How is the FIX Trading Community addressing the outages topic?
One of the things we’ve been doing is workshopping specific outage scenarios. We’ve had a look at some recent outages and analysed what information is needed, how it will be used, and what is missing. Asking those questions is important because you start to build more of a process workflow and ultimately, this isn’t just about the communication content, it is about an entire process of effectively managing an outage.
What we’re working towards is the standardisation of communication protocols. The industry as a whole has come together to work on this. The main point is to make sure that the information that is shared following an outage is relevant to the audience it is being sent to, which requires a bit of thought and is something that firms should be focusing on as they start to grapple with these kinds of problems. It’s going to require a little bit of thought, but I do think it’s ultimately doable.
What are the main hurdles when looking at building a standardised communication system for outages?
One factor is that each firm does things differently, which makes it quite hard to understand what’s going on if something’s not working properly. With communication currently, you effectively end up with a bespoke process for every single firm, which obviously is hard to deal with.
I think the other thing is that, as we’ve gone through the work of asking the question ‘what should go in an electronic communication about a market outage’, we’ve learned about the process and realised that it isn’t just a question of issuing a communication which is then disseminated on various other electronic channels – there is more to it. You have to look at it from the perspective of whether the industry is even getting the information it needs the most.
When is communication most important?
I think what we’re talking about here is more the ‘heat of battle’ type messaging, as opposed to postmortems. The idea being that these messages in theory could be fed directly into trading infrastructures, which in turn could react directly. We should be conceptualising the idea that, if you get a message from a venue about an outage, there is a plan b.
When something goes obviously wrong, in many ways this is the easiest type of problem to solve. Trickier problems are when it’s not obvious what’s going on and you must wait for events to unfold. Within the trading ecosystem, this usually happens because of complex infrastructures, even at individual firms. This is why standardised communication is important, you’re not just communicating what is wrong, but also the reassurance that it’s being investigated. It works like a yellow alert. These are useful because you can at least keep an eye on things, and be ready for it to go green or red. As firms will have had a chance to prepare, the potential impact should in theory not be as severe, which in turn contributes to market stability.
In terms of more detailed explanations following the resolution of an outage, there is currently no standardised postmortem communication protocol. As the disruption is no longer being felt, firms have more time to deal with and digest the information, rendering the standardisation of postmortem communication protocols arguably less important.
Who should these announcements be going to?
This is a great question because it’s something that the FIX group has been asking. If you wind it back to what the regulators are particularly interested in, I think they basically see it as communication from trading venues to their participants. But, when you look at it more broadly, and based on industry feedback we’ve received, you start to realise that , understandably, everybody wants to know everything. So even in a particular case where a venue is trying to tell their participants, the participants themselves might want to communicate this to their own clients and of course the message they will be sending their clients can be different. They’re not simply going to just forward the communication from the trading venues because the implications to a broker’s clients might not be the same as the implications to the broker.
On the trading desk, if you have a problem with the system, often everyone starts phoning up all the clients simultaneously but being able to do that electronically would speed things up very, very quickly. Sometimes the message is all you need to help continue to trade and the quicker you can get that done, the better.
How long will it take to put this [the standardisation] in motion?
We’ve got something which is pretty close to an agreed model and we’re now about to start the process of drafting up FIX messages. We’re looking to have that done sometime this quarter and at that stage come out with the standard, the documentation, and overall, our view of how we think this should be addressed.
In terms of timeframes for implementation, this is still TBC. What we usually do is go to the major players, both those who are involved in this process already and then perhaps those who aren’t yet and walk through what we’ve done. That takes time. Also, there’s a regulatory driver and their timeframes are for now still unclear. But even the fact that the regulators are looking at this means that firms will be somewhat incentivised to look into this as well.
For now, the best approach is to approach the major exchanges, all the big banks, perhaps some of the technology providers in the middle and simply explain why and how having standardised communication for outages really helps – it just makes it a lot easier for everybody. It makes it much more likely that people will build something if they know that they’re building something which is the same as everyone else’s.