Abstract
The BoL has been implementing risk-based supervision (RBS) methods on a pilot basis. The RBS manual is now substantially complete, and the BSD staff are applying the RBS methods on a pilot basis. The BSD staff have drafted Institutional Profiles (IP) and Risk Assessment Summaries (RAS) for more than half of the banks; benchmarks and peer groups are being implemented; on-site reports of examination (ROX) have been utilized for several banks. The mission worked with off-site and on-site teams analyzing data for two pilot banks. Utilizing actual results for two pilot banks, the mission assisted BSD staff in identifying the risks, measuring and evaluating the impact on banks’ condition, and formulating appropriate conclusions and ratings.
I. Background
A. Introduction
1. MCM delivered a TA project on enhancing the capacity of the BoL in RBS in 2016–17. The project, which was financed by the Financial Reform and Strengthening Initiative (FIRST), consisted of three short-term missions. These missions assisted the BoL in developing a supervision manual, which was a first step to effective RBS.
2. The project enjoyed strong ownership. The BoL established two, five-person task forces for each on-site and off-site supervision. The task force format provided a useful means of knowledge exchange and hands-on training for capacity building. BoL staff demonstrated a willingness to engage with the missions and showed increased enthusiasm to move forward.
3. In November 2017, a follow-up TA was provided. The BoL had made great strides in preparing for implementation of RBS of banks. A new supervisory manual reflecting key principle of risk-based approach to supervision had been drafted. The manual includes templates of Institutional Profile (IP), Risk Assessment Summary (RAS), and Report of On-site Examination (ROX). The BSD drafted benchmarks for 18 Financial Soundness Indicators(FSIs). Enhancing on-site examination procedures had advanced.
4. Implementation of RBS remains a multi-year project. The adoption of new procedures should be followed by appropriate training of all BSD staff. Supervisors need to become more analytical and able to explain the rationale for their assessments to bank management and boards of directors. The legal and regulatory frameworks require a comprehensive review to support the implementation of RBS. The BSD should also ensure the adoption of proper corrective measures as well as the monitoring of how banks respond.
5. A new TA project was approved in December 2017 with the FIRST funding as a Phase II project in order to maintain the commitment of the BoL in enhancing financial sector supervision with a risk-based supervisory approach. The following areas have been identified as priorities:
Assisting in implementing RBS manual,
Assisting in strengthening the regulatory framework to effectively implement RBS,
Providing training and building capacity of supervisory staff through close interaction with BSD staff.
B. Progress Since the Previous TA Mission
6. The BSD has demonstrated favorable efforts to apply RBS concepts in practice. Following the November 2017 TA mission, BSD staff prepared IP and RAS reports for 50 percent of banks and also prepared ROX for several banks. Most narrative text of ROX tended to be a recitation of data volumes and changes rather than a risk-focused analysis of levels and trends; however, this is understandable for the initial efforts.
7. The off-site team is now reviewing a broader set of financial indicators when updating CAMELS ratings and addressing risk sources and levels. This is an improvement over previous practice. The addition of peer groups for domestic and foreign banks also strengthens review of financial indicators. Further work with RBS methods will enable more substantive comments regarding bank risk levels and risk management capacity.
8. RBS methods are being progressively applied for analyses and examinations. The RBS manual is proceeding through formal approval and adoption by senior management of the BoL. Meanwhile, BSD staff are applying RBS methods on an incremental basis to become more familiar with the process and develop their risk analysis and writing skills. This is beneficial to prepare staff for more effective capacity development at future TA missions.
