California CPA. June 2010

Asset allocation after the ‘Great Recession’

By Joel H. Framson CPA

Whether we manage the money ourselves or act as a sounding board, CPAs are often involved with investment decisions for clients—and it’s likely that we were profoundly impacted by the gyrations of the capital markets. We know our clients’ goals and we understand their fears. And if we’re to serve our clients well, we must understand what went wrong to modify our investment approach and advice.

Looking Backward

The first issue we must address is whether we agree that something went wrong in our understanding of risk and implementation of an effective asset allocation strategy. While we can’t control returns, the control of risk is a key element in designing an appropriate portfolio.

Modern Portfolio Theory (MPT) has been a cornerstone of portfolio building for decades. Magazine space will not permit a detailed analysis of MPT and related calculations, but we can briefly describe the theory and identify alternate approaches based on what we’ve learned and observed.

Harry M. Markowitz and William F. Sharpe advanced the asset allocation process by creating a multidimensional framework in attempting to create efficient portfolios. Markowitz posited that “optimal” portfolios could be developed that would maximize the return for a given level of risk, or conversely, minimize the risk for a given level of return.

As we know, the formula for optimizing the portfolio required a computer program to assess standard deviation as the measure of risk, correlations between asset classes and expected returns.

However, when these theories were being advanced in the 1950s, computers were much less powerful than today. Markowitz settled on a methodology that was somewhat simplistic, but the best that could be done at the time.

His optimization program focused on measuring risk (standard deviation) as the deviation of an asset’s return compared to its long-term average return.

This is a critical concept and a significant limitation due to the limited computer power at the time. The computation considered deviations above an asset’s long-term return to be equally significant to those deviations below the average return. In other words, the program looked at a return of 30 percent as being as risky as a return of negative 10 percent if that asset’s average return had been 10 percent. In each computation, the deviation was 20 percent—therefore the standard deviation computed to be the same.

Markowitz stated a very important difference—he would have preferred to use an approach that could focus on downside risk. Standard deviation views upside and downside risk symmetrically to calculate the variance from the mean. Investors, however, are much more sensitive to downside risk than they are to a variance from the mean on the upside.

Behavioral Finance and Asset Allocation for the Individual Investor

There are a few key underlying assumptions to understand that have been part of MPT theory:

• Investors behave rationally.
• Investors seek to optimize their returns.
• Investment returns follow a statistical “normal” distribution (bell-shaped curve).
• Standard deviation and correlation primarily define portfolio risk.
• Markets are efficient.

However, behavioral finance studies have concluded that either many of these tenets are not true or are not particularly applicable to individual investors. Consider:

• Investors often behave irrationally.
• Investors’ identification of their own risk tolerance changes over time, especially under different market conditions.
• Investors do not view upside and downside risk the same way because they are focused on objective-based goals.
• Market returns are often not “normally” distributed, exhibiting skew or “fat tails” (extreme events).
• Extreme market behavior happens more frequently than statistically predicted.
• Inputs (standard deviation, expected returns, correlations) needed to calculate MPT are not static and fluctuate over time.

For individuals, risk is much more than standard deviation. Risk is an emotional condition and includes fear of a bad outcome, loss or underperformance. But it’s likely that the biggest fear is that of failing to achieve a financial goal—not being able to retire, remain retired or assist children or parents.

Individuals require a minimum acceptable return (MAR) to accomplish their goals. This is not a new concept for those of us who provide financial planning.

We learned in 2008 that when extreme variances from the mean occur, many asset classes become positively correlated and seem to move in unison. Asset classes that we presumed from historic patterns would provide diversification all seemed to move in the same direction: down. Portfolios that were built utilizing negatively correlated asset classes to modulate risk and return failed to achieve their desired results.

We observed that the global economic system was on the verge of collapse due to any number of factors. While we are familiar with the concept of a world that has become more economically “flat,” perhaps never before did we experience how globally interconnected our financial affairs have actually become.

Buy and Hold vs. Tactical Asset Allocation

Investors and investment managers will always benefit from tools that assist them in building better portfolios. We all want two main things from such tools:

• To control risk.
• To attain the MAR to enable clients to reach their goals.

The challenge with utilizing software designed to produce optimal portfolios under MPT is that the programs don’t focus on downside risk, and they look historically for their inputs on returns, correlations and standard deviations. Markowitz never intended that MPT be based on historic numbers. In his view, asset allocation decisions would be based on forward-looking estimates of these inputs.

