Solving the EMIR Reporting data quality problem 

Solving the EMIR Reporting data quality problem 

SFTR insights provided by Jonathan Lee

Data quality – the two words that are perhaps most likely to annoy ESMA and the NCAs when associated with EMIR reporting. We all know that EMIR reporting data quality has been poor for some time – in fact, since the regime was introduced.  Our testing results confirm this is the case.  So what did the first EMIR and SFTR Data Quality Report (DQR) released by ESMA last week, tell us?  Well to be honest, not a huge amount we didn’t already know.  But for a regime now seven years old, the statistics make for some grim reading for regulators and firms alike.

What’s the background to the report?

Since September 2014, the Data Quality Action Plan (DQAP) has been in place between the NCAs and ESMA to try to improve the quality of the data. Part of the plan requires NCAs to perform a quantitative assessment of the data reported by counterparties in their Member State.  This is known as the Data Quality Review (DQR) and has enabled the NCAs to compare specific data of their own supervised counterparties, with the data of other counterparties based in other E.U. Member States. ESMA produced an NCA peer review in 2019 into this process across six regulators with a mixed review of their assessment.  The FCA in particular came under criticism from ESMA as not being proactive enough.  It is worth noting that due to Brexit, a lack of data sharing between the FCA and ESMA/EU NCAs will change future peer reviews. 

Pairing and matching rates

Pairing and matching rates have been low since EMIR’s inception in 2014. But what is interesting is that ESMA didn’t even publish matching rates in the DQR.  Its most recent published figures had matching rates at 29% so we have to assume it has the data  – why not include it?  The DQR puts pairing rates at 53% which is slightly lower than the 59% it has previously quoted. Does that mean data quality is going down, not up? Regardless, pairing rates should be a lot higher given that counterparties still only have to agree three fields[1] to pair.

So what does this suggest?

Confusion and communication issues between counterparties still exist when trying to meet their ongoing T+1 reporting obligation. ESMA also emphasises this view and that this further boils down to firms not reporting at all (!), not reporting in a timely, complete and accurate manner, as well as disagreements about the Unique Trade Identifier (still!).  Some other key components from the report are mentioned below:

  • Based on early 2021 data, around 7% of daily submissions are being reported late by counterparties;
  • Up to 11 million of open derivatives did not receive daily valuation updates;
  • According to ESMA estimates, there tend to be between 3.2 and 3.7 million of open non-reported derivatives on a given reference date during 2020.

What needs to happen for EMIR data quality to improve?

  1. More guidance to improve understanding of the obligations

Greater clarity needs to be provided by the regulators specific to the data fields. The trade bodies put together Best Practice Guidance for firms on how to populate fields last year.  We applauded this at the time and would also welcome further initiatives from regulators, especially where confusion still exists between Regulatory Technical Standards/Implementing Technical Standards and the ESMA Q&A. One example where we see consistent confusion amongst firms is the use ‘Broker ID’ field.

  1. Better communication within firms

Firms already have a packed working day in order to meet their existing obligations, but the fact that ESMA are drawing out issues over the UTI and not reporting at all provides concern. If there are ongoing issues, appropriate and documented escalation should be occurring within firms and where appropriate, if material issues are arising, escalation to your supervisory team at your NCA.

What will happen if things don’t improve?

Ultimately we could still see more fines being levied on firms for their EMIR reporting.  The main reason why firms get fined is due to ‘Management and Control’ related issues. This is especially prominent in any EMIR and MiFID fine we see. Kaizen can very much assist on this matter and identify why your reporting isn’t accurate, complete or provided in a timely manner. Once your reported data has been tested, we then provide you with several different data quality reports which can be used to present to your senior management/committees.

What about SFTR?

ESMA has been keen from the outset to bring SFTR governance, oversight and controls in line with EMIR reporting. To this end, SFTR is included in the 2020 DQR for the first time. At this stage, it is a rather token inclusion given the July 2020, October 2020 and January 2021 go-live dates for sell-side and market infrastructures, buy-side and non-financial counterparties respectively.

The report plots the growth in open SFTRs, with the subsequent decline in January 2021 as the UK drops post-Brexit. ESMA also plots the lion’s share of open SFTs being securities lending & borrowing transactions (with repos coming a distant second) and where investment firms and credit institutions continue to report the majority of SFTs.

In this first report, there is limited feedback on SFTR data quality given the lack of maturity of the regime. Comments and a chart are used to demonstrate the decline in rejection rates during the regime’s short life. At Kaizen, we would caution that trade repository rejection rates alone are a very poor measure of data quality. SFTR is a new, complex and challenging regulatory reporting regime in which our regulatory testing routinely flags a multitude of valid but wrong reports.   

If your EMIR or SFTR reporting isn’t up to best practice or if you’re unsure of its quality, we can help. Please get in touch for a conversation with one of our regulatory specialists to discuss your reporting challenges.

Our EMIR and SFTR reporting core training courses also cover all aspects of the reporting requirements to gain an excellent overall understanding of the scope, drivers, trading scenarios and key fields that make up an EMIR and SFTR report. Book your place now.


[1] Reporting counterparty ID’, ‘ID of the other counterparty’, and ‘Trade ID’