Construction Methodology for Optimal Portfolios - Part One
The method in which investors construct a portfolio is a process that has evolved over time as more research on portfolio theory becomes available to the general public. In the early days of investing when portfolio theory was still a relatively undeveloped field, portfolio construction was simply a matter of selecting and putting together a list of stocks that the investor thought had growth potential. Added to this was the well known fact that diversifying rather than holding all your eggs in the one basket was a good way of reducing risk and limiting the impact of poor performance in any one stock. Together, these two basic methods of stock selection and diversification became the first systematic form of portfolio construction.
As more research on the relationship between asset classes and efficient risk management became available, the more astute investment firms utilized the information published by finance academics to develop portfolios that gained a performance edge on the traditional portfolio construction method. Various construction strategies have emerged over time to capture the performance benefits associated with the relationship between different asset classes and characteristics of small, large, value or growth companies.
We have spent considerable time reviewing portfolio theory and the factors that determine investment returns. In doing so we have considered the academic research into portfolio management that has been published over the past six decades. The research has found overwhelmingly, that purchasing assets based on their predicted future prices in an attempt to outperform the market is highly improbable.
Given that the future price of a security bears no relationship to the past price of that security, the future price of a security is unknown. Moreover, given that one cannot predict the unknown, then it is not possible to accurately predict the future price of a security and ultimately this means that actively managing a portfolio provides no value, after fees and costs are taken into consideration.
However, some methodology needs to be applied in constructing a portfolio. If active management has been shown not to add value, what methodology should be used?
This question, led to the development of the passive management style of portfolio construction. This methodology aims at acquiring securities that replicate a specific index and then holding the portfolio over the long-term. However, to maintain close correlation with the chosen index, the portfolio would need to be adjusted to take into account the changing characteristics of existing companies, the emergence of new companies within the index and the fact that certain companies will no longer fall within the parameters of the index.
The passive style of investing essentially accepted market risk as a given, however, there are other elements to market risk which influences the variation of returns of specific securities.
The identification the ‘sub-elements’ of market risk and the realization that these risks can be captured more efficiently to influence the returns of a portfolio, resulted in the development of an alternative method of portfolio construction commonly referred to as the ‘Asset Allocation’ approach.
The seminal work of Brinson, Randolph and Beebower (1986 and 1990)[1], highlighted the extent to which the asset allocation decision influences the returns of a portfolio. This research, of 91 large pension plans in the USA suggested that the asset allocation decision is the dominant factor in determining portfolio returns.
Factors Influencing Variation of Portfolio Returns
Overall, the asset allocation decision determines 94% of the long-term variance of returns generated from an investment portfolio. The residual factors affecting long-term performance are market timing (2%) and security selection (4%).
Asset Class Selection refers to how assets are allocated in a portfolio amongst asset classes, Market timing refers to the shifting of portfolio assets in and out of markets or between asset classes, and Security Selection refer to the exclusion or inclusion of specific bonds, stocks or other securities.
Essentially asset class investing seeks to apply evidence-based methods to investing, just as they have been applied in other fields. Over the years, investment strategies have been developed based on solid academic research. This structured approach uses insights about risk and return characteristics, securities pricing and the asset class building blocks to develop a total portfolio solution.
In short, this approach to portfolio construction is based on probability not prophecy. Short-term past performance is considered irrelevant, as research has demonstrated this is an extremely poor indicator of future returns. Furthermore, this approach does not rely on forecasting because as has been mentioned, the future is by definition unpredictable.
Financial models are used to identify and understand the relationship between risk and return. The most compelling of these models is the leading research into the sources of share investment returns by Professor Eugene Fama of the Booth School of Business and Professor Kenneth French of Yale University[2]. Their ‘three-factor’ model indicates that portfolio returns have the following key dimensions.
► Market Factor: Shares have higher expected returns and risk than fixed income securities.
► Size Factor: Small companies have higher expected returns and risk than large companies.
► Book-to-Market (BtM) Factor: High BtM (or ‘value’) shares have higher expected returns and risk than low BtM (or ‘growth’) shares. Companies with a market valuation (or capitalization) close to book (or liquidation) valuation are generally characterized as ‘value’ investments. When market capitalization significantly exceeds book value, companies are broadly characterized as ‘growth’ investments. This is referred to as the ‘value’ effect.
The Fama & French research concluded that these three factors explain most of the differences in portfolio performance over the long term, as depicted in the following diagram. This is consistent with the earlier research by Brinson, Randolph and Beebower (1986 and 1990)[3], which suggested that market timing and security selection have minimal effect over extended periods.
Factors Explaining the Differences In Portfolio Performance
Most people would generally agree that the share market is riskier (more uncertain) than Government bonds and other interest bearing securities. Equities have a greater risk of capital loss than fixed income investments and so provide a higher level of compensation / return.
In similar fashion, most people would consider that smaller companies are riskier than larger companies. Smaller companies tend to be more susceptible to the commercial risks of operating a business and generally do not have the strong brand presence or balance sheet of larger companies. Smaller companies must offer a higher rate of return in order to attract capital.
The third factor, (the ‘BtM’ effect) is best understood by contrasting an excellent (‘growth’) company with an organization that is in financial distress (a ‘value’ company). A company in distress will have experienced poor growth and have a weak earnings outlook. Such a company is generally regarded as being a higher risk than a strong and healthy company.
In financial markets, value companies must offer a higher rate of return in order to attract capital. This is because a value company has a lower prospect of survival than a growth company. As a result investors tend to write-down the ‘going concern’ worth of the company and instead place greater emphasis on its liquidation (‘break-up’) value when pricing the company’s shares.
The most systematic and disciplined way to achieve a higher expected return over the long term is to increase the portfolio weighting in shares. In particular, to increase the proportionate amount held in smaller companies, value companies or both. This higher expected return will be commensurate with a higher degree of ‘risk’ (using the academic definition of volatility over time).
Ultimately, the return you get is the reward for bearing risk. Next week, we will examine how investors can consider these risks and review how their portfolio can efficiently capture returns given the degree of risk inherent in any portfolio.
[1] Source: Study of 91 large pension plans over 10-year period. Brinson, Randolph & Beebower “Determinants of Portfolio Performance” Financial Analysts Journal, Jul-Aug 1986 and “Revisiting Determinants of Portfolio Performance: An Update” 1990.
[2] Fama & French "The Cross-Section of Expected Stock Returns" Journal of Finance June 1992.
[3] Source: Study of 91 large pension plans over 10-year period. Brinson, Randolph & Beebower “Determinants of Portfolio Performance” Financial Analysts Journal, Jul-Aug 1986 and “Revisiting Determinants of Portfolio Performance: An Update” 1990.
Disclosure: Please note that all information provided is of a general nature and does not take into account your current financial situation, needs or objectives. Before acting on any of the information you should consider its appropriateness, having regard to your own objectives, financial situation and needs. The Prosperity Planners Pty Ltd t/a Prosperity Planners ABN 56 679 768 897 is an Authorised Representative No. 1289369 and Credit Representative No. 532716 of FYG Planners Pty Ltd Australian Financial Services Licensee and Australian Credit Licensee No. 224543 ABN 55 094 972 540 Level 2, 39-41 Alexander Street Burnie TAS 7320.