Non-Normality of Market Returns: A Framework for Asset Allocation Decision Making

Abstract

When market stress and disruptions hit portfolios especially hard, it is natural for investors to ask if anything could have been done differently. The turmoil of 2007–2009 is just one among many such financial crises over the past 30 years— rare and unpredictable events that can, and should, challenge conventional ideas about portfolio construction. And often, it is in such times of reflection that one arrives at important insights and begins to craft effective solutions. This article presents a statistical methodology that can incorporate three categories of non-normality; serial correlation, “fat” left tails (negative skew and leptokurtosis), and correlation breakdown (converging correlations). The results show that extreme negative events due to “non-normality” of asset returns are observed with much higher frequency than current risk frameworks allow for. Consequently, traditional asset allocation frameworks that are based on assumptions of normality in asset returns can significantly understate portfolio downside risk. The authors conclude that incorporating nonnormality can lead to more efficient portfolios.