| Air Bank a.s. Česká spořitelna, a.s. (Erste Group) Českomoravská záruční a rozvojová banka, a.s. ČEZ, a.s. Hypoteční banka, a.s. (KBC Group) Investiční kapitálová společnost KB, a.s. (Amundi Group) Komerční banka, a.s. (Société Générale Group) Kooperativa pojišťovna, a.s. (Vienna Insurance Group) Moravia IT, a.s. MONETA Money Bank, a.s. Nova KBM d.d. |
OKULA Nýrsko, a.s. OTP Banka Slovensko, a.s. PSJ holding, a.s. RWE Transgas Net, s.r.o. sanofi-aventis, s.r.o. SKB Banka, d.d. (Société Générale Group) Tatra banka, a.s. (Raiffeisen Bank International Group) UniCredit Bank Czech Republic, a.s. Západočeská energetika, a.s. Železničná spoločnosť Slovensko, a.s. |
ARM worked with LeasePlan Česká republika to assess the credit risk management processes. Specifically, ARM focused on assessing the methodologies related to the provision of leases to corporate clients for completeness, accuracy and appropriateness, and compliance with group methodologies.
The outcome of the collaboration was a report containing ARM's views on the following areas:
"The cooperation with ARM has met our expectations - we will use the report to further develop our methodologies and internal procedures. We appreciate that the report includes references to specific parts of the documents to which ARM's comments relate. We also found it very useful to be able to discuss partial conclusions with ARM on an ongoing basis." Michal Kožuch, Risk Director & Local Compliance Officer
ARM cooperated with J&T Bank a.s. in the area of client segmentation for concentration risk assessment.
Specifically, ARM analysed the interconnection of individual segments using internal and external data, i.e. the interconnection within a particular segment and the connection of the selected segments with each other. Based on the analysis, adjustments to the existing segmentation were proposed and discussed with J&T Bank representatives.
The output of the cooperation was a report containing a detailed description of the analyses performed and the final adjustment of the sectoral segmentation, including its justification and selected conclusions resulting from the analyses. The report was accompanied with documents containing a detailed analysis of individual segments and their interconnections not only in the Czech Republic but also in selected European countries.
"The cooperation with Advanced Risk Management met our expectations. Their innovative approach in setting up the sectoral segmentation enabled us to use the available data effectively and thus justify the chosen segmentation in an objective manner." Tomáš Hejduk, Director of Credit Risk Management
ARM has cooperated with Národní rozvojová banka, a.s. in the development of a methodology for determining Forward-Looking Information ("FLI") for purposes of determining loss allowances under IFRS 9.
Within the framework of the cooperation, procedures were developed which allow to take into account various scenarios of expected future development of the macroeconomic situation and to estimate their impact on PD and LGD parameters in the calculation of provisions.
The outcome of the cooperation was:
An updated methodology describing the procedures for determining the FLI parameter, the way it is taken into account in the calculation of provisions, the rules for updating the calculation and the related roles and responsibilities;
a calculation file in MS Excel for the calculation of the FLI parameter;
calculation scripts prepared in the Python programming language for estimating the statistical model, including a user manual.
"Thank you for your cooperation. I particularly appreciate the detailed yet clear format of the output and the responsiveness of ARM to requests arising from the joint discussions." Pavel Fiala, Member of the Board of Directors
ARM cooperated with SKB on the development of statistical rating models for the credit assessment of commercial clients and private individuals. The rating model use behavioural and financial information. The final rating is obtained by aggregating outputs from the behavioural and financial component.
The model development process has the following stages:
The model development process was aligned with the SKB and OTP Group methodology of model development and considered expert opinions provided by both SKB and ARM. The rating models together with detailed documentation was prepared using the R software. All models and their components were presented to the SKB representatives on the executive summary meeting.
