CS/CMI/VWMICAFFS/36/2013

Page | 1

Distr.

LIMITED

CS/CMI/VWMICAFFS/36/2013

August 2013

Original: ENGLISH

COMMON MARKET FOR EASTERN

AND SOUTHERN AFRICA

Workshop on Validation of the Manual for Implementing the

COMESA Assessment Framework for Financial System Stability

17-19 August 2013

Nairobi, Kenya

REPORT ON WORKSHOP FOR VALIDATION OF THE MANUAL FOR

IMPLEMENTING THECOMESA FRAMEWORK FOR FINANCIAL SYSTEM STABILITY

2013 (IZ-mkc)

  1. INTRODUCTION
  1. The COMESA Committee of Central Bank Governors in their 18th Meeting which was held in Kigali, Rwanda in December 2012, assigned COMESA Monetary Institute (CMI) to prepare a Manual for the Implementation of the COMESA Framework for Assessing Financial System Stability. The purpose of the Manual is to provide practical step by step outline of how to implement the Framework. In view of the decision, CMI prepared the following two manuals:

i)A Hand Book for Assessing Financial System Stability based on SHIELD Rating; and

ii)Manual for Foreward Looking Financial stability Reports and Stress Testing.

  1. CMI, also organised a workshop for validating the above mentioned manuals from 17-19 August 2013 in Nairobi, Kenya
  1. ATTENDANCE, ADOPTION OF THE AGENDA ITEMS AND ORGANISATION OF WORK

Attendance

  1. The workshop was attended by delegates from Central Banks of Burundi, Congo(D.R.), Egypt, Kenya, Madagascar, Malawi, Rwanda, Sudan, Swaziland, Uganda and Zambia. The list of participants is contained in Annex II of this report.

Adoption of the Agenda and Organisation of work

  1. The workshop adopted the following agenda items:
  1. Calling of the Meeting to Order by the Chairman;
  1. Adoption of the Agenda and Organisation of Work;
  1. Sharing Country Experiences in the Implementation of the COMESA Framework for Assessing Financial System Stability:
  1. Validation of the Manual of COMESA Framework for Assessing

Financial Stability;

  1. Validation of a Manual for Foreward Looking Financial Stability Reports

and Stress Testing;

6 Presentation by ATI on the Use of political and Commercial Risk

Insurance as a Risk Mitigation Tool for Banks and its Potential Impact

on the Capital Allocation Rule

7 Any other Business; and

8. Closure of the Workshop

  1. The workshop agreed on the following hours of work:

Morning: 9.00-12.30 hrs.

Afternoon: 14.00-17.00hrs.

  1. ACCOUNT OF PROCEEDINGS

Sharing Country Experiences in Assessment of Financial System Stability ( Agenda item no. 3)

  1. The delegates from member countries reported on the status of implementations of the COMESA Assessment Framework for Financial System stability as contained below in table 1. The report focused on the following:
  1. Establishment of a Financial Stability Unit;
  2. Establishment of a Financial Stability Committee & Terms of Reference;
  3. Development of in-country Strategic Implementation and Action Plans;
  4. Production of forward looking Financial Stability Reports incorporating, among other issues, Financial Stability Assessments and SHIELDS ratings.
  1. The workshop noted the following from member countries reports:

i)Most COMESA member states have set up Financial Stability Units.

ii)The member states are at various stages of setting up the Financial Stability Committees.

iii)Member countries are still working toward the conduct of SHIELDS ratings of financial stability once the requisite institutional and logistical arrangements have been finalised.

iv)The financial stability assessments and macroprudential supervision are new concepts, which have gained increased attention following the global financial crisis. As such, there are still a number of challenges in respect of resources and skills endowment.

