Course Outlines

Credit Courses

Basel IV Credit ratings, pricing and returns Bank Analysis Credit Analysis-Foundation Credit Analysis-Advanced

So,it is Basel IV now…

An introduction: setting the scene.

There was continuity in the previous Basel iterations.

Basel I’s primary objective was quantitative: capitalising the banking system. That objective achieved, Basel II could now build on Basel I and set a qualitative objective: a more risk-sensitive calculation of capital, which would improve the quality of risk management in the banking industry.

As the financial crisis painfully demonstrated, the Basel committee (BCBS) had been wrong: banks did not have enough capital, and the one they did have was of a poor quality. And liquidity was a problem too. So, Basel III, released in 2011, duly introduced measures to remedy these issues, which, by and large have now been implemented around the world.

What Basel III did not do however was change the “Pillar 1” rules, which still govern the calculation of risk weighted assets for credit and operational risk. These were really the innovative part of the Basel II framework.

The future Basel IV rules, by contrast, do change the Pillar 1 calculations, and in doing so go against, cynics will say renege on, the very premises, and promises, that underpinned Basel II.

Not that you would tell. The final document released by the Basel committee in December 2017 presents the new rules as the “finalisation” of the Basel III framework. The “revisions”, as they are known, “complement the initial phase of the Basel III reforms previously finalised by the committee”. Business as usual almost, just adding a few final touches.

As an aside, it is not only the understated nature of the approach that is slightly intriguing, its timing is curious too. As far as I recall, there was no mention of anything to do with the new rules in the years leading to the release of Basel III in June 2011. In fact, it is not until late 2014, some 7 years after the financial crisis, that BIS publications began mentioning the need for “revisions”.

Be that as it may, market participants quickly understood that something bigger was afoot, and dubbed the new rules “Basel IV”, not an official designation of course.

So, how did it come to that, and why do the new rules matter, practically and conceptually?

The BCBS explains: “A key objective of the revisions incorporated into the framework is to reduce excessive variability of risk-weighted assets (RWA). At the peak of the global financial crisis, a wide range of stakeholders lost faith in banks' reported risk-weighted capital ratios. The Committee's own empirical analyses also highlighted a worrying degree of variability in banks' calculation of RWA.”

So, to be clear, banks facing the same or similar risks end up with very different amounts of capital.

The culprits are the IRB banks, almost by definition the largest and more complex banks, whose internal models, as the BCBS sees it, pose problems at 3 levels: simplicity, comparability and risk sensitivity.

Simplicity: in a nutshell, the complexity of banking, and the mathematical models used by large banks have placed “increasingly high demands on a relatively small pool of supervisors with expert knowledge of advanced modelling methodologies”. And “as this complexity has increased” notes the BCBS, “it has also challenged the ability of a bank’s board to understand and oversee the way in which the bank manages its risks.”

What this is telling us is that the Pillar 2, “Banking supervision” created by Basel II is not working as it should and that, unless central banks ( and banks’ boards it seems) embark on a hiring spree of very scarce and expensive specialists, it never will.

Comparability: “A number of studies have found material variation in risk-weighted assets across banks. Complexity associated with the use of internal models, the degree of discretion provided to banks in modelling risk parameters and the use of national discretions have all contributed to this variation. This has highlighted the difficulty of comparing capital ratios.”

Again, central bank supervisors were supposed to prevent widely diverging assessments of (the same) risks. Pillar 2 is not working and as a result Pillar 3, “Market discipline”, isn’t either: “The failure of consistent disclosure to keep pace with these changes has eroded the comparability of banks and challenged the effectiveness of market discipline provided by Pillar 3.”

Risk sensitivity: the Basel II premise here was that internal models would measure risk more accurately. And that, as a result, banks would shy away from capital heavy, more expensive, weak credits and move instead to more rewarding, better quality, exposures.

In this instance, it seems that it is the BCBS’ faith in human nature that was misplaced-and disappointed-: “In principle, internal models allow for more accurate risk measurement. But if they are used to set minimum capital requirements (Read: to game the system) banks have unintended incentives to underestimate risk.”

And there is more. It is not always deliberately that banks miscalculate risks: “some asset classes are inherently difficult to model. This undermines the assumption underlying the current architecture that internal models are always more accurate. A number of empirical studies have suggested that simpler metrics are at times more robust than complex ones. This suggests that the blanket use of internally modelled approaches may not always measure and differentiate risk accurately and appropriately for all portfolios and risk types.”

Reading the above, you will be forgiven for thinking that the Basel II framework was, in some respects, built on some rather naïve assumptions. And that it took a while to understand how naïve they were.

The above however also makes the BCBS’s response fairly easy to understand: the IRB and Standardised Approaches will be more closely aligned.

Basel IV (1) enhances the risk sensitivity of the Standardised Approach, (2) places constraints on the use of internal models and estimates by IRB banks and, (3) just in case that is not enough, sets a “floor” to the capital output of whatever is left of the IRB Approaches equal to 72.5% of the capital that would be required by the same IRB bank if it used the new Standardised Approach.

In other words, the maximum benefit (capital saving) a bank can achieve by using the IRB Approach rather than the Standardised Approach, is capped at 27.5%.

Problem solved?

The new rules should certainly help to reduce the variability of capital outputs and restore the comparability and credibility of RWA calculations.

Striking the right balance, however, between simplicity and comparability on the one hand and risk sensitivity on the other, is a difficult thing to do. In fact, these objectives tend to be mutually exclusive.

Basel IV is, to a degree, an admission of failure, but things must be kept in perspective. Not all IRB banks have been gaming the system or producing erratic capital outputs, and the adoption of the IRB Approaches did incentivise, or force, many banks to raise their game and become better managed banks.

Future will tell, but there may well be a risk that, with less upside on the capital front, banks that could adopt the IRB Approaches, or use them already, may decide they are not worth the trouble after all. And those banks still determined to game the system, or simply unable to improve the quality of their portfolios, may conclude that the Standardised Approach, enhanced or not, is a much friendlier place to hide weaker credits.

Where does that leave the banks?

In a nutshell, IRB banks will have to calculate capital using both the IRB Approach and the Standardised Approach to calculate the capital “floor”. In fact, Advanced IRB banks, depending on their business mix, will probably have to use both the Advanced and Foundation Approaches. Three Approaches in total.

For these banks, the constraints on estimates and the use of collateral for capital relief, can only mean one thing: more capital.

Standardised banks face the prospect of more complex RWA calculations and reporting as more asset classes are created. Commercial real estate exposures, which are currently not an asset class, are a case in point. The magnitude of the impact on capital will depend on the individual banks’ business lines, but overall, the message is clear: more, not less, capital.

So, whether Basel IV achieves its objectives or not, at least the direction of travel is known: a heavier administrative burden and more capital. Now that is continuity.