But classic econ 101 would also say that if the above was true, then everybody who thought the shares were worth only $15 (plus spread) would sell, resulting in share price dropping back to around $15.

I think you're wrong. My reasoning is simple. If the IPO investors thought that the company was worth less than $24 (e.g. $16), they would have sold already, and the price would have dropped. If they thought the company is worth $24 (or more), they would have bought the IPO at any price between $16 and $24 (minus spread) as well.

Really, the only argument for under-pricing the IPO is as a favor to the banks/hedge funds. But the question is, did the banks ever return the favor?

Isn't there some law that all investors buying stock at around the same time have to be offered the same terms?

I'm not a lawyer but I think this is part of what's behind the trend toward convertible notes in company financing in SV, compared to the traditional priced equity round -- too hard to give a different deal to different investors, for what amounts to the same shares?

Would love to hear more from someone who knows more about this.

I think it's very hard to argue that day 1 or near-term price action on any IPO is solely based on value instead of sentiment.

And yet finance 101 disagrees.

Yes, market cap should be independent of number of shares, and the market cap = # of shares * $ per share formula still holds, which is why many prefer to think about market caps rather than accounting units.

But it's also the case that a firm issuing a share for the price of $15, increases book value of the firm by $15, and this directly offsets the resulting shareholder dilution. Arguing it was impossible to find buyers at 24 dollars a share is a bit silly, given that the market immediately did just that.

Let's say that HarryCo, my own personal widget company, was IPOing. I'm selling 100 shares in the IPO.

Goldman Sachs would like to buy 50 shares for $14 each. Morgan Stanley would like to buy 50 shares for $15 each. Barclays would like to buy 50 shares for $16 each.

Citizen Bob would like to buy 5 shares for $24 each.

So what happens? HarryCo IPOs at $15 (the maximum I can get to cover all 100 shares) and then immediately pops to $24 when Bob buys 5 shares from Morgan Stanley.

I did not leave $900 "on the table" in my IPO because there wasn't enough demand to sell 100 shares at $24.

A lot of your hypothetical hinges on GS / MS buying at 15 and holding at prices below 24. Which is a lot like saying they actually value it at 24, but were somehow allocated shares at a bid lower than that. IMO, Google did it right with the Vickrey / second price auction, and apparently this has been formalized OpenIPO: into https://en.wikipedia.org/wiki/OpenIPO

If we assume that GS and MS have a reserve price of 24, as evidenced by the market open, then their strategic 15 and 16 bids do represent money left on the table.

Fact is Goldman Sachs - and maybe other banks - were allocated shares with a priviledged price during the FB's IPO as they were working on the IPO while buying stocks for their own clients. I know this was a special IPO, but my guess is that it is likely to happen for others

The special IPO is the one where that _doesn't_ happen.

You missed the last part:

> And then immediately crashes back to $16 because there are no other buyers willing to buy at $24.

Which obviously didn't happen.

I'll grant that that's the commonly-accepted way to look at market cap, but I think it leaves out a crucial assumption: "Given the current supply and demand for/of the firm's equity".

But I think that's not always a great assumption, because sometimes things happen that don't affect firm fundamentals (profit, revenues, margins) but do affect valuation, like changes in interest rates, shifting investor sentiment on the sector, a lighthouse investor publicly stating their intention to buy/sell, etc -- all of these things change market cap without altering the business per se.

All this to say, I think the simple CAP=SHARES*PRICE formula is too simplistic. I think it's more accurate to think in terms of ranges of what the firm's equity might be worth, and realize that it's a highly subjective assessment based more on aggregate opinion, than some hard objective law of physics.

Indeed! This is the same reason that when companies get bought (like LinkedIn) the price is generally a decent premium over the last traded share price.