Status of Recommendations Last TA Mission (Nov-2017)
Status of Recommendations Last TA Mission (Nov-2017)
| Recommendations | (Priority) Status |
|---|---|
| General | |
| Finalize working version of RBS manual and supporting documents (IP, RAS, ROX) | (High) RBS manual and tools are substantially done; must now be formally approved by BoL. |
| Continue development of RBS analytical and examination skills among all BSD staff through appropriate training | (High) Being done through on-the-job learning-by-doing; formal sessions to train all BSD staff are proposed for future TA missions. |
| Conduct periodic reviews of the coverage and quality of the collected data | (Medium) To be done as new data is received from banks. |
| Review the legal and regulatory frameworks to support the risk-based approach to supervision | (Low) To be done as new laws, regulations are developed. |
| Begin developing internal BoL policy for applying appropriate regulatory responses and corrective/enforcement actions | (Medium) To be done Discussed further at this mission; policy development team to be appointed. |
| Off-site Supervision | |
| Prepare and utilize IP and RAS for all banks | (High) In process: Completed IP and RAS prepared for 41 banks and two banks reviewed in detail with staff |
| Further develop and refine benchmarks for FSIs | (Medium) In process: Two peer groups established and new benchmarks drafted for testing |
| On-site Supervision | |
| Prepare priority-based on-site examinations schedule for 2018 | (High) Done. 2018 plan reviewed during mission. |
| Apply RBS on-site examination procedures for all examinations | (High) RBS procedures used for several test banks; being applied incrementally for other on-site exams. |
| Formulate the scope and areas of focus for on-site examinations using IP/RAS/last ROX and discussing with off-site analysts | (Medium) In process. Examinations are being planned with input from off-site and on-site. |
Status of Recommendations Last TA Mission (Nov-2017)
| Recommendations | (Priority) Status |
|---|---|
| General | |
| Finalize working version of RBS manual and supporting documents (IP, RAS, ROX) | (High) RBS manual and tools are substantially done; must now be formally approved by BoL. |
| Continue development of RBS analytical and examination skills among all BSD staff through appropriate training | (High) Being done through on-the-job learning-by-doing; formal sessions to train all BSD staff are proposed for future TA missions. |
| Conduct periodic reviews of the coverage and quality of the collected data | (Medium) To be done as new data is received from banks. |
| Review the legal and regulatory frameworks to support the risk-based approach to supervision | (Low) To be done as new laws, regulations are developed. |
| Begin developing internal BoL policy for applying appropriate regulatory responses and corrective/enforcement actions | (Medium) To be done Discussed further at this mission; policy development team to be appointed. |
| Off-site Supervision | |
| Prepare and utilize IP and RAS for all banks | (High) In process: Completed IP and RAS prepared for 41 banks and two banks reviewed in detail with staff |
| Further develop and refine benchmarks for FSIs | (Medium) In process: Two peer groups established and new benchmarks drafted for testing |
| On-site Supervision | |
| Prepare priority-based on-site examinations schedule for 2018 | (High) Done. 2018 plan reviewed during mission. |
| Apply RBS on-site examination procedures for all examinations | (High) RBS procedures used for several test banks; being applied incrementally for other on-site exams. |
| Formulate the scope and areas of focus for on-site examinations using IP/RAS/last ROX and discussing with off-site analysts | (Medium) In process. Examinations are being planned with input from off-site and on-site. |
II. Implementation of RBS Approach
A. Formal Approval of RBS Manual
9. Formal approval and adoption of the RBS manual is anticipated by year-end 2018. The RBS manual is substantially done. It is now going through additional internal review process to harmonize terminology from other BoL departments. Final approval is anticipated by end-2018.
10. Some refinements to the manual may be needed to ensure effective RBS. As part of the internal review process for the RBS manual, some gaps may become apparent where minor refinements are needed to the manual following the banking law revisions.
Recommendation: Do not delay approval and full implementation of RBS methods pending approval of the amended banking law. The risk-based methods in the manual are flexible enough to accommodate the existing banking law, and any necessary revisions can be made after the amended banking law is enacted.
B. Practical Application of RBS Methods – Initial Efforts
11. The mission worked with off-site and on-site teams analyzing data for two pilot banks. Utilizing actual results for two pilot banks1, the mission assisted BSD staff in identifying the risks, measuring and evaluating the impact on banks’ condition, and formulating appropriate conclusions and ratings.
12. Sharing bank data allowed the mission to provide optimal advice using relevant data. BSD staff provided data without names for the two pilot banks. This was very helpful as a first step. More hands-on training is needed for the team members to achieve good risk analysis and report writing skills. Following that, training will be needed to develop similar skills for all BSD staff including those in the BoL provincial branches. This will require additional TA missions using actual bank data from off-site returns and on-site examinations.
1) Off-site Supervision
a. Implementation of IP and RAS
13. Bank supervisors have drafted IP and RAS reports for more than half of the banks. The off-site team continues to prepare these documents in support of initial transition to RBS. The pace with which the off-site team has drafted IP and RAS documents is acceptable for the breadth of information which must be summarized. All should be complete by the end of the third quarter of 2018. Once the IP and RAS have been drafted for all banks, maintenance of the documents will require only routine effort.
