毕博上海银行Risk Mgmt Capabilities DiagProp

《毕博上海银行Risk Mgmt Capabilities DiagProp》由会员分享，可在线阅读，更多相关《毕博上海银行Risk Mgmt Capabilities DiagProp（20页珍藏版）》请在装配图网上搜索。

1、 PROPOSAL TO SUPPORTBANORTE IN THE DIAGNOSIS OF RISKMANAGEMENT AND MEASUREMENT CAPABILITIES AND NEEDSOctober, 1996 CONTENTSI.Current SituationII. Project Objectives and WorkplanIII.Timing and StaffingIV.Oliver, Wyman & Co. MethodologyV.Experience StatementI.CURRENT SITUATIONBanorte has identified th

2、e need to establish a robust framework for measuring and monitoring the overall risk and capital profile of the institution. Such a framework will serve to reinforce sound capital management and protection principles at Banorte. Furthermore, as Banorte pursues various strategic growth initiatives th

3、is framework will serve to inform capital deployment decisions.To this end, Banorte seeks to improve its current business and product risk measurement through the development and implementation of a methodology for measuring risk in a comparative manner across all business types and attributing econ

4、omic or risk capital to each of its principal activities. Our risk measurement and capitalization methodology is based upon an elemental approach to translating default probabilities to economic capital at all levels and across all activities of the institution. As such it will provide meaningful in

5、put to the evaluation of the prospective consequences of specific business development initiatives and enable Banorte to achieve significantly higher levels of capital productivity.However, before embarking on such an initiative, Banorte has expressed interest in obtaining a perspective of its curre

6、nt risk measurement and management capabilities relative to the frontier of risk capabilities found in the most advanced institutions. Further, Banorte would like to gain an understanding of its current systems capabilities as required to support such a broad-reaching risk initiative. II. PROJECT OB

7、JECTIVES AND WORKPLANAs discussed, the first phase of our efforts at Banorte will be designed with the objective of assessing the status of each line of business relative to its risk capabilities. Included in such an assessment will be an overview of Banortes systems capabilities for supporting risk

8、 measurement and management. With this objective in mind, the diagnostic will include four major steps:1. Project initiation and introduction to Oliver, Wyman & Co. methodology2. Review of existing business line processes and underlying methodologies relative to the risk frontier3. Data assessment4.

9、 Identification and prioritization of efforts required to reach the frontier of risk measurement and management.1. Project Initiation and Introduction to Oliver, Wyman & Co. MethodologyThis step will include introductory meetings to communicate the overall objectives, goals, and work plan. In additi

10、on, at introduction, as well as at key points throughout the project, we will be sharing our perspective on risk measurement and management. This will take the form of discussions and presentations on the OWC risk methodology and how it relates to Banortes specific lines of business and overall inst

11、itutional needs.2. Review of Existing Business Line ProcessesA review of existing business line reporting and the methodologies used within that framework (transfer pricing, provisioning, cost allocation) will be conducted. Discrete risks within each business unit will be identified and methodologie

12、s for evaluating risk (e.g., grading systems, scoring systems, audit practices) will be reviewed. These processes and methodologies will be compared against the frontier of risk management in order to assess specific capabilities and potential gaps in current practices.3. Data AssessmentExamination

13、of the data currently available (or feasibly available within a short timeframe) to support subsequent project steps as appropriate. This will include sample data extractions and review in order to determine, not only the availability of data, but the quality and accessibility of data required to su

14、pport risk measurement efforts. This will be an assessment of data required to support risk efforts, and not an assessment of overall bankwide systems capabilities. At the end of this step, we will provide a summary of capabilities, and work with you to develop action steps to improve systems capabi

15、lities to support risk measurement efforts and subsequent project initiatives.4. Identification and Prioritization of EffortsDevelopment of a recommended prioritization of risk management project efforts in which the depth and sophistication of methodologies we will aim to develop is outlined and th

