Guest Article

Investment Selection and Monitoring: A Practical Approach to Best Practices

By Donald Stone, AIF®. Mr. Stone is President of Plan Sponsor Advisors, LLC, an investment consulting firm and registered investment advisor that works exclusively with qualified retirement plans. You may reach Don at: dstone@psaretire.com, call 312.214.1500 or visit their website at www.psaretire.com.

As investment consultants, Plan Sponsor Advisors, LLC deals with these questions every day. In meeting with plan sponsors and reviewing plans, their oversight process (or lack thereof) and their asset allocation, we have become convinced of the need for a best practices approach. A recent study by Elton, Gruber and Blake of 400 plans found that:

• The investment choices offered by 401k plans was inadequate in 62% of the plans studied - that is 62% of plans studied did not offer effective diversification across broad asset classes.
• For a given category (large value, for example) the 401k plans studied tended to offer fund options with more risk than the category average.
• Most plans that offer an S&P index funds had higher expenses for the fund than participants were able to get on their own outside the plan.

These statistics are particularly disturbing when you realize that assets in a 401k constitute the single largest financial asset for 50% of all plan participants. Any plan utilizing a best practices approach to fund selection and monitoring should avoid each of these pitfalls.

Below we have outlined what we believe is a best practices approach to investment oversight, including:

• the process a plan committee should follow;
• key considerations in developing an asset allocation; and,
• investment analytics to support fiduciary decisions.

Plan sponsors may legally be able to (and often do) do less. We are not lawyers, but as investment consultants we feel obligated to consider those elements that may mitigate fiduciary liability, head off potential costly litigation before it happens, and provide participants with a superior benefit plan.

What ERISA Requires

Plan Sponsors often struggle to understand what they need to do to fulfill their duties as a fiduciary. The Employee Retirement Income Security Act of 1974, better known as ERISA, says that plan fiduciaries are held to the standard of a "prudent expert" unless they hire a professional "with knowledge of such matters" to assist them (§404(a)). In §404(a)(1)(B) ERISA requires fiduciaries to determine a reasonable asset allocation and ensure that the investment furthers the purpose of the plan.

The Department of Labor (DOL) has made it clear that in enforcing ERISA they will not judge fiduciaries on the results they achieve but rather by their process they follow. But process is not static. What might have been a reasonable process in 1974 would almost certainly not be today. And best practices of just a few years ago, may not be today. Fred Reish, a nationally known ERISA attorney, recently commented, "the expectations of the performance of plan fiduciaries and committee members, is increasing." This evolving standard may catch some plan fiduciaries unaware and expose them to unnecessary liability because they have not reviewed and revised their process.

Background

At the time that ERISA was enacted, most retirement plans lacked professional management. Security selection was often an informal process conducted by company executives on a part-time basis. Even professional money managers lacked many of the analytical tools available today. ERISA's enactment jump-started the move to professional asset management, while the development of a broad range of software tools over the succeeding decades has created a much more rigorous and demanding process than was followed in past years.

It is hard to believe (even for those of us who witnessed each step along the way) that when ERISA was enacted there were no personal computers to crunch the numbers. The investment analytics that were available were computed by hand or on a mainframe computer. The Sharpe Ratio, developed in 1965, was not yet in general use and the ubiquitous nine style boxes devised by Morningstar were years away from invention. In the mid '80's when 401k plans were being formed, investment returns were reviewed, but true investment analysis was typically done only for defined benefit plans.

What we have seen in the 30 years since ERISA's enactment is an evolving process and a continuing growth of, and accessibility to, sophisticated investment measurement tools. The regulatory and litigation environment continues to evolve as well, most notably with the enactment of Sarbanes-Oxley and its governance requirements. Sarbanes' requirements for public companies are rapidly being adopted, in limited form, by many private companies, and the governance issues are applicable to the oversight of virtually all retirement plans. In addition, there are a number of high-profile on-going lawsuits pertinent to participants and their retirement plans, including Enron. Enron is establishing precedent on who is a fiduciary in addition to addressing issues around company stock in the plan. Whether a company is public or privately held, these are watershed events for plan sponsors of retirement plans. In short, what was considered a good process in 1980, or even 1990, might well be considered imprudent today.

A Best Practices Approach for Today

So if the standards have changed over the years, what is a best practice approach for today? That is, what combination of process and analytics will provide fiduciaries with the knowledge they need to make informed decisions, and what is most likely to withstand challenge by the Department of Labor or plaintiff's counsel? By best practices, we are including not only investment analytics that should be used to choose and monitor investment managers or funds, but also non-technical processes or actions taken by fiduciaries to fulfill their duty and thereby mitigate liability. That process is supported by the analytics, but neither part can stand alone.