14. Off-site staff made minor revisions to key ratios in the IP and RAS. Off-site management requested that a new credit risk indicator be added to provide a different perspective on NPLs.2 Off-site staff have also prepared data collection requirements for remaining maturity of certain assets and liabilities needed to calculate three liquidity ratios. New liquidity data should be available by year-end 2018.
b. Peer Groups
15. BSD staff members have developed two peer groups within the banking sector to improve benchmarking. Building on guidance provided during previous TA missions, off-site staff drafted benchmarks for two groups of banks, domestic and foreign. Some characteristics vary slightly within each group, but the parameters chosen are good. The use of peer groups enhances understanding of similar market participants, risk indicators, operating results, and capitalization.
16. Peer benchmarks for domestic and foreign banks are ready for testing. During the last year, off-site staff have progressed from drafting benchmarks to revising benchmarks based on peer groups. The mission helped off-site staff to make adjustments to some of the draft peer ranges to harmonize performance expectations between peer groups. Some of the foreign peer group ranges were adjusted and qualitative analysis emphasized for data sets with large standard deviations.
Recommendation: Incorporate peer group benchmarks in analysis and commentary for IP and RAS.
c. Supervising Different Types Banks, i.e. Domestic vs. Foreign
17. Foreign ownership structures and parent companies are important to evaluate as sources of strength or weakness. The mission discussed the need to fully understand the structure, ownership, management, capital, and liquidity support available from parent companies of foreign-owned banks, regardless whether they are branches of subsidiaries. As BoL implements RBS, examiners should incorporate consolidated and cross-border supervision techniques into their analysis.
Recommendation: Add consolidated supervision concepts into risk analyses.
d. Overall Assessment of the Banking System
18. BSD requested guidance on using RBS and FSIs to evaluate the overall banking system. Current practice for systemic reporting is a quarterly “composite CAMELS rating”3 plus a discussion of broad trends in lending and capital funds. The assessment is provided to senior management of BoL on a semi-annual basis.
19. The mission discussed ways to strengthen systemic reporting using FSIs. The mission reviewed with off-site staff the benefits and limitations of system average indicators, noting that qualitative analysis and commentary must accompany macro-level data. Reports should discuss the reasons for any significant changes, determine which institution(s) causing the changes, and comment on the underlying causes of systemic observations. The mission welcomed BSD off-site management’s plans to incorporate system-average FSIs along with a risk-oriented analysis supported by IP and RAS.4
20. The mission also discussed ways to balance CAMELS ratings and RAS reports as support for quarterly banking systemic analysis. As BSD becomes more proficient at assigning CAMELS ratings, individual bank ratings may be used to broadly assess and describe the overall condition of the banking system. The mission suggested that a list of banks by CAMELS ratings stratified by total assets would be very informative to BoL senior management, especially if supported by summary level discussions of risks and risk management in problem institutions.
21. The mission provided sample templates for reporting of banking system data to BoL senior management. The templates offered different ways to summarize banking system condition with risk analysis that supports conclusions and appropriate supervisory responses.
Recommendation: Include system-average FSIs and supporting risk-oriented analysis of institutions and root causes in banking system reporting. Use peer group averages to support analysis of performance of domestic versus foreign institutions.
2) On-site Supervision
a. Reports of Examination
22. BSD staff prepared reports of examination for several banks using the new format. The new risk-oriented format was used to write examination reports for several banks, but the content was largely a compliance analysis. This is normal for examiners new to RBS. The two examination reports provided to the mission did, however, provide a good basis for training and contrasting the old compliance to the new risk-focused style.
23. Risk-based examination reports must provide more analysis. The two pilot ROX provided to the mission lacked adequate analyses of risks and did not contain the recommended matrix summarizing major risks. The reports did contain multiple appendices of spreadsheets showing data for multiple years and multiple quarters. Examination reports can be significantly improved if only a few key ratios, thoughtful risk analysis, and concise year-end data is shown in the body of reports with detailed supporting data retained in the workpapers.
24. Content must focus on analysis of “condition, cause, and consequence”. In order for bank boards of directors, managers, and BoL senior management to effectively address problems, examination reports must focus on the most important risks, their severity and trends, and the likely consequences if left unchecked. Going forward, BSD has requested further assistance working with actual data in order to build capacity, strengthen staff analytical skills, and achieve the desired quality of examination reports.