"Our cooperation with ARM met all our expectations. The provided outputs are detailed, comprehensive and of a high quality. We greatly appreciate that ARM regularly discussed and confirmed the individual aspects of the model development with us." Dejan Vičič, Head of Risk Portfolio Monitoring
ARM has cooperated with SG Equipment Finance Czech Republic s.r.o. in the development of a behavioural rating model to regularly update the credit rating of clients with a lower level of exposures. The outputs of the model can be used to predict the probability of default at the client level over a one-year horizon.
ARM's cooperation included:
A Python script was also provided to automatically generate detailed documentation in HTML format. During the development process, the assumptions of the statistical procedures used were continuously verified. The model was developed to comply with both the general theoretical requirements and the requirements defined by the group procedures.
"We are very satisfied with the quality of the service delivered by ARM and with the smooth course of our meetings throughout the cooperation. I particularly appreciate the flexible and willing incorporation of our ongoing suggestions and comments." Jan Endrle, IRBA project
In the first phase of the project, ARM validated the calculation of the probability of default (hereinafter “PD”) used for determining provisions under IFRS 9 for small and medium-sized enterprises. ARM assessed the correctness of mathematical and statistical approaches for the calculation of PD, the legitimacy of the accepted assumptions and their suitability for the calculation of PD within the provision determination. At the same time, ARM assessed the accuracy and completeness of the methodology, the quality of the input data for the calculation of the PD estimate and the back-testing of PD estimates in the analysed historical horizon.
The output of the cooperation was a final report containing an overview of findings in the above‑mentioned areas and related recommendations. ARM also provided a technical documentation containing performed analyses and simulations of the impact of selected recommendations.
In the second phase of the project, ARM cooperated with ČMZRB on the implementation of the validation findings. ARM specified the modifications to software used for the calculation of the PD parameter and discussed it in detail with the programmers. ARM provided professional supervision over the implementation of individual recommendations and verified the correctness of their implementation. The verification also included quantification of the impact of the changes made.
"Advanced Risk Management, s.r.o. supported its findings by a detailed analysis of the PD estimate, which included not only verification of the accuracy of the current model, but also simulated the effect of possible modifications and extensions of the current model. The outputs are prepared very comprehensively. ARM further ensured that all findings were properly implemented. We are very satisfied with the work done by ARM." Ing. Pavel Fiala, Deputy Director-General
ARM refined a behavioural model for SKB banka d.d. describing behaviour of the non-maturing deposits of retail and corporate clients, in particular of current and saving accounts. The aim of the model was to capture not only the withdrawal behaviour of clients, but also the reaction of the bank when changing the client interest rates in context of changes of market interest rates.
The outputs of the model may be used for liquidity management as well as for measurement of interest rate risk in the banking book (IRRBB). Main modelled outputs:The model is detailly documented in the methodology that describes not only the individual parts of the model and all applied methods, but also all related processes (e.g. model operation and recalibration).
ARM delivered a model prepared in the programming language Python that automatically generates full documentation in HTML. The model also exports relevant results into MS Excel file. Another MS Excel file serves as source of input data and parameters of the model. The usage of the tool, its maintenance and structure of applied classes and functions in Python are described in the user manual.
"Solution developed by ARM was efficient, consistent with the previous model, robust to changing market conditions as well as user friendly. Thank you for all your support and guidance.“ Nataša Mohorčič-Zobec, CFO
ARM has participated in the development of the non-retail rating model for SKB. The model was designed for SME as well as standard corporate exposures. The model structure included financial and non-financial parts.
First, ARM performed validation of the financial model that included evaluation of logical and factual correctness of the model, assumptions of the model, approach to model calibration etc.
Further, ARM developed the non-financial rating model and rules for the aggregation of financial and non-financial models to produce the overall rating assessment of corporate and SME clients.
The non-financial rating model was developed based on data from questionnaires, behavioural data and financial trend data. The model was developed in the R software. ARM prepared the model formula together with complete model documentation. The documentation was prepared in the form of both detailed technical version and executive summary.
