Market Risk Management Framework July 28, 2012
Views or opinions in this presentation are solely those of the presenter and do not necessarily represent those of ICICI Bank Limited 2
Introduction Agenda VaR methodology Stress testing Group structure and framework 3
Market Risk: Definition Market risk is the uncertainty resulting from changes in market prices. Market Risks arises due to movements in Interest rate Exchange rate Equity prices Commodity prices Volatility Directional risks from taking a net long/short position in a given asset class 4
Market Risk: Example 1 Position : Bank bought a 7% interest paying Government bond with 1 year maturity trading at par (i.e. current yield =7%) Underlying risk factor : interest rate/yield There is a risk of increase in yield as it would lead to loss in the portfolio Valuation date 27-Jul-12 Maturity date 27-Jul-13 Yield 6% 7% 8% Price 100.94 100.0 99.07 5
Market Risk: Example 2 Position : Bank has 100 USD position ( equivalent Rs 5,600 assuming current USD/INR = 56.00) Underlying risk factor : USD/INR exchange rate There is a risk of Rupee depreciation as it would lead to loss in the portfolio Exchange r at e INR posit ion 50 5000 56 5600 62 6200 6
Market Risk: Example 3 Position : 100 equity shares of ABC company @50 (Initial portfolio value : Rs 5,000) Underlying risk factor : Movement in equity price of ABC There is a risk of decrease in ABC share price as it would lead to loss in the portfolio value Shar e Pr ice Por t f olio value 40 4000 50 5000 60 6000 7
Introduction Agenda Limits framework and VaR methodology Stress testing Group structure and framework 8
Risk measures Market Risk Value-at-risk (VaR) limit / Mark-to-market (MTM) sensitivity trigger Daily stop loss limit Cumulative loss limit / Cumulative loss trigger Greeks Notional size of investment Net Overnight Open Position Aggregate Capital Market Exposure Liquidity Gaps Liquidity Risk Aggregate/Individual Gaps Liquidity Ratios CS01 Credit Risk Single Issuer exposure Tenor-wise exposure Scrip-wise exposure Rating-wise exposure Interest Rate Risk in Banking Book Interest Rate Gaps DoE EaR Basis Risk 9
Value-at-risk (VaR) Value-at-Risk is defined as a loss level that will not be exceeded at some specified confidence level for a specified time horizon under normal market condition What loss level is such that we are X% confident it will not be exceeded in N business days? If our VaR for one day horizon and at a probability level of 95% is 10 million Euros, that means: There is a 95% likelihood that over the next 24 hours we will lose less than 10,000,000 Euros The likelihood that our losses will exceed 10,000,000 Euros over the next 24 hours is 5% 10
Main methods of computing VaR Analytic VaR Historical Simulation Monte Carlo Simulations 11
Analytical Method Assumptions about the probability distributions of return on the market variables are made e.g. Assuming the returns on risk factors are normally distributed. Volatility of each risk factor is extracted from the historical observation period. Historical data on investment returns is therefore required. The potential effect of each component of the portfolio on the overall portfolio value is analyzed 12
VaR for normal distribution 13
Advantages and Limitations Advantages Analytical VaR Relatively low computational Intensity Easy to explain Ease of carrying out root cause analysis Limitations Assumption of normal distribution Limitation with products having non-linear risks e.g. options 14
Analytical VaR Framework Assuming that portfolio returns are normally distributed: VaR z * V* σ t z: 1.65 (95%) or 2.33 (99%) V : Position σ t: : Standard Deviation( Volatility) t = Holding period (σ t: = σ 1d *sqrt(t)) Daily data: assume that expected return μ = 0 15
Example We have a position worth Rs 20 million in company ABC shares The volatility of Stock ABC is 1% per day We use holding period (t)=10 and 99%ile Confidence Interval, Hence z = 2.33 The standard deviation = 1% *20,000,000 = Rs 200,000 per day The standard deviation of the change in 10 days is: 200, 000 10 = $632, 456 16
Example Contd. Assuming that the change in the value of the portfolio is normally distributed Since z 99% =2.33, the VaR is 2. 33 632, 456 = $1, 473, 621 17
Historical Simulation Calculation of VaR based on simulated changes in the value of the portfolio The simulation is carried out using historical data of market prices 18
Historical Simulation Create a database of the daily movements in all market variables. The first simulation trial assumes that the percentage changes in all market variables are as on the first day The second simulation trial assumes that the percentage changes in all market variables are as on the second day and so on 19
Advantages and Limitations - Historical Simulation Advantages No need to assume normality No need to forecast volatility and correlation Effective in measuring non-linear risks Based entirely on historical data, objective Limitations Computationally more intensive vs. variance-covariance Implied volatility and correlation 20
