By Mike O’Hara, Special Correspondent.
As we approach the third anniversary of its implementation date, the impact of the EU’s MiFID II continues to be felt around the world as global players adopt key measures as a form of best practice. Part of its wide remit to recast the mould for European institutional financial markets was to set out a series of measures designed to improve OTC market transparency by bringing non-exchange traded securities onto venues.Since then, OTC markets – in the fixed-income marketplace in particular – have witnessed a shift toward ‘electronification’, as the drive for greater transparency saw the bond markets adopt some of the characteristics of the electronic equities markets. The regulation’s authors outlined a number of key measures in their attempt to promote more transparent markets, among them the requirement for firms to capture and accurately time-stamp all orders, quotes and trades (whether electronically matched or otherwise) to meet record keeping obligations, and the requirement to prove best execution.
Whereas all of this is now pretty much the norm in equities markets, there are significant challenges that make widespread electronification of fixed income trading an entirely different proposition. The global bond market encompasses vastly more instruments, many of which might trade only rarely, compared with the equities market. And unlike equities, a large proportion of fixed income trades, particularly in more complex or illiquid instruments, are negotiated bilaterally between counterparties.
Regardless, the shift towards greater electronification in fixed income continues, and regulation is not the only driver. Firms are also seeking greater efficiency, reduction in manual processes, improved workflow, more accessible and meaningful data, and – for the buy-side in particular – better access to liquidity, all of which electronification can potentially offer.
So in fixed income, how is the MiFID II premise, and the move towards electronification going? The answer, as so often is the case, is it depends.
A great deal of global fixed income trading already takes place electronically, with Bloomberg the dominant mechanism for electronic communication. In highly liquid, standardised instruments such as on-the-run US Treasuries, electronic platforms such as CME Group’s Brokertec and the DealerWeb platform operated by Refinitiv’s TradeWeb subsidiary dominate the inter-dealer market, whereas in the dealer-to-client (D2C) space, a multitude of platforms are available to facilitate electronic execution, through either an equities-style central limit order book (CLOB) or an automated request-for-quote (RFQ) model.
Much trading however, particularly in the corporate bond market, still tends to follow a more analogue workflow (phones, chat, use of e-mail and spreadsheets, etc). And although algorithmic and electronic trading is becoming more widely adopted, it is generally used for smaller trades. Large trades, for the most part, are still negotiated over the phone or via chat.
Another notable trend is the increasing adoption of ‘all-to-all’ trading platforms. OpenTrading from MarketAxess for example, saw its usage grow by 53% year-on-year in first three months of 2020. Other all-to-all platforms like Tradeweb, Liquidnet, and Trumid have also experienced increasing volumes in 2020. But all-to-all still only accounts for a fraction of overall fixed income trading volumes.
Impact of COVID-19
This year has seen a rapid increase in electronic trading in fixed income. A recent report from JPMorgan highlighted the fact that by June of this year, 77% of the D2C market in US Treasuries was conducted electronically, up from around 50% for the two years previously.
Much of this can be attributed directly to the unprecedented bond market volatility in March and April, which spiked massively during the onset of the COVID-19 pandemic. As dealers were unwilling or unable to take on potentially risky trades due to capital constraints, liquidity was reduced and spreads widened as a result. Traders therefore had to resort to electronic channels to execute their trades. Also, many of them were working from home, so they needed to be able to transact based on firm prices quoted on a screen.
However, a separate study from ICMA pointed out that in European corporate bonds, electronic trading actually broke down during the March/April peak volatility, which meant traders had to revert back to high-touch methods and pick up the phone to their dealers, a development that goes some way in illustrating the point that electronic trading is not necessarily the answer when markets are volatile, particularly if those markets are already illiquid.
Electronification is Not Just About Trading
Notwithstanding the nuances within individual markets and market situations, the fixed income markets are moving steadily towards greater electronification. But this trend is not limited to electronic execution and order matching. Other areas are also becoming more automated, including liquidity sourcing and aggregation, pricing, auto-quoting and automatically responding to RFQs.
In pre-trade, algos are increasingly being used in the decision-making process. Earlier this month, for example, electronic market-maker Virtu launched a proprietary pre-trade liquidity model to assist with execution decisions, liquidity management, and documenting best execution. More broadly, firms are also starting to adopt cloud-based AI platforms such as Overbond for pricing and quantitative analytics.
