Value at risk analysis of country balance sheets: Thailand in the 1990s

The frequency and severity of currency crises have increased significantly in recent decades, and the nature of crises has changed as well. In the past, crises were bounded by national borders; in recent episodes, crises are contagious. The most celebrated example is the collapse of the Thai baht in July 1997, which led to the collapses of other East Asian currencies.; This dissertation is motivated by recent episodes of currency crises in both developed and developing countries. Dornbusch (1998) argues we should look at a country's financial vulnerability to identify sources of risk and exposure, and to develop an effective supervisory system. This thesis attempts to assess the likelihood of a currency crisis by examining the financial vulnerability of Thailand before and after its financial crisis in 1997. The thesis starts by outlining the balance sheets of Thailand's central bank and three other consolidated sectors in the economy: non-financial, financial, and government. Since the balance sheets of these sectors are interrelated, deterioration in balance sheet in one sector can spread to other sectors and trigger financial crisis.; The concept of value at risk (VaR) was developed to assess the vulnerability of financial firms, but it can also, in principle, be used to assess a country's financial vulnerability. VaR is defined as “the expected maximum loss (worst loss) over a target horizon within a given confidence interval” (Jorion, 1997, p. 19). The hypothesis is that as the VaR of the sector or the country in question increases, the likelihood of insolvency increases and its ability to maintain its commitment declines. This thesis applies the delta-normal method to estimate VaR for various sectors in Thailand. Thailand is a “classic example” of a country that suffered from a currency crisis because of a prior banking crisis. Applying VaR to the balance sheets of various Thai sectors, allows us to examine the interdependence of different sectors and the usefulness of VaR in predicting potential financial crises. Sensitivity analysis is then performed to identify balance sheet items that may play a crucial role in determining VaR for various sectors. Finally, to evaluate the predictive power of VaR, we compare the results obtained from VaR are compared to traditional indicators such as the official reserves-monthly import ratio, the short-term debt-total debt ratio, the growth rate of exports and others.