GreySpark Partners recently published a report titled Trends in Repo Trading 2023. In this Q&A, Russell Dinnage, managing consultant, at GreySpark Partners shares the key findings of this report, including how corporate and investment banks are utilising technology to further automation in a move away from manual to “e-manual” trading and post-trade processing to support repo business growth and efficiency amid wider market trends and regulatory drivers.
Q. The report mentions that there are varying degrees of automation that Corporate Investment Banks (CIBs) rely on today in their repo trading and post-trade operations. Can you briefly share with our audience the differences between groups in terms of automation and what is driving the current level of automation (or lack thereof)?
A. The two groupings of CIBs highlighted in the report are (1) International Institutions (Tier I and Tier II-plus CIBs) and (2) Regional Institutions (Tier II to Tier III CIBs).
To summarise into layman’s terms the differentiating factors between each grouping that are highlighted in the report, the differences between them in terms of the extent to which they have automated their business process and workflow associated with repo trading are threefold.
The first differentiating factor is the volume of Dealer to Client (D2C) and Dealer to Dealer (D2D) repo trading the CIBs within each grouping are transacting quarterly / annually; all CIBs trade repo daily as an operational necessity, but the extent to which any CIB in either grouping is managing complex, long-dated positions in the D2D markets or is actively hedging client risk or the bank’s own, operational risks in the D2C brokerage environments can be a driver of trade automation in the sense that more trades (D2C or D2D) to handle daily means more credit / risk management to perform for the middle-office, means more post-trade work to process for the back-office, and so on. Automation naturally becomes a useful capability to draw upon to ease not only trader workload but also to streamline the overall costs associated with running the repo business or desk(s) and, ideally, to serve more clients globally in a more efficient manner or to service bank-internal funding / securities lending needs in a holistic fashion.
The second differentiating factor is the role of the CIB in the Triparty repo market; participating in the Triparty repo arena can be likened to a small hedge fund getting an ISDA agreement with a large bank to trade equity options using the institution’s single-dealer platform. The International (Tier I and Tier II-plus CIBs) are either the banks that ‘operate’ the Triparty repo market, or they are the ones most actively trading within it daily, and the trading therein is almost entirely automated / electronic in nature.
The third differentiating factor is down to the technology — either in-house built or vendor-provided or a mish mash of both — that a CIB front-office uses to manage repo trading and the extent to which straight-through processing is embedded into every stage or architectural area of that IT real estate. Whether it be for pre-trade purposes in terms of sucking in large amounts of different types of market / trade data for price formulation purposes, at-trade in terms of pricing different clients / counterparties in a tiered fashion across a multitude of different types of instruments or post trade in terms of the ability of front-office systems to seamlessly feed down positions and executions into middle and back-office databases, the ability to straight through process (STP) this data from front-to-back versus handling it manually — or semi-manually — is a frequently observed difference between international CIBs and regional institutions.
“Reliance on voice / chat is not necessarily representative of a decision by a CIB to either embrace or ignore automation’s benefits in repo trading — often, it comes down to the characteristics of the client base, what they want or how they are used to interacting with an institution’s repo trading franchise as well as the spend that a bank has invested in sec lending / sec finance / SBL operational technology.”
As to what is driving this automation (or a lack thereof) — The common factor that spans across the three competitive differentiators between international vs. regional CIBs is the extent to which those CIBs continue to rely primarily on voice / chat as the main mode of repo trade negotiation and execution on either a D2C or D2D basis. Reliance on voice / chat is not necessarily representative of a decision by a CIB to either embrace or ignore automation’s benefits in repo trading — often, it comes down to the characteristics of the client base, what they want or how they are used to interacting with an institution’s repo trading franchise as well as the spend that a bank has invested in securities lending / securities finance / securities borrowing and lending (SBL) operational technology.
Q. You mention regulations have recently driven automation in this space? How are these regulations continuing to drive automation among CIBs in this space?
