Nanex ~ 1-Apr-2014 ~ Refuting HFT Claims

1. But we provide liquidity!

Ask them to define what liquidity means. Here's the Chairman of the FIA Europe who completely gets it wrong. Here's an email exchange with an academic who helped write the SEC flash crash report who also, completely gets it wrong! Yes, HFT provides liquidity, but only if we are using the term liquidity to mean something else.

It gets worse. RBC created a product called Thor specifically to deal with the fact that HFT were cancelling orders at other exchanges when their market making software learned that a large order was in the process of executing. This means that during the execution of a market order, HFT are able to cancel their orders at other exchanges. That is not providing liquidity in any sense of the word.

Micro flash crash events occur when HFT suddenly pull their orders during the execution of large market orders. Micro flash crash events are a new phenomenon which first appeared right after Reg NMS when HFT began. Just a few examples:

• ITG
• The New Normal
• Mini Flash Crashes On The Rise
• Unclearly Erroneous
• Salesforce.com's $1.3 Billion Rise and Fall
• Facebook Liquidity
• The Most Valuable Stock in the World
• Coach Flash Crash
• Normalizing Deviant Behavior
• many more.

2. But we narrow spreads!

It's important to distinguish between electronic trading and HFT, one of those lowered costs, the other lowered ethics. Here's a chart from a pro-HFT camp that clearly shows spreads have not narrowed from HFT. Here's another chart showing the same thing, from the Virtu S-1 filing no less:

Just a few examples of data showing wider, or unstable spreads:

• NBBO spreads for all U.S. Stocks for every day from January 2007 through January 2012
• Quote Spreads in NMS Stocks
• Quote Spread Instability
• Tighter Spreads? Not at Market Open
• 1,262 Unique NBBO Spreads in 1 Second
• Quote Spread: 1 Cent or 1 Percent?
• IBM, KO, and MMM.
• many more here and here.

3. But we lower costs!

SEC Chairman Mary Jo White made this very misleading claim in recent speech (June 5, 2014), which we debunked here (using the same data and source!).

Many seem to forget that one could trade 100 shares for $8 electronically back in 1998. Thanks to the internet way back machine, these are easy to find. HFT proponents also never seem to mention the very real escalation in costs that HFT imposes on everyone - which are nicely summarized here.

4. But look at all these studies that show HFT is good.

Did you know that most (if not all) of those studies never looks at data sub-second and/or outside a very select group of stocks? Most focus on just one of 13 exchanges. And with few exceptions, are industry sponsored.

There's plenty of academic papers, that show harm from HFT, including this one that was published in Nature.

5. Those (insert expletive here) Nanex guys don't know what they are talking about. Why would HFT ever want to do X?

We get this a lot, for example, when we published articles about quote stuffing.

Here's an HFT lobbyist making that point, and our response.