Why there’s less high-frequency trading

By Felix Salmon
October 15, 2012
Nathaniel Popper arrives today with something that looks like good news on the high-frequency trading front: there's less of it!

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

Nathaniel Popper arrives today with something that looks like good news on the high-frequency trading front: there’s less of it!

Profits from high-speed trading in American stocks are on track to be, at most, $1.25 billion this year, down 35 percent from last year and 74 percent lower than the peak of about $4.9 billion in 2009, according to estimates from the brokerage firm Rosenblatt Securities…

The firms also are accounting for a declining percentage of a shrinking pool of stock trading, from 61 percent three years ago to 51 percent now, according to the Tabb Group, a data firm.

This is all true, and in fact it probably is good news, at the margin. But it’s not very good news, and it’s not as good news as it might look at first glance. Because while the number of trades is indeed going down, the number of orders is going through the roof. Here’s how I put it in my Radio 3 essay:

One reason that volumes are dropping is that the algobots are getting so sophisticated at sparring with each other that they’re not even trading with each other any more. They’re called high-frequency traders, but maybe that’s a misnomer: a better name might be high-frequency spambots. Because what they’re doing, most of the time, is putting buy or sell orders out there on the stock market, only to take those orders back a fraction of a second later, and replace them with new ones. The result is millions of orders, but almost no trades.

Call it the Stalemate of the Spambots: the HFT algos are all so sophisticated, now, that they just ping each other with order spam, rather than actually trading shares. Naturally, if you don’t trade shares, you can’t make money. But at the same time, anybody who does trade shares risks getting picked off by the very algorithms which are increasingly circling each other like prizefighters who never land a punch.

All of which is to say that just because HFT algobots aren’t trading as much any more, doesn’t mean that the waters are any safer for real-money accounts to re-enter. Indeed, the exact opposite is more likely: that the bots have poisoned the stock-trading waters so much that even the bots themselves fear to go in.

As a result, market regulators still have a huge amount of work to do, starting with a serious attempt to cut down on quote-spam. There’s no reason why regulators shouldn’t effectively ban the practice of putting in non-serious orders which disappear in the blink of an eye — although the risk, of course, is that if the algobots are banned from confusing each other with quote spam, then they’ll just revert to dominating trading instead. Which is why I still like the idea of a financial-transactions tax.

Popper says that “now that the high-speed firms are shrinking from the market, there are some indications that trading costs may again be rising.” This might be true, but it’s negligible: we’re talking here about a tiny uptick from 3.5 cents per share to 3.8 cents per share, after a long fall from a level of 7.6 cents in 2000. There’s no indication that this is either a trend or anything to be worried about.

In any case, let’s not assume that rising trading costs are always and necessarily a bad thing. Trading costs right now are incredibly low — low enough that they can, actually, rise a little bit without doing any visible harm. Fear of rising trading costs must not prevent us from continuing to prosecute the war on HFTs — especially if there are indications that we’re slowly beginning to win it.

10 comments

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/

If orders are the problem, why not a financial orders tax? If the goal is to discourage HFT, seems like that would disproportionately affect the HFT guys without getting much in the way of real money traders.

Posted by TalF | Report as abusive

Not sure I agree on the financial transactions tax. Some useful things they could do though is 1) put a requirement that orders can’t be canceled until a certain period of time has passed, 2) sub-decimal pricing, 3) make the maker-taker fee a function of the bid/ask spread (when the bid/ask spread widens either absolutely or relative to recent history, those who make liquidity should be paid more).

Posted by jmh530 | Report as abusive

+1 on financial transactions tax. Small, uniform, and universal. Slow ‘em down.

Posted by Eericsonjr | Report as abusive

All good ideas. If you can’t run your operation without spamming the book, it isn’t worth running. And a small fee won’t get in the way of legitimate trades — while providing clarity to the market activity.

Efficient markets require freedom of information. The present system is intentionally opaque.

Posted by TFF | Report as abusive

I love jmh530′s idea of forcing bids to stay good for some minimum period of time. Even if you made it something like five seconds, that would nuke much of the HFT nonsense, while remaining invisible to normal human beings.

Posted by Auros | Report as abusive

@Auros After thinking on it more, I would instead prefer something similar to a taker fee or something like that. So a quote that is out 1s or more (or something like that) has no extra fee, but the less time the quote is available before being canceled the greater the fee.

Posted by jmh530 | Report as abusive

This just shows that we have reached the limits of arbitrage on the time scale. I like Cowen’s idea of decreasing the size of the minimum spread from the (absurdly large) 1 cent, to something like .01 cents. Force the bots to compete on price instead of just time, and everybody wins.

Posted by qusma | Report as abusive

there is a very simple reason for less HFT activity and profits that seem to be lost on the author; the fact that the last 6 months have been the least volatile in 6 years. In times of high volatility, there is more demand for liquidity and as suppliers of liquidity there will be less demand for their services.
I find it quite unbelievable that the author still maintains that HFT is a bad thing when he actually shows a graph depicting a fall in institutional trading costs by almost 50% in the last decade! So where exactly are these negative effects?

Posted by wilbur2012 | Report as abusive

This and other topics that are relevant for speed traders and institutional investors will be discussed at High-Frequency Trading Leaders Forum 2013 London, next Thursday March 21.

Posted by EllieKim | Report as abusive

stylish, easy and restrained