The (not so) funny thing is that this has been a very well known problem for... ever. I'm reading "The House Of Morgan: An American Banking Dynasty and the Rise of Modern Finance" by Ron Chernow - highly recommended - and you can find the same mechanisms and scandals (with little variances) dating back at least to the 1920s.
Mixing trading and underwriting in the same firm is bound to create enormous conflicts of interest, no matter how tall the internal "Chinese walls" between departments are. This is imho definitely a place where strong regulations and strong checks are needed.
I've been on the buy-side for many years and the IPO market is a joke. Every time an IPO pops +20% it's irrefutable evidence that the banks are either incompetent and/or untruthful. I remain mystified that companies, particularly the hot issues like LinkedIn, ProtoLabs, et al, dont opt for auctions. It's the 8th wonder of the world.
The other solution would be to be to have an adjustable underwriting fee that declines depending on how much the stock is up using the ~5 day average closing price after the IPO. If the stock was up some percent over the IPO price, say 40%, the underwriting fee would drop from 7% to 1%. Over 50%? No fees. Something along those lines would keep the banks honest. You failed (purposefully or not) to price the offering correctly and needlessly transferred hundred of million in wealth from the company to trading clients? Fine, you dont get paid.
For the hot issues, trust me, there is still plenty of demand. Tons. You really think no one would buy Splunk at $35, that an IPO would fail? It happens millions of times per day already in the secondary market, so clearly there's more than enough demand. The difference would be that the hedge funds who generate utterly insane commission dollars wouldn't have an interest in the offering...it would be more buy-and-hold guys subscribing, and those folks just aren't that profitable for the bank trade desks. I used to work at one of these big funds and I've seen this all first hand and it's ridiculous.
The other solution would be to be to have an adjustable underwriting fee that declines depending on how much the stock is up using the ~5 day average closing price after the IPO. [...] You failed (purposefully or not) to price the offering correctly and needlessly transferred hundred of million in wealth from the company to trading clients? Fine, you dont get paid.
Except they still get kickbacks from their double dealing.
I remember Google's IPO vividly. Google's stock price didnt pop much. It priced at $85 and hit $100 the first day of trading. It hung around $100 for a few weeks and then started moving steadily higher throughout the fall, in large part because they reported an insane third quarter. The auction had other problems but the pricing was largely fine.
I'm talking about first day pops. These are situations that are clearly mismanaged. A stock that's supposedly worth $20 at 5pm on Tuesday is magically worth $35 at 9:30am on Wednesday despite nothing having changed. No supposedly functional market should be that inefficient. Unfortunately it's by design.
Reading this immediately made me think of what Facebook did right with their IPO. Uninformed observers criticized them for overpricing their IPO, while in reality (unlike LinkedIn, eToys) they succeeded in raising as much cash as they could (for more: http://blogmaverick.com/2012/09/04/facebook-handled-their-ip...)
Goldman didn't do what they are being accused of, and if they did there is nothing wrong with it. Besides, the best way to prevent this sort of thing from ever happening again is to not punish anyone, except possibly the guy who blew the whistle.
Also in the article, Goldman Sachs notes that these documents are a decade old. How could you trust (or even care about) a document from that long ago?
The following decade does add the amusing perspective that the "undervalued" shares were worthless a couple of years later. I wonder if they can dredge up documents finding that Goldman Sachs was aware that its grotesque undervaluation with respect to the market demand it was fuelling was simultaneously a grotesque overvaluation with respect to eToys' fundamentals. It would be funny if they tried to use "worried about the long term potential" as a defence.
I guarantee that if you look hard, you can find either an analyst or trader at GS who thought both of these things. That's because GS is a large bank full of many divisions, quite a few of whom are chinese walled from each other. (Admittedly, some of the chinese walling has happened in past 10 years...)
Banks are fragmented and individual portions of the bank don't always agree with each other.
Fun fact: a certain large bank (I won't say which one) is spending ~$10-90M on an internal matching engine. This is because they spend millions on unnecessary transaction costs - trader A wants to go long and trader B wants to go short, both of them hit the public markets rather than simply trading with each other.
Quite right, and a weakness of the original article is it doesn't make it clear how closely connected the "sales representatives" clamouring for repeat business were to the original IPO.
I'm intrigued by the possibility that an individual person closely connected with pricing the IPO might have expressed a view akin to: "we've got to set a sub-market price to ensure our institutional investors make a killing on the first day, because this stock will be in trouble once the hype dies down"
But still, that was over a decade ago, and I'm sure no investment banker would ever be so cynical today...
It goes towards their incentive to under price the IPO. If the incentive a decade old is the same as it is today, I doubt that things would be different.
