Causes of Results Achieved

+Regarding the internal RR assessment system, the bank has started collecting loss data since May 2007 on the internal system. In case the RR event involves multiple business units, the bank will determine which process step the event occurred in and which unit is responsible for that process step. At the same time, RRHD loss event data is collected mainly from mass media and other official channels. RCSA is deployed across all business units at the head office and branches.

+ For measuring and calculating capital for operating risks, VietinBank is using the Basic Factor Method (BIA) as prescribed in Circular 13, aiming to use the Standardized Method (SA) to calculate capital for operating risks. The Bank is able to classify (map) its current business areas and activities into 8 business areas of Basel II based on the roles, responsibilities and business areas assigned to each business unit.

- Regarding planning and management to ensure adequate capital, VietinBank has allocated capital to business units for RRTD. The capital allocation method is based on the Top down and Bottom up combination based on the feasible capital supply plan, growth plan, business plan, and RR control orientation.

- Regarding endurance testing, VietinBank has conducted endurance testing in measuring RR types according to the Scenario Analysis and Sensitivity Analysis methods, meeting the requirements in Circular 13 and Circular 41.

- About RR synthesis, economic capital calculation

Maybe you are interested!

+ The Bank has designed the process: determining the scope of units for calculating required equity capital; forecasting financial indicators, determining target capital; approving equity capital plan. The Finance Block, Risk Management Block, and business blocks are assigned responsibilities in RR synthesis. For process auditing, the Internal Audit Department independently reviews and reports to the Board of Supervisors.

+ The calculation and synthesis of economic capital is currently being carried out according to the regulations and instructions in Circular 13, similar to the simple aggregate method (basic method) in the synthesis of economic capital of Basel II.

- On capital use according to RR principle

+ Applying RR-adjusted profit assessment according to RAROC into banking business activities, VietinBank is gradually approaching the method of assessing customer and portfolio profits based on RR. Currently, RORA has been included as a criterion for evaluating portfolio performance and planning.

+ The credit portfolio RR limit is determined based on the analysis of the RR characteristics of the sub-portfolios and the overall bank portfolio including RR capital. The process of setting the RR limit is based on the portfolio optimization method that is being piloted. The results are for reference only.

- Regarding reporting, the bank has conducted periodic capital adequacy assessment reports. In addition, the bank has conducted an independent review of all capital management regulations, policies and procedures.

2.3.1.1. Reasons for the results achieved

VietinBank has the necessary conditions to implement ICAAP, specifically:

Regarding business strategy and RR strategy, VietinBank has built a clear and specific strategy. The implementation of ICAAP has received the attention of the senior management and most of the relevant departments and divisions.

Regarding the risk management system to manage the bank's key risks, VietinBank builds a risk management structure in accordance with international practices in general (according to the three-line control principle of Basel II) and ICAAP requirements in particular.

The RR policy framework has been officially issued and plays a guiding role in the risk management processes (risk identification, measurement, control, and reporting) in the Bank's functional units.

Regarding human resources, VietinBank has staff with experience in operating and managing capital safety according to current market practices and has the ability to continue to develop in implementing ICAAP.

Regarding data technology, VietinBank has deployed a number of systems to analyze big data related to capital and manage data in accordance with the requirements of the State Bank and Basel.

II. At the same time, this data system is also used to perform reporting work. Regarding KRI, the bank has developed a KVRR Statement issued by the Board of Directors, including

16 indicators and thresholds classified by key RR types as well as overall factors.

Regarding communication work, the bank has implemented it methodically and fully, spreading the role, meaning as well as the ICAAP implementation process not only to all staff in the bank but also focusing on external communication.

2.3.2. Limitations

a. On identifying, assessing key risks and determining risk appetite

- In the work of identifying and assessing key risks, VietinBank lacks an information system and analytical techniques that allow the Board of Directors to measure potential risks in all on- and off-balance sheet activities.

- Regarding customer and transaction RR assessment according to the internal rating system, VietinBank has not yet implemented specific definitions, processes and assessment standards. This has caused limitations in assessing material RR at the bank. In particular, the bank has not yet completed the system to verify the accuracy and consistency of the system, rating process, and estimates of all related RR factors.

b. About risk measurement method

(i) Credit risk

- Regarding the development and implementation of RRTD measurement models, VietinBank has completed the construction of PD, LGD, EAD estimation models, ensuring the necessary conditions to comply with the baseline approach (FIRB). However, the data to develop the LGD, EAD models needs more time to accumulate to meet the requirements of the advanced approach (AIRB).

