Getco cuts staff with trading volumes thin [Financial Times] http://www.ft.com/intl/cms/s/0/b6bf723e-b5a9-11e1-ab92-00144feabdc0.html The HFT world does seem to be shrinking, with staff cuts, firms putting themselves up for sale, etc. It's not just declining volumes -- it's also shrinking spreads, competition, etc. -- secular changes in market microstructure that should ultimately benefit longer-term investors, provided liquidity is available after all is said and done, of course. That said, brokers/PFOF partners trading against their retail clients (through internalization or PFOF deals) are probably doing OK as they have a captive audience -- unfortunately -- to the detriment of their clients, not to mention the market as a whole. But that's one for the SEC to fix -- if they can ever overcome the resistance.
I mentioned it in another thread, in 1.5 years there will be nothing left that was once known as ultra high frequency trading. Of course pretty much every bank and hedge fund will be colocated to run their algos but it will be the "new normal", meaning, regular execution of orders in order to get into regular intraday or longer-term frequency positions. Shops that made their entire pnl at the detriment of longer-term investors will sell themselves out or simply die because they cannot sustain their high fixed cost base anymore.
PFOF & internalization is the name of the game. Retail muppet spread paying market orders still get filled at NBBO. No worse. Wholesale gets price improvement, making that spread. And the the bottom-feeding screen based market makers get no action, love, or volume. Tough times for high speed nerds.
Technology is generally only paid for once, though. While speed matters, the incremental improvements in speed just won't be paid for anymore. The fixed cost coming from the labor (the programmers) are what will no longer be sustainable. You can't pay guys 400-500k for 2 more microseconds anymore. I'm only speculating as an outsider (link no longer works), but the reason for layoffs probably has to do with lack of growth, more so than an outright lack of profitability. Or maybe there is a decline in profitability, so there was some kind of refocusing on areas of growth, where some other areas got axed. Decline of overall volume should be a concern for everyone who is a trader, though. Not just HFT. We'll all be fighting over a smaller piece of the pie.
A few things are going on at Getco, not just trading volume. Getco is certainly not as profitable as before, but I think that's pretty much most of the HFT industry (i run a HFT operation myself, so I would know a little). Two, Dan Colemen bacame CEO of Getco in Feb, since then he has been getting some fairly senior ppl from UBS (Coleman was head of equities at UBS). Getting rid of ppl under the pretext of a layoff is a well known tactic for re-organization.
I'm expecting more HFT shops to give up, and to try to sell themselves to banks, as technology IP, modern trading platforms etc. Lots of old infrastructure & 1990s trading platforms in banks.
i've already said HFT has peaked. it has zero scalability and too many players doing the same thing = bad for profitability. There are no secrets in trading.
While I will admit to a severe lack of knowledge about the back-end part of trading, it seems to me that the pricing of an asset would need to be nearly the same globally ("the law of one price") or else something is very out of whack. If I as a retail trader buy or sell a futures contract at a given point in time at price X, are you saying that someone with access to a non-exchange-based market can buy or sell that same (or an essentially similar) asset for less than X? I guess I can see some block trades getting someone better pricing, but by how much? Again, my question may reflect more on my lack of knowledge than anything else, but I'd be interested to know if I'm interpreting your statement correctly.