Sell-side firms are on the cusp of a revolution in liquidity provision for Fixed Income markets. Mike O’Hara of The Realization Group and writer Dan Barnes spoke with a panel of industry experts including: Sonali Theisen, Head of Fixed Income Market Structure and E-Trading at Bank of America Merrill Lynch; Ahmed Heikal, Product Manager for Execution
Products and Services at Vela; Simon Linwood, Executive Director, UBS Bond Port EMEA at UBS; and BBob Mudhar, Partner at Citihub Consulting, to determine how these changes will impact the use of trading technology.
Sell-side fi rms are under pressure to refresh the technology that underpins their Fixed Income market-making operations. If they do not, they risk getting pushed out of client relationships by non-traditional liquidity providers. The Fixed Income markets are evolving step by step towards more automated models of pricing and execution. The driver is necessity. The traditional liquidity providers – banks and broker-dealers – have been less able to warehouse the risk of holding bonds post-financial crisis, as the cost of capital to support the risk they carry, and the liquidity ratios they have to maintain, have increased. Asset managers need to reduce their costs of trading wherever possible, as part of their overall reduction of costs. They also need to capture data more effectively than ever before to meet their compliance burden.
Standing in the way of progress is the fragmented liquidity picture for most Fixed Income instruments. First of all, there are multiple instruments per issuer. Secondly, there is no centralized venue for issuing corporate or municipal bonds and liquidity is periodic rather than continual, typically peaking around the points of issuance and redemption.
Yet pressure to change is increasing. Non-bank electronic liquidity providers (ELPs) are stepping in where possible. They are providing streaming pricing in either liquid assets or for indexed bonds, which can be priced based on the creation/redemption of exchange-traded funds (ETFs). Typically, they make markets in smaller sizes rather than the blocks banks can trade, however as comfort levels grow trade sizes are likely to increase.
Banks are also stepping up their efforts in this area. By offering streamed liquidity, banks are helping their buy-side clients to begin the process of automation on their trading desks. There is now potential for sell-side firms to gain a competitive edge through the smart application of technology to their trading workflows, in order to be at the crest of this wave of change.
The complexities of workflow automation Asset managers want more electronic trading, and where possible greater automation.
“These were conversations that didn’t happen in Fixed Income five or ten years ago,” says Sonali Theisen, Head of Fixed Income Market Structure and E-Trading at Bank of America Merrill Lynch.
A key driver for the adoption of more electronic execution is the consistency it offers, which helps to build certainty around the trading process. Under the revised Markets in Financial Instruments Directive (MiFID II), asset managers are mandated to provide best execution for their trading across all asset classes.
The best execution policy, demanded by MiFID II, has to be translated into the execution process. One of the advantages to executing electronically is the consistency it offers. Trading on an electronic venue also allows automatic data capture, which helps to support post-trade execution quality analysis. It also creates a consistency in dealer activity on the sell side, which allows execution quality to be more consistently evaluated.
Ahmed Heikal, Product Manager for Execution Products and Services at Vela says, “People are still grappling with how to build the best execution mechanism for Fixed Income. The understanding is that, in order for it to happen as a prerequisite, electronic trading is required. So that is pushing the market in this direction. There are challenges of course, being that there is no centralized exchange, there is no market data tape outside of the US, and therefore the market is fragmented.”
Delivering a more automated workflow in bond markets is made more difficult by the complexity of the instruments themselves, and the structure of the markets in which they are traded.
Fixed Income instruments are all yield bearing and time limited, but the type and number of issuers in a market, the liquidity profile and the costs of market making are enormously varied. Equally, the investment mandates and risk appetite of portfolio managers within asset managers constrain the markets that buy-side firms can trade.
Liquidity in bonds is typically episodic, and most focused when bonds are first issued or ‘on-the-run’. That affects pricing. If a bond has not traded for weeks, the last price at which it was traded may not be relevant today. Therefore, pricing reliability typically decreases from the point of issuance.
To overcome this uncertainty, banks provide indicative information, or axes, to show which way they are trading and in what size for a given bond, without committing to a price. That has been supported through the standardization and automation of axes by third-party platforms such as Neptune.
Trading protocols and data transparency
The most liquid bond market is US Treasuries. Dealer-to-dealer trading for on-the-run Treasuries is conducted via a central limit order book (CLOB), which is common in equity markets and is appropriate for continuous liquidity and pricing.
Dealers have been auto-quoting Treasury prices to asset managers for years. In 2017, Mihir Worah, managing director at bond investment giant PIMCO, told a Federal Reserve conference, “We are moving towards full automation whether we like it or not; we are 80% there today and of that 75% is auto-quoted.”
However, that market is an outlier. Also in 2017, Greenwich Associates reported that electronic trading of government bonds and investment grade corporate bonds was around 52%. Yet electronic trading is not the same in these markets as it is in equities.
The preferred electronic execution protocol in dealer-to-client (D2C) markets for corporate bonds is request-for-quote (RFQ), which is the same protocol used in voice trading, but managed via electronic messages and interfaces. This allows liquidity to be sourced episodically. But to create a more automated bond market, asset managers need to receive continuous, firm prices.
“Clients now want to be informed by data, on a scale that’s not yet been seen in Fixed Income markets,” says Theisen. “They want to be informed in their decision-making process before they even pick up the phone and decide who they engage with. They want more data at their fingertips.”
If data is more freely available and accessible on a more continuous basis, buy-side fi rms can support the development of algorithmic trading on their desks as they have done in equities. Yet typically it is not; if banks are to deliver more data they need to engender a change in market structure, service model or trading workflows. And, as Theisen points out, data is not the only thing missing. “Even if the data is there, you will not have buyers and sellers of 30 thousand CUSIPs every day,” she says.
Next steps to automation
As data is key to traditional brokers in supporting their clients on the path to automation, banks need to consider how market-making operations can be changed to deliver this support.
Given the existing capital constraints that limit their risk appetite, electronic liquidity providers (ELPs) have thrived in certain parts of the Fixed Income markets, through a combination of innovative business models and a smart application
of technology.
The fi rst group of these are high-frequency trading (HFT) fi rms, who process market data at very high speed in order to spot low-margin automated arbitrage opportunities. These excel in the US Treasury market where a CLOB provides
microsecond updates on trading, and where there is a broad spectrum of market participants trading a few active CUSIPs. However, in corporate bond markets where trading is infrequent, they cannot function.
The second group of ELPs are ETF issuers, such as Jane Street or Flow Traders, who use a combination of primary and secondary market activity to deliver prices to the buy side. By engaging in the creation or redemption of bond ETFs they can quote a firm price at which to buy or sell those bonds included within the index that the ETF tracks.
Buy-side firms who have an associated ETF business also provide a similar pricing point.
