Improving on the participant-directed 401(k) investment model...
Much of the retirement plan community has been focused for over a decade on participant-directed 401(k) plans. As competition for 401(k) plan business has intensified among plan providers, there has been greater emphasis placed on the marketing of various "bells and whistles" such as ever larger menus of investment options, 24/7 online participant account monitoring and investment direction as well as individual brokerage windows. The increased attention on daily market activity, trading of individual stocks (or narrow sector mutual funds) and the ability to trade participant accounts on a daily basis have all worked to frustrate what the authors believe should be the fundamental, underlying objective of a participant-directed 401(k) plan: to provide those conditions that allow plan participants the best opportunity for a successful investment experience so that they can retire comfortably.
The likelihood of investment success increases as the participant's involvement in investment decisions decreases. This article, therefore, proposes an alternative approach to the participant-directed 401(k) investment model in order to allow – as far as possible – the fulfillment of the primary objective of a 401(k) plan. Our proposal, which is the Non-Participant-Directed 401(k) Plan, helps satisfy two important objectives at once:
- Increase the odds that the retirement investment and savings needs of plan participants will be met.
- Reduce the fiduciary responsibility of fiduciaries.
The Current State Of Affairs
The underlying objective of a 401(k) plan should be to provide the conditions that allow plan participants the best opportunity for a successful investment experience so that they can retire comfortably. However, there is increasing evidence that participant-directed investing through 401(k) plans is far less successful than many plan sponsors, plan participants and regulators had thought. This evidence is based on a large number of academic studies and widespread anecdotal corroboration. In short, the practice of allowing participant direction within a 401(k) plan is generally not a good one for helping plan participants retire comfortably.
Studies that examine the ways in which 401(k) plan participants direct their investments show how inadequate the participant-directed 401(k) investment model really is:
- The more extensive or complicated a plan's menu of investment options is, the more likely participants are to be "overloaded" with choices and unable to properly invest their retirement accounts. [See "Offering vs. Choice in 401(k) Plans: Equity Exposure and Number of Funds" by Gur Huberman and Wei Jiang, Columbia Business School working paper, September 2004 draft.]
- Many participants remain invested – through sheer inertia – in the default investment option of their 401(k) plan.
- We have found that many participants spend more time planning their annual vacations than analyzing how to invest their 401(k) plan accounts.
- Participants who trade their own accounts through so-called "brokerage windows" may suffer particularly poor investment performance due to higher transaction costs and the greater risk of more active trading. [See, "Do Investors Trade Too Much?" by Terrance Odean, American Economic Review, Volume 89, December 1999, pages 1279-1298, http://faculty.haas.berkeley.edu/odean/papers/overconf/Dolinvestors.pdf. See also Chapter 14, "Self-Directed Brokerage Accounts Reduce Success" in Retirement Success by Dr. Gregory W. Kasten (Lexington, KY: Unified Trust Company, NA 2004).
- When left to their own devices, many participants make poor investment decisions.
The Poor Investment Performance Of Plan Participants And Residual ERISA Section 404(c) Liability
This last finding has been well documented for over a decade by DALBAR Financial Services, a Boston-based financial consulting firm. DALBAR has conducted annual studies since 1994, which consistently find that the average mutual fund investor [DALBAR Quantitative Analysis of Investor Behavior [QAIB] Glossary & Methodology "The average investor refers to the universe of all mutual fund investors whose actions and financial results are restated to represent a single investor. This approach allows the entire universe of mutual fund investors to be used as the statistical sample, ensuring ultimate reliability."] underperforms market benchmarks by significant margins. The 2003 study, [See, 2003 DALBAR QAIB] for example, found that during the period of 1984-2002 the average annual return of the S&P 500 index was 14.3% while the return of the average stock mutual fund was 11.5%. This gap is explained by the costs of investing in mutual funds, including explicit costs such as commissions and annual expenses, and implicit costs such as trading costs. The return of the average investor in a stock mutual fund, however, was just 4.2%, underperforming the S&P 500 index by about 10 percentage points per year. Note that nearly all of this period included the most sustained stock market boom of the 20th Century.