16、e data sources to be used are specified. This will effectively delineate an approach to lead the entire institution to the frontier of risk measurement and management and capital allocation based upon the needs and issues developed during the diagnostic, including systems capabilities for supporting

17、 such an effort. This plan can then be used as a blueprint or point of discussion for future work, including strategic decisions, organizational needs, and systems plans in addition to risk initiatives.III.TIMING AND STAFFINGTimingWe estimate that the work outlined within Section II will take approx

18、imately three to four months to complete, depending on the speed of data acquisition. This estimate is based on our previous experience on similar projects. The Project Initiation and Introduction to Oliver Wyman & Company Methodology will be primarily occur in the first week to two. However, additi

19、onal discussions on methodology may be required throughout our work with you, and for this reason, we have highlighted this activity as on-going throughout the project. The Review of Existing Processes will require in depth interviews with key management personnel throughout Banorte. This work typic

20、ally requires about two months to accomplish and will be dependent on the availability of those personnel identified for interview. Because the Data Assessment work will require in depth discussion with your systems staff and actual data extractions, we find this phase to be of greatest uncertainty

21、with respect to timing. In our experience, even sample data extractions require a month to two months to acquire. The ultimate timing of this activity will depend on your systems capabilities. Finally, in the last two weeks of our stay with you, we will be summarizing our findings and discussing wit

22、h you our conclusions and plans for next steps required to reach the frontier of risk management.MONTH1234Project Initiation and Intro. to MethodologyReview of Existing ProcessesData AssessmentIdentification and Prioritization of EffortsOliver, Wyman & Company StaffingBob Kopech, the Director who le

23、ads our Emerging Markets practice, will take overall responsibility for the successful conduct and timely completion of this assignment. Additionally, Oliver, Wyman & Company will commit a full-time team consisting of a job manager and two consultants for the duration of the project. Furthermore, we

24、 will draw upon the full breadth of specialists within our Risk Management and Retail Financial Services Practices as necessary. Our team will work on-site at Banortes offices in Monterrey and Mexico City, as required.Banorte StaffingIn order to assist these efforts and provide appropriate navigatio

25、n through your organization, we suggest that you form a Project Team for the assignment. This group would ideally consist of some central, overall authority with representatives from each major business unit, as well as Credit, Finance, Treasury, Planning and Systems. Working collaboratively with su

26、ch a group will not only speed progress but also ensure communication with an eventual wider audience. Individual members of this group should plan to dedicate significant time to support this projects goals in their particular areas of expertise.We recommend that you designate someone to act as our

27、 principal point of contact for the duration of the work. Additionally, we will require some dedicated project support: either one full-time dedicated individual, or two half-time individuals. Ideally, these would be individuals who would have ultimate responsibility for risk management throughout t

28、he organization. These dedicated resources are critical to ensuring a smooth hand-off and facilitating discussions at the end of the project. Finally, you should anticipate some additional time commitment from people in systems and financial units to help us with data collection, meeting arrangement

29、s and information validation as required.Other ResourcesFrom an infrastructure perspective, we will need appropriate space for our team: three desks with international phone lines; at least one lockable file cabinet; access to a fax machine, copier and a conference room; 3 PCs with access to your ne

30、twork (the configurations to be specified) and a laser printer.III.OLIVER, WYMAN & COMPANY METHODOLOGIESThe assignment of capital is a major objective of Oliver, Wyman and Companys risk measurement methodologies, and is fundamentally a major linkage amongst all forms of risk within the institution.

