Good slide.

Alibaba Could Buy Yahoo for Free

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

Here is a dumb trade:

  • Alibaba acquires Yahoo for a mix of cash and stock.
  • Specifically, 0.39 Alibaba shares and $7.25 in cash per Yahoo share.
  • By my math, that's worth about $42 per Yahoo share, about an 8 percent premium to the current price.
  • That's it.

So that trade gets Yahoo shareholders out of their stock at a premium. A smallish premium but still. On the other hand, for Alibaba, it costs precisely nothing. I mean, Alibaba has to put up 384 million shares of stock and $7 billion in cash. But Yahoo actually has 384 million Alibaba shares and $7 billion in net cash. So once Alibaba owns Yahoo, it'll get those things right back. So it costs nothing.

Like, here is the math:

It is dumb math, but there it is. Alibaba gets Yahoo for nothing. Yahoo shareholders are made better off than they are now. It is a purely Pareto-optimal trade. Also Yahoo owns some other stuff, like a stake in Yahoo Japan worth around $8 billion. Alibaba would get those for free. Also it would get the rest of Yahoo for free! Perhaps it could do something with that.

This is a good trade. Yahoo and Alibaba should do this trade. Why won't they do this trade? One reason is that it is a little too obvious that Alibaba would be getting Yahoo (and Yahoo Japan for that matter) for free, and if you are a management team and board of directors and someone comes to you and says, "I want to buy your company for nothing," then your inclination will be to say no. Your company has to be worth something! Like, you're coming to work every day, doing stuff. Concluding that the stuff you're doing is worth nothing, or negative $8 billion depending how you count, is a little too much like staring into the abyss for most corporate executives.

Also for all I know they're right and the stuff is worth more than zero. And since Alibaba probably doesn't want to own Yahoo -- it has its own business to run -- it is not likely to be the highest bidder for core Yahoo or Yahoo Japan. So Alibaba would go to Yahoo and say, "Hey, we'd like to buy you for negative eight billion dollars" and Yahoo would say, "How about positive seven billion dollars?" and Alibaba would say no but maybe someone else would think that Yahoo is actually worth positive seven billion dollars, and so you'd have a pretty wide bid-ask spread.

But this shouldn't obscure the essential point of the math, which is that Alibaba is in some pure sense the highest bidder for the Alibaba shares locked up in Yahoo and that those shares are the biggest thing going at Yahoo. If Yahoo just sells those shares to Company X, then Company X will pay it $34 billion for those shares, and then Yahoo will pay like $13 billion in taxes, and then it will have only $21 billion. Similarly, anyone else who buys Yahoo will need to pay those taxes to move the Alibaba shares out of the Yahoo corporate entity, so they'll value those shares at $21 billion after-tax.

But Alibaba, uniquely, won't care. If you or I buy Alibaba stock, we want to have Alibaba stock; we want to be able to sell it one day, and generally have it be interchangeable with the other Alibaba stock in circulation. But if Alibaba buys Alibaba stock, it doesn't want Alibaba stock for trading or investing purposes. It has all the Alibaba stock it needs. It can actually print more Alibaba stock if it wants to. If Alibaba buys Alibaba stock, it's to retire it. It wants to take that stock out of circulation. If the stock sits in a U.S. subsidiary, and can't be extricated without paying taxes, that doesn't matter. From Alibaba's point of view -- particularly, from the point of view of measures like shares outstanding and earnings per share -- issuing 384 million shares to Yahoo, in order to acquire a corporate husk that owns 384 million Alibaba shares, is a nothing. 384 minus 384 is zero.

So if this isn't going to happen, what can you do with that information?

The trade that I'd love to see, as a financial engineer manqué, is one in which Alibaba intermediates a sale of Yahoo. So Company X wants to buy Yahoo's actual business. Say it's willing to pay $15 billion for core Yahoo plus the Yahoo Japan stake. If it also had to buy the Alibaba stake for $34 billion (its value), and then sell it for $34 billion, and then pay $13 billion in taxes, then it should value that stake at $21 billion net of tax. Throw in the $7 billion of cash, and it should be willing to pay about $43 billion for Yahoo -- roughly what I have Alibaba paying at a zero valuation of the business.

But you set up a three-way trade where Alibaba acquires Yahoo for cash and stock, for a total price of $47 billion, and then sells core Yahoo plus Yahoo Japan to Company X for $12 billion (call it $8 billion after tax). That way:

  • Yahoo shareholders get $47 billion instead of $43 billion: +$4 billion.
  • Alibaba pays $47 billion, gets back $42 billion from the Alibaba shares and cash as described above, and gets back $8 billion after tax from selling Yahoo: +$3 billion.
  • Company X pays $12 billion for Yahoo, which it values at $15 billion: +$3 billion.

Everyone wins! Except the IRS, ha ha ha, but be careful there. There's $4 billion of capital gains taxes paid in this scenario, versus $13 billion of capital gains taxes paid in the scenario where Yahoo sells its Alibaba stake on the market, sure, but versus $0 of capital gains taxes paid in the scenario where Yahoo just hangs around holding its cluster of weird stocks forever.

I wouldn't bet on this trade (who is Company X? ), though I would enjoy it. But there's another trade that might get a similar result, sort of in the opposite order. It's in two parts. The first part goes like this:

  • Yahoo puts a little business -- say the search business in Malaysia or something -- into a subsidiary, call it Yahoo Weird Proxy Inc. for reasons that will become apparent.
  • And it puts the Alibaba shares into Yahoo Weird Proxy.
  • And it spins Yahoo Weird Proxy off to its shareholders in a tax-free transaction.