The conscientious sponsor of a 401(k) plan can actually do something meaningful about closing, in this example, the approximately 3 percentage point gap between the return of the benchmark index and the return of the plan's average stock investment options: simply offer low cost, passively managed mutual funds which include index funds and asset class funds. In a world where two of the most successful long-term professional money managers, Sir John Templeton and John Neff, outperformed the market annually on average by 2 to 3 percentage points, eliminating 50% (i.e., 1.5 percentage points) of an approximately 3 percentage point gap by merely offering a menu of low cost, passively managed investment options in a 401(k) plan can sometimes make all the difference between a comfortable retirement and a difficult retirement.
Reducing investment costs increases investment performance. Reducing costs can, over time, generate large additional amounts of money in a 401(k) retirement account. For example, assume an 8% average annual return over a 35-year period and costs of 3%. A 1-percentage point decrease in investment costs (to 2%) increases retirement benefits by 39%. Using the same assumptions, a 2-percentage point decrease in costs (to 1%) increases benefits by nearly 94%. We think that most people would prefer to have, say, $1,000,000 rather than $500,000, or $500,000 rather than $250,000 as they enter retirement. It is important to remember in this context that participants in 401(k) plans have no power to reduce the costs in the investment options offered to them; only plan fiduciaries have that power by offering low-cost investment options.
What about the much more significant 10 percentage point gap found by the DALBAR study between the return of the benchmark index and the return of the average participant-investor? Even the most conscientious plan sponsor cannot do much about that gap. The relatively small gap between the benchmark return and the return of the average fund, as noted, can be reduced by offering low cost 401(k) investment options. It has become increasingly clear, however, that bombarding plan participants with ever more investment education is not going to make a serious dent in the relatively large gap between a benchmark index return and the return of the average participant-investor. The conclusion of the 2003 DALBAR study pinpoints why: "Investment return is far more dependent on [investor] behavior than on fund performance." As the cartoon strip character, Pogo, said: "We have met the enemy, and he is us." As a result, many plan fiduciaries, participants and their advisors, as well as regulators, have come to a harsh realization: plan participants generally are not successful investors.
Even though a 401(k) plan may comply fully with all the numerous requirements of ERISA Section 404(c), that fact alone does not relieve the plan's fiduciaries and its investment committee that selected the plan's menu of investment options from fiduciary responsibility for the prudent selection and monitoring of such options. There is an ongoing fiduciary obligation on the part of plan sponsors and fiduciaries, which generally goes unnoticed, thereby engendering ongoing fiduciary liability. The astute plan advisor will continually remind plan sponsor and fiduciary clients that even full compliance with the 404(c) rules by itself will not fully insulate them from liability.
Our Proposal: A Non-Participant-Directed 401(k) Plan
Those who agree with our view that the participant-directed 401(k) investment model does not do a very good job of either ensuring participant investment success or protecting plan fiduciaries from potential ERISA liability should consider the advantages of eliminating participant investment discretion.
Under our proposal, which we call the "Non-Participant-Directed 401(k) Plan," participants would no longer be responsible or able to direct the investment of their retirement accounts. Instead, the plan fiduciary would retain a bank, trust company, insurance company or registered investment adviser (RIA) to manage all the assets of the plan as an ERISA-defined "investment manager."
At first blush, our proposal seems to be a return to the good old days when most plans were invested on a "pooled" basis either by a discretionary trustee or by an ERISA investment manager, which was typically an RIA. Unlike the approach that would have been taken in the past with a pooled profit sharing plan, in our proposal the investment manager does not invest the entire plan account utilizing a single asset allocation as though there were only one participant or only one appropriate asset allocation for all participants.
Avoiding a pooled investment approach is crucial given the great importance of portfolio asset allocation to each plan participant. The importance of asset allocation was well documented in the 1980s and 1990s by the landmark Brinson studies. The 1986 Brinson study examined the investment performance of 91 large pension funds for the ten-year period of 1974-1983. The study concluded that more than 90% of the variance of a typical fund's investment returns across time is explained by its asset allocation. [See, "Determinants of Portfolio Performance" by Gary P. Brinson, L. Randolph Hood and Gilbert L. Beebower, 42 Financial Analysts Journal, July/August 1986, pages 39-44. See also "Determinants of Portfolio Performance II: An Update" by Gary P. Brinson, Brian D. Singer and Gilbert L. Beebower, Financial Analysts Journal, May/June 1991, pages 40-48 and "Asset Allocation Revisited" by William E. O'Rielly and James L. Chandler, Jr., Journal of Financial Planning, January 2000, pages 94-99.] William F. Sharpe, a Nobel Laureate in Economic Sciences, acknowledges the critical importance of asset allocation: "It is generally agreed by theoreticians and practitioners alike that the asset allocation decision is by far the most important made by the investor." [See, "Asset Allocation" in Managing Investment Portfolios: A Dynamic Process edited by John L. Maginn and Donald L. Tuttle (Boston, MA: Warren, Gorham & Lamont, second edition, 1990), page 7-3.]
