Speech by Stefan Ingves at the IIF Annual Membership Meeting
Speech by Mr Stefan Ingves, Chairman of the Basel Committee on Banking Supervision and Governor of Sveriges Riksbank, at the 2015 IIF Annual Membership Meeting, Lima, 9 October 2015.
As prepared for delivery
Good afternoon and thank you very much for the opportunity to provide you with an update on the work of the Basel Committee. This is also a good moment to reflect on the Committee's future work agenda. The post-crisis period has been very busy for policymakers, supervisors and banks as we adjust to the fallout from the financial crisis. On the policy front, much has been achieved and it is important that these reforms are now implemented across jurisdictions. But as always, there is more to be done. Looking ahead to the next 12 to 18 months the Committee's policy agenda continues to be shaped by the following broad themes:
(i) First, we need to finalise the post-crisis regulatory reforms;
(ii) Second, we will continue to work on revising the risk-based capital framework. The aim is to strike an appropriate balance between simplicity, comparability and risk sensitivity; and to reduce the excessive variability in risk-weighted assets (RWAs) under internal models; and
(iii) The third theme is to monitor the impact of the agreed reforms. The goal here is to assess that the standards we have agreed are implemented consistently, and to strengthen supervision.
I will focus my remarks today around the first two of these themes. The last theme, which is monitoring implementation of agreed standards and strengthening supervision, is of fundamental importance. I expect work in these areas will continue into the medium to longer term. I hope to say more on this third theme on a future occasion.
Post-crisis regulatory framework
I will start by saying a few words about our work to finalise the post-crisis reforms. Pre-crisis, the global regulatory framework was based on a single metric - the risk-weighted capital ratio. Today's global framework is strikingly different to the one we had in place prior to the crisis as it is based on multiple metrics rather than a single one. The risk-weighted capital ratio - the cornerstone of the pre-crisis framework - has been fundamentally strengthened by raising the quality and quantity of capital and by improving risk capture. Moreover, the risk-based framework has been enhanced by adding a strong macroprudential focus. Most notable is the introduction of a capital surcharge for systemically important banks and the countercyclical capital buffer framework. A number of other metrics now reinforce the risk-based regime. We now have:
An internationally consistent measure of leverage that accounts for differences in accounting practices across jurisdictions. It also includes leverage arising from off-balance sheet items and derivatives.
A global liquidity framework that includes a short-term stress metric, the Liquidity Coverage Ratio (LCR), and a longer-term structural measure, the Net Stable Funding Ratio (NSFR). This global framework replaces a patchwork of nationally based approaches; and.
An international large exposure framework that will replace a range of national approaches. This will limit the interlinkages across banks, especially systemically important banks, and thereby promote globally consistent regulation and supervision.
This system of multiple metrics recognises that no individual metric is perfect. Each has its strengths and weaknesses. For example, while a leverage ratio on its own might encourage banks to invest in higher risk-weighted assets, such incentives are curbed by the risk-based framework. Equally, the leverage ratio is there to constrain over-optimistic assessments of risk-weighted assets (either those set by banks or regulators). Similarly, the LCR and NSFR standards discourage any tendency to trade off low credit risks against high liquidity risks. The post-crisis regulatory framework therefore needs to be seen as a package of constraints that mutually reinforce prudent behaviour. Indeed, some banks may find the new system is more challenging to meet. More importantly, regulatory arbitrage of a system with multiple constraints will also be inherently more difficult than the arbitrage of a single risk metric. Again that should not be a surprise - it is a fully intended outcome of the multiple metrics framework.
The shape of the post-crisis regulatory architecture is now clear. Moreover, the various elements of the framework have either been finalised or are close to it. In terms of the various regulatory metrics, the LCR, NSFR and the large exposure regime have been completed. For the leverage ratio, the main piece of remaining work is final calibration. At present, Basel III's leverage ratio is set at 3% of Tier 1 capital. In terms of the risk-weighted capital ratio, the definition of capital and overall calibration were set when we finalised Basel III in 2010.
This brings me to the area which is the Committee's current focus, and which will likely remain so in 2016. This is the denominator of the risk-weighted capital ratio, and in particular concerns about excessive variability in the RWAs. I would like to focus the remainder of my remarks on this issue.
Risk-weighted assets - balancing simplicity, comparability and risk sensitivity
In November 2014, the Committee set out a multifaceted plan to the G20 for reducing excessive RWA variability. In the past year, the Committee has made substantial progress on this front but more work is needed. We continue to reflect on whether the calculation of RWAs strikes the appropriate balance between simplicity, comparability and risk sensitivity.
Let me give you a high-level overview of what has already been achieved or is close to finalisation. When it comes to addressing the weaknesses in the RWAs framework, we can distinguish between three broad areas.
The first of these is policy measures that directly limit the degree of RWA variability. This could be done by placing greater emphasis on standardised measurement approaches. Another way is by limiting the flexibility banks have in determining internal model-based estimates of RWAs.
Around the end of this year, the Committee expects to consult on a package of proposals that will enhance the comparability of RWAs that are calculated using internal ratings-based (IRB) approaches for credit risk. The package will include:
(i) constraints on credit risk model parameter estimates, such as restrictions on loss-given-default (LGD) estimates for low-default exposures;
(ii) simplification and harmonisation of the credit risk mitigation framework; and
(iii) alignment of the definitions of exposures under the IRB and revised standardised approaches;
We have already consulted on proposed revisions to the standardised approaches for credit risk, market risk and operational risk. I expect that the Committee will issue the second and final consultative proposals on the standardised approaches for credit risk and operational risk by the end of this year;
The Committee will also finalise the market risk standard (both the standardised and internal model-based approaches) by year-end.
