Thanks for coming, I'll be here all week.

Alibaba Had Some People Over for Lunch

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

A thing that Alibaba did today was convince something like 800 people show up at the Waldorf and stand in a huge line to ride an elevator to eat a boxed lunch and listen to Alibaba tell them to buy its stock. One conclusion you could draw from this is that Alibaba's initial public offering is underpriced. The preliminary price range on Alibaba's offering is $60 to $66 per share, and at that price range the demand seems to exceed the elevators. So make the shares more expensive! Capitalism 101 people.

But this is probably misleading. If you're standing in line for this lunch, it's probably because you're not big enough to rate a one-on-one meeting with one of Alibaba's two traveling color-coded management teams over the next 10 days. And those big meetings are the focus, since Alibaba is in the business of raising a comically large amount of money. At the high end of the $60-$66 range, it's looking for some $21 billion, but you can't just sell $21 billion of stock into $21 billion of demand and call it a day. If you figure Alibaba's bankers won't feel good unless then can get at least three times as many orders as they have shares, then Alibaba will need to convince people to buy $63 billion of stock. That's more than all U.S.-listed IPOs so far this year, combined.

If every single person at today's lunch today put in an order for $75 million worth of shares, that still wouldn't be enough. And probably some of the people at today's lunch don't even manage $75 million. So Alibaba needs big tickets:

It isn't enough for the company merely to find interested investors: It wants to identify fund-management companies and pension funds who will buy more than $1 billion worth of shares.

So the dynamic is not quite "create frenzy, sit back and watch the orders roll in." It's more a series of negotiations with big investors about price and value and governance. Today's kick-off meeting was not valuable so much in itself -- as a way to reach the sorts of investors who are willing to stand on line for an elevator -- as it is to impress the investors who would never stoop to that. "Look," Alibaba can tell Fidelity, "we've got 800 guys in the lobby who will buy this deal at $66 if you won't," whether or not anyone believes that. But the target is Fidelity, not the 800 guys in the lobby.

Here meanwhile is Aswath Damodaran, a finance professor at New York University who's an expert on valuation, and who previously valued Alibaba at around $66 per share:

I have a few emails in the last few days asking whether I feel vindicated by the fact that Alibaba's bankers seem to have arrived at a number very similar to mine and my reaction is mixed. First, given what I think about the valuations that emerge from investment banks in general, I am queasy that I am in agreement with their assessments of Alibaba's value.

Ha! Stupid bankers.

Second, and repeating a point I have made on prior posts about IPOs, bankers don't value companies in IPOs; they price them and the fact that they are thinking of pricing the company close to my estimate of value is, in my view, more coincidence than a reflection of consensus.

The art of pricing an IPO is not about correctly estimating the value of the company, but about correctly estimating what people will pay for it. There is no a priori reason for those things to be the same, but I guess there's some reason to expect them to converge on a big complicated deal that needs to rely on lots of big orders from sophisticated investors. Alibaba's not going to get lucky with one or two big excessively bullish orders here. It will need dozens.

The other weird thing for Alibaba is its profusion of bankers. There are six bookrunners -- Credit Suisse, Deutsche Bank, Goldman, JPMorgan, Morgan Stanley and Citi, listed in that oh-so-close-to-alphabetical order -- with a weirdly egalitarian division of roles. You don't normally see deals run by six banks, because usually there's some value in a tightly controlled, carefully coordinated process. I ... I would not have thought that a $20 billion deal is the time to give up that control? On the other hand, I guess each bank has different relationships, and with a deal this size you might as well make use of all of them.

If you're one of those banks, what are your incentives? Obviously you're getting paid to do your job. You're not getting paid very much on a relative basis -- it looks like under 1 percent to me -- but, I mean, the lead banks are mostly making around $30 million, it is not terrible work, though I guess the conditions are rather crowded. On the other hand, each dollar of additional price that you get for Alibaba -- worth $320 million to the company and its selling shareholders -- is worth only a few hundred thousand dollars to each underwriter. The incentives there would seem to be to play it safe, make sure that the deal is not an embarrassment, and not do too much to push the price up. Getting the last dollar for Alibaba makes you a little bit of money, at the risk of enraging your clients if the deal goes the way of Facebook.

That's probably good for Alibaba, too. It's raising money, sure, but like seemingly every internet IPO, it doesn't seem desperate for the cash: It's using it "for general corporate purposes" and had over $4 billion in operating cash flow last year. So it's not desperate for every last dollar. Meanwhile, none of its selling shareholders are selling more than a quarter of their stakes. Yahoo, the biggest seller, is going from 523.6 million shares to 401.8 million; Jack Ma, the founder, is going from 206.1 million to 193.4 million; SoftBank, the biggest shareholder, isn't selling anything. Their incentives are very much to have the deal go well for investors, so that their future sales are well received. Underpricing this offering could be a loss leader for future sales.

On the other hand, if you're one of six banks on this deal, wouldn't you be a little tempted to stick your neck out and tell Alibaba to be more aggressive on pricing? Alibaba is acquisitive, and at least one of its selling shareholders (Yahoo) has a whole lot more stock to sell in the medium term. So there's a lot of M&A and underwriting work here for a bank that can distinguish itself from the herd -- and getting as close as possible to the right IPO price, without going over, might be a way to do that. In some ways, the six banks on this IPO are in a very long and lucrative audition for future business, and being aggressive on pricing is one way to try to win that audition. Alibaba's main job in this IPO is to pit investors against each other to compete for the shares it's selling, but it can't hurt to pit the banks against each other a little too.

  1. Facebook's IPO seems to have been 4.5 times covered, and yet was a debacle.

  2. Source: Bloomberg function IPO <go>, filtered for U.S. exchanges, deals priced year-to-date, IPOs only. I see $46.4 billion of deals.

  3. From DealBook:

    Each bank team, led by senior deal makers, focused on several major tasks. Credit Suisse and Morgan Stanley, for instance, worked on the prospectus and will supervise the “lockup” period after which current investors can sell shares, while JPMorgan worked on the company’s valuation and offering structure.

    Goldman was awarded the role of stabilization agent, meaning it is responsible for making sure that the early trading after the I.P.O. goes well.

    I don't know what Deutsche is doing. Despite its out-of-alphabetical-order placement, Citi got to host the dry run (the presentation for the salespeople).

  4. My math is just:

    • The prospectus lists net proceeds of $7,643 million and expenses of $45.6 million, meaning gross proceeds less underwriting fees of $7,689 million at the $63 midpoint.
    • Gross proceeds to Alibaba at the midpoint are 123,076,931 shares times $63 equals $7,754 million.
    • So the gross spread -- the underwriting fees -- at the midpoint are $65 million on Alibaba's primary shares, or about 0.84 percent.
    • Multiply that by the full deal size -- including the shares that Jack Ma and Yahoo are selling -- and you get about $170 million of fees.
    • At $66 per share, the fees are up to $178 million.
    • If you include the 15 percent greenshoe, the total fees are about $205 million at the high end.

    Others seem to have higher numbers. Here is DealBook on Friday:

    The banks are expected to share in an underwriting fee of about 1 percent, according to people with knowledge of the matter, or just over $200 million if the stock sale prices at the high end. All but one lead underwriter would reap about $31 million.

    And here is the Wall Street Journal in March saying it was expected to be 1 percent to 2 percent.

  5. Like, $1 x 320 million shares x 0.84 percent x one-sixth = $450,000ish. A bit more with the greenshoe.

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:
Toby Harshaw at tharshaw@bloomberg.net