Stephen Millbrook jumped up from his desk, dashed across the trading floor and looked out the window. The Empire State Building was still there.
On the streets, people were walking calmly through midtown Manhattan, dressed in thick coats and covered with scarves to fend off the near freezing February air. No one looked panicked. There were no signs of another attack.
The evidence for an attack was confined to Millbrook's trading screens. In a matter of minutes, the Dow Jones Industrial Average had dropped 200 points. That was on top of a 346 point decline that the market had already registered. A 546 point drop meant something had gone seriously wrong. We're talking 9/11 wrong. On the first day of trading following the 2001 attacks, the Dow dropped 684 points.
Shares of Goldman Sachs had cratered. Goldman Sachs! Millbrook dialed a friend's number at Goldman just to see if Goldman was still there. The guy on the other end of the phone — Thank God there was a guy on the other end of the phone! — asked him if everything was alright in midtown. Goldman traders located at the southern tip of Manhattan were wondering if someone had attacked Times Square.
Later Millbrook would learn that the sudden drop was due to a "glitch." Something had gone wrong with the computer systems of the New York Stock Exchange, triggering a flash crash. This was on February 27, 2007, however, and no one had invented the term "flash crash" yet.
Traders on the floor of the NYSE would wind up having to keep their books open past the official closing bell as the exchange struggled to put things in order. None of the traders had ever been called upon to keep trading open after the bell. One trader told me over drinks that night that this marked "the death of the God of the closing bell."
In the end the Dow closed for a decline of 416.02 points. That was the still the biggest point drop in the market since it had reopened after the September 11, 2001 attacks. It was the seventh ever biggest one-day decline.
Earlier in the day, trying to limit the declines following a 200 point drop, the NYSE had imposed trading curbs. The effect, however, was to give traders more time to worry.
The talking heads on television blamed the crash on Chinese economic data and comments about a possible recession from Alan Greenspan. Most ignored something that was arguably far more important — the announcement by Freddie Mac that it would stop buying subprime loans.
The housing boom had ended two years earlier. Freddie's subprime exit indicated that the mortgage bust was upon us. Within a few months, subprime lender New Century and a pair of Bear Stearns hedge funds focused on investing in subprime would go down. Over the summer, highly unusual market movements would trigger massive losses by quant hedge funds over the course of weeks that became known as "the quant bloodbath." The housing bust was quickly transforming into a financial crisis.
The numbers 416 and 546 were indicators that something was seriously wrong. Millbrook (obviously not his real name) was right to panic. But few of us understood this on that cold day in February.