“I would like to thank ARM for the cooperation in non-retail rating model development. The cooperation met all our requirements, as all outputs were delivered in time and in the required quality despite the tight deadlines.” Dejan Vičič, Head of Risk Portfolio Monitoring
ARM cooperated with Poštová banka, a.s., in the development of activities and processes in the area of risk modelling. The cooperation consisted mainly in the preparation of a framework for the developmental methodology of credit models and the development of a template scoring model in one selected retail clients‘ segment.
The methodology of credit models covers mainly:
Along with the methodology, scripts in the Python language were also prepared, which enable the calculations and estimations described in the methodology to be performed. The scripts also enable automated preparation of the analytical and technical documentation of the developed model.
"The delivered outputs (methodology and the scripts) will be very helpful to my team in the development of other credit models. Cooperation with ARM was again at a high level of professionalism." Director of Retail Risk Management
ARM updated for Pražská energetika, a.s., (hereinafter “PRE“) the methodology for managing the credit risk of B2B customers. ARM updated the methodology in the section focused on internal models for credit risk management, especially in the areas of:
At the same time, based on the updated model for predicting the clients´payment behaviour, ARM revised the algorithm used to calculate the expected number of days past due.
"Cooperation with ARM in updating the methodology for credit risk management met our expectations. ARM prepared a proposal, which we discussed together. Based on the proposal and discussions, ARM designed the final methodology. We appreciated the knowledge and experience of ARM experts with credit risk management and internal models. " Jan Jeřábek, Head of Sales Support PRE
ARM performed expert verification of retail scoring models used in Poštová banka, a.s.
The verification covered both statistical models as well as models based on expertly set rules, and covered in particular the following areas:
The outcome of the cooperation was a report describing the findings and formulating ARM’s recommendations. The findings and recommendations were classified into several categories in terms of their importance and impact on the overall functioning of the bank. The report also set out the basis for verifying the models. The overall evaluation of the verified models was presented in the managerial summary.
"ARM's recommendations were clear, comprehensible and well-structured and they helped us identify priority areas for further development of scoring models." Director of Retail Risk Management
ARM has developed a rating model for the Ezpada s.r.o. company to assess the credit risk of large corporate counterparties. The rating model, based on the input data on financial and non-financial characteristics, assigns the counterparty an internal rating, which represents the level of counterparty credit risk and the probability of its default in one-year term. For the purpose of developing the rating model, ARM prepared a list of financial and non-financial indicators that were tested in terms of their ability to differentiate the level of credit risk of each counterparty. The most appropriate indicators were included in the model. ARM also created an internal rating scale and calibrated the developed model on this scale.
"Cooperation with ARM met our expectations. The rating model provides a robust tool for managing counterparty credit risk.", Michal Papež, Head of Risk and Compliance
ARM together with the internal audit department of Česká spořitelna, a.s. (hereafter „ČS“), participated in audit reviews of validation protocols on rating models and risk parameters PD, LGD a CCF. The main aim of the cooperation was to assess the compliance of the validation protocols of ČS and Stavební spořitelna České spořitelny, a.s., with the rules defined in the relevant group methodologies. ARM assessed the structure of the validation protocols as well as the way how the individual validation methods were assessed and the accuracy of the evaluation. In its review, ARM especially focused on the validation of the following aspects:
ARM prepared a final report summarizing all audit findings. ARM classified the severity of each finding reflecting its impact on the validation results and also provided recommendations on how to remove deficiencies.
ARM has performed a Gap analysis for SG Equipment Finance Czech Republic, s.r.o., (hereinafter SGEF) within the project to implement the Advanced IRB Approach. The Gap analysis assessed the current state of SGEF with regard to current and upcoming regulatory requirements laid down in CRD, CRR, related technical standards and guidelines and proposed a set of steps to close the identified gaps. The assessment was performed in the following areas:
Based on statistical analysis of anonymized external data and data of UniCredit Bank Czech Republic and Slovakia, a.s. (hereinafter referred to as “UCB”) concerning particular contractual parameters of clients, ARM has created a prescoring model for assessment of riskiness of clients within the process of providing pre-approved consumer loans. ARM created the model based on logistic regression. In order to achieve an optimal model quality, the model was developed on a development sample and its predictive ability was back‑tested on a validation sample. A part of the development of the model was its calibration to a target central tendency and additional expert modification of the initial model was performed after the consultation with UCB. The output of the cooperation was a protocol in English language which contained information about development of the model and its specification. At the same time, the results of the model were presented not only for the development and validation sample, but also for actual sample of clients. Technical instructions for implementation of the model were an integral part of the protocol.
