Hacker News Reader

Comment by seizethecheese | Hacker News Reader

Others have mentioned that a soaring price after IPO means that the company was undervalued. What I don't understand is why the mechanism for IPOs hasn't changed to better capture the correct value. Why couldn't a company auction blocks of shares instead of having to set a specific price? My understanding is that companies have a guaranteed floor on valuation by the bank helping them go public, but this could certainly co-exist with some sort of auction format.

View on HN

Dutch auctions have been tried before, with mixed results, most notably by Google.

The article below reviews the GOOG IPO. It's not entirely accurate in all the details (e.g., that GOOG would have paid a 7% fee (pg 429) - in reality, it would have been much lower - GOOG was a large and hot IPO). Nevertheless, it's a good summary of the process.

http://scholarship.law.berkeley.edu/cgi/viewcontent.cgi?arti...

There's no longer a bank valuation "guarantee" for IPOs though, and hasn't been for many years. If an equity raise (IPO or otherwise) isn't going well, the offering price will be reduced until there's sufficient demand, or the issuer will pull the deal.

Google's auction was not actually a Dutch auction, but a second price auction. And despite the thought that it's the "perfect" way to do an IPO, the auction setup actually cost Google millions of dollars.

Here's a link to a more detailed explanation - http://optimalauctions.com/designing-a-better-google-ipo-auc... Tldr; demand is a curve, not a point (Note - I'm an auction design expert)

My suspicion is that this is currently more art than science; the banks/IPO underwriters want the stock to pop, aside from even the guaranteed floor you mentioned. They want the psychological effect that comes with that "pop", that shows up on potential investors' radar.

On top of that, I imagine it's a little tricky because an IPO is when the early investors cash out, so they're the largest stake-holders in the company, and creating that "pop" in the hope that people buy-buy-buy to drive up the price creates a good return for those investors/stakeholders. I don't like it, since it can essentially screw the company out of cash, but that's my guess as to why such a thing exists.

On the other hand, a decent number of companies going public these days are going to screw over the investors anyhow (cough Zynga, GoDaddy, King), so if that means they get slightly less cash in the bank, I guess I don't really care.

I believe the early investors can't completely cash out since they're probably subject to a lock up period on the majority of their holdings.

I know employees & officers in the company are, but are the investors, as well (honest question)?

Yes, it's very uncommon (but not unheard of) to have large blocks of stock that aren't locked up in an IPO.

For SHOP specifically, 99.9% of the Class B stock (i.e., stock outstanding pre-IPO) is locked up for 180 days. (See underwriting section of the prospectus)

http://www.sec.gov/Archives/edgar/data/1594805/0001193125151...

> I don't like it, since it can essentially screw the company out of cash,

I'm not saying you're one of these people, but there's a general meme of non-financial people who don't understand that justifications for raising capital can indeed be to liquidate investors (or founders) positions in companies. This is very normal. Investors should know (I believe they're legally obligated) what their capital is being used for. If they don't believe their capital will increase the value of the business, then they simply wouldn't invest.

The market should correct itself in this case (if not leaving enough cash in the company is a problem).

Google did auction blocks of shares. They ended up getting less for the shares than they were expecting to sell them for, and the price still rose by 18% later in the day.

On paper the Vickrey auction method still makes far more sense, but bankers don't like it because they can't give preferential treatment to their favoured clients.

The modest 18% increase is likely due to the fact that demand for Google's shares increased since they held the dutch auction. Compare this to Netscape which jumped well over 200% on its first day.

The 18% increase was a result of the auction designs's flaws. Buyers of course have a demand curve but for this auction they were required to pick a single demand point. So, googles IPO price was $85. If I was a big buyer who actually bid 100,000 shares at $120 for the auction, I would have received my 100,000 shares at $85. But my demand is a curve - I would probably be willing to buy 125,000 shares at $100 or 175,000 shares at $85. So, the price increase on the first day was more due to bidders filling their demand at the IPO price.

EDIT: here's a link to a more detailed explanation, also linked elsewhere on this page - http://optimalauctions.com/designing-a-better-google-ipo-auc...

There is a devastating flaw in that proposal: it fails to funnel money into the pockets of the well connected.

Investment banks have a cozy relationship with VC, often as the source of significant portions of their funds. VCs have relationships with the startups that are most likely to go public, often with controlling stakes by the time they actually go public. It's a standard case of reciprocal back patting. It takes a well bootstrapped and successful company to buck the trend, and those are rare.

Because institutional buyers want to get something in exchange for agreeing not to sell for a long time and buying a risky issue that hasn't traded before.

Reply on news.ycombinator.com