Morgan Stanley was one of two banks with whom the in-person appointment had to be cancelled, so again feedback played a big part along with our previous observations. The first thing to note is that Morgan Stanley was picked out by several users who observed the algo strategies had been enhanced. In many ways this is not, and should not, be a surprise given the lasting excellence of the bank’s QIS team which has for more than a decade now been at the forefront when it comes to execution analysis; however, we have noted – many years ago admittedly – that there had been a tendency on the part of some clients to use the bank’s analysis tools and another provider’s algos. That appears now to have changed once and for all.
A big part of that shift could be how algo strategies, market access and smart order routers have become somewhat homogenised – not for the first time we would observe that the real differentiator in the algo world is the quality of the internal pools it can access, although it has to be noted that several algo users report using dark pools or mid-books a lot more than they used to. If, however, the world is about understanding the market dynamics, then this speaks to a long-term strength of Morgan Stanley.
The bank also has another factor in its favour, Fusion Edge, the solution that allows clients to not only access internal flow, but also that from other clients. Morgan Stanley has taken a very bespoke approach to the issue of creating liquidity pools for its clients and the word is it is one of the most efficient client-to-client mechanisms on the street. One of these initiatives has been the development of a mid-book for clients, meaning even less flow goes to the lit market, meaning, in theory, less market impact. Throw in the ability to front and rear load the orders, to effectively build into a strategy and pre-empt market impact in the case of the former, and you have a very popular solution for clients.
A big decision was taken at Morgan Stanley a few years back when it decided to reverse what looked to us, and many others in the FX world, as a shift to bringing equities and FX together through the adoption of a quasi-agency model. The reversal means that the bank has a strong FX business that stands on its own, and in turn this has meant that the IP is shared between the principal and agency businesses. It’s becoming a familiar story on the street where banks have woken up to the power, and often uniqueness, of the internal pools – and the subsequent need to have a strong principal offering, but by having the work of QSI to fall back on, Morgan Stanley seems not to have been hit as hard as some others by what now looks like a mis-step in the middle of the last decade.
The driver of this resurgence in the algo business, and indeed of the success of Fusion Edge, has been QSI React, a tool that builds workflow efficiency and simplifies what often looks at the complex matter of strategy selection. Not only that, but it offers protection in terms of the predictive models under the hood that look for potential obstacles on the execution path, such as a drop in liquidity.
The last year has seen the bank further enhance React with more visualisations after the algo has been submitted, as well as the introduction of a Sharpe ratio, which further enhances understanding of the performance of the algo by giving the clients another metric to use. QSI React is about more than just giving the client the tool and letting them go, however, the performance score provided in real time helps them understand how well the algo is going and, importantly, how the bank thinks it will continue to progress.
In terms of analytics, the visualisations have always been strong in Matrix, to this day it can often be forgotten how this was the first single dealer platform to take what was the radical step of having a “dark” view, and the last year is no different. It is easy for the client to understand how the liquidity regime looks, we also like the pressure index which gives a good indication of where the motivated (ie, aggressive) trades are coming from. Of note, the bank also provides analysis that offers different views of liquidity regimes across client-defined periods of time – and again the graphics make this complex data easily understandable.
There is still work to be done; our feedback is that the Matrix GUI can still look a little “busy” when too many tiles are on the screen, and some of the QSI work needs to be better integrated into the client workflow – this is being addressed now. Overall though, Morgan Stanley has achieved no small feat in bringing its execution tools even closer to its superb analytics – the details of which are exposed to the client. This builds trust and transparency in the process, which could be a key factor in the popularity of the bank’s execution product suite.
Awards for e-FX Excellence