Risk measurement under extreme events. An in-context methodological review

This paper reviews the basic methodologies for the estimation of Value at Risk (VaR) that are currently in use in international stock and financial market regulation and portfolio management. The main shortcomings of these methodologies are exposed and the direct consequences of ignoring these limitations are analyzed, the latter highlighted by the recent global financial crisis of 2007-2009. In addition, methodologies for the estimation of expected tail losses, based on the Extreme Value Theory, are examined. Both kinds of measures are contrasted with the aim to create a ‘risk ranking’ of different stock markets around the world. The study explores the major Latin American markets and some in developed countries. The lessons widely highlighted in the academic literature related to the shortcomings of VaR when markets face extreme events are corroborated. The paper concludes by stressing the need for more robust measures focused on the tails of the distribution in the measurement of risk.