of shareholders actually can't buy bonds or warrants; they are equity mutual funds that are limited to buying stocks. The natural way to sell bonds is to bond investors. Sears Holdings' rights are transferable, so the shareholders can sell them to bond investors, but the rights structure, besides being a bit shameful, makes it harder to market to bond investors, do investor meetings, negotiate terms, etc.

So why do it? Well, Sears Holdings has two big shareholders of particular note, ESL Investments and Fairholme Capital Management. ESL, Edward Lampert's investment vehicle, owns 48.5 percent, and Lampert is the chairman and chief executive officer of Sears Holdings; Fairholme owns 23.6 percent. ESL intends to exercise its rights in the offering, as do at least some Fairholme funds. And if others don't exercise their rights, then ESL intends to buy any bonds and warrants that others don't. It's safe to say that these guys want to buy this deal.

Of course, Sears Holdings could have done a regular marketed bond deal, and ESL and Fairholme could have put in orders in that deal. What would that look like? Sears Holdings has senior secured bonds outstanding, due in October 2018, rated CCC+ and trading at a yield of around 9 percent. The new notes are unsecured, so they rank lower in the capital structure; they also mature later. Figure they should come at a yield of, I don't know, 10 percent? 11? More? This is the sort of nosebleed area where you can't feel all that confident offering $625 million of sub-CCC+ bonds to high-yield investors. They might just say no.

So maybe you give them warrants to sweeten the deal. Like in this offering: Sears Holdings is offering warrants with an exercise price of $28.41, pretty in-the-money versus yesterday's close of $36.99. (The stock was up this morning.) I get a theoretical value of around $16 per warrant. And you get 17.6 of them with your $500 bond.

Here's a stab at the value proposition for this rights offering:

Hoo boy is that not investing advice! If you expect to receive Sears Holdings rights in the mail, please consult your investment adviser or psychologist or oracle before deciding whether they represent an attractive proposition. All I'm saying is, using some reeeeeeeeasonably conservative assumptions, you could see how someone might think that it's an attractive proposition. If you look at that chart, and you shouldn't, it seems to say that you're paying $500 for $725 worth of stuff.

That's nice! For you! Obviously it's less nice for Sears Holdings, precisely in proportion to how nice it is for you. I mean, it looks like Sears Holdings is selling $725 worth of stuff for just $500. But it all balances out because it's selling it to its shareholders. Sure, Sears Holdings is getting a bad deal, but its shareholders are getting a good deal, and they're all in it together. And sure, some of the shareholders won't want bonds and warrants and so will have to get rid of their rights. But that's fine; they can sell them, at a price that should roughly reflect how good a deal this is. (Which might be $2.64 per current Sears share! Though probably not!)

[Update: To be clear, the math above gets a value of about $225 per right. You get one right per 85.1872 shares, so the $2.64 is the rights value per share. I see a trade for 30 rights at $195 at 10:01 on Friday morning, or about $2.29 per share.]

So a story sort of emerges about why Sears is doing this deal rather than a more regular bond offering. If it did a regular offering, it might be able to get a better deal than this: There might be bond buyers willing to buy these bonds at lower yields, or to take fewer warrants, or whatever. That would be good for Sears Holdings. But it would be bad for the investment vehicle of Sears Holdings' CEO, who intends to buy around half of the bonds and warrants in this offering. If he bought half of a public offering, and agreed to take whatever deal the market price set, he might end up with a worse deal (for him) than this one. And he wants this one.

On the other hand, if Sears Holdings went to the bond market, it might not be able to get a better deal. It might only be able to get this deal, or an even worse one. That would be bad for Sears Holdings, but it would also be awkward for Lampert: He'd be buying half, or whatever, of what looks like a sweetheart deal, without offering it to other shareholders.

So I think that roughly explains why this is happening. Sears needs money. It is at the point where it can only raise that money on fairly embarrassing terms. The most natural provider of money on embarrassing terms is its CEO, but it's a little awkward to take money from your CEO on embarrassing terms, and anyway he drives a hard bargain. The only way to make that look OK is by offering the same embarrassing terms to everyone. And if not all of them are able to accept, because they're not natural holders of bonds and warrants, then so much the better. The CEO will step up. There's your deal.