STATUS OF IMPLEMENTATION OF THE COMESA FINANCIAL STABILITY ASSESSMENT FRAMEWORK BY COMESA MEMBER COUNTRIES

No. / Task / Activity / Completion Date / Status of Implementation by member countries
Financial Stability Unit / Establish Financial Stability Unit / 31 January 2012 / Burundi, DRC, Egypt, Kenya, Mauritius, Rwanda, Sudan, Madagascar, Malawi, Swaziland (WIP), Zimbabwe, Uganda
Financial Stability Committee / Establish a multi-disciplinary Financial Stability Committee / 31 March 2012 / Egypt, Kenya, Mauritius, Malawi, Rwanda, Sudan, Uganda, Madagascar (WIP),Swaziland (WIP) Zimbabwe, DRC (WIP),
Burundi
Strategic Plan / Develop in-country Strategic Implementation Plan covering institutional structures and chosen methodologies and/or modalities for:
  • Information gathering;
  • Data analysis;
  • Interpretation of the results;
  • Policy recommendations; and
  • Policy implementation.
/ 31 March 2012 / Burundi, Egypt, Kenya, Mauritius, Malawi Rwanda, Uganda, Swaziland (WIP), Sudan (WIP), DRC (WIP), Madagascar, Zimbabwe(WIP)
Action Plan / Develop in-country Action Plan for Implementation of COMESA Framework covering:
  • conduct of consultative meetings and/or meetings with other stakeholders;
  • establishment of technical subcommittee where necessary;
  • issuance of customized guidance on on-going surveillance, diagnostic assessment, and policy actions;
  • production of Financial Stability Reports;
  • Review of legislation where necessary; and
  • Roll-out of the Framework.
/ 31 March 2012 / Kenya, Mauritius, Rwanda, Sudan (WIP), Uganda, Swaziland (WIP), DRC (WIP), Malawi (WIP), Madagascar (WIP), Zimbabwe(WIP), Burundi (WIP), Egypt (WIP)
Financial Stability Assessment / Implement Financial Stability Assessment Framework incorporating:
  • Financial Stability Report (FSR) in Prescribed Format;
  • Financial Stability Assessment Matrix;
  • GLYOR five-tier coding system;
  • Financial Soundness Indicators (FSIs);
  • Macro-prudential analysis;
  • Stress testing;
  • Household Sector Survey Results; and
  • SHIELDS Rating System
/ September 2011 / Kenya, Mauritius, Swaziland (WIP), Sudan (WIP), DRC (WIP), Malawi (WIP), Rwanda (WIP), Madagascar (WIP), Zimbabwe (WIP), Uganda, Egypt (WIP), Burundi (WIP)
Financial Stability Assessment Reports / Electronic submission of Financial Stability Assessment Reports to the Secretariat for posting onto the COMESA website / October 2011 / Kenya, Mauritius, Rwanda (WIP), Uganda, Swaziland (WIP), Sudan (WIP), DRC (WIP), Malawi (WIP), Madagascar (WIP), Zimbabwe (WIP)
Burundi (WIP)
Egypt (WIP)
Financial Soundness Indicators / Electronic submission of Financial Soundness Indicators (FSIs) to the Secretariat for posting onto the COMESA website / 60 days after end of each quarter with effect from December 2010 / Burundi (WIP), Egypt (WIP), Kenya, Mauritius(WIP), Malawi (WIP), Rwanda, Uganda, Zimbabwe (WIP), Swaziland (WIP), Sudan, DRC (WIP), Madagascar (WIP)
Monitor compliance with the 25 Basel Core Principles for Effective Banking Supervision / Member countries conduct self-evaluations / March 2011 / Burundi, Egypt, Kenya, Mauritius, Malawi Rwanda, Sudan, Uganda, Zimbabwe, Swaziland (WIP), DRC, Madagascar (WIP),
*WIP – Work in Progress

Validation of the Manual of COMESA Framework for Assessing

Financial Stability ( agenda item 4)

