Liquidity metric monitor
This tool is designed to demonstrate some of the metrics we calculate using the information from FSA047 and FSA048. It is provided to firms on a ‘for information’ basis. The materials must not be used to submit regulatory returns required by our rules.
The tool also provides estimates of the ‘Basel III’ Liquidity Coverage Ratio (‘LCR’) and Net Stable Funding Ratio (NSFR) as announced in December 2010 (though see caveats below).
LCR and NSFR elements
- The LCR and NSFR calculations performed by the LMM from the FSA047/048 data set are estimations, intended to provide a useful guide to firms, but are not a substitute for a firm’s own assessment of their position with respect to these metrics. Errors will arise due to mismatches and gaps in the FSA047/048 data, which is based on FSA definitions, and not on the definitions required to calculate the Basel LCR and NSFR.
- Both metrics are also subject to an observation period to address unintended consequences, if any, and are therefore potentially subject to change.
- Firms should note the metrics specify minimum run-off rates and national supervisors have discretion to set higher run-off rates; they also contain off balance sheet categories where run-off rates are not specified, but where national supervisors are expected to set appropriate, but as yet unspecified, run-off rates.
- The metrics are also intended to be applied at a consolidated group level and do not, therefore, specify the treatment of intra-group loans and deposits. Where firms complete the FSA 047/048 on the basis of multiple legal entities the outputs will not capture ‘trapped’ liquidity in the context of calculating the consolidated metrics.
- Mismatches in the data and calculations include, but are not limited to, the following:
- The LMM uses a one month (modified following convention) stress horizon, whereas the LCR uses a 30 calendar day horizon
- FSA 47 and 48 report principal flows only and do not include interest paid or received or other outflows, such as dividends.
- Retail and SME outflows are calculated using the minimum run-off rates of 5% (stable) and 10% (less stable), individual jurisdictions are asked to set their own outflow assumptions and are "expected to develop additional buckets with higher run-off rates as necessary to apply to buckets of potentially less stable retail deposits". Supervisors may also assign an outflow to fixed-term retail deposits if, for example, they believe such deposits would be withdrawn in a stress.
- We have assumed a 1:1 map between a firm's Type A and Type B splits and the "stable" and "less stable" definition of the LCR, but as the former are defined by the firm itself, it would clearly be for the firm to judge. Firms should continue to report using FSA 047/048 definitions.
- FSA 047/048 do not differentiate between “stable” and “less stable” SME deposits. We have assumed a 50/50 split between "stable" and "less stable" and therefore assumed a 7.5% outflow for this category.
- In calculating the amount of Level 2 securities we have used a credit quality step 2 cut-off as per FSA 047/048 data definitions; whereas the LCR uses a credit quality step 1 cut-off. FS047/048 only differentiates between bank and non-bank corporates, and therefore we also cannot exclude non-bank financial institution bonds as well.
- The LMM calculates the buffer position at one month as a portion of the cumulative one month cash flow requirement whereas the LCR uses an adjusted buffer after rollover assumptions versus a 30 day cash flow requirement (adjusted current stock versus net cash flow). Both approaches are consistent, although the LMM calculation results in the 75% inflow cap applying to reverse repo of some level 2 assets (where FSA 047/048 does not separate these flows); it should only apply to reverse repo of non-level 1 or level 2 assets.
- The LMM applies the 40% level 2 cap to the requirement, whereas the LCR has a dynamic limit that “adjusted level 2 securities” cannot exceed 2/3 of “adjusted level 1 securities”; only if the LCR equals 100% are the calculations identical.
- The LCR calculations in the LMM assume that firms have no ‘operational’ balances as defined in the Basel text that would receive a 25% requirement instead of a 75% (non-financial) or 100% (financial) requirement. As such, Type B deposits from larger enterprises have been treated the same as type A deposits. Whilst this would typically be true, given the definition as it stands in the text, there will be certain business models that may have significant deposits in this category.
- Under the Basel proposals any short covering trade maturing within 30 days should be assumed to rollover to cover the short and not give rise to a cash inflow or a stock inflow. FSA 047/048 only capture overall shorts in any asset class.
- The NSFR is an accounting balance sheet test and the FSA047/048 uses a “cash balance sheet” using market or nominal values where appropriate, and ignores the accounting treatment.
- Any FSA047/048 balance sheet imbalance is assumed to require 100% stable funding (100% RSF) as these imbalances are typical fixed assets or investments in subsidiaries /associates.
- FS047/048 does not capture unencumbered loans maturing beyond one year that attract a 35% or lower risk weighting in order to apply the lower 65% RSF factor. This would predominantly capture unencumbered high-quality mortgages, so for some banks this would represent a significant difference.
- As with the LCR, the LMM assumes the firm’s definition of Type A and Type B retail deposits matches the Basel definition and cannot differentiate between stable and less stable SME deposits. In the latter case, we have assumed a 50/50 split and therefore assumed an 85% ASF factor (90% stable / 80% less stable).
- In applying RSF factors for securities, we have had to make assumptions as FSA 047/048 does not distinguish between securities owned outright and those borrowed under repurchase agreements, whereas the accounting balance sheet does.