Hedge-Fund Manager Wants to Rearrange GM's Stock
If you own a share of General Motors Co. stock, what do you really own? Common stock is a slippery beast, but you can write a list of its main attributes. It goes something like this:
- Dividends: You get a 38 cent dividend every quarter, unless GM decides not to pay the dividend, or to cut it.
- Dividend growth: If GM decides to raise the dividend, you'll get that too.
- Mergers and liquidation: If GM liquidates, or sells itself to another company, you're entitled to a share of the money, after all of the creditors are paid off.
- Buybacks: If GM's management decides it has more money than it needs, it might give some back to you by buying back shares, though there are no guarantees.
- Voting: You get to vote on stuff, like directors and mergers and whatever.
None of that exactly captures the essence of what you own: Facebook Inc., for instance, has never paid a dividend, has said it doesn't plan to, is too big for anyone to realistically buy, and your voting rights are pretty worthless. (It does do some stock buybacks.) Nonetheless its common stock is worth more than $400 billion. There is a mysterious residue to the value of common stock that comes from somewhere other than the lawyers' list of its formal rights.
Still, there that list is. Five things: dividends, dividend growth, mergers/liquidation, buybacks and votes. It feels sort of random, doesn't it? There are good historical and financial-theory reasons for that particular list -- it's the way to describe the "residual claimants" on a corporation, and to incentivize and monitor managers to take care of those claimants -- but it is just a list of stuff that people have thought up, not an invariable law of nature. You can monkey with it. Lots of companies don't have dividends. Snap Inc. went public with no voting rights. People bought its stock anyway.
If you think too hard about this, another idea might occur to you. If you can omit some items from the list, why can't you split up the items on the list? Like, instead of common stock being a package of (1) dividends, (2) dividend growth, (3) merger/liquidation rights, (4) buybacks and (5) votes, what if it was instead two packages, of, say, (1) dividends and (2)(a) dividend growth, (b) merger/liquidation rights, (c) buybacks and (d) votes? Would those two packages be worth more, or less, or the same as the one package? Would changing the numbering in that sentence create value, or destroy value, or is the notion that the numbering of a sentence could affect the value of a car company completely absurd?
Every four years or so, David Einhorn, the founder of Greenlight Capital LLC, rediscovers the concept of preferred stock and publishes a big presentation about it. 1 I think it is about the most moving and beautiful recurring event in finance, like the migration of the Demoiselle cranes over the Himalayas, but for preferred stock. His current idea is that General Motors should split its common stock into two classes. Here is Greenlight's proposal. One class of stock -- the "Dividend Shares," ticker GMD -- would get the current dividend of 38 cents a quarter. The other class -- the "Capital Appreciation Shares," ticker GM -- would have most of the other rights of shareholders; in particular, they'd be the ones that participate in any future growth, in the form of stock buybacks or dividend increases. 2 (They'd also have 10 times the voting power of the GMDs.) Each current GM share would be split into one GM share and one GMD share, and then everyone could trade so the shares could find their natural homes. "The Dividend Shares will be attractive to yield-oriented investors," says Greenlight, while "the Capital Appreciation Shares will be attractive to growth-focused investors."
The normal way to describe this would be to say that the GMD shares are "preferred stock" -- entitled to fixed dividends, but not growth or (much in the way of) voting rights -- while the GM shares are "common stock." Einhorn avoids those terms, possibly because "preferred stock" sounds sort of old-timey and fuddy-duddy, or possibly because it's weird for a big industrial company to have preferred stock equal to more than half of its common equity market capitalization, 3 or possibly (most likely) because "preferred stock" has some negative implications for credit ratings.
But that terminology aside, here is the takeaway of Einhorn's presentation: The Dividend Shares should trade like preferred stock with a dividend yield of 7 percent to 9 percent, for a value of about $17 to $22 per share. 4 (The common stock's current dividend yield is about 4.4 percent. 5 ) The Capital Appreciation Shares should trade like common stock with a price-earnings ratio of 5.6 to 8, for a value of about $26 to $38 per share. (The common stock's current price-earnings ratio is about 5.6, which seems like an odd floor for the new Capital Appreciation Shares' P/E ratio, but whatever. 6 ) The package, then, should be worth about $43 to $60 per share. The stock closed yesterday at $34.71. "Our plan will unlock between $13 billion and $38 billion of shareholder value through appropriate valuation of GM’s dividend and earnings potential," Greenlight says.
By doing nothing. 7 Right now, GM shareholders get a 38 cent dividend and a 5.6 P/E ratio. Einhorn's plan would give them a 38-cent dividend over here, and a 5.6 P/E ratio over there. If they think about those two things at the same time, they think the combination is worth about $35. If they think about the one thing, then meditate for a bit to clear their minds, then think about the other thing, they will think the combination is worth at least $43, or as much as $60. Or so Einhorn argues.
Of course he could be wrong. GM has rejected his plan, pointing out that the valuation impact is "unproven and entirely speculative," while splitting the shares would create some real annoyances. ("Material governance challenges arising from two classes of stock with divergent objectives," for one thing, and potential liquidity difficulties, for another.) My favorite objection is "that Dividend Shares would result in loss of GM's investment grade credit rating." Right now GM pays a 38 cent dividend every quarter on its common stock. Under Einhorn's plan it would pay the same 38 cent dividend on the same number of shares of GMD stock. But because it would call those shares "Dividend Shares," the rating agencies would treat them sort of like debt, as an obligation that had to be paid and that limited GM's financial flexibility. 8 Einhorn's plan is to change nothing substantive, but to pronounce the word "dividend" with a bit more of an emphasis. He thinks this will make investors notice the dividend more. GM's objection is that it might make the rating agencies notice the dividend more.
