Managing credit risk

William Mikhail* discusses proposed amendments to the Basle Accord on credit risk and capital adequacy in the banking sector
The Association of Egyptian Banks ruled, a couple of months ago, that banks should devise and implement new credit policies aimed at facing and reducing risk in bank credit. This has coincided with the prime minister's call for discussing a programme proposed for speeding up the process of merging some private and public- sector banks into larger and stronger entities. The two issues are, of course, not independent and are both related to the problem of adequacy of bank capital.
The 1988 Basle Accord provided guidelines for bank regulation that were allegedly adhered to by banks in over 100 countries for over a decade. Essentially, the accord has required the banks and the financial institutions in the participating countries to maintain a minimum specified ratio of total capital to risk-weighted assets and other risk-exposed items, whether or not these items were on the banks' balance sheets. The accord distinguished between tier-one (core) capital and tier-two (supplemental) capital and classified assets according to four categories of risk.
There is no doubt that the accord has played a significant role in regulating bank credit in many parts of the world over the past decade or so. However, financial markets have developed considerably in the last few years and many banks, especially the larger ones in developed economies, have started to construct and calibrate their own models of risk assessment and risk management. These internally constructed models became more and more sophisticated and proved to perform quite well when based on accurate information, appropriate assumptions and robust methods of estimation.
The good results obtained from these models have triggered the debate between banks and regulators over the pros and cons of continuing to depend on the crude and arbitrary ratios of capital requirements stipulated in the 1988 accord. While banks sought to optimise capital levels, the regulators and the Basle Committee aimed to minimise risk and to develop an international financial system that is stable, efficient, and equitable. It became apparent that some reconciliation had to be reached and that the accord needed to be modified in some parts and amended in others so as to make it more judicious and more useful. This has prompted the Bank for International Settlements (BIS) to prepare a draft for a new agreement, to be referred to as Basle Two, and to invite consultants, central banks, and financial institutions from all countries to study and comment on the proposed amendments as articulated in the numerous documents released since last January.
At the same time, the developments which have taken place in the Egyptian banking sector over the past four years have been cause for serious concern. Before the liberalisation policies of the 1990's, financial institutions in Egypt were considered by some analysts to be too heavily regulated given that economic policy encouraged investment and promoted expansion, and that in the nineties economists were calling for deregulatation. Central Bank rules appeared at times to be rather stringent, and the requirements of the Basle Accord seemed restrictive. However, these seemingly restrictive rules on credit policy in its relation to capital adequacy were not in effect so restrictive, and in later years it became obvious that there were many loopholes in the system that rendered this regulation ineffective. In practice, the application of the rules depended, to a large extent, on the banks' portfolios of assets and liabilities, on core and supplemental bank capital, and on the way risk-weighted assets and the other risk-exposed items were defined and calculated. Evidently, this has made room for loose definitions and inaccurate calculations, which resulted in the instances when capital requirements were not adhered to. Consequently, while some of the Egyptian banks have actually abided by the rules and held capital charges even in excess of the specified ratios, other banks have found ways to dodge the real application of the rules, and have in some instances presented certain financial instruments as subordinated debt when dealing with the tax authorities and as part of tier-one equity when dealing with the Central Bank of Egypt (CBE) in the assessment of capital adequacy.
The cross-reference given above between the proposed amendments to the Basle Accord aimed at mitigating risk effects and stabilising the global financial system on the one hand, and the critical situation which took place in some of the Egyptian banks on the other, immediately raises the question of whether a suitably amended Basle Two would be of particular benefit to Egypt and whether it might, in the longer run, bear the remedy to the credit ills of our banking system.
All drafts for Basle Two are now being revised in light of the huge amount of suggestions and remarks which the committee received from banks and consultants from all parts of the world, and are expected to be finalised before next January. The proposed new accord aims at getting a better evaluation of risk-weighted assets and other risk- exposed items, using a far more elaborate model than that of Basle One. It stands firmly on three pillars to which the committee feels strongly committed. These are: (i) a minimum capital requirement based on a capital assessment technique that would be considered suitable for a particular bank, (ii) a vigilant and efficient supervisory review process to be considered as an integral and critical component of the capital framework, and (iii) a strong disciplinary role for the capital markets through timely disclosures of detailed information on the techniques used for the assessment of regulatory capital and the control and management risk. The proposed model divides into many more categories of risk, which makes it more sensible and more representative of the degree of liability to danger. It introduces the concept of operational risk and attempts to account for it. Depending on the size of the institution and the availability of personnel trained in quantitative analysis, Basle Two encourages the construction and estimation of in-house portfolio models, which would of course need to be updated and re-estimated at close intervals.
Operational risks are supposed to include system failure, losses from inadequate staff, legal threats, fraud, and other forms of unexpected risks, or from external events. The necessity of introducing the concept of operational risk became apparent from the recent developments in the modeling and estimation of measurable risk (credit risk and market risk). Thus, while the arbitrary capital ratio in Basle One was meant, in effect, to represent the capital charges that were considered as an adequate buffer against all risks, it became necessary in the new accord to separate the first type, especially credit risk, from other kinds of risk. One main reason for this is the committee's concern about making the regulatory capital requirement more risk sensitive, having come to realise that the assessment and management of each type of risk would depend on the nature of the risk involved, on the formation of expectations on market prices, and on the effect of covariances among banks' assets. For instance, it was not recognised at the time of preparing the first Basle Accord that the covariance effect, or the degree of diversity in a bank's portfolio, is an important factor in reducing credit risk. Failing to realise this in due time has resulted in the discrepancy between regulatory capital and the 'economic capital' estimated internally by many banks.
