(Using the Value-at-Risk measure for the banking system in Ghana)
Value-at-Risk (VaR) is a measure of the maximum loss on a portfolio, expected from 'normal' market movements over a specified period of time. For example, a portfolio whose current market value is 100 million cedis has a 10-day VaR of 2 million cedis, at a significance level of 5%. This indicates that over the next ten days, with 95% certainty, the largest loss that should be expected is 2 million cedis. This estimated VaR should vary directly with the Capital Requirements for the bank who owns that portfolio of assets, with the coefficient of variation determined by the Central Bank. This coefficient of variation will embody a host of factors, which will reflect the domestic conditions within which banks operate in Ghana. For example, the real maximum loss expected on the portfolio might not be 2 million, due to the prevalence of corruption. If there is a possibility that, say 2.5 million cedis more cannot be accounted for, due to corruption, this might not necessarily be captured in market risk. This will be better captured in some form of a social loss function, and thus, should be reflected in the coefficient of variation.
VaR embodies the market risk of a portfolio, or indeed of a whole firm, in a single number. VaR was first used in the 1980's by large financial firms to measure the risk of their trading portfolios. In 1993, the “group of thirty” (a group of 30 prominent US financiers, bankers and academics) commissioned a report that recommended the use of VaR for all banks to measure the risk of their derivatives positions. VaR gained popularity following the launch by the global financial firm JP Morgan & Co. in 1994, of Riskmetrics, an integrated package to enable companies to measure VaR. In 1996, the Basle Committee on Banking Supervision (BCBS) required banks to set capital requirements according to “internal models” of market risk, leaving the choice of model to each bank, subject to certain stringent criteria. Under BCBS recommendations, the estimated VaR for a bank varies directly with the capital requirements that the bank must meet. Depending on the effectiveness of the VaR model used, the bank may be required to increase their capital requirement or vice versa.
VaR has become the most important measure of risk. The simplicity of the VaR concept has directed many to recommend that it become a standard risk measure, not only for financial establishments involved in large-scale trading operations, but also for retail banks, insurance companies, institutional investors and non-financial enterprises. The benefits of using VaR by Bank of Ghana, as a measurement tool to estimate the capital requirements of banks can be significant. VaR is a much more dynamic measure, which allows each bank to be assessed based on its own strengths and weaknesses, as manifested in the types and scale of risk it has to deal with, in its operations. Banks in fairly stable operations with relatively low risks should be allowed to keep a much smaller capital requirement, as will be determined by their VaR estimates. This will free up more capital for private sector investments and thus result in job creation, reduced unemployment, GDP growth, and increased welfare for society. Also by increasing their available business capital, the profitability of the banks will increase.
A difficulty with the VaR concept, however, is that it is calculated statistically without analysis of the daily changes of surrounding financial, economic and social conditions. VaR cannot be observed, since its estimation is contingent on the properties of a historical distribution. So the measure is an estimate of the true VaR of the portfolio and hence subject to estimation error. VaR is also as correct as the estimation of the true properties of the underlying distribution of the data of variables. If the variables under consideration are serially uncorrelated and homoscedastic, then we can assume the statistical properties of a normal distribution in calculating the VaR. Otherwise, other methods like using Monte Carlo simulation can be applied to unravel the true characteristics of the data. Kwabena Owusu Ampong BSc Engineering, KNUST, Ghana Master's in International Business, NHH, Norway Candidate for MSc in Financial Economics, BI NSM, Norway Views expressed by the author(s) do not necessarily reflect those of GhanaHomePage.