Risk management and control

Risk management in main risk types

Market risk

Market risk exists wherever the group has trading, banking or investment positions. Major exposures to market risk occur in markets served by formal financial exchanges and over-the-counter markets. These exposures arise from customer-driven business and from proprietary positions.

Approach to managing market risk

Market risk exposure on trading positions and capital funds

Market risk exposures as a result of trading activities are contained within Corporate & Investment Banking’s trading operations. The board grants general authority to take on market risk exposure to the group ALCO. Group ALCO sets market risk standards to ensure that the measurement, reporting, monitoring and management of market risk associated with Corporate & Investment Banking across the group follow a common governance framework. Each bank within the group has ALCOs to monitor compliance with these market risk standards. Both the Africa ALCO and International ALCO report into the group ALCO, chaired by the group chief executive.

The group manages market risk through risk limits. The group uses a range of risk measurement methodologies and tools to establish limits, including Value-at-Risk (VaR), stress testing, loss triggers and basic risk management measures.

The group generally uses the historical VaR approach to derive quantitative measures, specifically for market risk under normal market conditions. While VaR, calculated daily, provides an indication of possible losses under normal market conditions, the group supplements VaR with stress tests. The stress testing takes into account likely events that characterise the markets in which the group operates.

The group back-tests its VaR models to verify the predictive power of the VaR calculations. Back-testing compares the daily profit and losses under the buy and hold assumption with the estimates our VaR models had forecast.

Loss triggers are designed to contain daily, monthly and year-to-date losses for individual business units by enforcing management intervention at predetermined loss levels. Other basic risk measures specific to individual business units are also used. These measures include permissible instruments, concentration of exposures, gap limits and maximum tenor.

The table below shows the aggregated historical VaR calculations for Corporate & Investment Banking in the markets in which the group holds trading positions. The minimum and maximum VaR amounts show the bands in which the values fluctuated during the periods specified. The group calculates historical VaR with a holding period of one day and a confidence interval of 95%.

Market risk management units – independent of trading operations and accountable to business unit ALCOs – monitor market risk exposures due to trading activities. These units monitor exposures and respective excesses daily, and report monthly to ALCO and quarterly to the group risk management committee.


Trading book value-at-risk analysis (Rm)
      Normal VaR2           Stress VaR3    
Market variable Maximum1   Minimum1   Average   31 Dec   Maximum1   Minimum1   Average

2005

                           
Credit derivatives 43,3   12,3   26,0   39,3   266,8   66,3   138,3  
Equity – other 0,8   0,3   0,5   0,6   4,4   1,3   2,7  
Foreign interest rate and foreign exchange 32,1   11,0   18,4   20,0   182,4   64,9   108,8  
Energy 9,3   0,6   3,1   2,4   49,3   3,1   16,3  
Interest rates – SA 16,5   6,9   11,2   14,4   214,0   64,7   129,1  
Equities – SA 27,5     3,4   4,7   87,1     22,9  
Base metals 8,0   0,1   3,0   5,7   42,1   0,3   15,8  
Precious metals 12,9   1,4   6,2   1,9   78,1   9,5   40,0  
Commodities 1,9     0,9   0,6   4,9   0,1   2,1  
Diversification benefit4         (37,1)   (42,4)           (245,8)  
Aggregate 49,4   22,9   35,7   47,0   342,8   146,8   230,1  

2004

                           
Credit derivatives 46,0   10,5   33,3   29,0   223,0   96,4   156,4  
Equity – other 38,4   1,9   10,8   2,2   202,6   10,3   57,1  
Foreign interest rate and foreign exchange 32,4   6,3   17,9   11,0   122,2   39,2   86,9  
Energy 25,5   0,1   5,4   19,0   134,9   0,8   28,3  
Interest rates – SA 25,1   5,6   12,1   5,8   171,6   23,7   67,1  
Equities – SA 12,3     0,9   5,0   97,9     25,6  
Base metals 9,0   0,3   3,5   2,4   47,5   1,8   18,7  
Precious metals 8,5   2,7   5,3   4,4   69,2   13,8   37,2  
Commodities 1,7     0,2   0,1   5,1   0,1   0,7  
Diversification benefit4         (45,9)   (37,8)           (257,9)  
Aggregate 65,3   28,4   43,5   41,1   333,2   141,7   220,1  
1The maximum (and minimum) VaR figures reported for each market variable did not necessarily occur on the same days. As a result, the aggregate VaR will not equal the sum of the individual market VaR values, and it is inappropriate to ascribe a diversification effect to VaR when these values may have occurred on different dates.
2Normal VaR is based on a holding period of one day and a confidence interval of 95%.
3Stress VaR is based on a holding period of between 10 and 20 days and a confidence interval of 99,7%.
4Diversification benefit is the benefit of measuring the VaR of the trading portfolio as a whole. That is the difference between the sum of the individual VaRs and measuring the VaR of the whole trading portfolio.

