Basel Proposals on Credit Risk and Capital Floor Measurements Will Better Reflect
Bank Risks
On 22 December, the Basel Committee on Banking Supervision (BCBS) published two proposals aimed at
amending the current standardized approach (SA) for measuring credit risk and the capital floor framework
applicable to banks using the internal ratings-based approach (IRBA) to determine capital ratios. Combined,
these proposals of a more risk-sensitive SA (a refinement of the current Basel II SA) and the implementation
of a stricter floor (based on the SA) will better reflect banks’ risk, which is credit positive.
The proposals increase the comparability of capital ratios computed under both the IRBA and SA methods,
materially reducing the variability of risk measurements between banks, a principal factor that has
contributed to a lack of confidence in risk-weighted asset (RWA) measurements. Therefore, the BCBS
proposal fulfills its commitment to enhance Basel III’s reliability, consistency and comparability.
Globally, a majority of banks rely on the current SA for computing their capital requirements, but IRBA
builds on banks’ models and is the standard most used by large banks. Over time, RWAs on some exposures
such as mortgages fell out of sync with the actual risks banks incurred. Therefore it was necessary for the
BCBS to provide up-to-date guidance on how SA-calculated bank risks are measured. In other words, the SA
was too simplistic, which called for corrective action.
The revised SA seeks to remain straightforward and uncomplicated, given its target population of small and
midsize banks, and also provide the benefits of a more risk-sensitive framework. For example, RWAs on
residential mortgages will no longer be assigned a single risk weight (i.e., 35% of the nominal exposure).
Instead, RWAs would range between 25% and 100% based on the mortgage’s loan-to-value ratio and the
borrower’s ability to service the mortgage.
As another example, banks’ exposures to other banks will be risk-weighted at 30%-300% of the exposure
based on two metrics, a bank’s common equity Tier 1 ratio and its level of net nonperforming assets, instead
of relying on the external rating of banks or their sovereign. The revised SA will also greatly simplify the way
in which risk-mitigation techniques (e.g., collateral) are taken into account by prohibiting the use of and
reliance on banks’ own estimates.
All these revisions will improve the risk sensitivity of RWAs under the SA and will likely result in much
greater capital requirements at many banks, thereby improving their resiliency against shocks. The revised
SA is important for banks that will compute their capital charges against this standard. It also has
implications for banks using IRBA because their capital floor calculations will be based on the SA.
Under the capital floor, an IRBA-based bank’s capital requirement cannot fall below a percentage of the
capital requirement computed under the SA, regardless of the model-driven outcome. Whether or not the
floor limits the benefit of the IRBA method will depend not only on the IRBA capital requirement, but also
on the calibration of both the SA parameters and the floor itself, both of which will be determined during a
quantitative impact study this year. For example, a capital floor set at 80% of the SA would take effect in
case the capital requirement under the IRBA was less than 80%, thereby limiting the benefit from the IRBA
standard to 20%.
The BCBS takes the view that banks’ internal models are not fully reliable, and hence the BCBS must limit
the benefits stemming from their use. By the same token, the BCBS weakens the significance of capital
ratios that are not floored. Going forward, banks will likely be required to adjust their public disclosures to
communicate about the capital requirement reflecting the floor.