Consultation on the design of a capital floor was completed this year. Work on its calibration is ongoing and closely related to the finalisation of the overall package of reforms;
The second policy measure to address excessive RWA variability is enhanced disclosure. This includes the greater use of standardised templates. The aim is to improve understanding of the drivers of RWAs across individual banks and over time.
In January 2015, revised Pillar 3 disclosure requirements were published. These requirements included the introduction of harmonised templates to improve comparability and consistency of banks' disclosures. Ongoing efforts are focused on consolidating all Basel disclosure requirements into a comprehensive, coherent and up-to-date Pillar 3 framework. To be completed by end-2016, this work will include the Total Loss Absorbing Capacity (TLAC) standard, market risk, operational risk and capital floors.
Finally, the Committee relies on its ongoing data monitoring exercises that help us to address RWA variability. These exercises quantify the magnitude of the variability, identify the drivers of excessive variability and facilitate supervisory follow-up actions with outlier banks.
The Committee has completed analysis and monitoring of the drivers of variability of risk-weighted assets in the banking book and trading book. These exercises have now covered all trading book models and all significant portfolios in the banking book. The analysis has resulted in policy proposals to address excessive variation in RWAs, and also provided important input for supervisory follow-up actions at individual banks.
Let me now provide a few more details on the progress made on policy measures that will limit the degree of RWA variability.
I will start with the trading book. Revisions to the market risk rules have been going on for far longer than anyone cares to remember - this work started before the financial crisis. As the never-ending search for technical perfection has to end, pragmatic decisions need to be made to finalise the framework by the end of this year. A lot has been achieved in revising the rules, and the time has now come to move to implementation. Most notably:
The standardised approach for market risk has been completely revised and will serve as a credible fall-back and floor to the internal models approach.
The revised standardised approach is more complex than the Basel II standardised approach, but is also considerably more risk-sensitive. In part, the complexity of the standard arises from the complexity of instruments in the trading book, and the need to account for short positions and hedging.
To be frank, my personal preference would have been for something simpler. In part, this is related to my observation that, in aggregate, the market risk framework accounts for only around 6% of total risk-weighted assets. I recognise that this number is greater for trading-dominated banks. For the vast majority of banks, however, the significance of the changes (in terms of capital required) is likely to be quite limited.
By contrast, credit risk accounts for the vast majority of risk-weighted assets in most banks. So it was no surprise that when we consulted on revising the standardised approach for credit risk, we received around 180 comment letters. The response was both vigorous and clear: not many of you liked the proposals. But few came up with any better ideas.
With that as the starting point, the Committee is now well advanced in revising the proposals. We will consult on revised policy proposals by year-end and conduct a quantitative impact study in early 2016. In this case, striking the right balance between simplicity and risk sensitivity involves fairly clear trade-offs. What we proposed last December aimed to introduce additional risk sensitivity. At the same time, we tried to avoid undue complexity. The general feedback we have received could be sorted into two categories:
First, the proposals were viewed as too simple to accurately measure risk. That is, it was argued that additional risk drivers are needed to measure risk with a reasonable degree of accuracy.
The second and opposite view was that the proposals were already too complex and burdensome relative to the existing Basel II standards. In other words, the costs of additional complexity outweighed the benefits of increased risk sensitivity.
In choosing a way forward for the revised approach, I expect the Committee will follow the path of simplification rather than increasing complexity. This is likely to include re-introducing a role for external credit ratings into the credit risk capital framework.
The third major risk area is operational risk. The Basel Committee consulted last year on revising the standardised approach for operational risk. We discussed operational risk at our meeting a few weeks ago and we considered changes to last year's proposal that I believe will go a long way to addressing the weaknesses identified in the public consultation. By year-end, I expect that the Committee will publish for consultation a revised standardised approach as well as a proposal to remove from the regulatory framework the Advanced Measurement Approach for operational risk.
As we foreshadowed in our November 2014 report to the G20, in the Committee's judgement the benefits of the AMA are not proportionate to the related costs and complexity. The revised "Standardised Measurement Approach" to be proposed will significantly enhance the risk sensitivity of the current standardised treatments of operational risk, and would apply to banks of all sizes and complexity. In addition to public consultation, we expect to conduct a quantitative impact study of the standardised measurement approach in early 2016.
I would like to conclude by reiterating my message: the Basel III policy response to the financial crisis is largely complete and the overall architecture of the regulatory framework is now clear. While some work will remain outstanding at the end of this year, its direction is well defined and the Committee has publicly consulted on the key policy issues. Additional time, however, may be needed to refine existing proposals in order to, among other things, take into account comments received during the public consultation process and the outcomes of quantitative impact studies. The review of the standardised approaches for credit and operational risk are examples of key policy proposals that fall into this category.
Further, the post-crisis regulatory framework is comprised of multiple metrics. As a result, the Committee will continue to carefully review the coherence and calibration of the framework as we finalise the outstanding proposals.
The Committee's ongoing policy work is grounded in trying to find a better balance between the simplicity, risk sensitivity and comparability of the risk-based framework. Although I have not touched upon it in my remarks today, the Committee will also be in a position around the end of this year to articulate the outcome of its strategic review of the capital framework. This will provide further clarity on how the Committee intends to address the issue of excessive variability in risk-weighted assets.