Is there a reason more companies don't do what Google did and simply sell the initial shares by dutch auction? Even without the perverse incentives, I can't see any possible advantage to hiring a wall street bank to guess a price and allocate it to only a fraction of the market.
Google didn't do that, sadly. Their initial offering price was set by their underwriting bank, and most of the shares were distributed in the usual way. A small chunk of the shares were auctioned with the help of, IIRC, W.R. Hambrecht.
This is the 21st century. Startup CFOs need to start realizing that this is unnecessary and the democracy of the internet is more than enough to "raise your brand" legitimately.
Google unfortunately got punished by the financial cartel for doing that. Remember the IPO price was initially at $120-$130. And then during the quiet period, where a company can't defend itself, Wall Street commenced its hazing ritual. People don't remember this now, because Google is seen as successful, but there was all sorts of negative press doubting its business model and saying it was overhyped, and so forth.
Google was then forced to lower the IPO price to $80 or so. It's hard to say for sure, but that probably wasn't a fair price, as the shares were trading for more than double within 6 months.
Although they were also trading for 4-5x not too shortly after that, since the company was in a spectacular growth phase at that time, so it is hard to say.
Someone needs to build out the auction infrastructure where all market participants (institutional buyers, high-net-worth individuals, retail investors) can participate. Until recently Wall Street investment banks had a stranglehold on high-net-worth individuals, but that's changing.
SecondMarket is probably in the best position to implement this - they already have access to trading activity of mature private companies, and while they don't do auctions, they do run order books for large lots.
A common thing you read in articles like this (and I think Michael Lewis has mentioned it a few times) is that when clients go to Goldman Sachs, they know that Goldman is going to screw them somehow, but that the outcome in the end is still better than what they would get from any other investment bank.
Well, it's actually fairly complicated. Banks underwrite the sale of equity in an IPO, meaning that in one way or another they are providing a guarantee to the issuer - i.e., in the price of the sale, in the amount sold at a certain price, etc. So when it comes to the price of the guarantee (implicit in the terms of the arrangement), the issuer and the bank are sitting at opposite sides of the table.
As a consumer, you probably don't go directly to insurance companies to get underwritten for something (health, life, home, etc). You go to to a broker, who has a legal obligation to act in your best interest.
The investment banks are playing the line between moral and legal obligations. They will win, of course. And then hopefully we will see "IPO brokers" and corresponding laws appear.
Reading this brought me back to all the comments surrounding the Facebook IPO.
There were a bunch of people using the fact that the stock didn't pop and that they had to use the greenshoe to prop up the price as a sign of failure when in fact it was really a way to approximate the dutch auction process by overselling and then pulling back to get to market equilibrium.
Then there were all the investors saying they were hurt by the IPO because it didn't pop and they couldn't sell it a few hours after buying it, so they took a loss. They then argued that this was a bad move because now Facebook stock had a bad name in the markets which would hurt more long-term than if they had just submitted to common practice by giving them some free profits for doing no work.
The old days of making 4x returns in a few hours of busy work on a trading terminal must have been really nice.
Naive question from someone who knows next to nothing about IPOs:
Why not just divide the shares up amongst the existing shareholders, and just let them sell them normally on the market? Then the company/founders/vcs can just sell shares at whatever pace they want to, at whatever fair market value is.
It seems like the only possible outcome of a single day sale is information asymmetry, which means someone will always get "screwed".
Well, the reason that wouldn't work is because the primary objective of the IPO is to raise $$$ for The Corporation and not the "existing shareholders". The "I" in IPO is the critical part of the equation here since it is this initial exclusivity that lures large institutional investors from wanting to get in before suckers like you and I do. In turn, they swallow up very large amounts of shares guaranteeing the issuing company a large amount of money on the day of their IPO.
Furthermore, the Founders/VCs/Employees just don't have the reach or the technical know-how to appropriately price the stock to maximize their potential gains - the investment banks have both but, as you read in this article, they often use their expertise to their own advantage.
Lastly, the Founders/VCs are in fact able to sell off their shares during the IPO for their own personal benefit so long as there is no clause in their IPO terms stating that they cannot. This will enrich them personally BUT it is highly advised that they don't because the signal they're sending to the market is "cash out early!"
There is no such thing as information symmetry between any person, or in any trade. No two people on the planet have the same information.
> which means someone will always get "screwed".
Both parties always benefit from the trade, or the trade would not have taken place. Unless someone is deliberately misrepresenting information (fraud), then nobody gets "screwed" by trade.
They are not necessarily related. Information symmetry just means that both parties know exactly the same information and neither knows more or less than the other. In a scenario where one or both parties have information that is not accessible to the other they can still engage in a trade in which each party, at least believes, it is better off given their information. A neutral/third-party observer may not come to the same conclusion but that is not relevant to this scenario.
