Google IPO questions

I’m a financial journalist, but I’ve never pretended to understand the stock

market. Bonds, yes; stocks, no. A recent Reuters

story, for instance, includes this bizarre segue:

“In a deal like that where it’s priced for perfection, anything that occurs

that isn’t right on the number, you get hammered,” said Jim Huguet, chief

executive officer of Great Companies LLC. The Florida firm manages $230 million

in technology shares.

Of the thousands of U.S. public companies in the United States, barely more

than a dozen have prices above $100 per share and trade at least 10,000 shares

a day. As of mid-afternoon Monday, none of the Nasdaq-100 stocks (.NDX) or

the components of the Morgan Stanley High Tech Index traded over $90 a share.

The US stock market is obsessed with dollar price – you often hear people

jumping up and down saying that such-and-such a company just rose or fell $5

on the day, without bothering to mention what the bloody thing is selling at.

This is just another example of the same syndrome: the journalist clearly reckons

that being "priced for perfection" is more or less the same thing

as being priced above $100 a share.

The way I see it, pricing the shares in the triple-digit realm is basically

a way of deterring speculators and trying to ensure, as much as possible, that

the people who enter bids in the IPO are the buy-and-hold investors that Google

is looking for, rather than small day-traders. It probably doesn’t make a whole

lot of difference, but I reckon that a $20 stock going to $24 in the course

of a week is probably slightly more likely than a $130 stock going to $156 in

the same timeframe. So if you want to decrease volatility, price high.

The key way that Google is ensuring low volatility, however, is through its

use of a Dutch auction. I’ve touched

on this before, but basically the price is set by investors, not underwriters,

which means that no one’s going to buy with the intention of selling almost

immediately. Daniel Gross says

in Slate today that "most professional investors will likely boycott the

offering" – but that doesn’t really matter: they’ll have to buy sooner

or later (and sooner rather than later, I think), and so will provide a natural

offset to the phenomenon of the "winner’s curse" which is often associated

with auctions.

But amidst all the concentration on the style of IPO that Google has chosen,

I get the feeling that people are overlooking some rather obvious questions.

So, here are two of my own; I’m sure there are more. If anybody would like to

hazard an answer, I’d be very interested.

1. What’s with the fees and the underwriters? Google has two

lead underwriters – Morgan Stanley and CSFB. Normally, IPO underwriters

have a lot of work to do: they have to value the company, set an issue price,

and market the shares to investors. In this case, they can basically sit back

and let the internet do the work for them: they simply issue the prospectus,

wait for the bids to roll in, and then use those bids to set the offer price.

Not a single phone call needs to be made to a single investor.

Daniel Gross says that "to add insult to the injury of the chastened investment

bankers, Google has decreed that it’ll only pay a 3 percent underwriting fee"

– but in an offering that could reach $3.3 billion, a 3% fee comes to

an extremely respectable $100 million. Does anybody really think that the banks

are doing $100 million’s worth of work on this deal?

Even more puzzlingly, Google has taken its two lead underwriters and saddled

them with 29 – count ’em – extra

underwriters, comprising pretty much all of Wall Street and then some. What

on earth are 31 underwriters supposed to do in this deal? They’re not drumming

up investors, so what’s their role, beyond earning fees?

2. Why is Google selling 14 million shares? A large part of

the reason why Google didn’t go public ages ago is that it has no need for cash.

It is minting money, actually – even after paying for what is probably

the largest, most expensive and most sophisticated server farm in the world,

it made $79 million last quarter, and is now sitting on $548 million in cash.

That’s enough money to buy one hell of a lot of Bloggers:

it’s hard to see what use the present half-billion is to Google, let alone the

$1.7 billion or so it stands to receive from the IPO.

There are, of course, good reasons for the IPO beyond raising money for Google.

Most obviously, Google was funded with venture capital, and venture capitalists

want an exit strategy. Existing shareholders are selling about 10 million shares

in total, which is much more than enough to give Google an unambiguous stock-market

valuation and, should it need it, a currency for further acquisitions. My question

pertains to the 14 million shares Google is selling over and above that number,

with the proceeds going, we’re told, "for general corporate purposes".

If Google is already profitable, what use has it for having $2.2 billion sitting

in the bank?

Raising equity, for a company like Google which is likely

to sell at more than 150 times its previous four quarters’ earnings, is cheap.

But even so, is the accumulation of an enormous pile of cash really the use

to which Google shareholders would like to see the company’s equity put? Or

is there (and I’m genuinely ignorant here) some kind of rule which says that

a company has to sell at least as many shares in an IPO as its existing shareholders

do? Absent such a law, I simply can’t see why Google is doing this.

This entry was posted in Finance. Bookmark the permalink.

14 Responses to Google IPO questions

  1. charles says:

    What kind of pissant speculator is deterred by a $100 share price, and do they ever move markets?

  2. charles says:

    And PS, if it were true that there’s less volatility in the prices of expensive stocks, given that investors are risk averse, every company listed would have $500 stocks, no-one would ever dream of splitting stocks and there’d be a few less bankers. Reasons for hoping its true, but I bet it isn’t.

  3. Felix says:

    I have to say that I have enough faith in efficient markets (!) to believe that share price makes bugger-all difference, really. But, Charles, you do seem to be implying that you do know why companies split their stocks — pray tell — why would a company prefer a $20 stock to a $120 stock? I’ve heard one theory that it’s to do with maximising the float, or the number of shares outstanding, in order to decrease the power of short-sellers, but I don’t really buy it.

