With a delay of almost a year, the Basel Committee on Banking Supervision (BCBS) finally issued the last part of regulations aimed at completing the reform of prudential rules initiated after the outbreak of the global crisis.
The package, that some refer to as Basel IV, is the result of a global compromise and will be much less painful for banks than originally planned. The last prudential rules of the Basel III were prepared in the course of grueling negotiations lasting over three years. The first consultation document on the planned adjustments to the standardized approach (SA) for credit risk assessment was announced by the BCBS in December 2015. The proposal, which was at that time subjected to consultations, primarily included a revision of the risk weights previously applied in the SA for various types of exposure.
It also included recommendations aimed, among others, at reducing the dependence of risk estimates on external credit ratings, at identifying the sensitivity of risk exposures, and at increasing the comparability of risk estimates. All this taken together – despite the correction of the BCBS proposal presented a year later – resulted in one of the most protracted disputes in the history of regulation.
Let us recall, that a year earlier the BCBS also presented a proposal for a comprehensive review of the trading book (FRTB). This reform, carrying quite significant consequences for the bank capital requirements, also in Poland, will soon be introduced into European law as part of the so-called small amendment of CRD IV and CRR. At the end of 2015, the BCBS proposed to the G20 its vision for the completion of the reforms.
It included the following aspects:
- increasing the quality and level of capital;
- enhancing risk identification (risk capture) by banks;
- constraining leverage and excessive concentration;
- adding macroprudential regulations;
- addressing the principles of liquidity risk management;
- guidelines for sound management;
- supervision of banks.
After a three-year discussion on the most important proposals of the BCBS a compromise was reached in December 2017. It was announced as the conclusion of the post-crisis reform of the banking sector known as Basel III.
“The package of reforms endorsed by the GHOS [Group of Central Bank Governors and Heads of Supervision, or the BCBS’s governing body – editor’s note] now completes the global reform of the regulatory framework, which began following the onset of the financial crisis,” said Mario Draghi, the President of the European Central Bank, at a press conference following the compromise, according to a BCBS press release.
Basel III in brief
The “Basel III” reform of prudential rules in the banking sector, initiated in 2010, was a response to the global financial crisis. When the crisis broke out and subsequent banks were bailed out by governments using taxpayers’ money, it quickly turned out that the previous capital requirements were very easy to circumvent. Some banks had core capital that could be used to cover losses at the level of only 1 per cent of the risk-weighted assets (RWA), even though Basel II established a requirement of at least 2 per cent.
The BCBS was supposed to close the loopholes that allowed the banks to circumvent the regulations. It was also supposed to establish rules ensuring that individual banks would be more resilient to the possibility of incurring losses. Capital and liquidity requirements were the first issues to be debated. However, it quickly turned out that closing the gaps requires more detailed regulations covering all areas of banking activities, such as hedge accounting, off-balance sheet exposures or the concentration of exposures.
While the determination to introduce the rules of Basel III into the global legal order was initially quite strong, it gradually faded over time, especially in the Eurozone, which was affected by a recession that lasted many years. The BCBS took the view that only a strong, well-capitalized and trustworthy banking system could supply the economy with money. But what could be done when banks are deleveraging and are not willing to open new exposures, the banks’ valuations are falling because the market is negatively valuing the declining returns from poor-quality assets, and the return on equity is lower than the cost of raising capital?
Raising the capital requirement (for Tier 1 it increased from 4 to 6 per cent), requirements regarding its quality, additional requirements for the world’s largest banks, introducing the liquidity standards, setting of leverage limits, countercyclical capital buffers – all these solutions were introduced by BCBS by the end of 2015. What were the results?
Banks in the world and in Europe significantly strengthened their capital, lowered their leverage ratios, and a large percentage of them already met the liquidity standards. The last, 12th consecutive monitoring of the implementation of Basel III recommendations, carried out towards the end of 2017 by the Bank for International Settlements showed that from mid-2011 the CET 1 core capital of the group of 105 largest banks in the world increased from 7.2 per cent to 12.3 per cent, and in absolute figures from EUR2,125bn to EUR3,023bn. In Europe, the core capital of the largest banks rose by 56.8 per cent to the level of 13.4 per cent.