“Cooperation with Advanced Risk Management, s.r.o. fulfilled our expectations, work on the project and its output was realized in accordance with the requirements, on time and in a good quality. I especially appreciate the ARM’s initiative in quality control and processing of external data used in development.” Manager of Rating and Scoring Models
ARM expertly assessed the scoring model that Moravský Peněžní Ústav – spořitelní družstvo (hereinafter “MPÚ”) introduced for the purpose of pre-approval of factoring credit lines for individual clients. The cooperation involved in particular the following areas:
ARM assessed processes and methodologies of UniCredit Leasing CZ, a.s., (hereinafter UCL) in terms of completeness, internal consistency and suitability for A-IRB approach. In addition, ARM evaluated compliance of internal processes and methodologies with regulatory requirements (CRD IV, CRR, Decree No. 163/2014 Coll. and EBA Guidelines for IRB, especially for default definition and for estimation of risk parameters). The cooperation involved following areas:
ARM assessed the methodology of UniCredit Leasing CZ, a.s., (hereinafter “UCL”) for client segmentation under the IRB approach. Additionally, ARM organized a workshop in UCL to present its findings and recommendations. The recommendations took into account the requirements of Regulation (EU) No 575/2013 of the European Parliament and of the Council of 26 June 2013 on prudential requirements for credit institutions and investment firms.
“We were very satisfied with ARM’s approach. They were willing to split the workshop into 2 subsequent parts, it helped us to better capture the identified shortcomings.” Head of Strategic Risk and Portfolio Management
ARM verified validation process of credit risk models. Subject of the audit were both validation protocols of each scorecard and validation protocols of risk parameters. Within the project ARM verified the following:
The goal of the cooperation between ARM and NKBM was the assessment of accuracy, consistency and correctness of NKBM’s revised methodologies on default definition and IRB segmentation. During the cooperation ARM verified, whether the rules specified in the internal methodologies on default definition and IRB segmentation are accurate and in line with generally accepted principles. Furthermore, ARM checked the compliance of internal methodologies with relevant regulation (CRD IV, CRR and relevant EBA Guidelines).
ARM cooperated with ČS’s internal audit on verification of risk margins calculation included into loan prices. During the realization of the cooperation, ARM focused especially on the following:
For the company LeasePlan Česká republika, s.r.o., ARM has created an expert score card for evaluation of operative leasing requests of natural persons – entrepreneurs. Part of the cooperation was identification of entry data necessary to be included in the operative leasing request. The entry data were separated between those inputted by client when creating the request in web interface, those automatically acquired from the database of Non-Banking Client Information Register (NRKI). The output of the contract was an expertly set score card reflecting risk appetite of LeasePlan company, detailed description of individual entry data, algorithm of score calculation and methodology for acquiring and evaluation of the information from NRKI.
The subject of cooperation between ARM and MPÚ was creation of new methodology for calculation of risk margin in the process of loan product pricing. Furthermore, ARM prepared for MPÚ a pilot calculation of risk margin in given client segment (based on prepared and mutually approved methodology) which is subsequently used in MPÚ as a basis for calculation of risk margin also in other segments. During realization of the contract, ARM used not only available historical data of MPÚ, but also own expert assessment of selected parameters necessary for calculation of client’s risk margin. Subsequent cooperation covered expert assessment of complete methodology of loan product pricing.
ARM prepared an analysis and independent assessment of the current operational risk management system for Exportní garanční a pojišťovací společnost, a.s., (EGAP). As part of the analysis, ARM prepared draft recommendations for the further development of methods and procedures used in operational risk management, for example in the area of key risk indicators, scenario analysis and reporting. The consultations also included the preparation of a draft procedure and recommendations for the acquisition of a new software tool for managing operational risk.
