How Facebook’s IPO was bungled by Nasdaq’s computers

http://qz.com/88987/sec-documents-show-how-nasdaqs-computers-bungled-facebooks-ipo/

by donohoe • 13 years ago

34 26 13 years ago

qz.com

downrightmike• 13 years ago

yeah, sure "the servers" did it

tantalor• 13 years ago

Gripe: Block quotes should have narrower margins than the body text, not wider.

tzs• 13 years ago

They should also actually be block quotes. Those aren't. They are images. I can't think of a good reason for this, becaus the PDF they came from is one that allows text selection and copying.

akozak• 13 years ago

I don't fully understand why NASDAQ ends up taking a position on stocks ("It just so happened that Nasdaq was shorting shares of Facebook at that time."). If it's a mere exchange, should it just be matching buy and sell orders? When does it take ownership of anything?

Is there a good explanation of how this works somewhere?

danbruc• 13 years ago

My guess is that its in their terms of service - they failed to consider all orders when they matched them and therefore have to account for that.

mikestew• 13 years ago

Because of the way they finally handled the snafu, they effectively ended up with a short position. NASDAQ did not set out to hold a short position, but because of how they covered customers who couldn't cancel their orders they ended up short FB. (They covered those customers because they quit trying to valid order matching by accounting for cancelled orders. They just took cancelled orders out of the equation.)

angli2• 13 years ago

My understanding is that they don't usually end up with stocks, but their handling of this specific situation created what amounted to a short position.

Essentially, the bug (whatever you want to call it is irrelevant from the financial angle) meant a bunch of people wanted to sell facebook stock and were unable to do it. They said that they were willing to sell at the current going rate, which was $42.

NASDAQ was unable to process the orders, so they weren't able to process the sales. The resolution they settled on was essentially becoming the other party to the transaction: they would provide the FB shares to the (possibly disgruntled) customers at the same price - as if nothing had gone wrong.

But, as this is happening, NASDAQ doesn't actually have FB stock - they have to buy it on the market later. Like a normal short, they benefit if the price goes down since they have to spend less to get back the shares. For example, if FB went down $2, NASDAQ only has to pay $40 for the share, and they make a $2 profit.

This is all a long-winded way of saying they didn't have a short in the traditional sense, and in normal cases this doesn't happen, but it happened as a result of this screwup.

gcb0• 13 years ago

still does not explain "if i want to sell, the exchange should match me with someone. they fail to match me with someone, but instead they buy my shares"

makes no sense. at all. unless exchange does not means what i think it means. and they should pay sales tax for shares as they are selling/buying them as a dealership buys and sells cars.

SeanDav• 13 years ago

The simplest explanation is that usually an exchange is essentially a conduit - they match buy with sell orders. If for some reason (i.e. computer glitch) they are matching only buy orders, they are creating short positions in their systems.

minimax2• 13 years ago

The exchange matches buy orders against sell orders. In this case they were matching buy orders against sell orders that had already been cancelled (which is not something that should ever happen). NASDAQ stepped in and took the other side of the buy orders, and so ended up with a short position.

DigitalSea• 13 years ago

I am tired of hearing this repeated excuse. When things don't go as expected everyone looks to someone or something to point their fingers at and blame. I seriously doubt this was a "server issue" and more of a deliberate delay to benefit a certain few.

mcnees287• 13 years ago

Do you really believe this? It would fit the narrative of the evil financial system, but I would be carful allotting credit for such a scheme. Sometimes shit happens.

NelsonMinar2• 13 years ago

Do you have any evidence for this fraud accusation?

NelsonMinar2• 13 years ago

I guess not.

fleitz3• 13 years ago

Yes, it was a cunning plan to make $10.2 million on a total market cap of $4.45 trillion.

pyvpx• 13 years ago

if you only had to press a button or two for $10MM and could walk away unscathed, wouldn't you?

fourstar• 13 years ago

And now the market has dictated its true value.

joezydeco3• 13 years ago

Time to repost the awesome Nanex analysis of the IPO.

"How HFT Caused the Opening Delay, and Later Benefited at the Retail Customer's Expense"

http://www.nanex.net/aqck/3099.html

And a great quote from the update posted within:

"This also brings another example of the dangers of placing a blind, mindless emphasis on speed above everything else. Algos reacting to prices created by other algos reacting to prices created by still other algos. Somewhere along the way, it has to start with a price based on economic reality. But the algos at the bottom of the intelligence chain can't waste precious milliseconds for that. They are built to simply react faster than the other guys algos. Why? Because the other guy figured out how to go faster! We don't need this in our markets. We need more intelligence."

fleitz3• 13 years ago

There's an incredible difference between a trading strategy and an investing strategy.

If you are on an investment timescale then really you should care less about HFT and trading. What is the brain if not an 'algo'?

Most of the people who complain about 'algo' are traders who are now obsolete, because computers can perform their 'technical analysis' faster and with less emotion, than a human can. They claim they offer 'intelligence' but don't offer much evidence that they actually do so.

joezydeco3• 13 years ago

If my investment strategy is "buy stocks and pray that some random HFT fuckup in the next 40 years doesn't wipe out my entire life's savings along with everyone else's", then the evidence is growing year by year that this strategy is going to be a losing one.

How Facebook’s IPO was bungled by Nasdaq’s computers

Related Stories