On 18 December 2018, ESMA published an updated version of its Q&As on the benchmark regulation (BMR). ESMA has added two new questions and answers which focus on the distinction between 'input data' and 'methodology' for benchmarks. There has been much discussion on this issue for some time and the ESMA BMR Q&As are an attempt to clarify the distinction (and therefore the different requirements) between the two.
As a reminder, requirements around the input data for a benchmark are set out in Article 11 of the BMR (such as what can constitute input data, accuracy of input data and controls in respect of input data). Requirements relating to methodologies for determining benchmarks are dealt with in Article 12 of the BMR. Article 12 requires an administrator to use a methodology for determining a benchmark that (amongst other things):
- is robust and reliable;
- has clear rules identifying how and when discretion may be exercised in the determination of that benchmark; and
- is resilient and ensures that the benchmark can be calculated in the widest set of possible circumstances
Yesterday's clarification from ESMA confirms what can be included in a benchmark methodology. ESMA has said that, a benchmark methodology can include factors that are not input data (input data being data in respect of the value of one or more underlying assets, actual and/or estimated prices, quotes, committed quotes or other values used by an administrator to determine a benchmark). ESMA goes on to say that, "[t]hese factors should not measure the underlying market or economic reality that the benchmark intends to measure, but should instead be elements that improve the reliability and representativeness of the benchmark. This should be considered as the essential distinction between the factors embedded in the methodology and input data".
ESMA is at pains to illustrate the difference between "methodology" and "input data" in its updated Q&As. ESMA says that, "[a] possible way to distinguish these parameters of the methodology (i.e. factors that are not input data) from the underlying input data is to consider the following difference between the two. Input data changes are taken into account by the methodology every time the value of the benchmark is to be updated, as they reflect the changes in the underlying economic reality measured by the benchmark. By contrast, changing values of the factors are not taken into account in every computation of the benchmark, but only in instances pre-determined by the methodology".
ESMA goes on to give the example of where the market capitalisation of an issuer is considered on a quarterly basis to determine whether its shares should be included in the benchmark; this would form part of the methodology of a benchmark and should not be considered input data. However, if the methodology of an equity benchmark, for example, includes the use of one or more FX rates every time the benchmark is updated, then the FX rates should be considered as input data.
The second question which has been added in the ESMA Q&As relates to regulated data benchmarks (i.e. benchmarks determined by the application of a formula from specific input data (set out in article 3(1)(24) of the BMR), including from trading venues, ARMs, APAs, etc.). Here ESMA has clarified that, the methodology for a regulated data benchmark can include factors not covered in article 3(1)(24) of the BMR only if those factors are not considered input data.
Fundamentally, ESMA sees the methodology of a benchmark as being elements that improve the reliability and representativeness of the benchmark; input data on the other hand sets out the underlying market or economic reality that the benchmark intends to measure. ESMA hopes that the publications of this update will help to clarify this distinction.
