In May 2016, the global Assets Under Management (AUM) in ETFs were broadly estimated at around $3 tn USD. Just over one year later, research firm ETFGI estimated the AUM at $4.8 tn USD. With a general trend of ~20% AUM growth per year, and every expectation this trend will continue, the rise of ETFs continues to astound. At Itiviti, we have observed the direct effects of this phenomenon in the arenas of proprietary trading and liquidity provision, as firms compete for new opportunities and the technological edge. However, with these new opportunities come new challenges, and following some targeted research and detailed market analysis, one challenge in particular resonates as critical to the success of today’s ETF trading operation – static data and composition management. But what problems exist, and how can they be overcome?
Despite being associated with reference data such as identifiers and listings, an ETF’s composition is anything but static. A multitude of factors can shape and influence the composition, from fund rebalancing and creation/redemption activities through to dividends and other corporate actions, meaning a fund’s holdings can change significantly day-to-day, or even intra-day. In a recent Itiviti survey, 92% of respondents cited static data and composition management as vital to their business, and it’s easy to see why – without timely and accurate data, all other trading system functionalities are undermined, from pricing to quoting to risk management. When coupled with the increasing numbers and complexity of these products, and the heightened regulatory environment as typified by MiFID II, the cost of failure is immense.
For this particular challenge, the solution is twofold – a timely and accurate source of data, and crucially, a specialized data model coupled with automated processing and associated workflows. On the first point, funds are compelled to release their full holdings on a daily basis, so this information is readily available. But considering the number of issuers involved, differences in publication time, medium and format, and the number of downstream integration points required, it quickly becomes a laborious process with multiple points of failure. Another option is to homogenize and persist the raw data into a purpose-built, in-house data store. This provides a single, auditable ‘golden source’ for multiple downstream consumers but still comes with a great deal of overhead, and does this cost provide any edge? Finally, there are the vendor options including IHS Markit and ULTUMUS, offering standardized, timely and comprehensive data as a service, but which come with costs of their own.
The source of composition data is one side of the story, but just as important are the processes, workflows and data models required to utilize it. Full automation is a given, but there are other aspects that are quickly becoming market expectation. For example, can your trading system detect suspicious composition changes across thousands of ETFs, and provide explicit and actionable feedback in real-time? Can your trading system rule out ‘false positives’ due to corporate actions like splits or reverse splits, and conversely, detect that a composition does not reflect recent corporate actions? Can your data model handle all required baskets for a product, intra-day composition changes and fragmentation across countries and currencies?
The industry is rising to meet this challenge – for example, ULTUMUS have rolled out their ETF and index data solution, and at Itiviti, we have recently integrated our Tbricks by Itiviti trading system with IHS Markit’s SOLA platform. Just like the data itself however, the industry is never static, with more and more demands being asked of vendors. New regulation, new product innovation and new trading strategies all drive data and functional requirements, with some firms now electing to employ multiple data sources overlaid by a comparison framework, for additional levels of resilience. These developments are a testament to the criticality of data quality in the realm of ETF trading, and the challenges involved in utilizing it – the latter of which, Itiviti is committed to solving.
Execution In this video, Lars Wiberg, VP Strategic Research Trading and Trade Execution, traces the history of the Order Management System (OMS) and its evolving versatility, with a focus on technology and key functionality. This is the first in a series of 4 videos that will explore the past, present and future of Order Management Systems […] February 15, 2018
Execution This is the first of a multi-part series that explores the past, present and future of Order Management Systems (OMS) and the closely related Execution Management Systems (EMS). We cover these trading tools mainly from a sell-side perspective. An Order Management System (OMS) is a software application for facilitating and managing the order and trade […] January 30, 2018
Trading In the financial industry, speed is everything. Taking advantage of trends and market opportunities, but also responding to new regulations, means a constant focus on developing systems with lower latency, snappier GUIs and faster event handling. But the development process itself also needs to be fast. Speed means being able to learn quicker – exploring […] January 23, 2018
Trading We take the concept of time for granted, but in truth it has a complicated history. It was actually only relatively recently that the idea of standard time became widespread and allowed us to synchronize time within different geographical regions. In the financial world, where thousands of events can take place in a single millisecond, […] January 16, 2018