This little journey through fund analysis history provides some background to modern portfolio analysis and reporting techniques. Although the work of these pioneers in our industry is up to 60 years old, it is still taught in universities and finance courses as the foundation of modern funds management.
Portfolio diversification
In 1990, William F. Sharpe was awarded the Nobel Prize in Economics, along with Harry Markowitz and Merton Miller “for their pioneering work in the theory of financial economics”. Markowitz, often referred to as the father of Modern Portfolio Theory, did not invent the idea of portfolio diversification, which can be traced back to Shakespearean times (see The Merchant of Venice, Act I, Scene I). His key contribution was the idea that while risk can be reduced through diversification, it cannot be completely eliminated. In his seminal 1952 paper 'Portfolio Selection', he was the first to formulate a mathematical model for the diversification of investments. This model demonstrated that what was important was not the riskiness of any individual investment, but rather the contribution that the asset made to the risk of the overall portfolio.
Sharpe built upon this foundation to develop the Capital Asset Pricing Model (CAPM) in 1964. This model utilised a linear regression approach with Beta as a measure of the sensitivity of an asset's return to the return of the market and Alpha as a measure of the excess return for the risk incurred. Both parameters were used to calculate a fair-value rate of return of a risky asset.
Like all good ideas in finance, it wasn't long before others contributed to improvements. Indeed, Sharpe was not alone in formulating such ideas as Jack Treynor, John Lintner and Jan Mossin had independently come to similar conclusions. In 1972 Fischer Black (of Black-Scholes option-pricing fame) improved the CAPM model and it became widely used as a model for pricing individual securities. Later the Black-Litterman model was introduced which incorporated an investor’s subjective views on the market trends which may influence return levels.
Attribution analysis
In 1988, William Sharpe took the ideas to the next level in the article 'Determining a Fund’s Effective Asset Mix'. In this article, Sharpe introduced the idea of ‘attribution analysis’, later dubbed returns-based style analysis (RBSA). In the CAPM model, we can look at the return of an asset with respect to the return of the market. If we consider that any single index is a representation of the 'market' for this 'style' of investment, we can thus calculate exposure to this market. If we then use a number of indices to represent different markets, we can assess the exposure to various markets based on the returns of our portfolio. From here, we can assemble a 'style portfolio' being a portfolio of our indices, weighted by the exposure we have determined that we have to each. The style portfolio will usually not explain 100% of the fund returns (as the fund might hold a mix of assets in each market different to that of each index), however a well-fitted style portfolio should explain the returns to a high level.
The immediate practical applications for this model were obvious, as it allowed an investor to analyse a fund’s underlying exposures. If the underlying holdings are unknown, this can help the investor understand what the fund is doing and where it is exposed. If on the other hand the holdings are known, the analysis can uncover ‘hidden’ exposures (for example the share prices of an Australian company that imports food could be exposed to the Brazilian market).
In 2002, Michael Markov developed and patented a new version of RBSA called ‘Dynamic Style Analysis’. This innovation was designed to improve the capture of changing weights in the portfolio (i.e. a fund will generally not have static holdings throughout the year, but will rebalance the portfolio in response to market conditions). Such changing asset weights add a further level of complexity to the style analysis, as we now have to explain not only which exposures are present in the portfolio, but also how these exposures have changed over time.
Building on past research
Although research continues in the development of new modelling techniques, such research is often based on the work done by those mentioned above. We now have available to us some sophisticated tools to gain an understanding of funds for which we do not have full disclosure of holdings, to perform due diligence of funds we are considering investing in, to break funds down to their component exposures and even to identify hidden exposures in portfolios.
In the words of Sir Isaac Newton, “If I have seen further it is by standing on the shoulders of giants.”
Joe Maisano is a Director and Dr. Alex Radchik is a Project Consultant at Trading Technology Australia (TTA). TTA is the Australian representative of Markov Processes International, the developers of MPI Stylus Pro™.