  1. Mr. Gift Chirozva made a presentation under this agenda item. In his presentation he highlighted the following:

i)SHIELDS rating;

ii)Theoretical and empirical justifications of SHIELDS rating;

iii)The rating and scoring methodology of SHIELDS rating;

iv)Quantitative and qualitative approaches to assess systemic stability; and

v)How to undertake overall SHIELD rating

SHIELD rating

  1. Under this topic he highlighted the following:

a)The proposed handbook provides guidance on how financial stability practitioners might determine the individual components of as well as the overall SHIELDS on the basis of a five tier GLYOR risk coding system notwithstanding the peculiarities of individual country circumstances.

b)SHIELDS is an acronym for Solvency Conditions; Home Economic Conditions; Institutional Quality; Earnings Conditions; Liquidity Conditions; Default Conditions; and Systemic Loss, factors which characterise the state of stability in a given jurisdiction as informed by developments in global macrofinancial conditions and event risk, the local financial system, and the domestic economy.

c)At a conceptual level, the COMESA Framework for Financial Stability Assessment is composed of three mutually reinforcing building blocks / pillars namely: (a) on-going macro and micro surveillance; (b) diagnostic financial stability assessment; and (c) policy options.

d)The SHIELDS rating system thus facilitates the integration of various quantitative and qualitative financial stability analysis into a standard framework, results of which are amenable to comparison over time with a given country and across nations.

e)The COMESA Framework for Assessing Financial Stability Assessments provides for systematic monitoring of the individual parts of the financial system (institutions, markets, and infrastructure); components of the real economy (households, firms, & public sector); global macro-financial developments; and event risk (e.g. catastrophes).

f)Whereas, there might be temptation to collapse some SHIELDS components into an assessment of banking sector it should be noted that systemic risk goes beyond the confines of the banking sector, hence SHIELDS treatment of factors such as liquidity and default conditions goes beyond the safety and soundness analysis of banks.

Theoretical And Empirical Justifications of the SHIELDS Rating System:

  1. He highlighted the following under this topic:

a)In Goodhart, Sunirand, &Tsomoscos (2004, 2005, 2006) and Tsomocos (2003) and Aspachs et al (2006)’s two factor model suggests overall Financial Stability can be summarised by probability of default (PD) of both banks and economic agents and bank profitability.

b)Episodes of financial instability are characterised by a combination of high probabilities of default and low bank profitability

c)Blancher et al ’s (2013, p7) tool kit of systemic risk monitoring points out systemic risk can be aggregated at different levels including individual that institutions and markets, transmission channels, and the whole financial system and the economy. In addition they also single out:

i)credit risk (a key source of risk in most financial institutions) as assessed by probabilities of default and loss given default;

ii)liquidity risk as denoted by liquidity ratios, market liquidity & financial institution’s funding conditions

iii)and market risk being sensitivity to volatilities in interest rate, exchange rate, or asset price shocks as well as latent vulnerabilities.

The Rating & Scoring Methodology of SHIELDS Rating

  1. The following were the salient points which were presented under this topic;

a)Intuitively, assignment of ratings calls for the adoption of a rating methodology. SHIELDS anticipates the multiple sources and causes of financial stress and hence rely on a wide spectrum of quantitative and qualitative evaluation factors. It provides a systematic framework, instrument, or single tool that summarises the degree of financial stability of the system into a common yardstick amenable to comparison over time and across nations.SHIELDS, like CAMELS, is usable in countries at various stages of regulatory sophistication.

b)A rating methodology should ideally identify relevant risk factors that would be monitored, tracked and analysed against set benchmarks. In SHIELDS, the benchmarks are calibrated on the GLYOR scale, which is an ordinal scale.

c)It was noted that there are two dimensions, which need to be considered regarding the choice of the benchmarks, namely choice of the indicator and the applicable threshold. Regarding the choice of indicator for financial stability, it was noted that the literature recommends that policymakers should have a wide range of indicators at their disposal rather than relying on one single indicator, bearing in mind that each indicator has its own purpose, merits and limitations. Turning to indicator thresholds, it was noted these could be derived from historical levels that signalled systemic stress in the past, or determined on the basis of best practices benchmarks set by international standards setting institutions as well as statistical methodologies. Determination of benchmarks should be established up front to facilitate easy interpretation of the results.