  4. simian says:

    To answer the question of whether (empirically) stocks with higher price have less volatility, I ran a simple historical analysis.

    Turns out that higher price generally does mean lower volatility (correlation is -.17 between price and standard deviation of returns), but the explanatory power of the relationship is weak (linear regression has an R squared close to 0). In particular, the relationship is concentrated at the tails: stocks with extremely high prices (>$100, like Berkshire Hathaway and about 5 other stocks in the top US stocks 1000 by market cap as of the beginning of the study, in early 2002) do have lower volatility – not surprising in this one case. Stocks with very low prices (

  5. simian says:

    Stocks with very low prices – less than $10/share, about 25 stocks – have significantly higher volatility.

    Before decimalization this made more sense, as the smaller tick increments were a larger percentage of price for low-priced stocks. It’s also related to distress – stocks trading at $5 generally didn’t start out that way and are likely to be more volatile, subject to short squeezes, etc.

  6. charles says:

    Felix– God knows why companies split stocks, I implied it was because it gave bored bankers something to do, maybe it gets free publicity, I dunno.

    Simian –there’s not much to explain if the R-squared is that low (and, I’m guessing the coefficient wasn’t significant, right)? But what little there is to explain may be due to underspecification –maybe very expensive stocks are more mature for example, and that may well correlate with lower volatility…

  7. Nathan Myers says:

    A cynic would say that starting over $100 means it is just that much longer until you get de-listed.

  8. Sterling says:

    Felix – everyone participating in this dutch auction is a speculator. It’s possible there may be settlement delays slowing down flipping (I’m not sure how they’re handling clearing) but flipped the stock will be. It’s easier to argue that some speculators will be put off by the novelty of the dutch auction, and will instead wait to jump on it until it hits the secondary market. It may very well see a significant pop.

    One possible reason Google is issuing all this stock is that it intends to buy it back at a lower price than its initial offer. The dutch auction shares are to have no voting rights, so they can let those shares sit out there as long as they like, and use the pile of cash to buy back the standard IPO shares (which do have voting rights).

    Google doesn’t want to be a public company, or it would have gone to IPO long ago. This may just be a tactic to generate a lot of cash, and to be able to continue to do business as usual.

    I don’t care for it.

  9. Felix says:

    Sterling — I’m not following you here, I’m afraid. Firstly, “everyone” in the auction is a speculator? Surely untrue: I’m sure lots of people are buy-and-hold investors who want to be able to look back and say that buying Google in ’04 was like buying Microsoft in ’84.

    Secondly, the way the auction is structured, the ONLY way there will be “a significant pop” is precisely if a lot of the speculators stay out of the auction, and then rush to buy in the secondary market. In other words, what you’re saying is that everybody who’s participating in the auction is a speculator betting that other speculators WON’T be participating. Seems needlessly complicated to me when it’s much easier to simply consider them to be investors buying a company at a certain price of their own choosing. Occam’s razor, and all that.

    As for the distinction between “dutch auction shares” and “standard IPO shares”, I think there is none. You’re right that the shares being offered have few (not zero) voting rights, but the real voting shares aren’t being offered on the market at all, as far as I can work out, and therefore can’t be bought back. Do let me know if I’ve got this wrong.

    You’re probably right, though, that Google doesn’t want to be a public company, and that that is a reason not to buy it. After all, it will be run for the benefit of a handful of controlling shareholders, not for the benefit of those who own the listed shares. I still don’t see why it wants a pile of cash, however.

  10. Sterling says:

    Felix – unless someone’s primary goal in purchasing a security is dividend income, that person is a speculator. Now, there are short term speculators and long term (or buy-and-hold) speculators, but everyone who buys Google is going to buy it because they have an expectation of receiving more money for the shares than they pay for it. And I can’t imagine that a lot of buy-and-hold investors are going to be that enthusiastic about Google on Day 1.

    I believe that Google is issuing two classes of stock with this IPO, one with voting rights and one without. (I’ll double-check this in the morning and report back.) They’re basically running two concurrent IPOs – one through traditional underwriting channels and one through the Dutch auction, and it would be impossible to do this with one class of stock, for obvious reasons.

    Now, on to the more complicated issue of participation in the Dutch auction. Most investors never get access to quality IPOs – underwriters with retail operations normally route the shares to their best reps, and to their best reps’ best clients. Given that most investors have zero familiarity with IPOs, and given that participating in something as exotic as a Dutch auction is ITSELF a barrier, I think that a lot of interested smaller investors will simply wait until Google is available on the secondary market. This may cause the stock to pop – both the A & B shares. I’ll do some more research on this tomorrow and let you know.

  11. MemeFirst says:


    Is Playboy hot again? Never mind the stories about hot young film stars partying with Hef at the Playboy Mansion; I’m talking about the magazine itself. The latest issue includes a big article on Google, which nearly derailed the highest-profile…

  12. Joe says:

    Felix I was wondering if you could help me understand why would a stock company buy back stock in the secondary market. Are there any advantages or benefits?

    Also what reasons would leveraging a business with preferred stock versus common stock have?

    I’d appreciate the help.Thanks!

  13. Very nice post, I was expecting something like this from you. keep up the good work.

  14. james lee says:

    I wish i would have gotten in to the google ipo. What a great stock that his has been.

Comments are closed.