Our proposal envisions hiring an investment manager such as an RIA to:
- Take on full discretionary investment authority with respect to all the assets of the plan.
- Create or manage sub-accounts which reflect appropriate asset allocations for various types of plan participants with differing risk tolerances and investment time horizons.
An RIA could structure its investment program to provide, for example, as many as five or six different managed portfolios to take into account the fundamental differences in risk tolerances and investment time horizons of the various participants in a 401(k) plan. In other words, each participant's individual account would have an investment manager, responsible for selecting investments that fit the needs of each particular participant.
Our proposal would, therefore, require the RIA to obtain some information from each plan participant (perhaps in the form of a questionnaire and/or an interview) to determine what investment portfolio would be most appropriate for the participant. Plan participants would not select the investment portfolios used for investing their individual accounts. Instead, we envision a professionally managed 401(k) plan, but one that takes the individual risk tolerances and investment time horizons of each plan participant into account.
More specifically, we envision offering an appropriate risk/return-balanced range of five or six model portfolios comprised of low cost, broadly diversified and automatically rebalanced passively managed mutual funds designed in accordance with leading academic research to be both prudent and diversified. In our approach, an ERISA-defined investment manager would select investments (with information provided by the plan participant). Plan participants will have the ability under our proposal to periodically provide input to the ERISA-defined investment manager on meaningful issues such as changes in their personal circumstances (i.e., risk tolerance and investment time horizon) over which they have control, as opposed to changes in the value of financial markets over which they have no control. On the other hand, the ERISA-defined investment manager is qualified to make informed investment decisions as long as it receives relevant information from the participants. Our proposal includes a mechanism for participants to update their individual information on an as needed basis. Imagine how powerful and efficient the process can be when participants document their own information and provide that regularly to the ERISA investment manager. The result is an investment for the participant that has a risk/return profile appropriate to the participant's own unique situation, including an appropriate risk tolerance and investment time horizon. Some participants would be placed in aggressive plan investment options and others would be placed in conservative investment options, while yet others would be placed in investment options somewhere in between.
A number of principles of Modern Portfolio Theory, a large body of academic and empirical work on investing, were incorporated into ERISA by Congress in 1974. The model portfolios that we envision take maximum advantage of this Nobel Prize-winning work. For example, the most efficient and effective way to diversify a portfolio, according to Modern Portfolio Theory, is to eliminate as much "uncompensated" risk from it as possible. [For a discussion of "uncompensated" risk and "compensated" risk, please see page 37 of The Prudent Investor Act: A Guide to Understanding by W. Scott Simon (Camarillo, CA: Namborn Publishing Co., 2002).] Passively managed mutual funds are the ideal investment vehicles for eliminating as much as possible the uncompensated risk from a portfolio. Such funds, by definition, diversify virtually all risk that it is possible to diversify. Within each proposed model portfolio investment option, there is nearly "perfect" diversification of each passively managed fund within the portfolio (e.g., "vertical" diversification) and nearly perfect diversification among all such funds across the portfolio (e.g., "horizontal" diversification). [There is a good reason why Modern Portfolio Theory is not termed Modern Investment Theory: the relevant unit of analysis must always be the portfolio. Yet very few 401(k) plans offer portfolios as investment options. Instead, they offer stand-alone mutual funds or individual stocks as investment options. Such options can be likened to the parts that make up a car. The participants are asked, in effect, to assemble all the parts (i.e., stand-alone investment options) on their own in order to manufacture a car (i.e., a portfolio). With a model portfolio, the parts are already assembled for the participant; all that it needs to do is provide the ERISA-defined investment manager with enough information so that the manager can select the appropriate car from the menu of five or six investment options that comprise the cars sitting on a showroom floor.] There is simply no better way to diversify the risk of a portfolio. [Although it's not possible to invest in a mutual fund (or funds) that entirely rids uncompensated risk from the asset class (or asset classes) comprising a portfolio, passively managed mutual funds can come very close. For example, the Vanguard Total Stock Market index fund eliminates about 99.5% of the uncompensated risk from the asset class of all publicly traded U.S. stocks. A passive investor who invests only in the U.S. stock market, therefore, incurs virtually no uncompensated risk in its portfolio and very nearly achieves the "ultimate goal of diversification" described in the Restatement and as set forth by Modern Portfolio Theory. See The Prudent Investor Act: A Guide to Understanding by W. Scott Simon (Camarillo, CA: Namborn Publishing Co., 2002), pages 37-38.]
