How has fixed income data use evolved in recent years?
The biggest evolutionary change in fixed income has been the reliance on data as part of the investment and trading decision making process. It’s not just a case of data being used on the trading desk anymore and I think Covid accelerated that. The connectivity that Neptune has achieved, driven by the buy-side demand, to be plugged into OMS, EMS, PMS, liquidity optimisers, signal generating systems, etc. is something that we couldn’t have foreseen seven years ago.
It means, in our area of expertise, axes are used from the smallest of trades, as part of the auto-ex decision making, to the very largest, where knowing who is a seller or buyer and limiting information leakage is vital for both sides of the market. Axes are no longer the preserve of the trading desk at asset managers. The use of this data by the portfolio management function means they can better represent their investment process by taking liquidity into account.
What still needs to change?
Dealers still need to get better at distributing their data and managing where it goes to. One of the restrictions on getting more insightful data from the sell-side to the buy-side is a concern around how to get the information there and what happens after that. Neptune has made a big impact on that in axes, however there are other areas, like runs, where we think a similar model may work. It means less reliance on the human aspect, but not total exclusion by any stretch of the imagination.
On a micro level, the dealers need to get better at making sure they’re sending the most accurate and best quality data. A lot of dealers are challenged by the amount of data they have to distribute. One of the key things is how do they get that out there in a systematic way that also feels safe? You get a lot of line traders who still send things out via email or spreadsheets because they feel that’s the safest way because they’re direct but there’s no controls around it. With regards to runs for example, a lot of dealers would admit that they’re not sure exactly where that information goes to and what happens on the back of it.
The way data is used in Europe and the UK is more advanced than in the US. At Neptune, some of our most advanced clients are in the states but in some cases the data hasn’t become the thing that it has in EMEA. I’m sure it’ll catch up in the same way that e-trading has. There is not a correlation to AUM. There are some very big funds that still operate in a fashion that more advanced firms in the US and here would consider out of date. The client dealer relationship in the US is not so data-driven because of the concentration around larger dealers in the US. There’s more fragmentation in Europe and the UK.
Is data cost something still worrying participants, how is this specific to fixed income?
All costs are a worry to fixed income participants. I’d suggest costs of electronic execution is a bigger concern than data, certainly for the sell-side, as it has a direct impact on a per trade bottom line basis. One of the main complaints we get from the sell- and buy-side is that data is a high margin basis and so it’s attractive for people to acquire as much data as they can and then sell it back.
Many entities follow the equities exchange model to generate that high margin business. Often this involves buying data where your data has formed a part of that information. For both the buy- and sell-side, there is more that can be done to ensure they receive value for their contribution to data sets.
Could a consolidated tape have any actual impact on fixed income trading?
Using TRACE in the US as the example, the consolidated tape will certainly have an impact on trading. The extent of that impact depends on where it settles, in terms of coverage and deferrals, and who the CTP is. I didn’t live through TRACE as I wasn’t in the US at the time but the feedback I got was that it was initially a big impact but it was more widespread in terms of what it captured. The CTP feels like it will be a softer launch than TRACE, in terms of the coverage and the deferrals so it is likely in the EU and UK to have less initial impact. It will have some impact undoubtedly but it depends on the extent to which it lands. In terms of axes, having post-trade data allows Neptune and our clients to better track the value of axe data and relationships.
How do you expect greater data access to impact traders’ future protocol selection?
Greater data access has already led to protocol evolution in e-trading. An area that was once dominated solely by RFQ now has significant flow in all-to-all and portfolio trading and data has driven much of this change. Without data, the buy-side can’t go in with confidence using these protocols. Likewise, the growth in algo pricing in credit, and the resulting expansion of algo axes, has been a strong area for Neptune in 2023.
One of the biggest growth trends we have seen has been in algos and algos are data hungry. An algo is going to be less sensitive to protocol types than a line trader may be. Up to a certain size if an algo gets hit with a request for market as opposed to a request for quote they’re going to much less sensitive than when there’s a human on the other side. Line traders are notoriously unwelcoming to request for market. Having said that, any new protocols, like new technology, has to have a value proposition. It can’t just be a different way to do what you can already do now. Sometimes people can get carried away with new things and lose sight of value add. It has to add something new. A lot of protocols don’t gain traction.
What is the next stage of innovation when it comes to axes?
Axes are what has driven Neptune’s success. It’s the most valuable pre-trade data you can get in the marketplace. Tiering is a big one for Neptune and ensuring the right information is going to the right clients, for the mutual benefit of the dealer and that buy-side entity. It isn’t a case of all information to all people, especially in credit or emerging markets. There are some trades in some sizes that you are just not going to trade electronically due to information leakage.
There are several projects in the pipeline primarily driven by the buy-side including matching axes directly to trading and also tagging axes less generically. By this we mean dealers letting select clients know more about the origin of the axes, in order to elicit a more meaningful engagement on the data. This can only enhance trust and the relationship between sell- and buy-side.