“Horizontal inequity”, or why a higher capital requirement when granular risk models are lacking, is suddenly a problem

The trend for the capital requirement is as follows: more capital, more transparency, combined with less complexity and less divergence. The big question is: aren’t the Basel Committee and the supervisors overshooting in this trend?

After several exercises with the so-called hypothetical portfolio benchmarking analysis (banks were asked to calculate the capital requirement of the same, hypothetical, portfolio) the Basel Committee concluded in 2013 that RWA calculations diverge widely for most of the credit portfolios. Sometimes there is a good explanation for these divergences, such as different jurisdictions exercising their national discretion or standardized versus internal rating based (IRB) approach. However, often the divergences are practice-based and there seems to be no reasonable explanation. One of the conclusions of the hypothetical portfolio benchmarking exercise is that capital ratios could vary by more than 2 percentage points from the benchmark in either direction – and this is for the same (hypothetical) portfolio. However, in most of the cases the differences fall within the +/- one percentage point range, which softens the 2 pp divergence conclusion.

Credit risk is by far the largest component of total RWA (source: BCBS, RCAP, Analysis of RWA for credit risk in the banking book, July 2013

Credit risk is by far the largest component of total RWA (source: BCBS, RCAP, Analysis of RWA for credit risk in the banking book, July 2013

Risk sensitivity: the rise…

The desire for simplicity and equality (“level playing field”) is nowadays stronger than the principle of risk sensitivity. However, when Basel II was introduced, risk sensitivity was key. Banks were stimulated to develop Probability of Default, Loss Given Default and Exposure at Default models so that the riskiness could be established at the loan level. A granular risk assessment was felt to be key for adequate risk management. Banks had to demonstrate that the risk measures were not only used for the regulatory RWA calculation, but also for internal risk management and risk containment purposes (the ‘use test’).

A whole new industry arose out of the risk sensitivity. Risk quants developed a wide array of statistical techniques ranging from pretty straightforward logistic regressions for causal PD models to neural networks. Software developers released statistical solutions for risk modeling and support functions such as risk data collection, backtesting and risk reporting. Model validation departments have been set up for an independent assessment of the risk models. Only models that have been validated by this department may be implemented for the regulatory risk calculations, after approval by the supervisor. Advanced tests need to be run in order to assess the model performance such as the accuracy ratio and the area under the curve.

Banks were rewarded for their sophisticated risk analysis by allowing them to use their own models in the IRB approach that typically led to a lower capital requirement than the standardized approach.

… and fall

This approach comes to an end. The Basel Committee is revising the standardized and IRB approaches and will bring their outcomes closer together. This is done by improving the standardized approach and streamlining the IRB approach. As a result the perceived, model-induced variation in capital ratios across banks will be reduced.

The most important technique that is suggested by the Basel Committee, is the capital floor. This floor will ensure that the capital ratio will not fall below a certain level. It differs from the leverage ratio in that the floor will be risk sensitive. This risk sensitivity will only be at a very basic level, without the complexity of full-fledged IRB models.

The capital floor is said to address a number of issues including RWA inconsistency (as determined by, amongst others, the hypothetical portfolio exercise) and ‘horizontal inequity’, which is, as far I can establish, a new term in the Basel domain. Horizontal inequity addresses the problem that the standardized and IRB implementations have led to significantly different levels of the capital requirement. The hope is that part of this inequality is redressed with the help of the capital floor.

This is remarkable since the difference between the outcomes of the two approaches was a desired outcome. It reflects the incentive that banks had to improve their risk measurement and management practices. I think that this argument still holds.

My conclusion is that the Basel Committee and supervisors need to be careful to replace the sophisticated risk sensitive approaches with blunt tools such as leverage ratios and capital floors.


Over Folpmers
Financial Risk Management consultant, manager van een FRM consulting department, bijzonder hoogleraar FRM

Geef een reactie

Vul je gegevens in of klik op een icoon om in te loggen.

WordPress.com logo

Je reageert onder je WordPress.com account. Log uit / Bijwerken )


Je reageert onder je Twitter account. Log uit / Bijwerken )

Facebook foto

Je reageert onder je Facebook account. Log uit / Bijwerken )

Google+ photo

Je reageert onder je Google+ account. Log uit / Bijwerken )

Verbinden met %s