The devil is in the denominator
The first phase of the implementation of the Basel III reform mainly focused on the calculation of capital ratios “from the numerator side”. In the next phase, the BCBS dealt with the calculation of risk-weighted assets, i.e. the denominator. The Committee decided to look at how banks measured the asset risk.
After many studies the BCBS concluded that banks on average “put aside” 75 per cent of their entire capital on credit risk. Of that, 15 per cent is associated with operational risk, 5 per cent – with market risk, 2 per cent is reserved for the credit value adjustment (CVA), and the rest is allocated for other risks. The BCBS came to the conclusion that operational risk, market risk and CVA are not sufficiently estimated by the banks.
The BCBS started with a review of the trading book (Fundamental Review of Trading Book, FRTB), i.e., how banks cover market risk (market position risk, currency risk, commodity price risk) with capital. Then came the turn of the CVA and operational risk, as well as the standards for securitization. And finally – again – it dealt with the credit risk weights used in the standardized approach and the comparability of measurement of risk using this method and using the IRB approach. Each such review (and only some of them have been listed here) resulted in recommendations that caused the capital requirements to grow larger and larger.
In the final document summarizing the latest introduced recommendations, the BCBS once again revised the ones regarding operational risk, assuming that its calculation before the crisis exposed the weakness of the approach used by banks. In addition, banks paid exorbitant penalties for their “bad behavior”. According to the calculations of the Boston Consulting Group, in 2009-2016 they amounted to USD321bn, of which USD118bn was paid by European banks and USD204bn by American banks. All these losses were impossible to predict using the internal models, therefore they cannot be used to estimate operational risk.
The capital requirement for operational risk is supposed to cover the risk of losses resulting from inadequate internal processes, human errors, improperly operating systems or external events. In order to determine it, banks will have to review their operational losses over the past 10 years. The requirement will be obtained by multiplying the ratio of past losses by the changes in pre-tax profit, corresponding to the increase in the size of the bank’s business.
The BCBS also raised the leverage ratio (LR) requirement for the largest banks in the world (global systemically important banks, G-SIBs). This ratio is the reverse of the leverage – Tier 1 capital is in the numerator, and the denominator includes on-balance sheet and off-balance sheet exposures (including derivatives, repo transactions and other transactions financed with securities). It was previously assumed that it cannot be lower than 3 per cent.
The additional leverage ratio buffer for G-SIBs is equal to half of the additional risk-based capital buffer. So, if the specific capital buffer for a given G-SIB was 2 per cent, the LR buffer will amount to an additional 1 per cent, which means that the Tier 1 capital of such a bank cannot be lower than 4 per cent of all its exposures.
The discrete benefits of internal ratings
“A cautious and reliable calculation of risk-weighted assets is an integral element of the risk-based capital framework,” wrote the BCBS justifying its subsequent steps. Its research confirmed that there is a large differentiation in risk-weighted assets in various banks and that it cannot be explained solely by the differences in the degree of riskiness of their portfolios. On the contrary, the risk of very similar assets is measured in various ways. And this means that banks are putting aside more or less capital, depending on how they measure risk. In this way, capital ratios become incomparable and it is difficult to trust them.
This variation especially concerns certain banks, mainly large ones, that use internal models for risk assessment. Theoretically, these models should allow for a more accurate risk measurement than the standard weights imposed by the supervisors. However, the use of internal models provides an opportunity to minimize risk weights for the capital requirements to be lower – observed the BCBS.