One other fun thing. Sears Holdings has long been a popular stock among short sellers. The stock exchanges show 17.6 million shares of short interest, out of 106.5 million total shares, though that short interest number is on a delay and may be different now. But it's a lot, especially when you consider that ESL and Fairholme between them own 78 million shares. To be short Sears Holdings, you need to borrow stock, which is a difficult and expensive proposition these days. But lots of people do it anyway, because they're really jazzed to bet on Sears Holdings' failure.

This deal should make their lives a lot harder. Sears Holdings has 106.5 million shares outstanding, but people own a total of 124 million shares -- 106.5 million that they've bought from Sears Holdings, and 17.6 million that they've bought from short sellers. I know, it's weird, it's the mystery of short selling, that's just how it works, people own more shares than there are. And if you were short a share of Sears Holdings yesterday, then today you are short a share of Sears Holdings and one of these new rights. So now people own more rights than there are.

That's bad enough, but it gets creepier on Nov. 18, when the rights expire. The rights, by my math, have value, so they should all be exercised. Or rather, more than all of them should be exercised. If the rights connected with 124 million shares (i.e. 1.46 million rights) are exercised ... what happens? Well, Sears Holdings isn't going to honor more rights than it issued (1.25 million, for its 106.5 million shares outstanding). It will honor them for 106.5 million shares. The other 17.6 million will be the responsibility of short sellers to honor. They'll have to borrow to do that, and will end up short something like $100 million of Sears Holdings bonds and warrants on 3.6 million shares. Bonds tend to be harder to borrow than stocks, since some of the obvious sources of stock borrow (retail investors in margin accounts, etc.) don't really work for bonds. If you think it's hard to borrow Sears Holdings shares now, wait until you have to borrow bonds and warrants -- especially when most of those bonds and warrants will be held by ESL and Fairholme.

Or you can close out your position in the bonds and warrants by buying them in -- but where will you buy them from? Again, ESL and Fairholme will own most of them, and seem like buyers rather than sellers. There are $100 million of bonds that need buying, and probably not that many that need selling. One gets the sense that the short sellers will feel at least gently squeezed.

I have no idea if this is an intended or anticipated result, and perhaps there is some simple solution that I'm missing and short sellers will sail through this rights offering untroubled. But it sure looks like a pain to me. Normally a rights offering drives down a company's stock price, both as a matter of arithmetic and because it looks a bit desperate, and so vindicates short sellers and makes them a profit. This one seems to have the opposite effect. Sears Holdings' stock price is way up from where it was when it first announced this offering last week. Sears Holdings has managed to turn what usually looks like a desperate move into one that actually looks quite strong.

Updates eleventh and seventeenth paragraphs to add trading price for the rights as of Oct. 31 and clarify difference between right value per right and per share.

  1. Not at the top. Things further up include: government bailout, debtor-in-possession facility in bankruptcy, etc.

  2. The way it works is, you have 100 shares outstanding trading at $20 each, and you give shareholders the right to buy one new share for every five shares they hold now, for $15 a share. A bargain! Except that if they all do it, then you've sold 20 new shares for $300, and instead of owning five out of 100 shares in a $2,000 company (five percent, worth $100), the shareholder now owns six out of 120 shares in a $2,300 company (five percent, worth $115). You haven't magically created any value. On the other hand, if a shareholder doesn't give you the $15 to exercise her rights, she's left with five out of 120 shares in a $2,300 company (4.2 percent, worth $96), so she's lost both votes and value. (Assuming all the rights are exercised, by the operation of oversubscription rights where if she doesn't exercise someone else can exercise her rights for himself.)

    All of this is just as true, by the way, of a regular-way stock offering, where stock is sold to the public for cash: That also dilutes existing holders, and tends to come at a discount to the previous trading price. But it tends to come at much less of a discount -- just enough to convince new investors to buy. Because a rights offering is being offered to all existing shareholders, there's less shame in bringing it at a big discount to the trading price. "You're all getting a great deal!"

