Nasdaq halts high-speed trading service after regulatory pressure
ft.com
ft.com
It sounds weird. Normally if you want to attract or profit from high frequency traders, you offer the differential product (fast market data / colocation / direct exchange connection) to whomever is willing to pay and then you take pains to equalise latency between those participants. That's usually what both the market and the regulators want.
If you assume collusion with trading firms, privately offering special services to a handful of participants also doesn't make sense - better to do deal with one, who would pay more for the exclusivity and be less likely to publicly expose you. For a top market maker, competing with 50 others isn't much worse than competing with 5 others, but they might pay millions a month to have zero competitors.
So, ruling out bad business or corruption, what's left? It sounds like it might just be a compliance failure by Nasdaq: they marketed the feature to the obvious firms, the ones who were already paying for the fastest service, and neglected to do the right regulatory filings to advertise it to everyone. Which is not nothing - regulators enforce transparency on this thing for good reason - but is pretty close to a victimless crime.
Yet, this is the first I'm hearing of it. It makes me wonder if there was some kind of intent, or if it was offered on an 'if they ask about it' basis.
With the way trading arbitrage matters down to the nanosecond, depending on the exact type of connections they allowed this for and for how long, millions of dollars could have moved from some firms to others. There's no real way to know the scale of the impact.
One could argue that the real harm it does, is that since trading is cheaper, it encourages gambling.
I'm sure some clever person can figure out a way to exploit that. I'd be interest to hear how that would work.
I guess another way would be to quantize the time of trades down to a certain time threshold and two things come in at the same quantum, make them play Scissors, Paper, Stone to get them to decide who came first.
They are successful in reducing, but not eliminating, the advantages of lower latency. It's not clear that they make the market more efficient or cheaper for unsophisticated users overall.
Level playing field my ass!