Loan Risk Management Term Paper
- Length: 6 pages
- Subject: Economics
- Type: Term Paper
- Paper: #73308912
Excerpt from Term Paper :
Portfolio Risk Management
In today's competitive banking environment, an important challenge is to ensure adequate diversification of revenue sources across products, market segments and market and credit risks (Sturzinger). Banks must assess their risk appetite and risk capacity as basic components of the budgeting and planning processes and identify their vulnerabilities through risk management techniques.
Risk is defined as uncertainty of returns from a portfolio (Credit-stress testing, 2002). The volatility of a portfolio's returns indicates the level risk and is influenced by many risk factors. Therefore, one of the risk manager's primary goals is to measure the influence of each risk factor on the volatility of portfolio returns and to manage the composition of the portfolio so that the volatility of its returns is reduced. The risk manager also has to measure the influence of the risk factors on each other. Determining the effects of multiple risk factors and quantifying the influence of each is a complex task, but portfolio risk management techniques can help. Some examples of risk management techniques discussed in this paper include: performance analysis, value-at-risk models, stress testing, Monte Carlo simulation, and heuristic controls, each having individual strengths and weaknesses.
Historical performance analysis provides insight of how a portfolio has performed over time. However, older data has limited value for forecasting risk because the structure of the portfolio and the market environment are constantly changing (Brooks, Beukes, Gardner, and Hibbert, 2002). Daily and monthly performance data, on the other hand, can be useful as explained by these authors. Risk managers can slice and dice performance data in different ways to identify performance problems and gain a better understanding of their cause and how perceived concentrations of risk are or aren't being rewarded. For example, an examination of a rolling sixty-day tracking error and relative performance plot can act as an early warning system provided that the risk manager does not over react to short-term spikes in tracking errors. This later tendency may cause excessive portfolio turnover and distraction from adding value over the long-term. Despite its usefulness as a portfolio risk technique, performance analysis is often overlooked by risk managers.
Value-at-Risk (VaR) Model
Value at risk is an estimate of the largest loss that a portfolio is likely to suffer during all but truly exceptional periods (Hopper). VaR can be used to assess the potential loss on a portfolio of assets generally or the user can specify any horizon and frequency of loss that fits a particular circumstance. As an example, Hopper describes a bank that specifies a horizon of one day and sets the frequency of maximum loss to ninety-eight percent. A VaR calculation might reveal that the maximum loss is $1 million. This means that, on average, in ninety-eight trading days out of 100, the loss on the portfolio will not exceed $1 million over a one-day horizon. However, on two trading days in 100, losses will, on average, exceed $1 million.
The method of calculating VaR depends on the horizon chosen and on the kinds of assets in the portfolio. According to Hopper, one method may yield good results with portfolios consisting of stocks, bonds, and currencies over a short horizon, but the same method may not work well over longer horizons such as a month or a year. And, portfolios that contain derivatives require methods that are different than those used to analyze portfolios of stocks, bonds, or currencies may be needed.
When properly used, VaR can give an institution an idea about the maximum losses it can expect to incur on its portfolio a certain fraction of the time (Hopper). Using results, an institution can judge how it should re-allocate the assets in its portfolio to achieve the risk level it desires. But VaR methodology is often improperly used, leading to poor risk-management decisions. This happens for one of two reasons: either the VaR is incorrectly calculated or the VaR is correctly calculated but irrelevant to the institution's real risk-management goals.
Risk statistics work well for estimating risk during normal market conditions, but they cannot predict the occasional, unexpected crises that result in extreme market shocks (Stress testing, RiskMetrics Goups). Stress testing allows portfolio managers to assess how badly things could go during a crisis and to assure that losses do not exceed their loss-tolerance level. Analyses of historical stress scenarios that have resulted in the largest losses for a given portfolio mix are a common form of stress testing. To illustrate historical stress scenarios, RiskMetrics Group provides examples for a portfolio consisting of sixty percent equities and forty percent fixed income that resulted in the largest one- and five-day portfolio losses:
Worse-Case Scenario Portfolio Example
Mex Peso Fallout
Source: RiskMetrics Goup
In addition to historical scenarios, another approach to stress testing is to invent an extreme scenario based on what the portfolio manager thinks might go wrong in the world.
Both historical scenarios and the invention of scenarios have weaknesses (Stress testing, RiskMetrics Group). The problem with historical scenarios is that history is unlikely to repeat itself while inventing scenarios is inadequate because no one has a crystal ball for predicting the future.
Monte Carlo Simulation
Monte Carlo simulation is a method by which portfolio managers can anticipate the probability of meeting specific financial goals at certain time periods in the future. This is accomplished by generating thousands of possible scenarios that investments might take. More technically, Monte Carlo finds the best approximate answers or distributions of probable answers to problems with many variables and/or many possible outcomes (Davidson). It requires many simulations with randomly valued variables to achieve accuracy. Because of the multiple simulations, the method takes time, especially for highly complex instruments or large portfolios, with a direct trade-off to be made between speed and accuracy.
In banking there are many situations when conventional portfolio theory does not allow the risk manager to fully understand the complete distribution of returns (Brooks, Beukes, Gardner, and Hibbert, 2002). Instances where Monte Carlo simulation might be useful are:
Analyzing portfolios containing instruments with asymmetric returns such as options.
Understanding credit migration impact on the distribution of portfolio returns, credit losses and defaults.
Studying the impact of different specifications for the time variation in share price volatility on portfolio returns.
Examine the impact of fixed trading rules such as a stop loss within a hedge fund set up.
Investigating the returns from performance fee structures.
The major disadvantage of Monte Carlo simulation is its complexity, but it does a good job of explaining risk exposures because it provides specific examples of events that are of concern.
Information about future potential losses and about the likelihood they will occur usually has to be put together for every individual case with a good deal of design work and risk managers avoid the use of heuristics (Schubert, 2003). However, risk managers can exploit previous work through the application of different heuristics or cognitive rules of thumb that assume the available information is incomplete and selective and that the probability estimates derived from it will therefore be distorted. Good heuristic controls recognize that experts may over- or under- estimate probabilities depending on their personality, background and experience, and on the way they formulate the problem. And, with any heuristic, simplification is achieved at the price of systematic error.
For many reasons, bank managers require adequate measures and assessment of risk. Risk management techniques such as performance analysis, value-at-risk models, stress testing, Monte Carlo simulation, and heuristic controls may provide banking institutions with better insight in identifying the sources of market risk, leading to a better understanding and analysis of how their…