The SEC May Stop IEX, But CNBC's Commentary Will Not Be Why

|
Includes: BATS, ICE
by: Kurt Dew

Summary

A comment found on CNBC, May 25, badly misinterprets the state of HFT technology.

This comment urges the SEC not to permit “speed bumps”.

Every major exchange already has a speed bump.

IEX is vulnerable to such poorly considered attacks.

Tell me lies, Tell me sweet little lies.

- Fleetwood Mac

It was with some surprise that I noted the qualifications of the authors of "Why the SEC should just say 'no' to IEX," found here on CNBC on May 25. They are professors at top US universities. Understanding the impact of High Frequency Trading (HFT), and the remedy IEX offers for its abuses, is a challenge for all of us. But at least most of us make an effort. The authors of this article have not made an effort.

For readers that have not been following the attempt of IEX to become an SEC-approved exchange, there is a terrific description of the struggle IEX is enduring in this Bloomberg article.

"The Argument" between CEOs of IEX and BATS, on CNBC.

The flawed "answer" the authors provide to the question:

Should the SEC permit the IEX' 350 micro-second intentional delay in the arrival of orders, known as the speed bump?"

is:

Deliberately adding any amount of latency [delay in the time from order placement to order entry] to the current market system would be a step backwards, especially when applied in an unequal way." (my insertion)

What is a "speed bump?" You can read about the speed bump and the purpose to which IEX puts it in my recent article, here. This "answer," provided by the authors is uninformed about current exchange practices and misleading about the rules of IEX. Let me explain.

1. All the major old-line exchanges [NYSE, a subsidiary of Intercontinental Exchange (NYSE:ICE), Nasdaq (NASDAQ:NDAQ), and the BATS (BATS:BATS) exchanges] deliberately delay orders using the same method as IEX, coils of cable through which the order's signal must travel. For a discussion by Nasdaq Chairman Bob Greifeld of Nasdaq's "delay coil" of cable and its purpose, read here. For a list of the different charges at NYSE for different latency speeds, see their latency rate prices here.

As these references make clear, for these other exchanges, the delay is for the purpose of introducing inequality into latency; the purpose is to create faster speeds for HFT customers who have paid for this privilege. All orders placed on the IEX, in contrast, go through one cable, and thus are treated equally, in that sense.

You cannot buy less latency at IEX. In the case of the three old-line exchanges, the HFTs' order arrives faster, the speed depending upon how much the HFT has paid the exchange. But counter to the impression left by the article, the purpose of the speed bump at IEX is not to create inequality.

But if intentional delays are not the answer, there is nobody with the answer. It is also of note that the other exchanges do not bother to ask the SEC for permission to delay their orders. They are entitled, it seems.

2. I was taken aback by the article's accusation that the IEX delay is applied unequally. All incoming orders to IEX cross the same speed bump. The article ultimately explains their case: "… hidden-pegged orders, which automatically reprice to preset benchmarks (such as inside bid and ask quotes), are able to bypass the delay when repricing based on market movements."

I believe this description of the "unequally applied" delay, rather suggestive of deception on IEX's part, refers to a second important innovation introduced by IEX in its application for exchange designation. The important innovation: the 'discretionary peg'. The Bloomberg article cited above, explains briefly why and how the discretionary peg was developed by IEX. I explain in more detail below.

The authors could not be more misleading. First, the orders that are able to benefit from the discretionary peg do indeed pass through the speed bump just as all orders do. But they are 'smart'. Thus, they may change automatically after they enter the deck, in response to the behavior of other orders in the deck, after they are placed and entered.

The discretionary peg has a patent pending. (The discretionary peg, nonetheless, has been imitated - ineptly and unethically - by the New York Stock Exchange, as I explain in an earlier article, here.) The order was a response to a discovery IEX made while studying the frequency with which orders were drawn away from IEX in response to better prices elsewhere - a hint that IEX prices might be becoming "stale" somehow.

The research that IEX was conducting was something I've waited to see. If IEX was ultimately nothing more than a speed bump, I did not think the exchange would meet the enormous potential created by its independence from the Big Boys who now dominate all financial trading.

It would be foolish for anyone, especially IEX, to think that IEX's speed bump is the end of the attack on IEX by HFTs. HFTs are going to think of more and better ways to race the delayed orders. And so they have, it seems.

The HFT traders were predicting the incoming orders before they arrived, IEX discovered, and their attack on the order was being made before the order was placed. So IEX responded cleverly. The exchange gave clients the ability to place an order that lifts itself when IEX's computers observe one side of the market begin to "crumble," as HFT traders pull the rug out from under orders placed at the midpoint of the bid-ask spread. (That is, when the HFTs began to lift their orders, intending to pick off that of a customer.)

You will not be immune from HFTs, even on IEX, unless IEX wins the race with HFTs to keep your order safe. The good news is IEX tries. The other exchanges ask the HFTs how they may help them manipulate the other customers' orders, and what the HFTs are willing to pay for the opportunity.

So yes, there are resting orders at IEX that can be executed faster than an order from outside. But every order passes through the speed bump. I don't mind if I share a market with that kind of order with any trader at IEX, since I have no plans to manipulate IEX prices.

If one fine day all the shenanigans of the crafty HFT traders cease - the dealers who feel entitled to change the rules of their captive exchanges at will, the dealers who can find experts such as these authors to explain the imaginary failings of IEX, the dealers who feel no need to request permission themselves from the SEC to manipulate the old-line exchange rules and policies - IEX has already pledged to remove the speed bump and the discretionary peg.

There will be more high profile articles dumping on IEX. Money talks. But if you want a thorough, carefully documented, but careful discussion of what IEX offers that is different from the other exchanges, you need look no further than the IEX comment letters on their exchange application to the SEC.

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.