d)The SHIELDS framework employs the following benchmarks:

i)Absolute benchmarks, e.g. CAR; EU Surveillance Framework; The Weighted CAMEL Framework.

ii)Z-scores: Deviations from the mean normalised by the standard deviation;

iii)Quintile Scaleis based on quintiles and involves determination of a relevant range, and the scoring between the best ("1") and worst ("5") are based on a sliding scale;

iv)Percentiles i.e. transform the variables in percentiles, using their sample cumulative distribution function; and

v)Signal Extraction Method whereby the thresholds are set at parameters which minimises the noise to signal ratio of the early warning indicator.

e)He drew the attention of participants to benchmarks that have been proposed by a number of authorities including:

i)The new EU Macroeconomic Imbalances Monitoring Procedure;

ii)USAID& Partners for Financial Stability;

iii)The Weighted CAMELS Framework;

iv)Lindgren, Garcia &Saal (1996);

v)Demirguc-Kunt and Detragiache (1998a);

vi)Laeven& Valencia Data Set 2012;

vii)Reinhart &Rogoff (2009);

viii)Kaminsky& Reinhart (1999) Signal Extraction Model

f)He explained that there are a number of other useful indicators that have been suggested in the literature including Financial Soundness Indicators (FSI) and several other indicators proposed in IMF working papers.

Quantitative Techniques to Assess Systemic Stability – Selected Examples

  1. Under this topic he explained that there are several quantitative techniques used to assess systemic stability, each of them presenting both advantages and limitations, examples of which are:

a)Stress-tests techniques which allow for the identification of potential shocks and estimate the financial system resistance, but not a comparative portrait of the level of stability over time and across nations;

b)Aggregate Financial Stability Index (AFSI), which facilitate comparisons over time and across nations;

c)Market Based Methods including Distance to Default, Merton Models, Contingent Claim Analysis, whose applicability in developing nations are often limited by data availability;

d)He informed the workshop that regarding the early warning systems, the SignalExtraction Approach developed by Kamisky and Reinhart (1999), is noted in the literature to be the most popular and involvesthe following steps:

  1. identifying historical crisis episodes,
  2. selecting leading indicators as predictors of crisis episodes,
  3. setting threshold values of the selected leading indicators,
  4. constructing composite leading indicators, and
  5. watching for red signals and responding appropriately.

e)Construction of an aggregate financial stability index (AFSI) represents, beside the early warning systems and the stress tests, one of the popular quantitative methods for measuring the stability of a financial system. In order to construct this, the following steps need to be followed:

  1. selection of individual indicators,
  2. selection of the method for normalization of the indicators and identification of a weighting method (which relies on the retained criteria and on the established weights, features of the system & data availability), and
  3. aggregation of individual values into the composite (or partial) index.

f)Validation of the econometric relation between the AFSI and a group of macroeconomic variable provides the possibility to perform a forecasting exercise in order to assess the future stability.

Qualitative Approaches to Stability: Example of Expert Opinion Index

  1. Under this topic he explained the following:
  1. Operationalization of this technique involves development of a comprehensive questionnaire covering all the relevant evaluation factors. The questionnaire is administered to experts in the field of systemic risk who are required to rate each attribute on a scale of “1” being the best, and “5” being the worst.
  2. As expected, for the results of this approach to be valid, the requisite sampling caveats such as the size of the sample, sampling methodologies and so on should be statistically valid. The results of this approach are aggregated into a financial stability index.
  3. A brief illustrative extract is as follows:

