Market Risk-Measuring(VaR, A.L.M and Balance sheet Management)

In several countries the monetary authorities are permitting usage of VaR models to measure Market risks on traded instruments in determining appropriate Regulatory Capital charges. In addition VaR models are used to determine Capital allocation and performance based compensation. Each process entail risk assessment. For market risk management, the answer is Risk Limits. In determining this some where the subjective human judgement should get integrated.

Simple VaR model is estimating the sensitivity of each of the components of a Portfolio to small price changes. Should there be 1% change in the interst rate or exchange rate it is assumed that the market price movements follow a particular statistical distribution. This would enable a risk Manager to draw inferences about the potential losses with a given degree of statistical confidence. For example on a given portfolio it might be possible to show that there is a 99% probability that a loss over any week period will not exceed say U.S.Dollar. 1 million

This model is further improved by a model called Simulation model. This is little more complex. It is also called Para metric VaR. This uses historical data. This is a more direct approach unlike the statistical model based on probability approach. This takes a given portfolio, revalues it directly at current and previous market prices measured over a given period. It then takes a more extreme observations that large simulated losses-as indicative of what theoretically could be lost on the portfolio

These VaR models have the advantage of point to the benefits of being able to summarise in a single figure as estimated level of risk faced by the bank in its trading activities. These are powerful management tools. These models can guide the management to size up the potential losses and their expected frequency in normal circumstances

It is quite interesting to note an example given in the write up (Finatial Engineering News, 1997) to drive home the point that there is a need to integrate subjective human judgement with a VaR model which is basically an objective tool

When we talk about the market risk it is meaning full when the market risk is quantitative. It is a process involving 4 steps > Define the Risk to be measured > Agree on a model for that risk-Specify a risk measure that is compatible with that model > Estimate the value of that measure implied by the model. The process could be-Risk(Market risk of a specified portfolio) > Risk model(Market variables are assumed to be jointly normally distributed with specified volatilities and co-relations) > Risk measure (Say one day 90% VaR > Risk estimate) Achieved with saysimulation model. The following case will clarify the conflict between the subjectivity and objectivity of the model and human judgement.

A forex trader, on a particular day takes on a Long Position say in Japanese Yen exceeding the limit set out. The trader knows the market and is aware of the combination of market factors and also perhaps according to the information he had that the central bank was intervening in the markets-that is going to drive the market for yen up in the short term. The trader considers the position appropriate. The risk manager however did not agree. He was not aware of central banks intervention.

All he knew was that the trader exceeded the limit. Reviewing the VaR number that indicated the limit violations. The trader revolted saying that the model was wrong. I knew the market and I also knew that the central bank will be intervening. I was in touch with the host of professionals all day long. This VaR model is just a bunch. The model does not know that the yen is going up, but I do. There is Zero risk in my long position. Any other position would be ridiculous under the circumstances.

Who was right? Who was wrong? The trader knew his job and the market-At the same time the risk manager, if being over ruled every time, what is the use of having him?

The answer is to build up a better VaR model that captures the traders intuitive understanding of the central banks intervention. Some may support this while some others may cling to the existing VaR model claiming that the efficient markets and no arbitrage conditions ensure its ultimate validity. Our risk manager and the trader had their legitimate difference of opinion. The right approach is to challenge the idea that every risk has a number-that there is a right model that will find the number and other models are wrong.

We must embrace the notion that the risk is subjective. We cannot manage the market risk by having risk manager forming and enforcing his own opinions about the riskiness of a traders position. This would be unfair to the trader. Instead, we implant an objective bench mark for risk in the form of a VaR model. Such a model should assume that the market variables are normally distributed despite other observers preferring the Log normal assumption. It may not capture market “LEPTO KURTOSIS"

It probably cannot understand the sticky violations. This is important that if we had a perfect model there is no need for a trader. A VaR model is therefore limited because it is objective whereas risk taking is subjective. If we deny the subjectivity we deny the role of human judgement. This leaves us with two inconsistent market views. That of the Model-and-That of the trader.

The Question is how we can use objective VaR model to manage risk taking process but not place arbitrary or even dangerous-restrictions upon the active traders? The answer is RISK LIMITS. Risk limits enable an organization to manage risk by limiting traders to taking positions within SPECIFIED RANGE.

The VaR models role is therefore to objectively define what that range is. The traders role is to select the optimal position within the range. In this context VaR is just a tool for delimiting a set of acceptable portfolios. We can call it Risk Measure if we like(COURTESY-Financial Engineering News)

Market Risk management - BALANCE SHEET Management. Banks carry a mixture of both fixed rate and floating rate assets and liabilities Many of them are subject to repricing when interest rate changes. There could be extreme market events or scenarios. There could be abnormal large price movements and it is these that pose the greatest risks to banks.

There could be sudden, steep fall or rise in the interest rates. Banks with short term liabilities funding the long term assets will lose if interest rate rises. On the contrary, banks who have funded short term assets with long term liabilities will also incur loss if interest rate falls.

A typical Balance sheet management lies in identification of maturity mis matches between assets and liabilities. An imbalance of assets over liabilities over a particular time or vice-versa would give a net liability asset position. This could be tackled by writing new liabilities or new assets with similar maturity or re-pricing arrangement. It is also possible that bank may leave the mis-match as it is and the risk generated out of this may result in potential loss or gain in the event of rate changes.

This is typically GAP analysis and this ofcourse gives only an imprecise picture of interest rate risk on the Balance sheet. Steady growth of banks interest income is one of the main goals of the banks A.L.M, besides Liquidity management. There are attempts to adopt more sophisticated method by simulation analysis. This will supplement the short comings in the normal, imprecise A.L.M.

The Simulation type involves detailed forecasting of the entire Balance sheet items say for 2 or more years ahead and subjecting all the fore cast cash flows, making the balance sheet to a variety of price shocks which may involve parallel shifts, twists or rotations of the yield curve. The resultant potential exposures are then measured often in terms of their impact on the banks net interest income. With information so obtained Balance sheet strategies can be worked out and leaving interest rate risk hedged or left uncovered.

In Balance sheet management often encounter treatment of assets and liabilities having no formal re-pricing dates. Ofcourse banks current account balances carrying no interest and of a short term nature may contain a core balance which serves as a hedging tool against some fixed assets. Early loan repayments can re open an unexpected interest rate position.

Policy of levying penalties for pre payment is a hedging tool Competitive pressures often results in banks not resorting to it. One of the contentious issues in the balance sheet management is the treatment of Capital. Capital serves as a buffer against potential losses with in a bank. It is a means of funding the asset side of the balance sheet. It is a scarce resource on which bank must generate an acceptable rate of return to the owners