This is a totally new angle for these awards and, we feel, a real benefit of the more flexible approach we have taken this year. It is hard to know where to put Deutsche’s work in market structure research within our previous framework, but luckily this year the problem is solved.
Execution quality remains a huge issue for a lot of clients, some, we have to stress, because of their own shortcomings and unwillingness to address certain issues in their models. That said, just as clients used to look to their banks for a price and maybe some market colour, now they look to them for insights into their execution style.
We have noted how BNP Paribas uses post-trade TCA to analyse a portfolio of trades, well Deutsche Bank takes it even deeper. Not only has it provided the data for a series of papers on aggregation and liquidity pool quality, it has allowed (with the client’s permission of course) live testing of the theories. It has developed real life studies of more intelligent execution.
It is impossible to quantify the value of this work, but importantly it is starting to show up in Autobahn as Deutsche builds upon the excellent start made in 2017 to its analytics package. There is plenty of high quality research available on the FX market structure, but little of it comes from banks, which is a shame because they are, probably, the largest data repositories in the industry.
Deutsche Bank has identified the need for greater engagement on these issues, and has done something about it – effectively it has found a new way in which to get the message out that it is at the cutting edge of market development. For that, its clients, and indeed the wider industry, should be grateful.
Following on from Deutsche Bank’s excellent work in market structure analysis the bank has evolved its market colour app from the original release some two years ago. Probably the key change has been how it has shifted from being data that algo users could utilise to something for all FX market participants.
Market Radar looks good, does not dominate the screen real estate and offers a flexible look at market conditions. The data can be viewed graphically on an historic or real time basis and, unlike most other apps available, it can offer volume at price data.
Users can access a range of looks, such as top of book firmness, depth of market, volume, spread, momentum indicators, technical analysis – all of which are underpinned by the bank’s alert framework that notifies users to impending or actual market events as well as changing market conditions.
Probably the real strength of the product is in how it can show multiple analyses in one window, it effectively offers a live market impact study as well as statistically driven market indicators. As an example of the latter, one of the range of studies that underpins the analysis is a mean reversion indicator, something the bank is likely to build further upon for a very interesting analytical study in the months to come.
For algo users, Market Radar offers great cost of execution graphics including predicted execution paths, thus helping them decide whether to use an algo strategy or to risk transfer the risk.
Market Radar represents what could be termed the complete solution for market analytics, however that would be to undersell it because not only is the technology framework scalable but the thought leadership work at the bank will not cease here. Over the years conversations with the client around market colour have evolved, but they have largely stayed in the voice world, albeit using data as the backbone of the discussion. Market Radar represents the future as it automates so much of the interaction between bank and client – it is the next stage of the electronification of the relationship.
The third leg of Deutsche Bank’s advance in the e-FX space over the past year has been the release of two new order types, both of which leverage the bank’s FX franchise. This makes sense for a bank that has, rightly, over the years prided itself on its risk warehousing and execution abilities.
Deutsche still declines to go the route of several other algo businesses in establishing a separate unit and as such it makes sense, as we noted when writing about Barclays’ development, to release algos that use the bank’s proprietary risk and technology services.
The Pi (principal interest) order is designed for traders that want to execute risk, for whatever reason, over a period of time. By tapping into and using Deutsche’s own pricing and execution technology they can access the bank’s client franchise and execute at the pace dictated by the market, thus reducing impact.
Users can select clip size; execution style (from aggressive to passive); the maximum skew to be shown (from mid); whether to use last look (and again we question why a customer looking to execute a hedge or enter or exit a position would reject trades); and a “smart mid”, which will allow the client to work the order at an offset to Deutsche’s mid.
The second order is eBest, which does exactly what it says on the tin – it replicates the traditional ‘at best’ voice order in the eFX world. In keeping with the nature of the order, the interface is simple, choose ‘buy’ or ‘sell’ and then ‘faster’ or ‘slower’ and the bank’s execution technology does the rest.
This is an order type that needs to access risk transfer and order-generated liquidity and as such doesn’t fit in the agency world. Perhaps more importantly in the current environment, by shifting this execution style online, the bank is providing full accountability around performance and outcomes.