The FCA’s mid-year report on MiFID II transaction reporting

School’s out for summer and financial firms are being handed their report card by the FCA. Is it an A for effort and a D for quality?

The FCA’s MRT has hosted a transaction reporting forum where they focused on data quality and completeness for MiFID II. The forum slides have been published and they contain some fascinating data on the extent and quality of data submitted in the first six months of the new reporting regime. During this period, the FCA ingested nearly 3.5 billion transaction reports, but also had to reject a large number of reports. The FCA produced the following list of its top ten rejections:

  1. Instrument is not valid in reference data on transaction date (CON-412)
  2. Underlying instrument is not valid in reference data on transaction date (CON-472)
  3. Country of branch membership is missing (CON-370)
  4. Commodity derivative indicator is missing (CON-640)
  5. Instrument classification identifier is incorrect (CON-430)
  6. No underlying reported for swap transaction (with direction indication CON-473)
  7. Net amount is missing (CON-351)
  8. Transaction report with the same transaction reference number has already been sent for the firm and not cancelled (CON-023)
  9. Transaction for cancellation cannot be found (CON-024)
  10. Transaction has already been cancelled (CON-025)

The top two rejection reasons both concern instrument validation and I don’t think these will be a surprise to anybody. It’s difficult to know who is responsible for these rejections as the reference data the FCA uses (the ESMA ‘FIRDS’) is not perfect and should not be considered a “golden source” – in Market Watch 54, the FCA reported that “ESMA states that it cannot take any responsibility for this information [FIRDS] being complete, accurate or up to date”. Firms need to report instruments that are reportable (!) but sadly FIRDS may still not be a perfect source to test against this obligation and instrument reference data remains a key concern for reporting firms.

Whilst firms are at fault for the majority of these rejections, it’s perhaps unfair to blame them exclusively for all the rejections caused by points three through to seven. We know firms have received rejections for missing ‘country of branch membership’ when the ‘venue’ field has been populated with a systematic internaliser (which would appear to be inconsistent with the ESMA validation rule). Some of the other rejections are dependent upon the CFI (instrument classification) being correct in FIRDS and FIRDS hasn’t always been up to the task. Some participants have also questioned whether some of the ESMA validations have been designed correctly. For example, ‘Net amount’ is required for all cash bond transactions, but this validation is determined by CFI code starting with the characters ‘DB’ which unfortunately captures instruments such as ETCs (exchange traded commodities like iShares Physical Gold ETC) which have a debt structure but aren’t the type of instruments that logically have an associated ‘net amount’ value. Similarly, some participants have questioned the relevance of the underlying direction indictor for CDSs and whether the validation is consistent with the example detailed in ESMA’s Guidelines document.

The FCA continues to provide excellent feedback on reporting issues and the importance of providing complete and accurate transaction reports. However, it would be a mistake for firms to assume that their reports are adequate just because they pass validation and are accepted by the FCA. Validations do not check whether the actual data contained within the reports is correct, something that our quality assurance testing is able to do. The accepted reports are just the tip of the iceberg and it is the issues that lie undetected underneath the water that we are able to identify with our testing.

Read: Mark Steward, the FCA’s Director of Enforcement and Market Oversight’s speech on MiFID II and the fight against financial crime.

Further reading: Ian Rennie on the issues we’ve seen so far in our MiFID II testing