11.19.2015

OPINION: Regulators Come Closer to Defining HFT

11.19.2015

Ever since the May 2010 ‘flash crash’, regulators on both sides of the Atlantic have adopted U.S. Supreme Court Justice Potter Stewart’s ‘I know it when I see it’ mindset towards high-frequency trading, rather than formally defining the practice themselves. However, the EU and European Securities and Markets Authority have painted themselves in to a corner with MiFID 2 and MiFIR.

Under MiFID 2, trading firms will need to synchronize their server clocks to coordinated universal time (UTC) so regulators can precisely reconstruct market activity from time stamps.

If a firm trades via voice or uses a request-for-quote (RFQ) system where human involvement is necessary, those trades need to be time-stamped in one-second intervals that can have no more than a one-second divergence from UTC time. High-frequency, algorithmic trades need to be time-stamped in a microsecond interval that cannot have more than a 100-microsecond divergence from UTC. Finally, any trades do not fall into the previous two buckets must be time-stamped with a one-millisecond increment and cannot diverge from UTC time by more than a millisecond.

Determining whether a trader used a voice system is simple. But determining which of the two electronically executed buckets a trade falls into, not so much.

European regulators have two approaches that they might use to define high-frequency automated trading.

The first, which is used by German regulators, would be to define a specific threshold of message rates. Its simplicity makes it attractive. But as soon as the regulators draw an arbitrary line in the sand and say “no faster message rate than that,” the limit would quickly wind up out of date as network message rates continue to increase as the technology evolves.

The second approach would involve trading venues identifying firms that have a median order lifetimes lower than the median order lifetime of all of the other orders on the trading venue. Once a venue identifies a firm using such a technique, that venue member would be treated as using the same technique on all other EU trading venues, according to regulators.

Again, the regulators do not get down into the weeds and say how much lower a firm’s median order lifetime must be compared to the rest of the orders trading on the the same venue.

Yet, if the regulators expect firms to meet the January 2017 deadline for MiFID 2 compliance, someone will need to bit the bullet and actually write down a number. If not, the EU is going to need to grant the implementation delay the industry has requested since no one will know the messaging limit and how granularly they will need to synchronize their server clocks.

Pension funds, sovereign wealth funds, endowments and other institutional asset owners are sitting on vast troves of data -- but extracting value from that data is more challenging than ever.

#AssetOwners #DataQuality

Technology costs in asset management have grown disproportionately, but McKinsey research finds the increased spending hasn’t consistently translated into higher productivity.
#AI #Fiance

We're in the FINAL WEEK for the European Women in Finance Awards nominations – don't miss your chance to spotlight the incredible women driving change in finance!
#WomenInFinance #FinanceAwards #FinanceCommunity #EuropeanFinance @WomeninFinanceM

ICYMI: @marketsmedia sat down with EDXM CEO Tony Acuña-Rohter to discuss the launch of EDXM International’s perpetual futures platform in Singapore and what it means for institutional crypto trading.
Read the full interview: https://bit.ly/45xRUWh

Load More

Related articles

  1. Execution algorithms are a growing share of global spot FX, particularly among buy-side firms.

  2. Demand for state-of-the-art execution algorithms in FX is growing rapidly.

  3. FCMs Promote Algorithmic Trading

    There was a 75% year-over-year increase in daily principal traded.

  4. FCMs Promote Algorithmic Trading

    SmartDark features prioritized routing to venues with larger executions sizes and better price stability.

  5. The two firms have published research on using quantum computers in pricing algorithms.

We're Enhancing Your Experience with Smart Technology

We've updated our Terms & Conditions and Privacy Policy to introduce AI tools that will personalize your content, improve our market analysis, and deliver more relevant insights.These changes take effect on Aug 25, 2025.
Your data remains protected—we're simply using smart technology to serve you better. [Review Full Terms] | [Review Privacy Policy] By continuing to use our services after Aug 25, 2025, you agree to these updates.

Close the CTA