Recommendation: Continue using the new ROX format, placing emphasis on analysis of risks rather than recitation of data and compliance; make reports more concise by limiting use of appendices and placing detailed reference data in workpapers.
b. Examination Scheduling
25. The mission discussed preparation of the 2018 on-site examination plan. At previous TA missions, suggestions were offered for developing a priority-based examination plan. A plan for Jan/Jun-2018 was developed by BSD utilizing four main criteria: (i) asset size in banking system, (ii) off-site CAELS5 rating, (iii) interval since the last on-site examination, and (iv) extent of unresolved issues. Progress on the plan was reviewed with staff and while there is room for improvement in defining the scope and intervals of examinations, it is seen that BSD is implementing risk-focused methods on an introductory basis.
26. An internal BoL policy for on-site examinations would be appropriate. The BSD staff are applying basic risk-focused methods for determining the schedule and scope of on-site inspections; however, it would be helpful for BoL senior management to set broad priorities by adopting an internal policy for the frequency and scope of on-site examinations. For example, an appropriate policy would specify that maximum desired interval between full-scope inspections for systemic/non-systemic banks and problem/non-problem banks. In addition, periodical reviews (e.g. semiannual) of examination scheduling could be introduced with the adjustment of plan on a quarterly or biannual basis considering the current issues and available resources.
27. Planning actual on-site examinations using RBS methods is seen as a valuable learning exercise. Planning/conducting on-site examinations using RBS methods has been discussed with staff in conceptual terms; greater benefit can be gained from “learning-by-doing”. In this regard, BSD may want to select one or two additional pilot banks for inspection. TA experts can work with staff to plan the examination and provide feedback after the examination by reviewing ROX.
3) CAMELS Ratings
28. Assigning appropriate CAMELS ratings needs further development. BSD staff assign specific ratings for individual CAMELS components based largely on benchmark ranges; the overall rating is still very preliminary with most banks rated 2 (“good”) or 3 (“fair” or “medium”). The existing desire is to assign ratings using a matrix of finite ranges. A more comprehensive set of key ratios has been adopted and benchmark data is still being refined and tested. In time, general ranges will be developed to provide a better starting point for assigning ratings, but more practice and guidance will be needed to improve examiners’ abilities to consider and assign ratings in a holistic manner.
4) Communication Between Off-site and On-site
29. BSD intra-departmental communication is improving. Off-site staff aim to prepare quarterly updates to IP and RAS for all banks – once the initial IP and RAS have been prepared for all banks. Off-site and on-site examiners are communicating more frequently before and after on-site examinations. Off-site management said that any significant change in risk profile would be communicated immediately, both horizontally with on-site and licensing colleagues but also vertically with the Director General of Supervision. The off-site and on-site team should cooperate with other divisions such as licensing and legal for the effective implementation of RBS.
5) Specific Requests from BSD
30. The BSD staff requested input on specific areas. BSD staff asked the mission to provide feedback on several issues that would help them explain to BoL senior management the rationale for actions and conclusions.
a. Bank Business Plan Evaluation
31. Banks’ business plans are evaluated as part of license applications. When application is made for a banking license, the business plan is evaluated by BSD. The primary focus should be ‘reasonableness,’ and the onus must be on the bank to substantiate reasonableness, not on BSD staff to prove otherwise. The fundamental issue is ‘reasonableness’ of assumptions.
32. For example, are projected growth and revenue rates reasonable for existing economic conditions? Or, are planned products and services already available in the proposed market area, and if so, how does the bank plan to compete against other banks? Analysis must also consider the experience and probable ability of bank management to execute the proposed strategy.
33. Business plans are also reviewed during on-site examinations or when material changes are proposed. During routine on-site examinations, operations and performance of a bank are evaluated against the business plan – primarily the annual budget – with bank management expected to explain material deviations. Similarly, if a bank makes a major change in activities or target market, such as a foreign bank that recently changed its focus from consumer banking to target corporate clients, the bank should explain the underlying rationale and the support for the new focus in advance to the BSD.
b. Loan Review Sampling
34. BSD requested guidance on selecting loan sample size and kinds of loans to review. The current practice of BSD for selecting loans to be reviewed relies on examiner judgment with an informal target to review not less than 50 percent for different categories of loans. For example, one-half of large loans6 would be reviewed, one-half of NPL’s, one-half of loans to related parties, etc.
35. Best international practice does not define a minimum sample size for all banks. Rather, it is important to review (a) loans that pose the greatest risk to capital, and (b) enough other loans to support accurate conclusions about the lending function and credit risk of a bank. Sound supervisory practice is to evaluate large loans and other concentrations of credit, non-performing loans (individually if large or collectively if smaller), loans to related parties (directors, executive officers, major shareholders), and a representative sample of new loans and new loan products. In addition, not only a bank specific perspective should be reflected, but also cross-sectoral trends and issues need to be considered.