A. The Securities Financing Transactions Regulation (SFTR) often gets looked at as simply a ‘post-trade reporting’ rule set. But, really, SFTR’s purpose is to create transparency in the corners of the capital market globally that were not heretofore covered by benchmark regs like Dodd-Frank or MiFID II, both of which were designed by regional regulators to enact the principles of the G20 nations’ 2010 decision-making within their respective jurisdictions (US and EU, respectively). SFTR, in the EU, is one of the final pieces of that puzzle, along with hundreds of other pieces of EU-wide and member state-specific rulemaking. Crucially, SFTR creates a hygiene standard for post-trade reporting and transparency that — in the case of the repo market — any buyside firm or sellside institution dealing with a client / counterparty domiciled in the EU must abide by; so, they take those factors back to their home market or region outside of the EU and apply them not only to the European clients, but to other client groupings as well (possibly) because it represents a degree of post-trade ‘best practice.’ Of course, the on-going costs associated with SFTR compliance for a bank and its repo trading business, specifically, can be immense depending on the size of the institution’s footprint in the D2C / D2D market driven by either client demand or internal, operational securities lending / securities finance need.
SFTR is driving automation continually among CIBs because it requires them to consider the types of collateral that they are taking on through the course of repo trading activity as well as the volume of complex, long-dated trades that they are willing to transact and how easy or hard it will be to, eventually, centrally clear those transactions and at what cost throughout the lifecycle. The Basel Committee on Banking Supervision (BCBS) Accords mean CIBs must be smarter about the so-called cost of capital, and so there is an emphasis placed on using cash for collateral management purposes above all else as well as on novation in the clearing process. As such, the BCBS Accords are not so much driving automation as they are pushing up compliance costs across the board, especially for securities lending businesses, and so CIBs are incentivised to automate other activities such as trading as a means of cost cutting elsewhere. Other market factors driving repo trading automation include the growing importance of D2C markets such as Eurex Repo or GLMX, which increasingly play a role in helping CIBs and their buyside clients to solve technology problems by building out STP-centric solutions or by providing post-trade services that rely heavily on STP.
Q. You mention 4 challenges firms face as they move from one mode (manual) to the second mode (e-manual) stat of repo trading and processing. Can you briefly explain these challenges and how technology can address such challenges?
A. In terms of how technology can address each challenge: Challenges 1 and 2 — Technology will not necessarily stop CIBs and their clients / counterparties creating unstandardised or complex repo instruments. But, if D2C / D2D brokerage venues can continue to expand the range of automated protocols that repo market participants can use to negotiate or execute on bespoke contracts, then maybe there could come a time when — partially because of regulation — a period of ‘peak repo’ could be reached in which every possible method for transacting any instrument has been recorded by the bulk of leading execution venues. This is highly unlikely, however. So, I think it might more likely be something like AI coming along as an application that CIBs can use to model the characteristics of the complex, long-dated repo that they wish to ‘make’ and then assess the viability of that instrument / product from an automated execution perspective.
Challenges 3 and 4 — There’s not much that technology can do here other than continue to grow in sophistication in terms of automated execution and post-trade STP, just to make everyone’s overtly manual at-trade methods less challenging in the post-trade pull through.
Q. What are the top three benefits a CIB would achieve in moving to an e-manual trading environment and harnessing the necessary technology to do so?
A. Lower cost of execution per trade using a linked-up pricing engine and credit / risk control function, assurance of regulatory compliance through post-trade STP, and the ability to take on more clients across more jurisdictions; to grow the business, effectively.
Q. Were there any findings in this survey/report that surprised you?
A. Probably the extent to which Tier I CIB traders whose institution(s) operate an astute, data-centric repo trading tech estate do not think so much about which type of brokerage venue that they are executing on (D2C vs. D2D); they just see a price and a client / counterparty / liquidity in the market and they hit it regardless of where it might be going. Also, the increasing popularity of GLMX as a hub for repo trading in the US; if GLMX eventually opens a venue in Europe, then that will be an interesting development.
Q. What do you think will be the main trend impacting repo trading in 2024 and why?
A. I think that there are several Tier II institutions that are looking to shed their existing in-house built repo trading technology estate in favour of a vendor-provided suite of services. Repo market connectivity in terms of gateways and the costs associated with managing them will still be an expensive, outsourced concern. But vendor selection is a popular topic now, and there will be a lot of IT operational complexity that will need to be resolved before any selected vendor solution can be fully integrated.
