Real estate market risk modelling

The global financial crisis towards the end of the last decade saw an increased interest in the role of risk management in the main stream financial investment market. Among other things, the measurement and management of market risk, credit risk and operational risk have become more pronounced than ever before. Value-at-risk (VaR), a tool which assesses the maximum possible loss of an investment, assuming a given confidence level, is widely used in the investment world to measure market and credit risk. This measure has however come under constant criticism as it only considers the maximum loss for a specific confidence level and ignores any losses beyond that threshold, which could arise from extreme events. Secondly, VaR assumes normal distribution of returns and yet this is not the case with most financial returns, which have the added complexity of being susceptible to the phenomenon of ‘fat tails’. (It should be noted however, that it is only the basic version of VaR that has the normality issue and this can be addressed through the use of Monte Carlo.) Thus, the credibility of VaR seems to be losing ground. Though derived from the principles of VaR, the expected shortfall (ES) is being forwarded as an alternative proposition due to its ability to overcome some of the shortcoming of VaR, particularly when it comes to dealing with tail risk. To this effect, the ES is being mooted as a tool for market risk regulation, replacing VaR in the banking sector as proposed by the Basel Committee on Banking Supervision. This said, the ES has its own challenges especially because it cannot be subjected to back-testing due to its non-listable attribute. Furthermore ES is also said to be quite sensitive to extreme values. In the real estate market, very limited research has been conducted on modelling market risk. This study therefore aims to investigate market risk modelling for real estate and assess whether, and / or the extent to which, the expected shortfall model offers a better alternative to VaR in terms of measuring market risk. Public real estate has been chosen as the focus of the study as it is more amenable to the application of VaR compared to private real estate.