And that's only a tiny part of the larger industry that makes the infrastructure for electronic transactions of all sorts. Here's a random company in that field. On that site, go find "Switching engine demo" (can't link directly, it's all Flash). Watch out for the sharks!
My ignorance of the financial world is nearly total, but it seems to me that stuff like this must be either a symptom or a cause of the recent collapse. The existence of hundreds of thousands of hair-trigger electronic automated trading rules all linked to each other through near-instantaneous networks seems problematic. What kind of dynamic system does this create? Doesn't it sound like you could get small-scale bubbles happening on a second-to-second basis?
It also sounds like another indication that the financial industry is a net economic drain. Think of all the people working on shaving milliseconds off of financial transactions. What possible benefit are they bringing to society? Presumably the only reason to care about millisecond-scale latency of transactions is because it brings some kind of advantage, in other words, you are racing to make your trade before the next guy can. The relation of this to any actual economic value escapes me.
I am enough of a believer in capitalism to acknowledge that the capitalist function of allocating capital productively is worthwhile. In fact I found my way to this topic via a casual conversation with a VC -- and whatever you think about Silicon Valley VCs, at least they are actual investors, not pure speculators; unlike the sorts of people who slice up shitty mortgages and resell them and unlike people who are slaving away to shave milliseconds off of transaction times. But 95% of the energy of the financial world seems devoted to things that have no visible relation to actual, real-world economic value.
Is there a way to bring the financial world closer to reality? One possible answer is a tax on financial transactions, which would put some dampers on whatever crazy feedback loops might exist in this system. Hm, Keynes had this thought in 1936:
"Speculators may do no harm as bubbles on a steady stream of enterprise. But the situation is serious when enterprise becomes the bubble on a whirlpool of speculation. The introduction of a substantial Government transfer tax on all transactions might prove the most serviceable reform available, with a view to mitigating the predominance of speculation over enterprise in the United States.
And here's a recent op-ed piece saying much the same thing.