Equity Valuation and Portfolio Management

CHAPTER 19

Portfolio Construction and Extreme Risk

Jennifer Bender, Ph.D.

Vice President MSCI

Jyh-Huei Lee

Vice President MSCI

Dan Stefek, Ph.D.

Managing Director MSCI

The events of the last few years have refocused attention on risk—particularly the risk of extreme losses. Fortunately, researchers have recently developed new models and analytics for understanding and managing extreme risk. In this chapter, we look at using such tools to enhance portfolio construction. Our goal is to adapt mean-variance optimization to produce active portfolios with less exposure to extreme losses than normal optimized portfolios. We do so by constraining the shortfall beta of the optimal portfolio. Shortfall beta measures the sensitivity of a portfolio to periods of extreme stress.

The empirical study we present illustrates the possible benefits of constraining shortfall beta. Using three common alpha signals, we compare portfolios constructed using optimization with and without a shortfall constraint. Interestingly, portfolios with shortfall constraints tend to fare better during turbulent periods and outperformed their mean variance counterparts over the period from January 1994 through June 2009. This is true even after we prevent the portfolio from timing the benchmark by requiring its beta to be one. While this study does not account for many of the costs and constraints managers face, the results are intriguing.