Looking at the latest changes regarding accuracy of business clocks, now reincarnated as RTS 25, I noticed a number of interesting points in the text. Most importantly, the absence of the usual clause “…applies from 3rd January 2017″, with just the generic phrase stating that the rules will “enter into force on the twentieth day following that of its publication”. Is this an accidental omission, I wonder? If not, given that the European Commission has three months to endorse the text, which could be delayed by one or two months if the European Parliament and Council oppose it, there’s a possibility we could see business clock synchronisation in force some time in Q1 2016. 

Also, ESMA sets the overall context for clock synchronisation squarely in the camp of market abuse detection, and in particular cross-venue monitoring and the use of post-trade data reports. The accuracy and granularity of timestamps applies only to trading venues and their members and participants, which suggests that buy-sides and non-members are potentially out of scope.

Thankfully, the infeasible nanosecond is out. The new maximum divergence from UTC mandated by ESMA is to a degree of 100 microseconds, or one millisecond depending on a venue’s gateway-to-gateway latency, and a more acceptable one second for manual or voice systems. The new corresponding timestamp granularity on reportable events for those lower latency matching engines is set at 1 microsecond or better.

So whilst the rules themselves look less onerous than many feared, they may well enter into force earlier than most of us expected!