I hope Jack Dorsey is tweeting in this picture.

Photographer: Yana Paskova/Bloomberg

Square Turns the IPO Corner

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

The initial public offering of Square was supposed to be a sign of doom for unicorns, a proof that you can't make money investing in pre-IPO tech companies any more. On the other hand, here is how much money Square's investors have actually made, by funding round:  

 

Those are paper profits, based on Square's $9 per share initial public offering price, and on a $13.50 per share trading price as of about 11 a.m. Here they are in table form:

If you invested in Square at any point, from its Series A round in 2009 through its IPO last night, you've made at least a 20 percent return on your money as of this morning, or about a 7 percent compound annual growth rate.  And if you got in early enough, or late enough, your returns were way better. In aggregate, Square's investors have tripled their money; its pre-IPO venture capital investors have almost quadrupled theirs. Everyone can breathe easy now! Unicorns are fine again.

[Update: Oops, not quite! In October 2015, after the financial statements reflected in the rest of this post, but before the IPO, Square sold $30 million more Series E shares to "one existing investor and one new investor" who waived the ratchet right described in the next paragraph. (See page 169 of the prospectus). Those investors have a paper loss of about $3.8 million even at a $13.50 stock price, though the "existing investor" may have made money on its overall stake. Everyone else is in the black.]

Obviously there were some bumps along the way. Last night's IPO priced below the Series D round, leaving the Series D investors with a paper loss. It priced even farther below the Series E round, but the Series E investors had negotiated for themselves a "ratchet," in which Square essentially guaranteed them a 20 percent return in the IPO. Their shares were worth only about $87 million at the IPO price, so Square had to give them another 10.3 million shares -- worth another $93 million -- to make up for it. 

The ratchet is controversial:

Wesley Chan, a managing director of Felicis Ventures, says he counsels company founders to avoid accepting ratchets.

“The only folks that win on this are the later stage investors who get the benefit of the ratchet and they do so at the expense of many of the employees and early investors who helped to build the company, who get diluted, often significantly, when the ratchet triggers,” Mr. Chan said.

But one shouldn't get carried away here. Here's that same chart without the ratchet :

 

By my math, the ratchet cost investors in the Series A through D rounds about $55 million, a rounding error on their $1.3 billion of gains. It cost other pre-IPO shareholders -- founders and employees and such -- another $68 million. But without the ratchet, the Series E would be in the red, even after this morning's IPO pop. And who wants that? The goal of an IPO is for everyone to make money. Everyone made money. (Update: Almost!) The ratchet worked perfectly.

I suppose the larger question is, how did the IPO work? The ideal IPO performance would be to price at the top of the marketed range ($11 to $13), and then see a nice 10 percent or 15 percent pop to reward IPO investors. Pricing $2 below the range, and then seeing a 50ish percent pop, is sort of a double embarrassment. The underwriters' marketing range, now, looks more or less perfect, if you ignore the last 24 hours. But if you don't, you'll notice that the deal priced catastrophically below the range, and then popped embarrassingly above the range. It turns out that there was plenty of demand for Square at the top of the IPO range. It's just that that demand wasn't in the IPO order book. Oops!

And so Square -- and Chief Executive Officer Jack Dorsey's charity, the Start Small Foundation -- sold shares worth $364.5 million for $243 million, leaving over $100 million on the table. Square left some more money on the table -- about $83 million, by my math -- because the underpricing made the Series E ratchet that much more valuable. That isn't ideal. And unlike some IPOs, Square could use the money; it has fairly significant negative net income, though its cash flow from operations is slightly positive so far this year. But again, there is no need to overstate the importance of $100 million here or there. Square had no particular plans for the money, and won't miss that $100 million too badly.  The main goal of this IPO, at this point in the unicorn cycle, was for Square to safely go public without incurring the hatred of the public markets or its private investors, and the IPO pop at least salvaged that goal. Everyone ought to be happy, or at least relieved.