Our Proposal Is Better For Plan Fiduciaries
There are two important reasons why our proposal is better for plan fiduciaries:
- They are no longer responsible for the selection and monitoring of plan investment options [It is important to remember that plan sponsors and fiduciaries are always on the hook for the oversight responsibility to ensure that investment managers and investment options are prudently selected and monitored.]
- They no longer need comply with the myriad rules of ERISA Section 404(c).
Our proposed Non-Participant-Directed 401(k) Plan, as noted, contemplates the appointment of professional investment advisors and money managers to invest plan assets. ERISA permits the use of professional investment advisors and money managers to be responsible for the investment of retirement plan assets. Specifically, ERISA Section 402(c)(3) states:
…[A] person who is a named fiduciary with respect to the control or management of the assets of the plan may appoint an investment manager or managers to manage (including the power to acquire and dispose of) any assets of a plan.
ERISA also provides fiduciary relief to the so-called "named fiduciary" if such investment responsibility is properly delegated to an investment manager. ERISA Section 405(d) states:
If an investment manager or managers have been appointed under Section 402(c)(3) [29 USC §1102(c)(3)], then…no trustee shall be liable for the acts or omissions of such investment manager or managers, or be under an obligation to invest or otherwise manage any asset of the plan which is subject to the management of such investment manager.
Although ERISA Section 405(d) specifically provides relief from "trustee" liability, the legislative history of ERISA, relevant regulations and case law support the conclusion that Section 405(d)(1) shields named fiduciaries from liability as well. In the "Joint Explanatory Statement of the Committee of Conference," the drafters of ERISA explained "as long as the named fiduciary had chosen and retained the investment manager prudentially, the named fiduciary would not be liable for the acts or omissions of the manager." [H.R. Conf. Rep. No. 1280, 93rd Cong. 2d Sess. (1974).] Additionally, Department of Labor regulations state that named fiduciaries can delegate managerial authority over plan assets to an investment manager, thereby releasing the named fiduciary from liability for the acts or omissions of the person to whom authority was delegated. [29 C.F.R. Section 2509.75-8, FR-14 Q&A.] Finally, the Second Circuit has held that "the obligations of named fiduciaries with regard to their duty of care… can be reduced by the appointment of an investment manager under ERISA Section 402(c)(3)." [Lowen v. Tower Asset Management, Inc., 829 F.2d 1209, 1219, (2d Cir. 1987).] This makes it possible for the person or persons who are normally responsible for investing a 401(k) plan to turn over primary authority for the prudent investment of the plan as a whole to an investment manager.
The named fiduciary must prudently select and monitor the activities of the investment manager. Prudent selection of an appropriate investment manager requires the named fiduciary needs to:
- Evaluate the investment manager's qualifications (i.e., experience, education, securities registration, references).
- Ascertain the reasonableness of the investment manager's fees.
- Carefully review the documentation regarding its relationship with the investment manager [See, Whitfield v. Cohen, 682 F. Supp. 188, 193, 1988 U.S. Dist. LEXIS 1655, 9 Employee Benefits Cas. (BNA) 1739 (DC SWY 3/7/88).
The named fiduciary also must monitor the activities of the investment manager by receiving reports of the investment manager's activities on behalf of the plan on a periodic basis. (Id.)] This is far easier to do than retaining responsibility for the selection and monitoring of a plan's investment menu, let alone complying with the other requirements of Section 404(c).
Can anyone be an investment manager? As you might expect, the answer is no. According to ERISA Section 3(38):
The term ‘investment manager' means any fiduciary other than a trustee or a named fiduciary, as defined in §402(a)(2) [29 USC §1102(a)(2)]–
(A) who has the power to manage, acquire, or dispose of any asset of the plan;
(B) who (i) is registered as an investment adviser under the Investment Advisers Act of 1940; (ii) is… registered as an investment adviser under the laws of the State… in which it maintains its principal office and place of business…; (iii) is a bank, as defined in that Act [15 USC §§80b-1 et seq.]; or (iv) is an insurance company qualified to perform services…under the laws of more than one State; and
(C) has acknowledged in writing that he is a fiduciary with respect to the plan.
