Basel's Second Round: From Rules and Regulations to Return
Most banks have already set a programme in motion for the timely fulfilment of the rules and regulations under the Basel II capital accord. Several developments have drawn attention away from Basel II, such as cost reductions, the realisation of IFRS and – lest it be forgotten – the fact that the accord itself was postponed twice. In this article we advocate giving the Basel II programme a higher priority by focusing attention on the return that is possible from the timely implementation of various elements from the Basel II capital accord. Here the emphasis lies on the added value of Basel II for the lending business and for controlling credit risk.
The Basel II capital accord is a collection of rules and regulations targeting improved risk control and the allocation of capital to risk. Article 601 of the Consultative Paper to Basel II forces banks to adopt methods to estimate their credit risks at counterparty and portfolio level. All of this requires access to certain data.
Basel II offers various approaches for estimating the capital required to cover credit risk. The Internal Rating Based approach makes credit risks more transparent and, therefore, easier to manage. The advantage is that capital can be freed up for the generation of extra profit. Recent calculations by various banks as part of the Basel Committee’s Quantitative Impact Study 3 show that retail portfolios are particularly ripe for substantial capital savings. Many banks have since plumped for the Internal Rating Based approach. As a result, they need to have the relevant historical data as soon as the new rules and regulations take effect in 2006. A good data system is necessary for testing internal rating models on the basis of results according to the requirements set by Basel II. Moreover, a proper data system is vital to proper risk management. The insert explains how the lack of a proper data system can lead to loan loss.
The advantage of a proper risk management mechanismOnly a few banks are presently capable of obtaining timely information on future amounts owed by their customers. This applies to both marketing and risk management. While their information is often sufficient at transaction level, it generally fails to extend to counterparty or portfolio level. The need for information on amounts owed at these levels has been recognised by various banks for some years now. The ‘One Obligor’ principle was thus used to keep track of the amounts owed to banks and their divisions by their counterparties and related entities. A few years after the ‘one obligor’ hype we see that very few banks have implemented this concept with real success. The resulting lack of information can lead to large and unexpected loan losses. Another example was one of the major banks where an account manager in Special Management found out too late that various people at a certain customer had continued to use their company credit cards even after their corporate account had been closed. The reason: inadequate information on the various products within a group of accounts. Even worse was the situation where the Treasury department conducted new money market transactions via the previously unidentified and non-terminated overdraft accounts of a customer which had since gone into default. In these cases much time is lost before the root of the problem is exposed. The earlier this information is obtained the more quickly and effectively the bank can take measures to prevent future losses. A proper risk management mechanism is therefore vital. The requirements of such a risk management mechanism are practically the same as under Basel II. |
The above shows why it is important to start collecting data straightaway. To this end, it is necessary to file data on each lending transaction. Records need to be kept for each transaction concerning the nature and size of the loan, its upper limit, its term and its interest margin. Details on the related collateral and the borrower are also important. A record will need to be made of the borrower’s financial position and, if relevant, its economic and legal relation with other customers which borrow from the bank. According to the Internal Rating Based approach, each transactions needs to be rated. Finally, it is important to keep track of the bad debt provision, if any. To create a proper history, all of this data needs to be regularly recorded, i.e. every month.
This level of data capture is necessary under Basel II for demonstrating the forecasting ability of the internal models by comparing the forecast internal ratings with the actual loan losses.
This also enables the bank to report on the development of its portfolio. By logging information at transaction level, the bank can conduct analyses and investigate trends.
The data need to be sufficiently accessible to the various users with all their different aims. Many software vendors have already recognised this need. Although no single software vendor has come up with a total solution for Basel – which would also be impossible given the gaps in the accord – various software vendors have already taken steps in the right direction to develop some useful solutions. Oracle and SAS Institute are two examples. Both vendors can tackle a substantial part of the Basel II requirements with their existing system retrieval, data warehousing, analysis and reporting software. While SAS Institute has laid a solid foundation for rating model development with its ‘Enterprise Miner’ data analysis software, the strength of Oracle lies primarily in a ready-made data model in its Oracle Financial Services Application. Other vendors such as Peoplesoft, SAP, Riskpro and Misys have also been focusing ample attention on Basel II.
Implementation of these solutions is more than just a software challenge. To achieve quick results, the future data-users need to be involved in the process as well. This is certainly true given the spate of data elements that still need to be retrieved from various production systems, which in turn span various portfolios for various users. I therefore recommend dividing implementation into the following phases:
Phase 1: preliminary analysis
Here several months are spent on in-depth analysis to determine which data elements are important for the implementation of Basel II.
Phase 2: structuring the data retrieval process
In this phase the data is retrieved from the various systems. At this stage it is important to show the initial results in the form of reports. These should be tailored to those who are responsible for the input with a view to creating team commitment.
Phase 3: improving the data retrieval process
The data retrieval process from phase 2 is worked out in further detail in phase 3. Given the labour-intensive nature of the process in phase 2 because of extensive manual work and control measures, phase 3 focuses on improving the data retrieval process.
Phase 4: streamlining the work processes
Here the work processes are simplified to improve the flow of information. One such application is included in the example on the provision process at ING.
Example: INGThis is an example of a Basel II solution with the emphasis on the information-users. It concerns the data warehouse of ING Risk Management in the Netherlands. Several years ago ING started retrieving data from the production environment for risk management purposes. From the very beginning, the primary information providers (i.e. the offices) were confronted with the results from this activity through reporting on the basis of key risk indicators. This made them keen to improve the source data. A greater value was placed on the reports to the management since this reliable information now served as a sound basis for risk management. The second phase covered the provisioning process. Up to that point, the provisioning process was very manual with lots of overlap. To support the transition to a more automated process, the file was cleaned up and the process further streamlined. Subsequently, all staff of Special Management were connected to the provisioning system that retrieved all the data from the data warehouse. In addition to saving significant time, the check on the accuracy and completeness of the data was also streamlined. Thanks to the great commitment of the various user organisations, the quality of the data is now so good that it can also be used for Basel. Nevertheless, Basel was never the aim unto itself behind the creation of this risk data warehouse. |
The above phases, with ample attention to creating team commitment, lead to improved risk control. The management is in a better position to manage risks using reliable reports and analyses. The front office can respond more quickly to negative signs. Such improvements lead to an immediate reduction in loan losses. It is therefore my opinion that a proper Basel II programme is one that aims not only to satisfy the rules in 2006 but also to boost current returns.