Those shareholders now own (1) a share in core Yahoo worth whatever it's worth plus (2) a share in Yahoo Weird Proxy, which, as its name implies, is pretty much a proxy on Alibaba. (Remember, the actual business in it is small.) This is perhaps an improvement -- the value of the core Yahoo stake is floored at zero, for one thing -- but it's not perfect. Yahoo Weird Proxy will not trade exactly like Alibaba, because you can't convert Yahoo Weird Proxy shares into Alibaba shares. In particular, if Yahoo Weird Proxy sells its shares, then it pays taxes. So Yahoo Weird Proxy should trade at a discount. Pretend it's a 10 percent discount.

But the second part of the trade goes like this:

  • Alibaba buys Yahoo Weird Proxy for stock.
  • It pays a premium to Yahoo Weird Proxy's trading price, but a discount to the underlying Alibaba stock (say a 5 percent discount).

Get it? Alibaba could buy Yahoo now, to get its hands on the underlying Alibaba shares. But that's messy: There's a lot of other stuff in Yahoo, and Alibaba is unlikely to want that stuff, or to agree with Yahoo on what it's worth. But if Yahoo can separate out the Alibaba shares into a company with less stuff, without paying taxes immediately, then Alibaba can eventually buy it much more cleanly. And in that trade, again, everyone wins: Yahoo Weird Proxy shareholders get to exchange Alibaba proxies at a 10 percent discount for Alibaba shares at a 5 percent discount. Alibaba gets to exchange Alibaba shares at a 5 percent discount for Alibaba shares at no discount.

This is powerful magic, and you need to be careful with it. The rules and incantations are difficult -- you need a business purpose for the spin-off, you probably can't do the two parts as a pre-arranged plan, you need to wait a year or two between the spin-off and the Alibaba acquisition -- and if you get it wrong the consequences can be dire. Also, good lord, I am not a tax lawyer, consult your own tax advisor. (I bet Yahoo is doing just that!)

But if this works, it gets Yahoo out of the Alibaba shares at considerably less than a 38 percent tax rate and then lets it run its core business on its own, free of annoying people running around talking about how it has negative value.

Go back to that trade at the top. It values Yahoo's business at zero, and values the $8 billion Yahoo Japan stake at zero, and still comes up with a price of almost $42 a share for Yahoo, based only on a commitment to tax ingenuity. Yahoo is currently trading below $40 a share. What does that tell you? Perhaps that I'm wrong about the feasibility of these schemes? Maybe. Or that Yahoo's board and management like the idea of a giant buffer of Alibaba stock and would hesitate to give it up, even if doing so would create free value for shareholders? That seems like sort of a scurrilous thing to think. Or perhaps that Yahoo has no commitment to tax ingenuity? I find that hard to believe. Everyone loves tax ingenuity!

  1. Assumes a $89.89 Alibaba stock price and a $38.65 Yahoo stock price as of today's close. Also:

    • Yahoo's cash, debt and share count come from the 10-Q.
    • Yahoo's Alibaba ownership and proceeds are based on the Alibaba prospectus, but don't quite match the numbers in the Selling Shareholders table because the greenshoe was exercised, and Yahoo sold an additional 18.3 million shares. So on my math Yahoo sold a total of 140 million shares at $68 a piece, for a total of $9.5 billion; I've assumed a 38 percent tax rate (and zero basis) to get about $5.9 billion in after-tax proceeds.
  2. I ignore their tax effects, though, I mean, it's a mostly-stock deal, it's probably not too bad. Or make it an all-stock deal and then do a buyback, with some complication of that simple math.

  3. I've ignored everything on the balance sheet except cash, short-term securities, and convertible debt, vaguely assuming that the receivables would pay off the payables etc. etc. Also the convert is trading at a premium so I'm being silly in treating it as paid off at par, but that is a fairly small effect.

  4. Again assuming 38 percent tax rate and zero basis.

  5. I think. If it owns 100 percent it should consolidate, and here is EITF 98-2:

    At the May 21, 1998 meeting, the Task Force reached a tentative conclusion that a subsidiary should account for an investment in the stock of its parent company in a manner similar to the manner in which it accounts for treasury stock and present that investment as a contra-equity account in its separate financial statements.
  6. I'm just assuming that the whole Yahoo and Yahoo Japan sale is taxable. Also ignoring shareholder-level tax, which depends on consideration.

  7. A good answer, sort of, is "Softbank." Softbank is a big owner of both Alibaba and Yahoo Japan, and is a sensible acquirer of Yahoo in the sense that it would enjoy owning both Alibaba and Yahoo Japan stock for a long time. So, sort of like Alibaba, it should be willing to hold those shares out of circulation and so can defer the taxes for as long as it wants to hold them.

    But if Softbank buys Yahoo it should do it directly, to defer the taxes on both Alibaba and Yahoo Japan. Buying through Alibaba avoids the Alibaba taxes forever, but incurs the Yahoo Japan taxes now. (Also Softbank probably actually wants the Alibaba shares, rather than having Alibaba intermediate them out.)

    This is probably the most economically sensible trade, but it is less beautiful than the defer-taxes-forever-through-Alibaba trade, so it lives only in a footnote.

  8. This structure has sort of a precedent in the work of the greatest of all tax structurers, John Malone of Liberty Media, who did something similar in spinning off Discovery Holding Company in 2005 and then merging it with Discovery Communications in 2008.

  9. Or perhaps: The market is discounting them because they'd take a while and only be kind of feasible. Or there are other explanations; for instance there is John Cochrane on convenience yields.

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