Why can't companies do a slow rolling IPO? So say you sell 1M shares on day 1 at $20 which is what your underwriters price you at. Then 2M shares on day 2 at market price. Then 4M on day 3 and 8M on day 4 etc.
> The problem is no underwriter would take an IPO for a million shares at say $10/share. It's just not something that they can make money on.
If you're offering a small fraction of the outstanding float, you don't need an underwriter because you've elected to reduce your risk. If markets are unfavourable, you raise less. Oh well. You've kept plenty of shares to try again next time.
We need to get over this "go big or go home" mentality.
I see your point, the problem is that if you don't do a road show, then no fund is going to invest in you, with google/facebook hyped companies excepted.
With out an institutional investor( Buy side) your stock will have a hard time getting any traction.
I guess it depends on your definition of allocated and offered, but to the exchange, they have minimum requirements for what is offered on their exchange.
If the stock won't trade, they won't let it list, period:)
No one will buy your IPO if its going to be at market price. You'll need to add a discount. Why buy from an IPO and do more paperwork when you can buy from the market instantly?
Also each time you sell shares in an IPO you need to have people make lots of calls to institutional and other investors to determine the demand and thus proper pricing and supply of new shares. Having to do this everyday will be expensive. You (as a bank managing the IPO) also probably don't want to pester fund managers too much (They have to take care of other stocks in their portfolio, too), otherwise you'll have trouble doing IPOs for other companies.
EDIT: I take that last sentence back, it seems like they really want those IPOs regardless if the bank is a vampire squid, or not.
No reason at all, except the finance business is not very innovative in areas like raising capital for companies. It would be easy to have a structure like an ETF where brokers could get more issuance from a company if they had excess demand for shares.
Underwriting does get you what it says, a guarantee that they will take the whole allocation, which was the original reason for underpricing. There have been a few cases where underwriters lost a lot of money (eg the BP floatation in the UK), but usually the underwriting fee is pure profit, especially if you can get the launch undervalued, and there are usually options that make it pretty hard to lose money (as with Facebook).
I'm not sure what this solves. If I know that the stock will have less competitive bidding 4 days from now, I'm waiting.
I'm a firm believer that releasing all the shares and allowing the market to decide what they'll pay for the stock, based on their own valuation method, is the most efficient method.
yes, this is the most efficient method for getting market value for something (its almost tautologous); however, thats not the problem identified. the problem is determining a fair price for the ipo so that the company being listed brings in as much cash as they should. releasing all the shares at once determines a fair price quickly, but that doesnt help the company bring in more cash
Why would a company agree to be "IPO'd" for such a low price? Doesn't the company and its shareholders need to accept some blame here?
Fair enough that the bankers may have nefarious intent and they shouldn't get away with it but agreeing to such a shitty deal out of ignorance is pretty stupid.
One of the reasons the company is taking on these underwriters of the IPO is because they don't actually know how much the company is going to be valued at in the market? GS would be in a way better position with their experience to know this.
If they didn't play both sides of this market it would be in their interest to price is as accurately as possible.
Should not the company hire some kind of professional(s) who will observe both sides(mostly the bank) and provide objective feedback? It seems like they were jumping into unknown territory without any knowledge.
I really can't believe all companies go IPO blind like I get impression from the article.
Most companies think they are hiring those professionals when they go with GS. Forcing GS to state very clearly that they do not have any fiduciary responsibility to their clients is kind of a bfd.
Look at these cases; there's a lot of guesswork, it's not rocket science. If you have no clue what your business is worth, and are deferring completely to a third party, why are you even going public?
When I read articles like this it's always centered around the bigger companies like Goldman and such, are there smaller companies who can handle taking an IPO public? I mean these guys are a business and need to make money, but feel like a smaller firm would be more transparent and willing to work with you and not for their other more well established clients.
In short, game theory strikes again: lazy auctioneers bid high blindly and hope that others will do the actual research to set the end price fairly, but those others have less incentive to do research because they are getting priced out by the lazy bums anyways. Another problem is that some bidders irrationally assume that a more popular stock is more valuable, so they too contribute to overpricing. As a result, sophisticated investors are staying away from auctions.
Are there any startups - or even startup concepts - that have a chance to disrupt entities like Goldman Sachs? How is it that a firm with such a reputation keep getting new business and very high fees?
Why isn't the price of a share set up to an unrealistically high value on the first day? The price would eventually go down to match the proper value of the share.
Mixing trading and underwriting in the same firm is bound to create enormous conflicts of interest, no matter how tall the internal "Chinese walls" between departments are. This is imho definitely a place where strong regulations and strong checks are needed.