Mike O’Brien, director of global trading at Eaton Vance
What are your biggest priorities and challenges from a fixed income trading perspective?
Mike O’Brien: One of the biggest areas of focus for Eaton Vance’s fixed income trading desk is use of technology to achieve greater levels of automation. I want my traders to concentrate on high value-add tasks, using the technology that’s become available in recent years to automate more routine tasks, in order to spend more time on larger or complex trades, as well as those in asset classes that are more difficult to trade. I don’t want them spending time on small-sized deals in highly liquid markets such as US Treasuries, as these can be automated.
What developments have enabled greater automation of trading in the fixed income universe?
MB: The gradual electronification of fixed income markets has been the biggest change in recent years, creating opportunities for automation for individual market participants. Of course, fixed income is not one market and runs the gamut from the most liquid, such as US Treasuries, to products that are still some distance from electronfication.
In parallel, we’re seeing the application in fixed income markets of tools originally developed for use in equities. This is a relatively recent development that we’re looking to take advantage of.
As fixed income traders spend more time on value-added trades, are you asking existing staff to develop new skills or are you looking to hire people with different skills to augment the in-house team?
MB: The role of a buy-side fixed income trader has changed pretty dramatically since the financial crisis, albeit the change has been more of a slow evolution over time than a big bang. Prior to the crisis, when liquidity was abundant, buy-side traders were primarily execution focused, achieving best price by calling round a handful of counterparties. That type of trader is no longer relevant. Today, buy-side traders need to develop a deep understanding of portfolio construction, market structure and bond valuations in order to contribute alpha to a trade. So, yes, we are looking for our traders to have different skill sets, but that is a recent phenomenon brought about by electronification. Our growing use of the tools that enable automation are the result of that change, not the cause.
How are you using transaction cost analysis (TCA) and related tools to achieve best execution in the fixed income markets and how do these tools and capabilities need to further develop?
MB: TCA is becoming a more important topic, both through the mandate of MiFID II and the development of analytics in a wide range of fixed income asset classes. Electronification of these markets will help their further development. As traders, we know which strategies we believe will achieve best execution, but we need to be able to quantitatively measure and verify outcomes that either confirm our expectations or, over time, help us to achieve better outcomes.
Whether we’re talking about the most liquid or most difficult markets, TCA in fixed income has already come a long way in the past few years – and still has a long way to go. That doesn’t mean we should not be using it yet. As long as you understand the limitations and assumptions inherent in the process, you can understand where there may be some margin for error in the results. That said, I’m still looking for industry developments to improve the process. It has evolved very quickly and I expect this to continue as more asset managers become focused on it.
What are the main ways in which you have evolved your relationships with sell-side firms and others to secure access to liquidity across the fixed income markets?
MB: In general, the buy-side needs to evolve and think about their trading strategies in different ways. With some of the new entrants into the fixed income markets, there are new ways to trade beyond RFQ. The buy-side needs to get more comfortable with those, understand how they work and how they can be used across the portfolio so that fixed income traders have multiple tools at their disposal to access liquidity. Buy-side traders need to understand how the market operates and how to value a bond sufficiently well to be a price maker than just a price taker. If someone’s looking for a bid on a bond, how should the buy-side trader respond? That’s not in the traditional buy-side skill set, but it increasingly needs to be. With so many protocols available across fixed income markets, the buy-side approach needs to change. As the skill set on the trading desk evolves, traders will become more comfortable with the newer protocols and have better access to a wide range of liquidity sources beyond the RFQ model.
The US Treasuries market has recently seen new platforms, protocols and liquidity providers. Is this an example of the kind of market in which fixed income traders must expect to trade differently?
MB: The most liquid markets are the ripest for new trading protocols to develop. On the other hand, they’re also the ones where the buy-side is most likely to rely on old habits. High levels of liquidity allow new trading protocols to gain traction quickly – but also permit the buy-side to remain dependent on RFQ. The RFQ works in the US Treasuries market: you are going to get responses back. In other areas of fixed income, it might not work so well, forcing firms to find other ways to trade. We’re looking at the tools and keeping an open mind on new ways of doing things. But that doesn’t mean the end of RFQ. We prefer to take more of blended approach.
A central part of my role is to give my traders as many tools as possible with which to execute trades. But they are in the best position to judge which tool is appropriate on a trade-by-trade basis. I need to be in the market, understanding what the various approaches to trading are and which ones could be relevant to us, in order for them to make decisions on individual trades.
This certainly extends to the new ways in which electronic liquidity suppliers are providing liquidity through some of the established electronic trading platforms in the more liquid fixed income markets. Any new source of liquidity is of interest. I don’t want us to get rid of the old protocols as there may be cases where they are still appropriate, and that includes sourcing new types of liquidity from existing platforms and protocols.
What kind of support are you looking for from technology and data providers to help your efforts to improve fixed income trading efficiency at Eaton Vance?
MB: I’m open to new and innovative ways of approaching a problem, but we don’t need vendors to reinvent the wheel with new technology. I’m always interested in removing sources of friction from the trading process as those represent transaction costs that I’m trying to minimise.
Technology and data are very useful tools to improve the efficiency of fixed income desks. Overall, technology is going to solve a lot of problems for fixed income desks. We’re just at the start of the journey with artificial intelligence but I think it is likely to add value in future.This article was originally published in the FILS in Philly Today magazine, produced by The TRADE, which was distributed to attendees of this year’s Fixed Income Leaders Summit USA conference.