The graph below shows the frequency distribution of daily income during 2005. It indicates to what degree the realised income and loss distribution deviates from a normal (symmetrical) distribution. In this case the distribution is skewed to the profit side. The graph details that income of R5 million – R10 million was realised on the majority of trading days (66 days).

Distribution of income
of trading units 2005
Distribution of income of trading units 2005

The graph below shows the value-at-risk analysis and actual income of trading units throughout the year.

Income of trading units and value-at-risk (Rm)
Income of trading units and value-at-risk (Rm)
Income of trading units and value-at-risk (Rm)


Market risk on equity investments

Equity management committees approve investments in listed and unlisted entities within an approval limit framework. Market risk on investments is managed in accordance with the purpose and strategic benefits of such investments, rather than purely on mark-to-market considerations. Periodic reviews and reassessments are undertaken.

Market risk exposure on banking positions

Banking-related market risk exposure is primarily due to structural interest rate risk arising from the differing repricing characteristics of banking assets and liabilities. Structural interest rate risk, which is the potential adverse effect of interest rate movements on net interest income, is transferred to and managed by the group’s treasury operations. Changes to the interest rate profile are achieved mainly by using derivatives, particularly interest rate swaps, where the shape of the yield curve and the group’s own view of interest rates are used as inputs to defining hedging strategies.

Asset and liability management (ALM) functions monitor exposures to interest rate risk. Banking-related interest rate risk in the group’s Domestic Banking operation is monitored by the ALM function in Johannesburg. Within the African operations, the in-country ALM teams monitor banking-related interest rate risk, with oversight by the ALM function in Johannesburg. Within Corporate & Investment Banking International’s treasury, banking-related interest rate risk, which is primarily in US dollars and sterling, is managed on an integrated basis together with the trading book interest rate risk, overseen by the International ALCO.

The primary banking positions of the group reside in the Domestic Banking operation. The main analytical techniques used to measure banking book interest rate risk are earnings-based measures such as forward looking dynamic scenario analyses, including Monte Carlo simulations, and static repricing gap analyses, which measure interest rate risk at a point in time, as well as valuation-based measures in the form of economic value of equity. The results obtained from these analytical techniques assist the group in evaluating the optimal hedging strategies on a risk-return basis.

The repricing gap for the Domestic Banking operation for the financial year end is shown below. All assets, liabilities and derivative instruments are placed in gap intervals based on their repricing characteristics. Assets and liabilities for which no specific contractual repricing or maturity dates exist are placed in gap intervals based on management’s judgement and statistical analysis, as applicable, based on the most likely repricing behaviour. Comparing the repricing gap as at December 2004 with December 2005, it is evident that the asset sensitivity of the Domestic Banking operation has increased. The extent of banking book interest rate exposure remains within limits set by Africa ALCO.

The Domestic Banking operation remains asset-sensitive and is therefore positioned to benefit from an anticipated rate hiking cycle. A 1% parallel increase in the yield curve is forecast to result in a favourable R415 million annualised net interest income impact, which represents 3,9% of net domestic interest income for a projected twelve-month period.

Repricing analysis of assets, liabilities and shareholders’ funds as at 31 December 2005 (Rm)

Domestic Banking operations Call   4 – 6   7 – 12   Over   Non-rate    
in South Africa 3 months   months   months   12 months   sensitive   Total

2005

                       
Total assets 297 541   12 666   7 987   14 617   94 305   427 116  
Total liabilities and shareholders’ funds 268 846   7 079   3 364   31 584   116 243   427 116  
Interest rate sensitivity gap 28 695   5 587   4 623   (16 967)   (21 938)    
Cumulative interest rate sensitivity gap 28 695   34 282   38 905   21 938      
Cumulative interest rate sensitivity gap as                        
   percentage of total assets 6,7%   8,0%   9,1%   5,1%          

2004

                       
Total assets 244 344   6 255   6 264   12 928   111 896   381 687  
Total liabilities and shareholders’ funds 223 693   5 798   3 804   12 070   136 322   381 687  
Interest rate sensitivity gap 20 651   457   2 460   858   (24 426)    
Cumulative interest rate sensitivity gap 20 651   21 108   23 568   24 426      
Cumulative interest rate sensitivity gap as                        
   percentage of total assets 5,4%   5,5%   6,2%   6,4%