Unless a proposed centralised reference data utility is able to shift ultimate responsibility for the potentially damaging impact of inaccurate data from financial institutions, the utility’s ability to save the industry money is limited, FIMA delegates heard.

In a heated debate on European Central Bank (ECB) proposals for a centralised data utility, the ECB’s external statistics division head Francis Gross said he didn’t foresee any change in the current liability assignation, which leaves financial institutions responsible for any client losses resulting from faulty or inaccurate data. A utility would act as a “dumb pipe”, serving as an intermediary between market participants, channelling standardised reference data from issuers to the markets; it would also serve as a standards provider. Gross said an attempt to alter responsibility would slow the process of bringing about the proposed utility and is not necessary to its creation, adding that standardisation of reference data would deliver benefits by itself as compared with continuing today’s state of data.

But audience members suggested that without such changes – existing reference data suppliers do not accept responsibility for data quality – institutions would continue to spend millions on ensuring the data they received from disparate sources was clean, normalised and consistent. Gross countered this by suggesting they would have to spend fewer millions with a utility in place.

The exchange came after a presentation by Soeren Christensen, director of Fincore, which is working on the project with ECB, outlining possible design and uses of the proposed utility. According to Christensen, there is no technological obstacle to developing the utility. He told the audience that the data model for the utility is already being used at ECB and has been for the past year.

Seeking to reassure data suppliers concerned about competition, Christensen said the utility would open up many opportunities for developers to build applications to run with the underlying data. He said the utility would provide the marketplace – yet to be defined – with both instrument and counterparty identifiers, and linked data covering instruments, counterparties, market data and corporate actions.

Using the existing data model, he said, it would be possible to build a host of value added applications. These might include the ability to create maps of complex counterparty and instrument relationships, or develop stock selection models that historically have taken days or weeks of effort to create.