Home > Library > Stable Times > Volume 11, Issue 2  

Wharton Professor Makes a Case for Stable Value

Wharton Finance Professor David Babbel pointed out three common misconceptions that make it harder for 401(k) investors to understand stable value and achieve retirement goals. He spoke at SVIA's Spring Seminar in Charleston, South Carolina. Correcting these misconceptions sets the stage to compare stable value funds to other asset classes used by 401(k) investors.

Financial Markets Have Changed and So Have Anticipated Returns

While most financial gurus urge investors to rely upon stocks, Babbel cautions that today's financial markets are very different from those of the past. He points out that the assumption that equities will continue to provide a sizeable return premium over the long term has been challenged on several fronts. In fact, the warnings in most prospectuses that the past history of returns is not a reliable indicator for future returns is equally applicable to the future returns of equities.

Most research on equities looks at 30 year-periods to provide historical returns, explains Babbel. Since 1926, however, when reliable data first became available, there have been only two non-overlapping periods of 30 years, which is not considered a meaningful statistical standard, says Babbel.

In addition, Babbel notes that in shorter holding periods, equities often exhibit lower returns and higher risks. These shorter periods are particularly important, because the 30-year periods are relevant only for workers who have at least 30 more years for their investment period. The number of workers who have a 30-years-or-more investment horizon is an increasingly small fraction of the workforce due to the aging of the population and increased employment mobility.

Babbel further explains that past equity returns reflected an economy that was markedly different from today's. That is also why most financial economists today expect an equity return premium that ranges from 2 percent to 3 ½ percent over Treasuries rather than the historical 6 ½ percent to 8 percent.

Using the Simplest Asset Allocation Model May Leave Savers Short

While mean-variance modeling may be easy to perform and simple for 401(k) investors to understand, Babbel cautions that reliance on this popular means of asset allocation does not take into account investors' aversion to loss. Mean-variance is also of limited use as it may be accurate for only short periods of time and is useful only for normal return distributions. In Babbel's analysis of common 401(k) assets--stable value funds, bond funds, money market funds, and stocks--there was less than one chance in a million that the assumption of the normal bell curve of returns was warranted. To the contrary, the returns on bonds, money markets, and stocks showed skewed returns and extreme returns that are not captured adequately by normal distributions, which underlie all mean-variance portfolio approaches.

Babbel elaborated that the shortcomings of mean-variance modeling are well known. Since Fred Arditti demonstrated in 1967 that investors care about risk and that return measures are not adequately described by simple mean-variance measures, economists have recognized that a more robust characterization of asset returns is necessary to make asset-allocation determinations. Mean-variance rankings of portfolio opportunities fail to distinguish some investments that would be selected by any and all rational investors because of the limited use of return-distribution information that is made with the mean- variance approach.

Stable Value Is Not Like Money Market or Bond Funds

Stable value funds are often grouped together with money market funds in surveys and studies, but there are significant differences. Babbel declares this assumption to be fundamentally wrong and explains that while both are intended to provide stability of principal, money market funds invest in shorter-term instruments. This results in lower and more variable investment returns.

By investing in a diversified portfolio of GICs and/or intermediate-term, investment-grade bonds, stable value funds achieve significantly higher average interest rates and provide consistent, predictable growth over the long term. "These are important characteristics to investors, especially those investing for retirement," emphasizes Babbel.

Stable Value Stacks Up

Babbel's analysis of how stable value funds compare with money market funds, bond funds, and stock funds in producing retirement wealth used three widely accepted methods in finance and economics: mean-variance analysis, stochastic dominance analysis, and multi-period utility analysis. Each method has its advantages and shortcomings, stressed Babbel. When used together, however, they form a powerful set of tests that can help determine which asset classes are most suitable for people accumulating retirement assets, explains Babbel.

Using the limited mean-variance analysis, Babbel found that stable value funds had superior risk-return characteristics to equities, money market funds, and bond funds. Babbel reports that no combination of money market fund investments, bond fund investments, or stock fund investments were able to achieve the same level of returns with such low risk that was achievable through stable value funds.

Babbel next examined the asset allocation strategies using a more robust technique known as stochastic dominance analysis. This technique was developed in the 1970s to address the severe shortcomings of the mean-variance approach, and it is relied upon to this day, says Babbel. It has the distinct advantage that the investment analyst need not know the particulars of the preferences of the investors. Rather, the analyst needs to know only the investor's preferences for wealth and risk.

Using the more rigorous stochastic analysis, there was no case in which investments in stocks, bonds, and money markets dominated each other, or dominated investments in stable value funds. However, the analysis did demonstrate that stable value funds dominated (i.e., was preferred to) money market funds and bond funds. This result held true for all possible forms of preference functions that an investor might have, as long as the investor preferred greater wealth to lesser wealth and lesser risk to greater risk. The study also considered investors who preferred positively skewed returns to negatively skewed returns, as well as investors who were averse to "fat-tailed" return distributions, under which extreme results (either positive or negative) are more likely to occur.

Lastly, Babbel used a multi-period utility investment framework to evaluate portfolio performance. This framework was developed at Berkeley, Yale, Wharton, and the Massachusetts Institute of Technology. Under this approach, Babbel explains, a family of preferences is used that has an attractive characteristic. It is termed "investor myopia." Investors that follow myopic behavior feel that it is consistent with asset allocation optimality over long horizons as well. Babbel explains that this family of preferences is the most widely used in the financial economics literature today.

The risk and return distributions of equity, bond, money market, and stable value investment funds were entered into the optimal asset allocation program for analysis. For almost all levels of investor risk aversion, the model found that the optimal asset allocation would consist either of stable value funds, equity funds, or some combination of both. There was no balanced fund, combination of equities and money markets, or combination of equities and bonds that beat stable value, notes Babbel. Furthermore, multi-period utility analysis found no empirical support for the life-stage funds that have become popular over the past three years.

Importantly, says Babbel, the third approach found that stable value funds enable investors to better plan for the future by providing consistent, competitive returns and principal stability. These stable value fund characteristics are also less likely to cause an investor to "turn off" saving in a 401(k) plan as negative and volatile earnings are minimized. Minimizing the "turn off" factor is an important consideration in selecting a default option for a plan sponsor who is trying to encourage a non-participating employee population to save in the 401(k) plan, noted Babbel.

About the Study

The study, which is sponsored by SVIA, is expected to be released this summer. It examines the risks and net returns of various assets. Stable value net returns were developed from data supplied by nine stable value managers who manage commingled funds, separate accounts, and full-service funds representing $189 billion in assets. The study was conducted by Doctors David Babbel and Miguel Herce. Doctor Babbel is a Professor of Insurance and Risk Management and a Professor of Finance at the Wharton School at the University of Pennsylvania and a Vice President and Senior Advisor at Charles River Associates International (CRAI). Prior to joining CRAI, Doctor Herce served as a Professor of Econometrics at the University of North Carolina at Chapel Hill.

Read Next: Editor's Corner