Journal of Risk

Predicting the next major financial crisis may be a precarious exercise, but the effort of doing so is obviously worthwhile. The present issue of The Journal of Risk contains a paper that offers an approach that attempts to identify early signs of impending crises. Another topic addressed here is a method of systematically constructing data sets based on open sources that are important for financial risk management. This issue concludes with two papers that focus on portfolio strategies: one for equities and one for bonds.

In “A three-state early warning system for the European Union”, Savas Papadopoulos, Pantelis Stavroulias, Thomas Sager and Etti Baranoff propose a three-state model to identify three precrisis periods. They use standard statistical tools – including multiple discriminant analysis, multinomial logistic regression and neural networks – for model fitting, and the total harmonic mean for out-of-sample validation. Based on empirical data for the European Union, the authors find that their model would have forecast the 2007–8 financial crisis between seven and twelve quarters ahead of time, with high accuracy.

One of the challenges faced by regulators and policy makers is striking the delicate balance between relying on financial institutions to provide data for systemic risk management and finding independent sources for validation. In this issue’s second paper, “A general framework for constructing bank risk data sets”, Xiaoqian Zhu, Lu Wei, Dengsheng Wu and Jianping Li propose a general framework for the classification and integration of publicly available data from a wide variety of sources. The authors illustrate their approach using data from Chinese and Austrian banks.

We now turn to the portfolio strategies papers included in the issue.

The first, “Covering the world: global evidence on covered calls” by Roni Israelov, Matthew Klein and Harsha Tummala, concerns one of the many low-volatility approaches that have risen to prominence: namely, covered index calls. These enable equity-like returns with substantially lower volatility. In their empirical study, the authors offer evidence that the source attribution for the performance of this strategy, previously documented with regard to the Standard & Poor’s 500 index, also extends globally.

Our second paper on portfolio strategies, and the last in this issue of The Journal of Risk, is “Balance-sheet interest rate risk: a weighted Lp approach”. In it, Lesław Gajek and Elzbieta Krajewska address bond immunization: a classical problem regarding bonds. In their approach, the authors develop a model that explicitly accounts for the gap between assets and liabilities in a manner that reflects their individual importance. In contrast to traditional approaches, which focus on proxy measures such as duration gap, the authors contribute a method that directly tackles the stochastic evolution of these financial instruments.

Farid AitSahlia
Warrington College of Business, University of Florida