It is easy and fun to be sarcastic about all of this, but that is mostly the wrong response. 9 This is what finance is. Creating value by raising money from investors and using it to build cars is great and all, but there is nothing particularly interesting about it as finance. It's just car-making. The magic of finance -- the reason it's a big industry, and the reason that hedge-fund managers are often a lot richer than car-company executives -- is that it can, in its finest moments, create value out of nothing.
There is a straightforward and unromantic way to explain that. Finance is about dividing up risks, and allocating those risks to the people most able and willing to bear them. People who want more risk cushion those who want more safety; people who want more safety compensate the willing risk-takers for that insurance; everyone is better off. Progress in finance, then, is about finding clever ways to slice risks more finely and market them to people who want them. 10 People who like General Motors's business but want safe steady income can just buy its bonds. People who like its business and want more upside in exchange for more risk can buy its common stock. But these are crude divisions, and a true financial visionary will see a category of risk -- Dividend Shares -- that doesn't exist yet 11 and try to fill that need.
But describing the mechanism in that way, as risk allocation, shouldn't diminish the magic of this effort to create value where there was no value before. The beauty of Einhorn's proposal -- whether or not it's right -- is the faith he puts in the power of the human imagination. General Motors has some stuff. Einhorn doesn't propose to change that stuff. He just wants us to look at it differently, to see the same facts in an altered light, to reconstruct the mental categories with which we think about the stuff. He thinks that effort -- the purely mental process of thinking about GM in a new way -- is worth at least $13 billion.
Einhorn's proposal, then, is to take all of the current elements of General Motors, add nothing, subtract nothing, but just arrange them in a subtly different and more pleasing pattern, one that excites people more, one that people will pay more money for. That's what art is. This is art.
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
In 2013, it was "iPrefs" for Apple Inc.: Apple had a lot of money, and while some people argued that it should give that money back to its common shareholders in the form of buybacks or dividend increases, Einhorn argued that it should instead create a new class of preferred stock, give the preferred stock to the common shareholders, and then give the money to the preferred stockholders. It was elegant, in its opposite of simplicity. Apple went with buybacks and dividend increases.
Einhorn's presentation doesn't correspond precisely to my list, and I don't know exactly how the GM and GMD shares would be treated in liquidation, etc. GM's counter-presentation says that the GMDs would have "no liquidation preference," but that doesn't fully answer the question. Maybe there's some fixed liquidation amount for the GMDs, and the GMs get any value above that amount (but they share pro rata in losses below that amount, rather than the GMDs having any preference). Or maybe they share pro rata in everything, though that would be a little weird.
Also buybacks of the GMDs would be "permitted, but not expected"; presumably they'd be like debt (or preferred stock!) buybacks, used as an economic trade (if the GMDs trade at a high enough yield) rather than a capital-return mechanism.
I mean, on a pro forma basis, using his numbers. His midpoint valuation is $19 for the GMDs and $33.11 for the GMs, meaning a preferred stock value -- counting the GMDs as preferreds -- of 57 percent of the equity market capitalization.
Seems right to me. I see a General Motors Co. bond with a 6.75 percent coupon, maturing in April 2046, trading at about a 5.5 percent yield. A perpetual preferred should trade at least 100-200 basis points wider than that, and the GMDs aren't even exactly "preferreds," so 7 percent to 9 percent seems reasonable.
Per Einhorn's presentation. I see it at 4.25 percent today, but the stock is up. Same caveat on the P/E ratio.
"We believe they will be valued based on a P/E multiple, and we value them conservatively at the current depressed P/E multiple," says Greenlight, somewhat sneakily. (The current multiple accounts for a dividend!) "But multiple expansion should occur because planned buybacks would buy more Capital Appreciation Shares than today’s common stock due to a reduced absolute share price."
"Our plan does not affect GM’s corporate strategy and will improve its financial flexibility," says Greenlight. "We are not advocating for any change to GM’s capital allocation policy, including capital devoted to balance sheet cash, dividends or share repurchases."
GM says: "We believe that Dividend Shares would likely be viewed as hybrid instruments with a partial debt component, leading to significant erosion of quantitative credit metrics."
By the way, the ratings agencies would be absolutely right to think that way! Current GM shareholders get a package of dividends and growth and whatnot. If GM falls on hard times, it can cut the dividend and say to shareholders, "well, this is the best way to preserve your chances for recovery and future growth." Some yield-oriented shareholders will grumble, but others will understand that it's the right decision for the company.
But if it splits into GM and GMD shares, and then the GMD shares all end up in the hands of conservative yield-oriented investors, it will be much harder for company to cut the GMD dividend. The GMD holders will get no upside from that decision; they'll just see their income suffer. And they will all be income-oriented investors anyway, so they'll really care about the dividend. Also the shares will be called "Dividend Shares," and it'll be embarrassing if they don't have a dividend.
All that means that the dividend on the Dividend Shares would be a lot more locked-in -- a lot more debt-like -- than the dividend on the current shares, so it really would limit GM's financial flexibility.
I mean, it's a little bit the right response.
Occasionally it goes wrong! One popular interpretation of the great financial crisis is that it was about misbegotten slicing of risk. That is probably right, but not a good reason to reject all risk-slicing. You still buy insurance. Stocks and bonds of the same company still have different risks.
I mean, preferred stock exists. But not at GM.
To contact the author of this story:
Matt Levine at email@example.com
To contact the editor responsible for this story:
James Greiff at firstname.lastname@example.org