Silvano Stagni, group head of marketing & research at IT consultancy, Hatstand asks if there is a data structure that eases the impact of regulatory change?
Any regulatory change has a substantial impact on data. Capturing information that did not have to be captured before, or requirements brought in by new regulation, will create new sources of data and an increase in volume.
For instance, implementing the countercyclical buffer* (CCB) in Basel III will create the need to capture specific information on the debtor/obligor. The CCB will not depend on the jurisdiction of the office where the credit was ‘booked’ but the one where the debtor/obligor is based. Therefore, it requires a set of data items to replace what previously might just have been achieved with a description.
The review of MiFID (MiFID II/MiFIR) will entail the introduction of best execution to non-equity and a new type of trading venue. It will imply standardised non-equity instruments traded in organised trading venues and the substantial curtailing of dark pool trading, if not an outright ban.
One of the principles of ‘best execution according to MiFID’ is the multiplicity of trading venues. The market reacted to MiFID I by creating a lot of multilateral trading facilities, some of which now trade larger volumes than many regulated exchanges. It is expected that the MiFID review will have the same effect for non-equity. Each new venue will create market data, and the move from dark pool trading to organised trading facilities will create more market data resulting in an exponential increase in volume.
Short of having a crystal ball, what are the requirements of a data structure that will be flexible enough to not only sustain the impact of regulatory change but also simplify the implementation of those changes? Here are some tips:
- A clean data structure is a good starting point. Duplication of data should be avoided but also duplication of structures (due to de-normalisation) should be resolved, or at least documented.
- Use of reference data, for example the LEI, allows a number of data fields associated to legal entities to become part of a reference data library using the LEI as a link.
- Review what market data you use and why. Market data does not just represent a cost – in itself a good enough reason to make sure you only use what you need – it will potentially increase in volume, creating bottlenecks.
- Audit your system and data architecture to highlight existing and potential future bottlenecks. Looking at possible ways to improve or eliminate them may not address a specific regulatory requirement but may make their implementation easier.
The four steps listed above will not future proof your data architecture, but they will create a slimmer and more flexible environment that will lessen the pain of implementing new requirements.
* The countercyclical buffer (CCB) is a pre-emptive measure that requires banks to build up capital gradually as imbalances in the credit market develop.
© Best Execution 2014