  3. I mean, Sears Canada is its own company, with its own rights offering, which is more normal in that it is for common stock, but less normal in that it is being made to Sears Holdings shareholders. It's like a weird spinoff or something. It is a whole other kettle of fish that is I think best left alone for now.

  4. Numbers from the rights offering prospectus. Here's Fairholme's Sears investment thesis.

  5. There are liquidating ESL vehicles that won't be exercising their rights, though the prospectus hints that they'd sell them to other ESL vehicles. You get the sense that both ESL and Fairholme will exercise as much as they possibly can, though the legal language is considerably more lukewarm.

  6. The 6.625 of 2018, CUSIP 812350AE6. Trade data from Trace via Bloomberg TDH and HP functions, and the Finra website. There are also some Sears Roebuck Acceptance Corp. bonds that are pretty ugly.

  7. Here you go:

    [imgviz image_id:izWY980ctqcA type:image]

    The conservatism is mostly in the 35 percent five-year implied volatility assumption, which seems absurdly low; one-year realized volatility is 60 percent. (At a 40 implied you get $17.22 per warrant.) I don't know if I'm operating the dilutive model correctly, or anything else really. I am specifically not putting in high rates for stock borrow, which is a bad idea; Sears seems to be a tough stock to borrow.

    By the way, if you run this using a spot price of $28.41 -- arguably fair since that was the value proposition when this deal was first announced last week -- you get a (conservative) warrant value of $9.63 per warrant. If you swap that in to the table in the text, you get a value per bond + warrant unit of $613, and a value per right of $1.33. Still attractive.

  8. This discussion is based on a Twitter conversation with IvanTheK.

  9. I mean, except insofar as they could have bought in a public bond offering too. But, like, retail investors wouldn't get a look at a bond offering. Bond offerings are clubby events.

    In addition to awkwardness, there are famously complex Nasdaq rules around when you can sell warrants on more than 20 percent of the company, which seems to be going on here. It's possible that you might worry about doing a public bond offering that looks like that, if insiders are taking down a big chunk of that offering. The rights offering seems to solve that issue.

  10. The stuff in the text is a weird fun short-term squeeze, but longer term this should also be a bit annoying for short sellers. The issue is that Sears Holdings is issuing warrants to buy 22 million shares. If you look at the math in footnote 6, those warrants seem to have an 80 percent delta. So theoretically they should be hedged by ... selling short 17.6 million shares. I think that number is a coincidence.

    Now, they won't be hedged like that. Most will be bought by Lampert and Fairholme, who will not hedge them. But some will end up in the hands of arbitrage funds that will own the warrants and short the stock as a hedge. Which will create further competition for stock borrow, and make fundamental short investors' lives more difficult.

  11. Actually the timing is a bit puzzling. The record date for the rights offering is today, as was announced 10 days ago. So I guess the shares went ex-rights three days ago? Maybe? (Bloomberg lists the ex date as "TBA." Maybe they're not ex yet? Some sort of due bill? I can't figure it out from the prospectus.) But the stock didn't really trade down? Like, theoretically, if you buy shares now, are you buying shares without rights attached? Anyway my use of terms like "yesterday" and "today" in the text is unscientific and approximate.

  12. This happens all the time, of course: Any dividend or rights offering or spinoff can lead to similar trouble, where more people are long the stock, and thus entitled to the dividend/rights/spinoff, than there are actual shares. So they have to look to the short sellers for their dividend/rights/spinoff. But usually the short seller just needs to "manufacture" either cash (for a dividend) or shares of a public company (for a rights offering or spinoff), which is relatively easy. Also I mean having 17 percent of your stock short, and 72 percent held by two big insider-ish holders, and then doing a rights offering is unusual too.

This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.

To contact the author on this story:
Matthew S Levine at mlevine51@bloomberg.net

To contact the editor on this story:
Zara Kessler at zkessler@bloomberg.net