Counterparty classification regimes, such as CRD IV and EMIR, give banks a good reason to centralise their reference data, and the BCBS’ Risk Data Aggregation Principles provide a clear framework for doing so.
From 1 January 2014, under CRD IV, firms will need to calculate CVA and hold additional capital on all derivatives contracts. However, there are several exceptions to this rule including corporate counterparties who only trade a small amount of derivative. This means not only that firms will have to classify their counterparties by this date but also that they will have to find new ways to transport new data to new places in the bank, all whilst dealing with a host of similar demands.
EMIR already requires banks to classify their counterparties according to whether they are financial counterparties (FCs), non-financial counterparties (NFCs) or non-financial counterparties who clear a certain amount of derivatives (NFCs+). CRD IV uses these same classifications but complicates the problem in two ways: 1) there are also a host of other classifications in the CRR, including CCPs, securities lenders, pension funds and public bodies, which all have to be treated differently; 2) CRD IV requires these classifications to be made available to new areas of the bank, specifically risk and treasury. For this purpose, data will need to be collected, validated and then aggregated, either by reconciliation of multiple silos or by a large-scale centralisation of the data.
When solving for this issue, banks should have in mind new requirements related to their risk data, in particular the BCBS Principles for Risk Data Aggregation. These Principles mark the first time that regulators have taken a clear and focused interest in how banks manage their data and the impact could be huge. In any case, G-SIFIs will have to have implemented these Principles in some form by Q1 2016 and the likelihood is that they will subsequently be extended to other institutions. Therefore, the case is there for FIs to make the necessary changes now, rather than revisiting their data systems in two years’ time.
Broadly, the Principles can be broken down into four major impact areas: governance, processes, infrastructure and architecture. Key to the implementation of CRD IV will be the architectural requirements. Banks must aggregate all data which is supportive of their risk function in a way that ensures accuracy, timeliness and completeness, and must move towards single sources for individual types of risk (market, credit, liquidity etc.) or else have automatically reconciled systems. All of this means that when banks come to consider possible target operating models for making counterparty classifications available to the back office, regulators are clearly trying to push them towards more holistic and less siloed solutions.