The output report describes the results of the analysis and details all the recommendations.
"Cooperation with ARM on the analysis of the operational risk management system helped us to formulate the main priorities for the development of managing this risk in the next period. We greatly appreciate ARM's experience with operational risk management. The prepared Output Report is a very good guide for us on how to proceed in the development of our internal system ", Silvie Goldscheiderová, Executive Director of Risk Management Division and Director of Risk Management Section.
ARM has participated in the transition of Equa Bank, a.s., from the Basic indicator approach (BIA) to the Standardized approach (TSA) under the CRD IV framework for operational risk. As a starting point of this project, ARM organised an introductory workshop which concentrated on the TSA approach, the methodology for the calculation of capital requirements, and the qualitative requirements on processes related to the calculation. In addition, ARM, together with the experts from Equa Bank and based on its accounting methodology, developed the methodology for the calculation of capital requirements by using the TSA approach and complying with CRD IV and related legislation. The methodology included the procedure for the assignment of bank’s activities to business lines.
“Advanced Risk Management, s.r.o., has helped us to prepare the methodology for the calculation of capital requirements exactly according to our agreement. Thank you for your help. We look forward to cooperation in the future.” Head of Enterprise Risk Department
ARM conducted an audit of the operational risk processes in J & T BANKA, a.s.. Based on the revision of methodologies in the area of operational risk, ARM assessed compliance with regulation requirements and with common practice. The output of the contract was an output report which contained findings and suggested recommendations.
"The output of the audit met our expectations and helped us to identify areas for further development of operational risk management.” Head of risk management, J & T BANK, a.s.
ARM performed revision of selected processes in the area of operational risk management of Moneta Money Bank (before GE Money Bank, a.s., "GEMB"). The cooperation was focused especially on the RCSA process and risk analyses realized in other areas beyond this process. ARM provided GEMB with know-how when proposing individual rules for the process of determination and assessment of GEMB’s risk appetite for operational risk. Individual principles and rules were designed to comply with other areas where the risk appetite is determined.
Primary the subject of the cooperation between ARM and GEMB was to:
ARM has implemented a project in GEMB for development of operational risk management regarding creation of gap analysis identifying compliance of internal methodology with requirements of currently valid CNB Decree No. 163/2014. The output of this part of contract was a description of individual gaps, including suggested adjustments in operational risk management system which would lead to their removal.
Furthermore, ARM has prepared for GEMB a workshop regarding possible ways of stress testing of operational risk and subsequently formulated principles of stress testing for internal GEMB directive.
In the final part of the contract, ARM prepared materials for regular training of GEMB employees in the area of identification of operational risk events. The goal of the training is to increase the awareness about operational risk in GEMB and provide concise instructions for identification and recording of operational risk events.
ARM provided Stavební spořitelna České spořitelna, a.s. with consultancy services in the development of behavioural models for the purpose of interest rate risk in the banking book (IRRBB).
Specifically, ARM provided advice on specific parts of the behavioural modelling of deposits, in particular on possible extensions to the original model in the area of segmentation, inclusion of expert insight and sensitivity analysis of market interest rates.
In the second phase, ARM assisted with the revision of the documentation describing the behavioural modelling of deposits, loans and other items in the SSČS portfolio for the purpose of measuring IRRBB, in particular:
"We were satisfied with the course of the cooperation and appreciate the smooth flow and quality of cooperation in updating the documentation." Miloš Lisý, Director of Risk Management
The main objectives of the assessment were:
Based on this assessment, ARM proposed regular monitoring procedures and controls to ensure better quality of input data and to continuously verify the suitability of the model for possible changes in the portfolio structure. During the cooperation, ARM also designed its own model for valuing the portfolio of investment wines, which is less sensitive to the structure of the portfolio than the assessed model.