- RR measurement system

Banks must have strict and effective processes to ensure compliance with the system of policies, controls and internal procedures related to the operation of the RR measurement system (BCBS, 2006). In fact, at VietinBank, the estimates are newly formed and need to be further tested and improved, along with extensive application in banking activities; There are no written instructions for each model and no standardization of management on model governance testing according to Basel II requirements; The distinction between collateral and valid guarantees according to Basel requirements and invalid collateral and guarantees is unclear.

- Approach

For many types of assets, the Commission has proposed two main approaches: the basic approach and the advanced approach (BCBS, 2006).

With the nature of being built according to the expert opinion method, the set of indicators has disadvantages compared to the RR PD/LGD/EAD measurement tools. The definition for each class has been documented, but the RR level in the scale of each customer group is not similar, the RR level has not been specifically quantified, leading to the process of planning credit policies using the class being subjective; The capital calculation according to the FIRB method has not been applied; The model according to the AIRB statistical method has not been deployed, so VietinBank needs to continue to accumulate more data to develop the LGD model to meet the requirements of the advanced method (AIRB).

- Implementation conditions

Provided that certain minimum requirements and disclosure requirements are met, banks eligible to apply IRB can rely on their internal estimates of the RR components to determine the minimum capital level for a given credit exposure (BCBS, 2006).

The data used to build the model is limited in length and sample size. At the same time, VietinBank has just started applying FIRB and has not yet conducted independent model testing.

(ii) Market risk

Some limitations related to RRTT identified by experts include:

- Currently, the bank has an independent price verification process, in which market prices are provided by the business unit and are verified in order to improve the efficiency of independent price verification in valuation. Independent price verification is being carried out by comparing prices provided by the business unit with a number of data sources. Market price valuation is carried out on a daily basis. However, the survey results further indicate that detailed standards for market data verification are not documented.

- The limitation related to perfecting the method and pricing model in the trading book mentioned by many surveyors is that the prices of foreign exchange forward contracts, foreign exchange swap contracts and bonds cannot be collected directly from the market.

Therefore, these instruments are priced using a marked-to-model on a daily basis. Due to some data issues, the bond yield curve without coupon yield needs to be reviewed before it can be officially used to price bonds and measure risk. Therefore, the risk measurement method at VietinBank is at the level of meeting the regulations in Circular 41 of the State Bank of Vietnam (standard method of Basel II), not meeting the capital calculation standards according to the internal model method of Basel II.

- The measurement of equity securities risk, commodity risk and option risk is currently under research for implementation at VietinBank.

- For market risk, banks need to have a method and model for pricing market risk in all situations and times (BCBS, 2006). The implementation at VietinBank shows that specific instructions or detailed standards for validating market data have not been unified and documented; due to some data issues, the yield curve of bonds without coupons needs to be reviewed before being officially used to price bonds and measure risk; the impacts or disadvantages of pricing according to the model (for forward contracts, foreign exchange swaps and bonds) have not been officially documented to help stakeholders in the bank clearly understand the impacts or disadvantages of the model. Currently, market prices (exchange rates, interest rates, gold prices) used in pricing have not been adjusted according to the level of market illiquidity or the holding period of positions. Stress VaR is under development, as the risks arising during the stress period as required by Basel have not been reflected or quantified; equity risk, commodity risk and option risk measurement have not been carried out. Additional tests for model validation have not been formalized and put into operation.

(iii) Operational risks

Limitations related to RRHĐ according to experts include:

- The method of measuring operating risk at VietinBank mainly focuses on arithmetic calculations of capital required for operating risk, and has not yet developed a component for calculating operating risk capital on the RWA calculation system (along with RRTD).

- Management and allocation of RRHĐ capital according to international practices is being studied.

research, not yet implemented in practice at VietinBank.

- Regarding the management of RRHĐ events and loss data collection, the bank has started collecting loss data since 2007 on the internal system. However, the quality and quantity of accumulated loss data so far is not much.

- Regarding the management of RRH events and the collection of loss data, VietinBank is well aware of the need to compare the losses recorded in the GL account with the collected loss data recorded in the internal loss database. However, the loss recording is currently involving many units, so independence is not guaranteed.

