Without trying too hard, you should be able to get to 100 microsecond tails (CPython and perl prototypes can hit 80 usec at the 90th percentile).
Generally people use 10G nics with some sort of DMA (e.g. solarflare nics support a solution called OnLoad).
Though it should be said that, over the past few months, the effect of a few microseconds is mostly irrelevant (though being slow by milliseconds can be calamitous)
How in the heck do these shenanigans add any possible value to the economy?
They also may increase trust that the current pricing makes sense. When you buy some stock, it is more likely that you're getting a "fair" transaction.
So even if you use customized, co-located hardware/FPGAs or even ASICs with the fastest interconnects and unlimited bandwidth and your engine responds in nanosec, still the message you get from/send to an exchange can sit there in a queue for something like eternity.