A named fiduciary, as a result, can largely delegate its overall investment responsibilities to investment professionals by properly retaining and monitoring an investment manager that accepts fiduciary status in writing in accordance with ERISA. In one fell swoop, then, two major objectives of a plan sponsor have been satisfied:
The plan fiduciaries with respect to plan investments are no longer responsible for the investment of a plan's assets.
Compliance with ERISA Section 404(c) is no longer necessary or appropriate since participants are not allowed to direct the investment of their individual accounts.
Our Proposal Is Better For Plan Participants
Our proposal is better for plan participants for two reasons. First, plan participants are relieved of the burden of trying to figure out how to invest in their 401(k) plan accounts. Although many participants would probably say that they like the idea of having investment control over their 401(k) accounts, most might be honest enough to admit that they are not the best qualified to manage such important investments. Second, based on widespread empirical data, most (if not practically all) plan participant accounts would generate greater investment returns with less volatility if the accounts are invested on a low cost and broadly diversified basis by a prudent investment manager who has agreed to accept full fiduciary responsibility in writing in accordance with ERISA.
What about the relative costs of the "traditional" participant-directed arrangement compared to the Non-Participant-Directed 401(k) Plan? Based on our analysis, we believe that the new approach will actually be less expensive – not to mention simpler. Here is why.
Our Proposal Is Better For Investment Consultants
Our proposal envisions a customized participant experience within the 401(k) environment that utilizes low cost, broadly diversified model portfolio investment options each of which is passively managed and appropriately tailored to a participant's individual circumstances. The four-step process of delivering this practical and cost-effective experience to participants is quite simple yet elegant in design.
- Gather participant-specific information, which the investment manager would use to select an appropriate model portfolio. Gathering this information could be automated through the use of technology or collected through enrollment meetings. Imagine how much simpler an enrollment meeting would be if the majority of the time is spent instructing plan participants to document their own personal situations instead of trying to explain to them concepts such as the "efficient frontier" and "standard deviation."
- Incorporate the plan participant information into the process of selecting an appropriate model portfolio for each participant. The investment manager would document the participant's information and the process used to select the portfolio.
- Send a communication to each participant reviewing the information provided and informing each participant of the model portfolio investment option that has been selected.
- Provide ongoing communication to each participant and a mechanism for the participant to interact with the investment manager when their personal circumstances change.
This process provides the investment manager with the necessary information to select the appropriate investment strategy, while eliminating the single most destructive element from most participant-directed accounts: participant investment discretion. According to the DALBAR study, as noted, about eight of the ten percentage points of underperformance suffered by the average investor-plan participant can be attributed to the poor investment decisions made by the participant.
Some may argue that gathering this information would be too costly or difficult. The fundamental objective of this process, though, is to provide a higher probability of investment success for each participant and to replace the burdensome 404(c) compliance process. Meeting these objectives to help increase the long-term benefits for plan participants seems to be a very fair trade-off against the amount of work required.
Once the emotion and guesswork of participants is taken out of the investment decision-making process by transferring the responsibility for selecting and monitoring participant accounts from participants themselves to an ERISA-defined investment manager, the great majority of the gap in performance between the average investor-participant and the appropriate market benchmark can be eliminated. What about reducing the less significant, yet still very costly, gap between the average stock mutual fund and the appropriate market benchmark? Providing low cost investment options for a 401(k) plan is a prudent answer for a very simple reason: every basis point reduction in investment costs translates directly into increasing the return of a portfolio with no added risk whatsoever! The use of low cost, passively managed model portfolio investment options in a 401(k) plan provides ample opportunity to reduce costs and/or to shift costs to more appropriate areas where additional services may be employed or needed.
