This story is interesting. BATS is the third-largest stock exchange in the USA. It tried to have an IPO on Friday. It was a high-profile flop.
When a stock is listed for trading, one exchange is chosen for the listing, although shares may trade anywhere. For example, Coca-Cola (KO) is listed on the NYSE, but shares may trade anywhere.
BATS chose to list the shares for their IPO on their own exchange.
Trading was supposed to start on Friday. Unfortunately, BATS’ computers didn’t work properly. There was a surge of orders to BATS for the IPO. After their server couldn’t handle the load, they halted trading. They also rescinded their IPO!
This quote was interesting:
The software that failed Friday “was an area that was tested very rigorously over months,” Mr. Ratterman said. The software had a bug in handling orders the way they came in, he said: “That combination hadn’t been seen in the hundreds of tests we’d run before this.”
According to that quote, the reason for the flop was that BATS had defective software. There was a lot of volume due to the BATS IPO. This caused a previously-unnoticed bug to crash their server.
I recently worked at a large financial institution. I did QA on a big new project. I found lots of obscure bugs in their code. For example, they weren’t properly calculating coupon dates and accrued interest for Treasury bonds. “February 28″ plus 6 months is not “August 28″, it’s “August 31″. That’s one type of obscure bug I found.
BATS has offices in NYC. If they really wanted to, they probably could have hired me!
Most large financial institutions focus on hiring a lot of mediocre people, rather than hiring people who really know their stuff. People who are really productive don’t fit into the dysfunctional financial institutions. If BATS had hired one high-performing programmer, they might have avoided their IPO flop.
The BATS IPO was a miserable failure. The reason was a software bug. That’s one example of someone losing a ton of money due to bad software.