A best practices approach lends itself to a natural breakdown into three areas: 1) procedural process, 2) asset allocation, and 3) on-going monitoring analytics. Procedural process deals with how the fiduciaries go about the process of selection and oversight. Asset allocation deals with the actual selection of asset classes for the plan and how they relate to each other. On-going monitoring analytics focuses on how funds or managers are actually measured over time.

Procedural Process

A best practices fiduciary process should include each of the following:

• A written investment policy statement, we believe, is the only way that an Investment Committee can demonstrate a thoughtful process. This is an area that has clearly evolved over the years. When ERISA was enacted probably fewer than 20% of plans had a written policy. This has been an area of some confusion as ERISA requires an investment policy but does not explicitly state that it be written. Today 68% (see Grant Thornton/Plan Sponsor Advisors 2004 survey of plan sponsors) have a written policy. Plan fiduciaries realize that it may be very difficult to demonstrate that there is a policy, and that it is consistently followed, unless it is in writing. Those that do not would appear to be taking an unnecessary risk. Having and following a written policy demonstrates procedural prudence over investments.

  The policy should have sufficient detail so that someone looking back years later (the DOL perhaps or plaintiff's counsel) could understand how a decision was made. Clear standards should be included for how investments are chosen, how they are measured and what will trigger placing a manager on a watch list. The roles of all interested parties should be clearly spelled out. At a minimum this will ensure that members of the Committee will understand what is required of them.

  The policy should be reviewed annually and modified as necessary. Common changes might include the addition of an asset class or a change in the appropriate benchmark for a fund or manager.

• Minutes should be kept of each review. What should be reported are: who was in attendance, topics covered, key questions or discussions and decisions made. Putting it in writing every time will demonstrate a thoughtful process and alleviate the need to rely on memory perhaps years later.

• Utilize an independent third party investment expert, unless the plan sponsor has someone or a staff that has the time, the tools and the expertise to perform accurate and independent analysis. Too often plan fiduciaries rely on their vendor to provide reporting and recommendations for analysis, placing funds on watch or replacing funds. Many vendors have reporting programs that are impressive on the surface, yet analytical tools do not remove the inherent conflict of interest of vendors reporting on proprietary funds, sub-advised funds and even nonproprietary funds where long-term business relationships and revenue agreements entwine with the investment decision process.

• Quarterly reporting and continuous monitoring should be the standard for all but the smallest plans where this may not be financially feasible. Quarterly reporting is usually sufficient since retirement plans are long-term investments and the focus should be on trends over a 3 to 5 year period. Continuous monitoring ensures that if something material happens between reporting periods (a regulatory problem for the manager, for example) that requires immediate attention, the Committee will be apprised of it and can take appropriate action.

• Reporting should tie directly to your investment policy - not just provide fund performance information as is so often the case. We believe that this is one of the most important elements missing from the monitoring of many plans. The investment policy is the road map to achieving the objectives of the plan. Every Committee action should directly reference back to the policy and monitoring should tie to specific metrics in the policy. This will demonstrate a clear and coherent process.

• Reporting should include a broad range of analytics. Fiduciaries should review a broad range of quantitative as well as qualitative measures for each fund or manager. In addition, fiduciaries should analyze how the complete investment menu fits together to accomplish the purposes of the plan. It is the third party's job to educate the fiduciaries and remind them what the analytics mean for the plan and its participants.

• Funds that don't meet the monitoring criteria should be replaced. The decision of how much time to allow an under-performing fund to improve its performance is not fixed and should take into account a number of factors. But many fiduciaries simply continue to shy away from making decisions about poorly performing funds, often choosing to simply add additional funds in the same asset class. At best this is a sloppy process that can send mixed signals to participants. At worst it demonstrates the fiduciary's unwillingness to perform the duties imposed by ERISA.

Asset Allocation

Too often plan fiduciaries simply view the 404(c) requirement to offer a broad range of funds as simply a numbers game - offer enough funds, the thought goes, and fiduciary liability is mitigated. From an investment best practices point of view, we believe that this approach is flawed. Simply offering a lot of funds does not necessarily mean that participants are being offered a broad range of funds in the sense that broad connotes funds fulfilling different points on the risk-reward spectrum. In fact, based on the study by Elton, Gruber and Blake cited earlier, most plans may not offer a broad selection of funds in the investment sense, even though the average plan offers 12-14 funds according to most studies.

The process of analyzing investments is not a disjointed exercise that only evaluates each manager or fund independently. Rather, the process is a more elaborate construct that evaluates not only each manager/fund against a number of relevant metrics, but also evaluates how the various managers/funds fit together as a coherent whole (or not).