Financial Stability Assessment Questionnaire

Solvency Conditions / 1 / 2 / 3 / 4 / 5
a. / Insufficientregulatorycapital / 0 / 0 / 0 / 0 / 0
b. / Counterpartyexposures / 0 / 0 / 0 / 0 / 0
c. / Investmentinilliquid assetportfolios / 0 / 0 / 0 / 0 / 0
d. / Over-relianceonleverage / 0 / 0 / 0 / 0 / 0
e. / Lackof countercyclical capitalcushions / 0 / 0 / 0 / 0 / 0
f. / Reliance on ST financing / liquidityrisk / 0 / 0 / 0 / 0 / 0
Health of the Economy / 1 / 2 / 3 / 4 / 5
a. / Unemployment / 0 / 0 / 0 / 0 / 0
b. / Currentaccountdeficit / 0 / 0 / 0 / 0 / 0
c. / Change inConsumerPriceIndex (CPI) / 0 / 0 / 0 / 0 / 0
d. / GDPgrowth / 0 / 0 / 0 / 0 / 0
e. / Ratioofconsumercredittopersonalincome / 0 / 0 / 0 / 0 / 0
f. / Personal incomegrowth / 0 / 0 / 0 / 0 / 0
g. / Housingprices / 0 / 0 / 0 / 0 / 0
h. / Consumerconfidence / 0 / 0 / 0 / 0 / 0
i. / Stock Market Equity values / 0 / 0 / 0 / 0 / 0
Institutional Quality
a. / Regulatory uncertainty/implementation / 0 / 0 / 0 / 0 / 0
b. / Global sovereign risk—“debt crisis” / 0 / 0 / 0 / 0 / 0
c. / Insufficient risk management practices / 0 / 0 / 0 / 0 / 0

Overall SHIELDS Rating

  1. Under this topic Mr. Chirozva highlighted the following:

a)Each component of SHIELDS is assessed / rated using various metrics & qualitative considerations.A similar approach is used in the CAMELS system.Each component rating may be assigned a rating on a scale of “1” to “5” with “1” being the best possible and “5” the worst. A composite financial stability rating (systemic stability) is derived from the seven component ratings.

b)The SHIELDS framework for assessing and maintaining financial stability has inbuilt flexibility in that it does not rely on any particular algebraic formulations, but pragmatic characterisation of the fundamental properties of financial stability. Quantitative models used should inform the policymaker rather than being automatic determinants of the ultimate decision. Accordingly, qualitative analysis, based on informed judgement, still plays an important role in financial stability.

c)A practical exercise was based on the Weighted CAMELS Rating Framework.

d)Prospective users of the draft handbook should bear in mind that it is intended as a general reference and is not intended to replace the need for professional judgement and advice relevant to the particular circumstances.

Discussions

  1. In the discussions that followed the presentation,the delegates appreciated the amount of work done to date. However, they observed that the Manual does not contain practical and clear step-by-stepexamples on how the SHIELD ratings will be determined.
  2. It was therefore agreed that Mr. Chirozva would revise the Manual to make it short on background materials and specially the definitions, and put more emphasis on the descriptions of the scoring and rating methodologies in order to provide practical guidance that is easy to follow.
  3. Regarding the call to have step-by-step guidance similar to EVIES Manual, Mr. Chirozva advised that it may be erroneous to have such expectations, as in prudential matters the most important issues are how to exercise judgement in the decision making process. He said that economic policy matters cannot be reduced to a tick-box type of a manual. He also pointed out that prudential risk management guidelines do not follow a “click “ and “tick” approach but provide parameters or rules regarding how the decisions should be made.

Recommendations

  1. The workshop made the following recommendations:
  1. Mr. Chirozva should revise the SHIELDS rating section of the Manual by providing numerical examples. He should elaborate the steps that should be followed to assign the ratings and undertaking of the SHIELD Rating System Assessment.
  2. The revised manual will be validated via electronic means, following circulation to Central Banks by CMI.
  3. Member countries are encouraged to expedite implementation of the COMESA Framework for Financial Stability Assessment, as well as customise the in-country benchmarks.

Validation of a Manual for Forward Looking Financial Stability Reports and Stress Testing ( Agenda itemv)