“I would like to thank ARM for the cooperation. The performed analysis meets all our requirements, and it also includes additional recommendations for model operation and designing an alternative model.” Custody and Depository Team Manager, Corporate Banking, Česká spořitelna, a.s.
Subject of the cooperation between ARM and Erste Asset Management was proposal of approach for measurement of currency risk of investment fund. ARM proposed several possible approaches to the issue corresponding to client needs. Further ARM elaborates each approach including a summary of their advantages and disadvantages and a practical example. The methodology included also proposal of assessment of the volume of investment fund’s currency risk.
ARM provided methodological support to ČEZ company in the area of regular revision and update of model EBITDA@Risk, which is used for calculation of impact size of changes of commodities market prices on the economic result of the ČEZ company. The methodological support was mainly provided in following areas:
Additional area of cooperation was providing consultation to the calculation of Value at Risk for credit risk of retail and corporate exposures.
ARM provided the company ČEZ with methodological support in the area of revision and updating of the model EBITDA@Risk, which is used for calculating the size of the impact of changes in market prices of commodities on the economic results of the ČEZ. The methodological support was mainly in the areas of:
"In cooperation with ARM, we are systematically improving the risk management system, both in the area of risk measurement and management. Mutual discussions on individual areas of risk management and ARM's experience help us achieve our plans." Ing. Zdeněk Stejskal, Head of the HSE Department
ČEPS, a.s. (ČEPS) has been working with ARM to update and strengthen its risk management system.
ARM, in cooperation with CEPS, proposed the main elements of the risk management system:
"ARM's lead consultant, Mr. Radek Křička, presented the main elements of the CEPS's risk management system efficiently and also prepared a draft internal methodology. Outputs processed by ARM have fully met our expectations. Discussions regarding the various elements of the risk management system and detailed solutions to specific procedural steps were very inspiring and helpful in terms of further developments of the risk management system." Head of Internal Audit
ARM has provided a series of consultations on the methodology development regarding interest rate risk arising from the banking book (hereafter “IRRBB”) to J&T Bank. Within these consultations, ARM assessed drafts of the updated internal methodology on IRRBB from the perspective of completeness and appropriateness of applied approaches, and their compliance with the respective regulation. ARM assessed the compliance of the internal methodology with the current regulation, and also its compliance with the draft regulation stipulated especially in the EBA document EBA/CP/2017/19 and in the proposed amendments to CRD and CRR. ARM’s findings and related suggestions on how to adjust the draft methodology were discussed in several workshops together with additional questions raised by J&T Bank.
“The Consultations provided by Advanced Risk Management, s.r.o., helped us to further develop methodologies on the management of interest rate risk arising from the banking book. The consultations were in particular beneficial in terms of suggestions regarding some challenging issues” Director of risk management.
The scope of the cooperation between ARM and Stavební Spořitelna České Spořitelny, a.s., (hereinafter „SSČS“) has been engaged in the assessment of risk appetite and related documents. ARM reviewed internal methodologies with respect to SSČS’s business activities as well as with respect to related regulatory requirements. A dedicated workshop has been organised to discuss both quantitative and qualitative internal limits established by SSČS within its risk appetite framework. The output of the cooperation was personal presentation of individual recommendations.
“We have been satisfied with cooperation and the outputs was of expected high quality.“ Director of Risk Management
ARM cooperated with Moravský Peněžní Ústav (hereinafter "MPU") on the assessment of the implementation of the CNB's corrective measures for the Internal Capital Adequacy Assessment Process (hereinafter "ICAAP"). ARM assessed the ICAAP methodology from the perspective of implementation of specific measures, as well as in a wider context related to the particular measures.
The output of the cooperation was the Final Report in Czech language, which contained ARM’s findings and recommendations related to the verified areas.
The scope of cooperation between ARM and Air Bank providing of project management and expert opinion in the following areas included:
In cooperation with the bank, ARM proposed ICAAP revision and prepared related internal regulation which contained description of the whole ICAAP concept, including the following:
While creating regulation, ARM also took into consideration the consolidated view on the group and the requirements of the parent company.