By Antoine Rescourio
Within the last few years, firms operating in the equities and exchange-traded derivatives (ETD) markets have placed an increasing amount of focus on the area of pre-trade risk. Rules such as SEC 15c3-5, the ‘Market Access Rule’ introduced into the US securities markets in 2010, and the ESMA Guidelines released in Europe the following year, have led the way in requiring firms to run ‘in-line’ checks and controls around their electronic order flow.
Although these regulations initially placed something of a burden on market participants when they were first introduced – particularly sell-side firms offering direct market access (DMA)-type services – the market quickly adapted. Today, one would expect that every firm offering low touch electronic market access to its customers has the requisite controls in place to be able to monitor order flow and to reject orders that could breach pre-defined limits.
Implications of G20 reforms on pre-trade risk
In mature electronic markets such as equities and ETDs, all of these pre-trade risk checks can now be performed with negligible impact on latency, particularly if they are run ‘on-the-wire’, i.e. on dedicated FPGA (Field Programmable Gate Array) hardware, programmed specifically to perform the task.
But what of other markets, which are less mature from an electronic trading standpoint? Following the G20 regulatory reforms, the electronically-traded, order book-driven, centrally-cleared model is now starting to become more prevalent across a much wider range of asset classes, including Fixed Income and OTC instruments. And this trend will only continue.
As this shift occurs, firms need to be prepared from a pre-trade risk perspective and to implement the necessary controls that will not only satisfy the regulators, but will also protect themselves – and the market – from malicious, rogue, erroneous or unauthorised trading. And as these markets become more automated, the most sensible approach is to integrate the capacity to perform pre-trade risk checks on day one – at a minimum cost – rather than as an afterthought later on down the line. But what works in one market will not necessarily work in another.
Broadening the scope
The methods currently used in the Equities and ETD markets are a good starting point, but pre-trade risk factors for nascent electronic markets will differ according to the asset class. For example, in the Fixed Income (FI) markets, ease of access to available liquidity is likely to be a key factor, so pre-trade risk controls for FI markets might encompass (for example):
These are just some examples, but there could be many more.
Having the ability to do all of the above at speed, while accessing the necessary data to enable more complex checks to be performed than might typically be the case for Equities and ETD markets, requires a new approach to pre-trade risk. One that is both adaptable and ‘horizontally scalable’, i.e. able to work across asset classes.
This performance/complexity aspect is important because some markets will have more dynamic risk checking requirements, whereas others might require more data points. The horizontal scalability factor comes into play because best practices are likely to be ‘borrowed’ from one asset class (e.g. Equities) and expanded upon by another (e.g. Fixed Income), or one regulatory regime (e.g. US) and another (e.g. Europe).
Adaptability here is key, because new markets and instruments are likely to introduce slightly different ‘in flight’ risk checks requiring ongoing modifications that would be hard to make with more rigid approaches.
The trend towards electronic markets is only going in one direction. And regulators will continue to push for comprehensive pre-trade risk controls in those markets. We are already seeing evidence of this with the MiFID II/MiFIR requirements to improve market transparency and investor protection.
Firms therefore need to be ready to adapt and evolve, which is why it is essential for them to work with partners who can provide flexible platforms that will allow them to perform iterative calculations and to adapt their risk checks across asset classes and products – without impacting the overall latency performance of the trading platform – in a scalable way.
A hybrid design, one that includes both hardware and software, can help to better address what is – and what is not – latency-sensitive within the MiFID II/MiFIR requirements, so that machine resources and network protocols can be efficiently utilised.
Having such a ‘framework’ to build on, as opposed to a rigid approach tailor-made for just one market (or even one client type within a market) will enable firms to rapidly react to new regulatory and risk requirements in the electronic markets of the future, and thus gain competitive advantage.
Read the original post on The Trading Mesh.
(Title quote courtesy Neil Armstrong)