31、It is this methodology which will ultimately allow the institution to view its businesses and products as a portfolio with risk and return measurement capability and the ability to evaluate and make strategic decisions based on an equivalent comparison amongst operating units consistent with traditi

32、onal portfolio management techniques. Aggregate portfolio position as well as individual sub-portfolios can be evaluated relative to the efficient investment frontier.The eight guidelines of capitalization described below function as a foundation for all of the Oliver, Wyman & Company risk measureme

33、nt and management work. With these guidelines, we can work with you to determine the span and depth of the risk measurement and capitalization tools required for your institution.1.Comprehensive Risk Coverage: Proper risk-adjusted performance requires consideration of all sources of risk within the

34、bank. We view risk as any phenomenon that creates potential volatility in the value of the firm. This definition embraces three major forms of risk: credit risk, or unexpectedly high loss rates; market risk, or potential volatility in the value of assets or liabilities from market price movements (e

35、.g., interest rate changes); and operational risk due to uninsured errors and omissions or changes in general business conditions (demand, competition, regulation, etc.)2.Two-Dimensional Risk Adjustment: Two risk adjustments are required. Expected losses - the mean rate of loss expected from a portf

36、olio - should be captured and reflected in the business units income statement. Unexpected losses - earnings variance due to the sources of risk described above - create the underlying economic requirement for a capital cushion. For business line evaluation to be fully risk-adjusted, RAROC (Risk Adj

37、usted Return on Risk Adjusted Capitl) measures must appropriately capture both the expected and unexpected components of risk.3.Economic Capital Attribution: Capital attributed to each business line should proportionally reflect the contribution of that unit to the banks overall return volatility. T

38、his approach inherently captures the diversification benefit in a portfolio of businesses as each business is capitalized based upon its contributory volatility rather than its stand-alone volatility. Economic capital assigned in this way will differ from both BIS regulatory capital and actual book

39、capital, as our methodology does not allocate capital, it attributes it. Analysis of the differences among regulatory, book and economic capital can lead to both strategic and financial structuring insights.*Our perspective in this subject is elaborated in the attached Oliver, Wyman Report The Uses

40、of Economic Equity. 4.Reference Point Anchoring: While relative unexpected loss contribution guides the assignment of relative capital levels to business units, attributing a specific absolute amount of capital requires linkage to an external reference point. The reference level we usually employ is

41、 a target credit rating from Moodys or S&P or some other appropriate source. The amount of capital to be attributed can then be determined as the amount which reduces the probability of default for the bank, given its risks, to the selected reference level. Note that the target rating reference poin

42、t should also be used as an anchor for selecting the yield curve used within the transfer pricing system and the required hurdle rate return on the capital attributed.5.Top-Down/Bottom-Up Integration: The conceptual methodology for capitalization should be usable at the line-of-business, product, re

43、lationship or customer levels and allow capital data to be aggregated and disaggregated without problems. Accomplishing this requires that the diversifying effects from pooling risk need to be captured appropriately at every level of capital assignment.6.Forward-Looking Measures: Both expected and u

44、nexpected loss measures should be forward-looking, allowing line-of-business evaluations to anticipate future risks. That is, instead of capitalizing a unit based on historical volatility patterns, it is desirable to capitalize it based on an assessment that reflects shifts in the mix of business th

45、at may have already occurred in the unit and which therefore justify a different prospective level of capital. In our experience, past volatility is only infrequently representative of future volatility at the line-of-business level.7. Adaptive, Sell-Improving Design: The methodology developed shoul

46、d be readily adaptable to changes in organization structure and allow new businesses to be integrated easily. Additionally, the methodology should be designed in a manner which allows subsequent empirical testing of key estimates so that the accuracy of the capitalization figures should improve as u

47、nderlying data and systems are enhanced.8. Client Tailoring: The specifics of the capitalization methodology should reflect the strategic priorities, existing tools, measurement data availability and systems constraints of the client for which it is designed. That is, project efforts should be prior

48、itized so that more sophisticated methodologies are developed for the larger and/or growing sources of risk, and reasonable approximations are used where the risks are smaller and/or data is not available in the near-term.For the purposes of our analysis, we will focus on an evaluation of risk capab