Monte-Carlo Simulation Calculation of VaR based on distribution of the simulated changes in the value of the current portfolio under randomly generated market scenarios based on estimated distribution for various parameters 21
Advantages and Limitations - Advantages Monte-Carlo simulation No need to assume normality Forward looking Less dependency on historical data vs. historical simulation Limitations Computationally most intensive May require subjective assumptions and hence issues with respect to acceptability of results 22
Back-Testing Tests how well VaR estimates would have performed in the past We could ask the question: How often was the loss greater than the 99%/10 day VaR? 23
Backtesting methodologies Backtesting may be based on: Hypothetical P&L The portfolio as on today is held constant for the next day and revalued using the next day s rates The difference in the actual portfolio value and revalued portfolio value is the backtesting P&L Actual P&L The difference between the portfolio values on two subsequent days is the backtesting P&L 24
Internal models approach - Background RBI circulated draft guidelines on migration to internal models approach (IMA) in December 2009 Banks advised to take a decision on migration to IMA with the approval of their Board Capital to be maintained at close of each business day and to be denominated in Indian Rupees Capital based on normal and stressed VaR RBI issued final guidelines on migration to IMA in April 2010 25
Internal models approach Capital requirement under IMA will be sum of: Value-at-Risk (VaR) based general market risk charge comprising: Normal VaR measure and Stressed VaR measure Specific risk charge computed as per standardised measurement method (SMM) Portfolio to be migrated to IMA: Investment portfolio in HFT book All derivatives Available for sale (AFS) portfolio to continue on existing SMM approach 26
General market risk capital under IMA General market risk capital required (c) = max {VaR t-1 ; m c * VaR avg } + max{svar t-1 ; m s * svar avg } Normal VaR measure to be the higher of: Previous day s VaR (VaR t-1 ) Average of the daily VaR (VaR avg ) on each of the preceding sixty business days, multiplied by a multiplication factor (m c ) Stressed VaR measure (svar) to be the higher of: Latest available stressed-var number (svar t-1 ) Average of the stressed VaR numbers over the preceding sixty business days (svar avg ), multiplied by the multiplication factor (m s ) 27 Floor for m c and m s set at 3
Introduction Agenda Limits framework and VaR methodology Stress testing Group structure and framework 28
Stress Testing Framework A diagnostic tool for assessment of impact of extreme but plausible scenarios on the portfolio Purpose To understand the risk of the portfolio Adequacy of internal capital Historical Setting limits Asian Financial Crisis October 1987 S&P Index crash World Trade Center Terror attack Standard 1% parallel movement in yield curve 25% change in Exchange rate volatility 9% change in Exchange rate
Stress Testing Framework Identify the risk factors Value the portfolio for the day Apply the scenario changes on the risk factors Revalue the portfolio Change in value of the portfolio is the stress loss
Introduction Agenda Limits framework and VaR methodology Stress testing Group structure and framework 31
Risk Management: Governance framework Committees of the Board Board of Directors Credit Committee Risk Committee Audit Committee 32
Market risk management framework Overseeing Authorities Risk Committee of the Board Asset-Liability Management Committee Other Committees Policies ALM Policy Investment Policy Business units Balance Sheet Management Group Proprietary Trading Clients desk 33
A Trade Life Cycle Trade execution and deal entry Deal validation/compliance with internal, regulatory norms Confirmation with counterparty Settlement of cash flows Accounting of cash flows and MTM Valuation of outstanding deals Reporting of P&L to senior management Risk position and limits monitoring
Role of Risk and Middle Office Teams Market Risk Middle Office Operations Process Formulation of policies, limits structure Valuation methodologies Risk assessment methodologies (VaR, stress testing) Reporting of key risk indicators to Risk Committee Daily and weekly reporting on risk positions Deal validation with term sheets, VRM etc Confirmation with the clients Settlement of the deals
Roles of Various Teams Teams Limits monitoring Middle Office Accounting / Valuation Process Setting up and monitoring of limits Reporting cases of over-utilized limits and expired cased Margin calls to clients Accounting of all treasury transactions Valuation in accordance with regulatory/internal guidelines Periodic p&l/position/risk reporting to front office
Thank You 37