A Scenarios Approach to Asset Allocation

Abstract

A number of different approaches to asset allocation are used by practitioners, including purely qualitative assessment, simple mean-variance analysis, and more complex multifactor modeling. Since Markowitz published his seminal paper in 1952, however, approaches that rely on the selection of particular parametric return distributions, on summary measures of risk, and on historical data as an indicator of the future still remain widespread. Little doubt exists that such reliance has resulted in serious mismeasurement of risk and misallocation of assets. In this article, Gosling proposes an alternative approach that is important in its implications for investment philosophy and practice. The approach makes more complete use of the information available about the future and virtually forces serious consideration of different time frames, alternate outcomes, and tail risk. The depth of information provided about risk and diversification is also a principal benefit of the approach. The information is not provided by forecasting the future, but by describing what could happen. These changes have the potential to make a significant difference to long-term investment outcomes.