Quantitative Portfolio Construction and Asset Allocation

Abstract

Ultimately the analysis of assets and asset classes must lead to some decision by the portfolio manager as to the composition of the appropriate portfolio for the circumstances. The two terms ‘asset allocation’ and ‘portfolio composition’ will be used interchangeably throughout this and following chapters. In this book we will not be going into detail about the determination of the exact impact of the initial asset allocation on ex post portfolio performance. Performance attribution is a very comprehensive topic by itself, and should require a book on its own. However, a well-known and often quoted empirical study by Brinson, Singer and Beebower (1986/91) indicates that as much as 90 per cent of a portfolio’s performance can be explained by the initial asset allocation. This means that even if the asset manager has hired and paid for exceptional portfolio managers possessing consistently superior stock-picking skills, this will not automatically lead to exceptional performance. This is because the initial asset allocation may have been far off the mark in terms of forecasting where large returns would materialise, and furthermore the portfolio may be seriously constrained by asset allocation limits that do not allow the portfolio manager to fully exploit the opportunities available to him. Given the fact that such a large proportion of portfolio performance is explained by the asset allocation decision, it could be argued that the asset allocation decision is the most important decision in the life of a portfolio. For this reason we shall now move on to establishing a quantitative asset allocation framework, which we can then extend and apply to various other purposes later on in this book.