This distinction between the various kinds of risk is probably also more suitable for developing countries in the same situation like Egypt. It is clear that operational risk has been of significant proportion in Egypt in recent years, and the failure to account for it has severely damaged the financial situation of some of the Egyptian banks and has even affected the financial system altogether. However, it should be mentioned that the estimation of economic capital needed as a buffer against operational risk is a tedious undertaking, since it depends on the availability of consistent predictions of the bank's future earnings, with confidence bands that would accommodate the volatilities in those earnings. This, of course, makes a good argument for the committee to encourage banks to use their own internally-based models for the determination of capital requirement, under adequate supervision, and indeed this 'customised' approach is what makes economic capital a meaningful concept for banks. The banking system in Egypt has also suffered from not accounting enough for market risk, and not taking into consideration 'residual risks' in credit risk mitigation techniques, like guarantees and collateralised transactions.
The effects of the new accord reach far beyond the immediate impact on regulating bank credit. Apart from improving risk management, the regulation as envisaged by Basle Two may help developing economies grow by having healthier banking systems and more sound credit policies. The new rules, suggested in the proposed modifications, would probably make it more difficult for the weak firms in emerging markets like Egypt to borrow from the banks, but are expected to make it much easier for the financially sound firms. This, again, would allow the banks to hold less capital as cushion against credit risk and, therefore, enhance the solvency of the banking system as a whole.
However, in order for the new rules to be effectively enforced, the banks will have to disclose more information, in line with the third pillar, about their risk management policies and the possible effect of these policies on capital requirements. At the same time, the CBE, as the principal regulator, will need to intervene more positively through a group of highly qualified officers, adequately trained in dealing with the intricacies of the new accord, which has been described by many analysts to be elaborate and quite sophisticated. The disclosure of substantive data and the proper supervision and intervention by the regulators are necessary ingredients for a strategy aiming to ensure that robust risk management policies are pursued, and also to increase the awareness of the banks that their procedures and policies are being closely watched and scrutinised by the regulators and the capital markets alike.
It is realised that the practical implementation of the new rules could be fraught with difficulty. Indeed, the Basle Committee recognises that there will be considerable resource problems, especially in relation to the availability of personnel whose education and training would enable them to master the complexity of the models and the other tools needed for assessing risk and determining capital adequacy. Other reservations relate to the robustness of the risk assessment techniques proposed for smaller firms, to the danger of underestimating the riskiness of lending that may result from providing incentives to banks using their own models of risk, to the problem of how to keep central cank regulators watchful and alert to the sources of violation of the rules, and also relate to many economists' apprehension that moving to a more risk-sensitive framework may well increase the liability to more pronounced cyclical volatilities.
Despite the complexities of the proposed techniques, and the reservations raised by many banks and some institutions including the IMF which supports an accord that would be appropriate to all banks in all countries, the Basle Committee seems to be strongly committed to the adoption of the new proposals. Apart from the elaborate models and techniques designed mainly for the larger banks, the committee is also working out a consistent set of less complicated rules for the smaller banks. These rules would be primarily concerned with forestalling crises in the banking system similar to what happened in Asia and Russia and, to a lesser extent, Egypt, in the late nineties, but would not have the other distinct advantages of the proposals for the larger banks. It is therefore suggested that the banking regulators in Egypt make it mandatory for the larger banks and those which follow relatively sophisticated banking practices to adopt the risk-sensitive approach to the determination of capital requirement, based on the three pillars mentioned above. The other group of banks may be allowed to use the simpler approach of calculating appropriate ratios, again under strict supervision from the regulators, but should be encouraged to move to the first category with incentives similar to those suggested by the Basle Committee. These incentives will of course be designed to result in reduced capital charges for the various types of risk.
Obviously, a thorough understanding and appreciation of these complexities by the CBE and the large banks in Egypt is called for, and a considerable improvement in the quality of regulation and the training of regulators is needed. This should be a long and continuous process, rather than a one-time undertaking. Although the new regulations are not planned to be implemented until the beginning of 2005, the CBE, in view of the above, may find it appropriate to urge the domestic banks in Egypt, both public and private, to start looking into the suggested details and study the possible difficulties they may encounter, and aim not only to implement the new regulations ahead of the proposed date, but also to keep abreast with the studies under way and the possible additions and modifications to Basle Two that may crop up in the interim period.
Finally, it should be emphasised that bank regulation and credit control should not be taken as being against encouraging investment and promoting expansion. Bank regulation in Egypt and developing countries may reduce the number of failed banks and financial service firms, promote more stable financial markets, mitigate the effect of inefficiency, reduce the degree of favoritism, and guard to some extent against corruption.
* The writer is professor of econometrics at the American University in Cairo.
Recommend this page
© Copyright Al-Ahram Weekly. All rights reserved
Send a letter to the Editor