A Fresh Look at Portfolio Construction: Getting Started

Being wrong when making investment decisions isn’t a crime, but not employing a comprehensive methodology for developing and implementing an investment strategy for clients could be. Under FINRA and ERISA regulations, advisors are required to evaluate a client’s risk profile, invest assets in a minimum of three asset classes and balance the client’s risk with an expected rate of return.

The Financial Industry Regulatory Authority rule on suitability of recommendations by advisors and their firms to customers and prospective customers took effect in July 2012. The rule dealt with a number of important issues, one of which is the use of the wording “risk-based approach.” In one instance the notice mentioned a risk-based approach in documenting compliance with supervisory staff, and in another the phrase was used regarding the supervision of recommendations and investment strategies.

In other words, advisors must use a consistent and measurable “risk-based approach” when analyzing client and prospective client portfolios and investment strategies and making any specific security recommendations. Simply put, advisors must be able to establish a client or prospect’s risk profile and they must be able to analyze the risk in a portfolio relative to the client’s/prospect’s risk profile.  

Virtually all pension sponsors and many money managers require a statement which lays out the investor’s objectives and assumptions as well as a clear workflow and methodology for achieving those objectives. Should the advisor’s development and implementation of client investment strategies be any less rigorous? Advisors who are oblivious to the potential liability arising from a failure to have a well-defined and repeatable method of determining where client capital should be invested do so at their own peril.  

Let’s assume that advisors are interested in developing and implementing investment strategies based on a method of investment analysis, portfolio design and performance evaluation expressed by quantitatively evaluating risk (client and investment) and its relationship to investment return. This method focuses attention on the overall composition of the portfolio rather than the traditional method of analyzing and evaluating the individual components.The investment manager is therefore able to examine and design portfolios predicated on explicit risk-reward parameters and on the identification and quantification of portfolio objectives.  

Financial professionals need to consistently apply these clearly defined, practical and repeatable methods of solving very sophisticated investment and retirement objectives.

Applying these methods will help provide solutions to:

1) better risk management

2) the design of investment portfolios with a high probability of achieving a client’s objectives

3) the selection of appropriate investment vehicles and managers.

Such a methodology looks like this:

Click to enlarge

This methodology rests upon four basic premises:

  1. Risk Aversion
    Investors are inherently risk-averse, unwilling to accept risk except where the level of returns generated will fairly compensate for that risk. It is probably reasonable to assume that investors are more concerned with risk than they are with rewards.  
     
  2. Efficient Markets
    Most academic and industry research supports the concept that markets, at least in the broadest sense, are reasonably efficient. The nature of an efficient market is such that all participants have the same information regarding the markets in general, and specific issues in particular, at the same time, although they may come to opposite conclusions as to an appropriate price for individual securities. It’s ironic that the sophistication of money managers and their virtually instantaneous access to information today creates greater unparalleled efficiency in the marketplace, thereby making above-average returns extremely difficult to achieve. 
     
  3. The Portfolio As the Determining Factor
    Perhaps the most important premise is that the focus of attention should be shifted away from individual security analysis to the consideration of portfolios as a whole predicated on explicit risk-reward parameters and on the identification and quantification of portfolio objectives.Today it is more likely that the efficient allocation of capital to specific asset classes will be far more important than selecting the "right" components of any asset class.
     
  4. Portfolios Can Be Quantitatively Optimized
    The fourth premise is the optimality of portfolio returns vis-à-vis portfolio risk, i.e., for any level of risk that one is willing to accept, there is, mathematically, a rate of return that should be achieved. Quantitative methods are used for measuring risk and diversification, making it possible to create efficient and theoretically optimal portfolios.

There are essentially two methods of optimizing portfolios:

1) Using historical performance.
This method will invariably find that mix of assets that did, in fact, provide the highest rate of return for any comparable level of risk.

2) Using forecasted rates of return and standard deviations
In many cases theoretically optimal portfolios are not acceptable simply because the historical rates of return are not in line with what the advisor and client believe to be realistic. Therefore, it is incumbent upon the advisor to forecast rates of return for each asset. 

For example, in 1987 historical mean rates of return for EAFE companies were far greater than were reasonably expected for the near future. In many cases prices were 90 times earnings—remarkably similar to domestic security prices in 2000. Using those historical mean returns in portfolio modeling would have led to portfolios that were extremely over-weighted in EAFE type securities and resulted in poor performance. While correlation and covariance characteristics don’t tend to change over short periods of time, there is absolutely no reason to assume that historical returns are likely to repeat themselves in the relatively near future. 

Both options may require some modifications by establishing holding constraints for individual assets. In effect, the entire process is as much art as it is science. 

The extent to which knowledge of one asset return provides information regarding the behavior of another asset is measured by the correlation of returns. Are they moving in the same or opposite directions at the same time? 

Measurements of the risk and return characteristics of individual asset classes and/or investments are inadequate in explaining what happens when investments are combined in portfolios.

The true measurement of diversification between assets is called the covariance of the assets. Covariance measures the timing, direction and momentum of the movement of two variables. By calculating the covariances and expected returns for all of the assets in any given portfolio it is possible to calculate the optimal portfolio mix for any degree of risk.

Each portfolio on this "efficient frontier" will generate the highest possible rate of return for any specific level of risk as measured by the standard deviation of returns. Any other portfolio which exhibits the same standard deviation (risk) will generate lower returns and will therefore be considered inefficient.

Click to enlarge

In other words, the investor must increase the expected return by the maximum amount for each additional unit of risk he is willing to take. The increase in return vis-à-vis any increase in risk is measured by several functions including the Sharpe Ratio, Treynor Ratio, Sterling Ratio and others. 

The number of assets in the portfolio is less important than the relationship of those assets. It is a misconception, albeit a widely held one, that investors must accept higher levels of risk to achieve higher returns.  

In subsequent posts we will look at Steps 1, 2 and 3 of the Investment Strategy Methodology.

Page 1 of 3
Single page view Reprints Discuss this story
This is where the comments go.