The resulting conclusion was that it was necessary to limit the tendency of lowering the capital requirements by banks applying the IRB approach. The BCBS said that banks applying this method for credit risk assessment should not benefit from this by putting aside lower amounts of capital for similar exposures. Initially, the Basel-based regulators implied that the comparability of risk weights with the standardized approach, i.e. the output floor, should be set at the level of 90 per cent. This means that a bank using the IRB approach could “obtain” a capital requirement that is 10 per cent lower than those using the standardized approach.
“The BCBS pointed to the need to make banks more sensitive to risk, to simplify methods and to introduce comparability,” says Paweł Flak, a partner at the consulting firm EY.
The latest reforms also introduced limitations to the estimation of risk parameters by banks using the IRB approach. An advanced approach (advanced internal ratings-based approach, A-IRB) cannot be applied to exposures towards financial institutions and large enterprises. A bank also has to perform its own due diligence when granting a loan to a large enterprise. Banks using the IRB approach also have to adopt the minimum levels of the probability of default by their counterparty. Meanwhile, the IRB method cannot be used for equity exposures at all.
Dispute over risk weights
The constant growth of capital requirements amid a continuous decline in the sector’s profitability and the weakness of the euro area economy led to concerns that further increases in requirements could harm the financing of the economy. In Europe, the financing of the economy is much more dependent on the banking sector than, for example, in the United States, where the capital market dominates.
Additionally, the banks in Europe keep the mortgage loan risk on their balance sheets, while the American banks transfer it to the government-run agencies Fannie Mae and Freddie Mac. Supported by the banking lobby, the European Union firmly rejected the BCBS proposals, although the latter was willing to reduce the output floor to 75 per cent. EU Commissioner for Financial Services Valdis Dombrovskis even said that the output floor was not needed at all. At the end of 2016, the European Parliament adopted a strong resolution opposing the new standards. “A solution we could not support is one which would weigh unduly on the financing of the broader economy in Europe,” said Valdis Dombrovskis in September 2016.
A compromise was necessary, however, because the goal was to calibrate the rules in such a way that it would be possible for them to be universally adopted and applied around the world. Otherwise the banks would benefit from regulatory arbitrage and the rules would be meaningless. It cannot be overlooked, however, that the European Union’s resistance against higher risk weights for mortgage loans to small and medium-sized enterprises and against a higher output floor occurred at a time when the European economy was clearly entering a growth cycle. The future may yet prove this resistance to be short-sighted. The disputes – mainly between the European Union and the United States – were also prolonged due to the resignation of Daniel Tarullo, the Fed member responsible for supervision. The negotiations lasted well over a year.
It was ultimately established that the calculation of risk-weighted assets made using internal models cannot jointly constitute less than 72.5 per cent of risk-weighted assets calculated using the standardized approach method. This means that a bank using the IRB approach can achieve benefits of no more than 27.5 per cent of the capital requirement.
A greater variation in the risk weights in the SA method was also approved. In the case of exposures towards small and medium-sized enterprises, the risk weight was lowered – in line with the European Union’s expectations – to 65 per cent from the initially proposed 85 per cent. However, the risk weight for venture capital exposures was increased to 400 per cent.
The compromise also included a significant extension of the transition period before most of the rules become applicable – they are supposed to enter into force at the beginning of 2022. The output floor is supposed to be phased-in gradually from the level of 50 per cent in 2022, to the full value of 72.5 per cent only in 2027, however, during this period the local supervisory body may reduce it to as low as 25 per cent.
Has the BCBS finally concluded the reform known as Basel III? This is probably the case. It will now assess its effects and perhaps suggest some corrections. Contrary to popular opinion, the prudential reform has not been ended at all. And it should not be ended, which does not mean that it should not be adjusted.
Simultaneously with the announcement of the completion of the reform, the BCBS announced a discussion paper on the treatment of sovereign debt. This is a key issue for the security of the banking system, as demonstrated by the debt crisis in the euro area. And it is an issue that has been swept under the rug since then. Moving away from zero risk weights or imposing sovereign debt exposure limits could break the vicious circle of links between governments and banks. Perhaps this is the beginning of Basel IV.