In a defined benefit plan this allows the construction of an efficient asset allocation to meet the needs of the plan. In a self-directed plan [401k, 403(b)] it provides participants with the raw material to construct a reasonable asset allocation across the risk-reward spectrum. This helps ensure that participants will not inadvertently increase risk by choosing multiple funds with the same securities, or a different asset class than they expected. It also provides participants with the opportunity to choose funds that are not highly correlated with each other, mitigating volatility in the participant's portfolio. Utilizing mostly funds from one fund family should generally be avoided for this reason. A 2004 New York University - Emory University study found a very high correlation among funds for many of the largest mutual fund families.

Plan fiduciaries should construct an investment menu (or review the current menu) to see that:

• The menu of choices is not highly correlated;
• The menu covers the spectrum of risk and reward; and,
• The menu does not inadvertently tilt toward a particular asset class exposure by offering multiple fund offerings in one or two asset classes, simply because the funds were available.

Investment Analytics

We have identified ten measures that fiduciaries should evaluate on an on-going basis as a best practice. We have also indicated the frequency for monitoring each. As noted above, each of these measures should be tied back directly to the investment policy. None of these should be seen as absolutely determinative as to whether a manager should be replaced. If one member of a team of investment managers leaves a fund, the fund should be placed on a watch list and further investigation performed. Similarly, if a fund under-performs for one or more quarters, the reasons for the under-performance need to be understood.

Qualitative

• Regulatory oversight: Each investment manager must be a regulated bank, an insurance company, a mutual fund organization, or a registered investment adviser to comply with ERISA. Continuous.

• Stability of the organization: There should be no perceived organizational or significant legal problems. The same portfolio management team should be in place for at least three years. Continuous.

Quantitative

• Asset Allocation: Often neglected in the process of analyzing funds is how the entire group of managers or funds chosen relates to each other. An appropriate allocation is one that, looked at in its entirety, has asset classes with low correlation to each other. That is, asset classes are chosen that tend not to move in tandem with each other. Additionally, underlying securities should be reviewed for overlap. Funds or managers with more than 20% of the same securities indicates that the funds in question may not provide adequate diversification. Annually.

• Correlation to style or peer group: The product should be highly correlated to the asset class of the investment option. This is one of the most critical parts of the analysis since much of the remaining due diligence involves comparisons of the manager to the appropriate peer group. We look for a correlation of 80% or higher. Quarterly.

• Performance relative to a peer group and appropriate benchmark: The product's performance should be evaluated against the appropriate benchmark and peer group's median manager return, for 1-, 3- and 5-year cumulative periods. Funds or managers chosen should typically be top quartile performers for all periods, and existing funds in the menu should be in the top two quartiles. Quarterly.

• Performance relative to assumed risk: The product's risk-adjusted performance (standard deviation and Sharpe Ratio) should be evaluated against the peer group's median manager's risk-adjusted performance. Funds or managers chosen should typically be top quartile performers for all periods, and existing funds or managers should be in the top two quartiles. Quarterly.

• Minimum track record: The product's inception date should be greater than three years, and the managers should have a minimum of three years managing the proposed asset class (if not the specific product) with a verifiable track record (composites are acceptable). Quarterly.

• Assets under management: The product should have at least $75 million under management (large plans may need to set a higher minimum, and plans should avoid comprising more than 5% of any fund). Quarterly.

• Attribution analysis: The screened product should have no more than 20% of the portfolio invested in "unrelated" asset class securities. For example, a Large-Cap Growth product should not hold more than 20% in cash, fixed income and/or international securities. Quarterly.

• Expense ratios/fees: The fund's expense ratio or manager's fees should not be above the median of its peer group (exceptions may be made for funds or managers that consistently provide superior performance). Quarterly.

Conclusion

In a brief paper such as this, it is impossible to cover in detail how investment oversight and selection should work in a best practices approach, and this paper is not meant to be exhaustive. Even so, we recognize that what we propose may represent a significant change from the current practices for many plan fiduciaries, and many will be reluctant to change.

But in a decade that may well see single digit average returns, plan fiduciaries face increasing scrutiny from participants and regulatory agencies. Implementing a best practices process for investment selection and on-going monitoring can mitigate fiduciary liability. Just as importantly, a sound investment monitoring process will ensure that participants have a better chance of reaching their retirement goals.

###

401khelpcenter.com is not affiliated with the author of this article nor responsible for its content. The opinions expressed here are those of the author and do not necessarily reflect the positions of 401khelpcenter.com. This article is for informational and educational purposes only and doesn't constitute legal, tax or investment advise.

About | Glossary | Privacy Policy | Terms of Use | Contact Us

This work is licensed under a Creative Commons Attribution-NoDerivatives 4.0 International License.