Numerous studies and mountains of empirical evidence all have come to the same conclusion: past performance provides little (or no) help in predicting future results. There is, however, a large body of evidence linking cost to performance. [See, "On Persistence in Mutual Fund Performance" by Mark M. Carhart, Journal of Finance, March 1997, pages 57-82 and "Returns from Investing in Equity Mutual Funds 1971 to 1991" by Burton G. Malkiel, Journal of Finance, June 1995, pages 549-572.] This evidence shows that the higher costs associated with active investment management tend to be a major drag on performance. While no one can predict future performance, it is possible to accurately predict future costs through the examination of past and current costs. In fact, the expense ratios of mutual funds have had a strange yet very predictable tendency to remain the same – or even increase - despite the much larger economies of scale that have occurred in the mutual fund industry over the last 20 years.
A number of sources have placed the average cost of a retail equity mutual fund from about 150 to 160 basis points (1.50%-1.60%) per year. Through the use of collective investment trusts and institutional platforms it is possible to provide a prudent menu of broadly diversified model portfolios with investment costs ranging from 30 to 45 basis points (0.30%-0.45%) per year. This represents a potential cost savings of anywhere from 100 to 120 basis points (1.00%-1.20%). These portfolios implement sophisticated investment strategies comprised of multiple asset classes that even include stocks in companies that are traded in foreign markets and emerging markets.
What does our new approach mean in terms of additional retirement savings? Naturally we cannot guarantee the specific or individual results of our proposal. But our analysis (using some of the empirical data gathered by other researchers, professionals and academicians) indicates that the average participant in a 401(k) plan may achieve a significantly greater account balance through relatively small savings in costs as shown below:
$6,000 Annual Contribution
Accumulated Portfolio Values
The bottom line is that the participant-directed 401(k) plan investment model has not worked very well for many plan participants. As we have shown, the current approach of giving plan participants more and more investment options and encouraging more and more active trading is not necessarily a recipe for success.
In addition, far too much emphasis has been placed on complying with the rules of ERISA Section 404(c). Not only are those rules relatively expensive and difficult to comply with, but they really do not provide the kind of protection from liability that most plan fiduciaries are seeking.
Sponsors of participant-directed 401(k) plans should seriously reconsider the way in which their plans are currently offered to plan participants and how the participants invest. Our recommendation is to hire a bank, trust company, insurance company or RIA to serve as an ERISA-defined investment manager for a plan's entire pool of assets. The investment manager would have to agree to manage all the plan assets taking into account the varying risk tolerances and investment time horizons of as many as five or six classes of participants. The responsibility of the investment manager would be to manage the assets of the entire 401(k) plan and provide sub accounts (identified by class of participant) that reflect participant-appropriate portfolio asset allocations.
Taking the responsibility for investing their own 401(k) accounts out of the hands of plan participants would reduce most of the performance shortfall they experience. Performance shortfall could be further reduced by implementing our suggestion to use low cost, broadly diversified model portfolios of passively managed mutual funds. In other words, eliminating the participant from the 401(k) plan equation has the potential of eliminating 8 of the 10 percentage points of underperformance identified by the DALBAR study while adding low cost passively managed model portfolios can eliminate much of the other 2 percentage points of underperformance.
Our proposal is offered against the backdrop of often unintentionally self-destructive participant investment behavior, most of which cannot be affected by any amount of investment education from whatever source, and the difficulties and uncertainties that plan sponsors and fiduciaries experience in complying with the 404(c) rules. Our proposed Non-Participant-Directed 401(k) Plan could make a significant difference in increasing overall retirement savings for plan participants. If it can accomplish that, and at the same time significantly reduce the liability of ERISA plan fiduciaries, then it is well worth considering.
This article was co-authored with W. Scott Simon and Gary K. Allen.
W. Scott Simon, JD, CFP®, is a principal with Prudent Investor Advisors, LLC, a registered investment advisory firm in Chico, CA, and is an expert on the Uniform Prudent Investor Act and the Restatement 3rd of Trusts (Prudent Investor Rule). E-mail: firstname.lastname@example.org; phone: (760) 390-7874. Gary K. Allen is a principal with Prudent Investors Advisors, LLC, a registered investment advisory firm in Chico, CA, and has been a business editor and talk show host for radio station KNBR 680 (http://www.knbr.com) in San Francisco for nearly nine years. E-mail: email@example.com; phone: (530) 895-0636.
This article was republished with permission from the Journal of Pension Benefits, Summer 2005, Vol. 12, Number 4, © 2005, Aspen Publishers, Inc. All rights reserved. For more information on this or any other Aspen publication, please call 800-868-8437 or visit www.aspenpublishers.com.