LSEG to Introduce Decimalisation on Matching…With a Difference

LSEG is to introduce a form of decimalisation to a limited number of currency pairs on its spot Matching CLOB in early October as the first stage of a process that could see the initiative expanded to other pairs in 2025.

The initiative will commence with six pairs, AUD/USD, EUR/GBP, EUR/SEK, EUR/NOK, USD/SEK and USD/NOK. In the first two pairs, participants will be able to place bids at a tenth of a tick increments. In the Scandi pairs, the protocol will see dealers allowed to place bids and offers at one pip increments, compared to the current 10 pips. Currently dealers have the ability to place orders at half-pip granularity in AUD/USD and EUR/GBP, thus LSEG is seeking to ensure the change is not overly-dramatic.

“We have been talking to our customers about this for some time and their feedback was to start in those pairs where there is already a degree of granularity,” explains Paul Clarke, head of FX venues at LSEG. “We have also heard from our Scandinavian customers that they want the ability to submit orders and trade inside the current increments in order to be more competitive for their domestic clients, so it makes sense for these to be our first group of pairs.

“We are very sensitive to minimising disruption to the market when we make changes,” he adds.

Any time a primary venue makes changes, especially around price granularity, there is naturally some unease in the market, and normally it involves changing behaviours as data becomes more granular, however Clarke believes LSEG has hit upon a good compromise – it is not going to change how it publishes its market data. “We didn’t want to go to the extra granularity in terms of publishing the market data, because we know there are worries about behaviour around that, such as tagging and pipping,” he says. “Particularly in pairs that are tick bound, where other markets may be trading inside, we want to give customers the opportunity to trade at those granular levels on Matching, but at the same time avoid making the top of book too noisy.  The intention is to enable market participants to improve the price to trade but minimise market impact.”

While it looks like the granularity will be ‘dark’, that is only the case with the market data, for all trades will be reported at the correct level. Users of Matching will also know that there may be a price improvement available from a change in the amount on either bid or offer. If, for example, the market is 10-11 in one million, this data will be published by LSEG, but if a bid is placed at a granular level inside at 10.1, it will continue to publish 10-11, however the amount on the bid will change. What won’t be known, crucially from a behavioural standpoint, is exactly where that improved bid sits or whether it has joined top of book at 10.

“We want to give customers the opportunity to trade at granular levels on Matching, but at the same time avoid making the top of book too noisy”

Equally, if two improvements enter the market, one selling, one buying at, to use the previous example, 10.4, the trade will be automatically matched at that level. If someone offers below the bid, or vice versa, the aggressor will receive the improvement, as per current practice.

When the bid is hit, however, LSEG will publish the deal, at 10.1 in the above example, for all to see. “This way, anyone crossing the book will get an automatic price improvement, but it is much harder for other players to ‘tag’ inside the bid,” Clarke explains. “The change is very much aimed at helping those customers wanting to price inside to get filled, while at the same time minimising the kind of behaviour we don’t want to see, such as ‘tagging’ or too much volatility in the top of book rates.”

Clarke also points out that LSEG already has a similar environment for manual traders, many of whom don’t want the noise associated with half pips and as such only receive pricing in whole pips. They currently receive price improvement if they hit a price where a half-pip improvement is available. “This has worked well for our manual traders, some of whom were very wary of the change to half-pips,” Clarke observes.

Another potential challenge with decimalisation on a platform of record such as Matching is how it interferes with current practices around highs and lows, but Clarke says this won’t change. “We don’t want to tell the market how to use the data,” he says. “Our high/low rules will remain the same – we will publish an absolute high/low alongside a market traded high/low. The latter is published when certain criteria around time and amount traded are fulfilled, and all that will happen is that it may involve a tenth of a tick rather than a half.”

While the change will involve little work for some traders, there will, Clarke accepts, be some complexity for liquidity providers or those placing multiple orders into the market. “LPs will want to monitor the market data and check their own trades against that data. They need to apply some logic to round out the rate from their trades or order to see where it sits in the market data,” he explains. “Our intention is to publish reports that help them – and us – understand the impact, if any, of the change. For example, do we see an increase in order placement when a top-of-book amount changes, indicating a potential incremental bid? We think these reports will help everyone understand any behavioural changes and fine tune how they interact on Matching.”