C. Training BSD Staff and Informing Bankers
36. Training all BSD staff in RBS methods is needed. Now that the RBS manual and tools are largely final and being tested on pilot basis, training must be given to all BSD staff. The mission conducted an introductory training session on RBS for most BSD staff, and BSD managers requested more training for future TA. In this regard, optimal results can be achieved working with actual data and building capacity through ‘learning-by-doing’.
37. RBS methods and expectations must be explained to bankers. After the RBS manual is formally approved by BoL, BSD plans to conduct an information seminar for bankers to explain how RBS will be applied in practice, what will be expected of bankers in managing their risks, and how the BoL will respond if poor risk management practices or excessive risks are identified. The mission welcomes this out-reach initiative of BSD staff and will assist in preparing materials for the seminar.
Recommendation: BSD staff should prepare a draft presentation for the bankers’ informational seminar. The next mission can offer input for content and clarity.
III. Supporting Framework for Effective RBS
A. Banking Law and Prudential Regulations
1) Amended Banking Law and Updated Regulations
38. BoL is revising the legal frameworks. The BoL is revising/updating the legal framework with TA from the World Bank. The initial focus is on the central bank law, and then on the banking law, other financial sector laws, and regulations. BSD staff have recognized the need for amendments to the banking law and regulations to properly support RBS and that RBS is a necessary prerequisite for the Basel-II capital framework.
Recommendation: The legal and regulatory frameworks should be properly reviewed and drafted to ensure support the RBS.
B. Appropriate Responses and Corrective Measures
39. Risk-based supervision identifies much more than compliance to rules. RBS methods reveal not only risks that exceed limits but also actual and potential risks that may be within regulatory limits but are excessive for a bank’s capital strength and managerial capacities. RBS also reveals both condition and cause so that effective responses can be applied.
40. Responses of BoL will determine the ultimate effectiveness of RBS. Once excessive risks are identified, whether in an individual bank or system-wide, it is important to respond promptly and proportionally. To be effective, responses and corrective measures must be firm, fair, and prompt.
41. An adequate feedback loop between examiners and banks’ management is crucial. The major findings should be discussed with top managements of banks, but also medium significance findings ought to be brought to their attention. Adequate follow up for systematically important issues has to be addressed on the level of banks’ management on a timely basis.
42. A formal policy for response and corrective measures is an appropriate next step. Along with implementation of RBS methods and improvements in BoL’s management information systems which will provide more ‘real time’ data, BSD staff recognize that a formal policy for applying response and corrective measures will be essential to ensure the efficacy of RBS. In addition to enforcing compliance to regulatory requirements, a formal policy will help ensure that BoL’s response measures are applied in a prompt and consistent manner.
Recommendation: The BSD should develop a formal policy for applying response and corrective measures. The policy and a supporting manual should be underpinned by the amended banking law.
IV. Capacity Building and Future TA
A. Continued “Hands-on” Training Using Actual Data
43. Senior management of BoL expressed a desire for more TA on a peripatetic basis. Transition to RBS from compliance-based supervision requires extensive capacity building and hands-on training for all BSD staff, especially during the early implementation stages. Up to now, TA has focused on developing the concepts, manual, and tools; going forward, emphasis must be on practical application, learning-by-doing, and informational out-reach.
B. Future TA Needs
44. BSD managers requested and the mission agrees that the following are areas where TA is appropriate:
Provide input to BSD for developing an effective reporting framework to BoL senior management incorporating FSIs and risk assessment summaries
Assist with identification and monitoring of systemic risks based on RBS principles and using FSIs
Provide further ‘hands-on’ training for preparing well-written ROX
Provide appropriate training for all BSD staff in the use of RBS methods including consolidated supervision concepts
Assist BSD staff in drafting an appropriate policy for response/corrective measures
Assist BSD staff in developing content for a bankers’ informational seminar on RBS
BSD staff provided data from the two pilot banks on a “sanitized” basis with no names or other confidential data identifying the bank or customers.
Written-off NPLs/(NPLs + Written-off NPLs)
CAMELS ratings are most useful as summaries of an individual bank’s performance at a point in time.
These tools offer snapshots of risk on an institutional basis.
CAELS stands for Capital, Asset Quality, Earnings, Liquidity, and Sensitivity to Market Risk
A ‘large loan’ is defined as an exposure that is 10% or more of capital.