- For RRH event management and loss data collection, in case the RR event involves multiple business units, the bank will determine which process step the event occurs in and which unit is responsible for that process step. In particular, specific criteria for allocating loss data from an event occurring in a centralized department or from an activity involving multiple business units, as well as from related events in a period are being researched for future application.

- In operational risk management and loss data collection, operational risk event data is mainly collected from mass media and official channels. Meanwhile, the requirement under Basel II is that external loss event information is recorded and tracked as much as possible. A systematic process to identify situations where external data is required and a method for integrating the data needs to be developed (AMA).

- To improve the effectiveness of the RCSA RR control self-assessment process, RCSA is deployed across all business units at headquarters and branches. Business units have not yet prioritized RCSA.

- Regarding the application of KRI in assessing the status of business activities, the bank's ability to generate RR and KPI, VietinBank builds KRI at 3 levels (whole bank, department block and branch level). Deploying many business systems on a large scale makes it difficult to synthesize data. There is a lack of personnel capable of establishing effective KRI.

- Internal RR assessment system

The bank's internal risk assessment system must include systematic monitoring of risk data, including major losses by business segment. The risk assessment system must be closely linked to the bank's overall risk management process (BCBS, 2006).

At the bank, the quality and quantity of cumulative loss data to date are still below expectations; loss recognition is multi-unit, so independence is not guaranteed; there are no specific criteria for allocating loss data from an event. A systematic process for identifying situations where external data is required and a method for integrating data has not been developed (AMA). Business units have not yet clearly recognized the importance and priority of RCSA.

- AMA method

The Basel Committee recognizes that the rationality and comprehensiveness criteria in the AMA methodology allow a high degree of flexibility for banks to develop an operational risk measurement and management system. However, in developing these systems, banks must have and maintain rigorous processes for the development of operational risk models and the independent approval of those models (BCBS, 2006).

Currently, VietinBank has not developed more advanced capital calculation methods than SA and AMA. The capital required for RRHD is only calculated numerically, but how to use this number in actual business operations to maximize efficiency has not been implemented.

(iv) Liquidity risk

In perfecting the cash flow forecasting model, the bank is currently applying the cash flow model to measure RRTK. The model can measure and forecast the cash flows of the bank's assets, liabilities and derivatives under normal conditions and under some stressed conditions. The limitation that experts have identified here is that the bank has not considered factors such as sensitivity to events, activities and strategies that can significantly reduce the ability to mobilize internal cash in RRTK management. Accordingly, the bank has not implemented hypothetical situations, reviewed corresponding decisions as well as handled crises.

Regarding the measurement and estimation of capital for RRTK, experts commented that the bank is still researching and implementing it.

For RRTK, this is a vital factor for the survival of every bank. The capital position of a bank can have an impact on the liquidity position of the bank, especially during times of crisis (BCBS, 2006). The implementation situation at VietinBank shows that the bank has not considered factors such as sensitivity to events, activities, and strategies that can significantly reduce the ability to mobilize capital in RRTK management.

(v) Interest rate risk

- In order for the revaluation gap model to be implemented and to address specific elements of the banking book, the bank has a ∆EVE/capital model based on Basel II requirements and an IRRBB coefficient according to the agreed limit with the strategic partner. However, most respondents said that the bank has not determined whether the level of detail and complexity of these models meets the requirements.

- For RRLS in the banking book, the RR measurement process needs to take into account all the bank's material interest rate positions and take into account all maturities and repricings (BCBS, 2006). The implementation practice at the bank shows that it is not yet determined whether the level of detail and complexity of the repricing gap model satisfies the requirements. The beta sensitivity coefficient of different products to interest rate fluctuations may also not have been studied accurately. These are RR model issues. The current method of predicting customer behavior for early withdrawal, renewal, and prepayment is based on statistical models that are difficult to fully capture the factors to explain customer behavior.

(vi) Other risks

The bank does not have a process for managing reputational and strategic risks as well as other material risks as stated in pillar 2 of Basel II.

c. Risk appetite and key risk indicators (KRI)

At the same time, the current KVRR does not fully reflect the relationship between RR, capital and income. The Bank has not yet completed the Stress Testing Tools and Process to

Comment