As we’ve now experienced, the dynamics of risk have changed and so has our understanding of how risky assets behave. At a minimum, we have seen that price movement correlations between asset classes do not always conform to historic patterns of correlation. For example, with the interrelated nature of the global economy, it is no wonder that the price movements of domestic asset classes are no longer as positively or negatively correlated with international asset classes.

In creating a philosophy of asset allocation, there has always been a major schism between proponents of buy and hold as a style compared with a more tactical approach to portfolio management. There are probably as many proponents of the more passive style as there are of more active management.

A number of recent articles suggest a risk-based portfolio construction, which differs from a traditional asset class approach in that it does not focus on the asset class, or style box, definition of an investment. The focus would be initially on the expected future risk level and secondarily on the expected asset class return. Building portfolios with this approach can help build portfolios with better risk-adjusted returns.

The potential downside is that more subjective work is required to determine the level of risk and expected forward-looking returns. But since those are exactly the assumptions needed to correctly utilize MPT, tactical asset allocation may have been what the Nobel Prize committee had in mind when conferring this honor on Markowitz and Sharpe in 1990.

Tools and Techniques for the Future

CPAs are often called upon to review client portfolio returns and determine whether the portfolio return is reasonable. This analysis might include a comparison of the portfolio rate of return to an index or to a blended benchmark influenced by such factors as the age or risk tolerance of the client.

These factors are certainly important and are, sadly, all too often overlooked in the strategic portfolio design. A red flag should immediately be raised if the portfolio composition doesn’t seem logical when considering these initial factors.

However, more tools are required to properly advise the client. The asset allocation techniques that were built on the efficient markets theories must be challenged due to today’s market environment. A static allocation to the variety of traditional asset classes historically utilized can be a dangerous trap to client retirement stability.

For example, a traditional portfolio allocation might have looked like the following:

• 60 percent equities, including large-cap domestic stocks, small-cap domestic stocks, international developed country stocks, real estate and emerging markets stocks.
• 40 percent fixed income, including U.S. government bonds, municipal bonds, international bonds and high-yield bonds.

It is critical today to focus on the downside risk of a portfolio and classify investments on the level of downside risk, not just the definition of the asset class it belongs to. With the extreme volatility recently experienced and the tendency of correlations to perform differently from historic norms, this risk classification is an appropriate exercise in our attempt to create more stable portfolios.

We know that prior returns are not guarantees of future performance. Yet portfolios will be fraught with construction flaws if they continue to be built using

past returns.

It is essential to create a forward-looking matrix of estimated returns that will protect client portfolios and build them with the combination of investments that can achieve a client’s MAR. The inclusion of asset classes and specific investments that may not have been included in the above example also will be important when building portfolios that can withstand the modern economic stresses.

Rethink Portfolio Construction

One method for rethinking portfolio construction is to categorize investments based on a realistic estimate of the risk level. Using this approach might result in the following portfolio:

• 60 percent “lower-risk” investments, including corporate bonds (investment grade), municipal bonds (focusing on revenue sources), tactical/strategic bond funds, private investments with stable income and stable principle, equity market income-oriented tools with low volatility and short-term bond funds to provide liquidity.
• 40 percent “higher-risk” investments, including high-yield bonds, emerging-market bonds, emerging-market equities, domestic and international equities (selected based on forward-looking growth estimates), precious metals and other commodities, real estate and global multiasset tactically managed funds.

Note that the higher-risk allocation category includes investments that historically would have been treated in the fixed-income allocation. However, by assessing the risk level of those investments, they are more logically included in the higher-risk component. This investment allocation results in a portfolio more closely aligned with the clients’ risk tolerance.

This example flips the percentage allocation between equities and fixed income to arrive at a 60 percent allocation to fixed income and other lower risk investments. In doing so, it is important to estimate how each investment will contribute to the required MAR. While each adviser might reach different conclusions on the forward-looking estimated returns, it is possible that a lower-risk model could be created to achieve client goals. After all, isn’t the goal of a well-designed asset allocation to achieve client goals with a minimum of risk?

Bob Dylan’s lyric, “The times they are a changing” is an apt description of the economy’s impact on portfolio design. New thinking and new portfolio construction techniques are needed to help clients meet their investment goals.

Joel H. Framson CPA/PFS, MBT is president of Silver Oak Wealth Advisors, LLC in West Los Angeles. You can reach him at (310) 207-4800.