Another area experiencing significant growth in electronification is portfolio trading (where large baskets of corporate bonds are traded in a single transaction). Much of this is becoming more automated, and rather than traders relying on spreadsheets and manual input, they are following a more electronic workflow. TradeWeb’s electronic portfolio trading volumes for example, grew from $2 billion in January 2019 to $100 billion in June 2020.
One notable effect of the continued fragmentation from the plethora of FI trading venues and platforms now available, is an explosion of data. This has led to firms making increasing use of data aggregation tools, not only to help navigate the complexity and improve the search function when finding the right counterparty for a trade, but also to improve analytics, build credit curves and achieve more accurate bond pricing.
However, data from such a fragmented market can often be poorly structured and of variable quality, particularly for the more opaque, illiquid instruments. And there are still no standardised protocols for example to identify dealers, perform TCA (transaction cost analysis) or prove best execution.
The challenge therefore remains, how to source, process and aggregate data from multiple sources – electronic or otherwise – in order to achieve faster pricing and liquidity measurement?
The Evolution of OMS/EMS and Smart Order Routing
One area of technology that is evolving rapidly is the order management and execution management systems (OMS/EMS) segment. Whereas traditionally these have been two distinct systems – with the OMS used for portfolio construction, order management, compliance and operational controls, and the EMS taking care of data aggregation, decision support and analytics, automation and order routing/trade execution – the two are increasingly coming together and becoming more functionally rich.
Leading fixed-income trading platforms and OMS/EMS vendors such as Bloomberg, Axe Trading, Charles River, FlexTrade and SimCorp/TradingScreen are increasingly offering seamless integration of OMS and EMS functionality, enabling more automation and aggregation, providing firms with the ability to initiate RFQs directly, analyse responses, and route orders automatically. It is worth noting, however, that two of the dominant players in this space, ION and Broadway Technology, recently reached agreement with the UK Competition and Markets Authority (CMA) to spin off Broadway’s fixed income operations, as the proposed terms of the acquisition would have given ION an unacceptable 25% share of the bond trading platform market.
The Holy Grail for the buy side is to be able to analyse historical data (to see which dealers have had axes in specific bonds in the past, for example), process streaming prices from counterparties, and assign automated execution pathways for RFQs or orders depending on the liquidity characteristics of the security. In effect, enhanced smart order routing (SOR).
In order to approach the levels of SOR that are now seen in the equities markets, firms will need to be able to evaluate real-time market conditions around price and liquidity, and route executions accordingly to the appropriate venues or dealers. This is one area where firms are starting to look at how AI and Machine Learning can help, by learning from historical activity and analysing and adapting to current conditions in real time. The challenge of course is that it is extremely difficult to get accurate, real-time liquidity figures from across the market. Electronic venues are obviously a lot easier for a SOR to process than the bulk of the OTC market.
What Next for the Electronification of Fixed Income?
While it is clear that the market is moving forward in many ways, a number of challenges remain. As stated at the outset, fixed income and equities are entirely different markets, so it’s unlikely that fixed income will follow a similar electronification trajectory to equities, at least in the short-to-medium term.
It is interesting to note that nine Wall Street banks have recently come together to launch a new platform, DirectBooks, to optimise the communications process and workflow for the primary issuance of corporate securities, which has always been a manual, labour-intensive process. With such a powerful consortium of Tier 1 banks behind it, this could potentially be a gamechanger for the corporate bond market.
It will be interesting to see where mergers and acquisitions occur in the fixed income technology space next year, and how they might impact the market. London Stock Exchange Group, for example, is now in discussions with Euronext about the sale of its Borsa Italiana Group, which operates the MTS bond trading platform. Once again, competition regulators’ concerns about market dominance drove this transaction in the wake of LSEG’s agreement to acquire Refinitiv, owner of TradeWeb.
Another trend worth noting, which might act as something of a catalyst towards greater electronification, is the growth of non-bank electronic liquidity providers (ELPs) in the fixed income space. Firms such as Citadel Securities, Flow Traders, HRT, Jane Street, Jump Trading and Virtu are now highly active in on-the-run (and some off-the-run) US Treasuries, not just in D2C but also in the interdealer market, providing streaming quotes and automated response to RFQs.
Non-bank ELPs such as these – which are experts in managing and internalising risk on their own books – are not restricted by the capital constraints of banks, which means that they can continue to trade in volatile markets. Most of them performed well during March/April 2020, for example.
As the market continues to evolve, it is likely that these non-bank ELPs will play a larger part, across a wider collection of instruments, and stimulate the electronification of the fixed-income markets even further.