Still, if this IPO proves anything, it's that valuing companies is hard. For all the talk about how private-market valuations are disconnected from reality, it's worth noticing that Square's $9 a share IPO price, and its $15.46 a share nominal valuation when it did its Series E round, were both pretty far off from where Square is trading now. Pricing a private fundraising round is a mystery, but pricing a public offering is also a mystery. Observing a public trading price isn't mysterious at all, but on the other hand it isn't proof of a correct valuation either. It's just another price that investors put on Square. After all the other prices they've put on Square, though, you could reasonably be suspicious about whether this one is right too.

(Corrects to add new fourth paragraph indicating that at least one investor lost money, in article published Nov. 19.)

  1. I am mostly relying on capitalization information in Square's prospectus (pages F-32 to F-34) for the values (measured by liquidation preference, not net proceeds), share amounts, etc. of Square's fundraising rounds. For dates, I am relying on CrunchBase, which has some slightly different dollar amounts. Math, and mistakes, are my own.

    I have also ignored the greenshoe, though I assume now that it will be exercised.

  2. My compound annual growth rate numbers are pretty approximate, basically fudging funding dates based on the CrunchBase data. I haven't bothered to calculate a CAGR for the 50 percent return in the, like, 16 hours since the IPO priced, though it is high.

  3. Incidentally, all of the preferred stock series have some form of ratchet. The prospectus (page F-34) says that "In the event the Company issues shares of additional stock, subject to customary exceptions, after the preferred stock original issue date without consideration or for a consideration per share less than the initial conversion price in effect immediately prior to such issuance, then and in each such event the conversion price shall be reduced" in a ratchety way. And since the IPO priced through the Series D price, you might expect the Series D to ratchet up too.

    But it turns out that the "customary exceptions" include an initial public offering: Those ratchets are only aimed at dilutive private share sales. (See Section IV(A)(4)(d)(i)(B)(4) of Square's certificate of incorporation.) So Series D's ratchet was never engaged. Only the Series E had a ratchet even in an IPO, so only Series E got more shares.

  4. My math here is just to hold market capitalization constant -- $2.95 billion at the IPO, $4.43 billion at $13.50/share -- but delete the extra 10.3 million ratchet shares and re-assign the value to the other shares. This is a little lazy in that it assigns some value -- $8 million at the IPO price, $12 million at $13.50 -- to the common stock sold in the IPO, which in this math would have been sold at like $9.29.

  5. More seriously, the ratchet encourages later-stage investors to invest even if they're worried about valuation. "If you manage to negotiate a can't-lose investment, you're more willing to pay up to get in," says Bloomberg Gadfly's Shira Ovide

  6. That's because the Series E ratchet was struck off a $9 price instead of a higher one. If the deal had priced at $13, there would be only 4.1 million ratchet shares instead of 10.3 million, saving Square 6.2 million shares worth $83 million at the $13.50 price.

  7. From the use of proceeds:

    The principal purposes of this offering are to obtain additional capital, to create a public market for our Class A common stock, and to facilitate our future access to the public equity markets. We plan to invest the net proceeds in short-term, investment-grade, interest-bearing securities such as money market accounts, certificates of deposit, commercial paper, or guaranteed obligations of the U.S. government. We currently intend to use the net proceeds received by us from this offering primarily for working capital and general corporate purposes. We may also use a portion of the net proceeds from this offering for acquisitions of complementary businesses, technologies, or other assets. We have not entered into any agreements or commitments with respect to any specific acquisitions and have no understandings or agreements with respect to any such acquisition or investment at this time. We cannot specify with certainty the particular uses for the net proceeds from this offering. Accordingly, our management team will have broad discretion in using the net proceeds from this offering.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

To contact the author of this story:
Matt Levine at mlevine51@bloomberg.net

To contact the editor responsible for this story:
James Greiff at jgreiff@bloomberg.net