49、ilities which will reflect the three types of risk mentioned above: market risk, credit risk, and operating risk. A. CREDIT RISKCredit risk management should be a core competency of any bank. Inadequate credit risk management typically represents a banks greatest vulnerability. Conversely, superior

50、credit risk management represents a unique opportunity to achieve competitively distinctive performance levels. Risk management processes should be designed for more than just loss avoidance; they should help line management identify opportunities to improve the fundamental risk-return positioning o

51、f their portfolios.Central to the development of effective risk management is accurate and consistent assessment and quantification of risk. Our risk framework uses two measures for this purpose: Expected Loss represents the long-run average loss rate expected for a portfolio based on its current cr

52、edit quality. Expected Loss is therefore not true risk in that it represents anticipatable losses. Rather, Expected Loss is best thought of as a necessary cost of doing business and should be incorporated in RAROC as a specific charge to the income statement. Unexpected Loss represents the uncertain

53、ty that losses will, in fact, differ from Expected Loss. Unexpected Loss is therefore reflects true risk and is incorporated in RAROC as the primarily driver of risk-adjusted capital.This two dimensional view of risk is depicted graphically below.The methodologies used to capture each risk measure a

54、re discussed in turn below.Expected LossCritical to the calculation of Expected Loss are accurate credit rating or scoring systems, together with quantitative parameters that link the ratings or scores to Expected Loss. To arrive at the most accurate and consistent view of Expected Loss the rating s

55、cheme needs to be two dimensional-for example, a credit score to reflect the expected default frequency (EDF) of the borrower; and a second measure to reflect the likely loss in the event of default (Severity), given the characteristics of the facility. This is illustrated below, using hypothetical

56、scales for each dimension:Expected Default Frequency is simply the probability that a company or individual will go into default within one year, and is principally determined by the risk of the obligor. For these purposes, default can take on any of several definitions - for example, actual inciden

57、ce of loss or placement in non-accrual status. The analytical framework can accept alternative definitions of default that are most suited for , as long as Severity and Loan Equivalent Exposure are defined in a consistent manner. Severity represents an estimate of the actual losses incurred on a def

58、aulted loan as a percentage of the outstanding balance at the time of default. Severity rates are formulated to be comprehensive of all economic costs of default, including cost of carry and recovery expense. Severity is found to be driven principally by the collateral used to secure a facility, or,

59、 in the case of mortgages, a comprehensive history of recovery experience.For closed-end products, such as mortgages, Loan Equivalent Exposure is the amount of each loan that is at risk should default occur, incorporating any expected amortization. Loan Equivalent Exposure is very similar to current

60、 outstandings.Unexpected LossUnexpected Loss represents the uncertainty that actual losses will differ from Expected Loss and is the primary determinant of the economic capital that will be attributed to each business union and loan. Unexpected Loss is calculated through an equation derived from pro

61、bability theory and is driven primarily by Expected Loss and correlation:As can be seen, Unexpected Loss increases with higher levels of Expected Loss, although the relationship is not linear. In addition, for a given level of Expected Loss, Unexpected Loss is higher with increasing levels of correl

62、ation. Correlation represents the extent to which loss on an individual loan is correlated with losses for Banorte as a whole. For example, a loan with a high degree of correlation is one in which its propensity to generate losses for the bank is significantly higher when Banortes aggregate loss rat

63、e is high.It is necessary to embody correlation into the Unexpected Loss calculation to measure contributory risk, or the incremental volatility brought to the bank by each individual exposure. In this manner, risk (Unexpected Loss) measured for each loan can be summed to equal the total risk facing

64、 Banorte, including portfolio effects and diversification. Once these building blocks of credit risk measurement have been established, we can extend the methodologies to support and drive the processes necessary to manage the portfolio for optimal risk-adjusted return and soundness. As described below, each methodological segment allows the organization to establish additional management capabilities. Ultimately, the full compliment of methodologies provides for the capability of establishing a fully integrated