A last look at Last Look? Barclays and FX market structure

Celent will help qualify your requirements and introduce you to the vendor
Spotted a missing vendor? Use this form to alert a vendor to the Celent service
Create a vendor selection project & run comparison reports
Register to access this feature
Click to express your interest in this report
Indication of coverage against your requirements
Vendor requires PRO subscription to activate this feature
Requires research subscription, contact Celent for more info
23 November 2015
Brad Bailey
The FX market is trying to digest the latest large FX fine and the impact on market structure. According to last week’s press release from the New York Department of Financial Services (NYDFS), Barclays was fined US$150 million for “automated, electronic foreign exchange trading misconduct.” The order goes on to detail this additional fine is a result of using its Last Look system to automatically reject what Barclays determined would be an unprofitable trade within the time window created by the system’s defined Last Look window. This brings foreign exchange fines by NYDFS against Barclays to $635 million. After spending time this weekend reading the order, it is clear that the issue at hand is not a Last Look issue, but rather improper customer notification and trading practice. In this case, Barclays was abusing both the intent and scope of Last Look. While Last Look gets a lot of discussion recently, it is a byproduct of market making in principal markets, such as FX and fixed income from a pre-electronic age, and the translation of those markets into electronic trading. As FX became more electronic, the European banks were early innovators in mapping principal trading functionality into electronic trading. In the case of FX, it became necessary to create a means to quote to thousands of customers through various channels (i.e. single dealer portals, multi dealer platforms, aggregated feed channels) at acceptable bid/ask spreads. Given that there are different types of clients, it is necessary to be able to quote different types of clients with very different risk profiles, technology ability, and holding time frames in different ways. A liquidity provider looks at an HFT counterparty much differently than a large asset manager putting on a hedging FX trade. Last Look is not inherently a bad practice, but it is a practice that when performed needs to be clearly mapped out to users of a platform. Platforms that incorporate Last Look functionality will be ensuring that guidelines are clear, functionality is sound, and procedures are well documented. In an OTC principal market, liquidity provision is not free. In the current FX market structure, Last Look is a necessary tool for many liquidity providers. Over time Last Look will become a less important component of FX trading. It is not clear that can be regulated away without disrupting the ability of market makers to provide liquidity in the current market structure, and given the global nature of FX. At the same time, concerns around Last Look are changing the calculus of liquidity between disclosed liquidity and anonymous liquidity. In my latest report FX Trading 2.0: Technology and Platforms, I explore the evolution of FX and how the market will incorporate all the forces at play. In many ways, after building on incremental change over decades, the FX landscape has shifted abruptly recently. The venue landscape has brought together once disparate centers of liquidity within the same firms. From the perspective of identifying the ideal venue to interact with, the landscape has become more challenging. Many of the FX platforms are separate or partially separate pools of liquidity within the same firm.

Insight details

Content Type
Blogs
Location
Asia-Pacific, EMEA, LATAM, North America