Reinventing Retirement Income in America

Traditional defined benefit pension plans, which are managed by employers and which promise workers a specific monthly payment on retirement, are disappearing. Instead, more than 42 million workers now participate in defined contribution retirement plans, primarily 401(k) plans, which specify the annual contributions to an employee's pension fund. A worker's contributions to these plans, the employer's vested match and any interest or dividends reinvested in the plan are the worker's property. Workers are responsible for choosing among investment options, and the account can be moved when the worker changes jobs and can be passed on to heirs.

As now constituted, however, defined contribution plans are performing poorly. The consulting firm Watson Wyatt found that among 503 employers sponsoring both a 401(k) plan and a defined benefit plan during 1990-95, the defined benefit plans averaged an annual return that was 1.9 percentage points better than the 401(k) plans. Even the 401(k) plans sponsored by firms in the financial services industry have often had below-average returns.

• An examination of plans sponsored by five leading financial services firms reveals that from 1995 through 1998, none had returns that matched a simple index of 60 percent stocks and 40 percent bonds.
• Although these companies offer investment advice to the public, the investment choices of their own employees underperformed the market index by 3.2 to 10.5 percentage points.

Companies offer little assistance in making wise investment choices, fearing they will open themselves to lawsuits charging them with responsibility if employees have investment losses. Many employees with limited understanding of investing make uninformed investments that reduce their retirement income. For example, in examining a number of large plans we found:

• Within plans, regardless of years studied or geographical location, the lowest-paid 20 percent of participants receive the worst returns and the higher the participants' average pay, the higher their returns.
• Almost two-thirds of the money in the lowest-income quintile was in a money market fund or bonds.
• By contrast, about 85 percent of the money in the highest-income quintile was in equities.

If designed properly, defined contribution retirement plans offer the best hope for a comfortable retirement for most workers. The changes that are needed are both politically and economically achievable. To remedy the flaws, this study proposes that employers be encouraged to offer a new type of plan: the American Freedom 401(k) plan. In exchange, employers would enjoy "safe harbor" protection from certain kinds of lawsuits. To take advantage of the American Freedom 401(k) plan, employers would have to:

• Give participants the opportunity to have their funds invested in efficient portfolios (e.g., market index funds) or in portfolios managed by investment professionals.
• Automatically enroll all employees (both new and current) unless they opt out.
• Set a minimum contribution rate of 4 percent to 6 percent of income by the participant – an amount that could prudently be expected, when combined with the employer contribution, to provide a reasonable retirement income – unless the employee specifically opts for a smaller amount.
• Require plan sponsors to pay all plan fees and expenses, and to disclose them fully.
• Prohibit cashouts of 401(k) accounts by the plan or the employee following termination of employment and allow all funds to be rolled over into another qualified plan or to remain in the previous employer's plan if the new employer does not have a plan.
• Make vesting 100 percent and immediate.
• Prohibit loans or hardship distributions from an individual account but allow "hardship loans" from the plan's trust fund.

The American Freedom 401(k) plan would maintain the advantages of defined contribution plans, while producing higher returns and encouraging more saving for retirement.

[page]

Today, millions of working Americans participate in the fastest-growing pool of retirement assets ever created, defined contribution retirement plans. These plans allow workers to save for retirement in accounts invested in corporate equities (stocks) and bonds. Contributions, interest, dividends and growth may compound tax free until distribution in retirement. The principal form of these accounts is 401(k) plans sponsored by firms for their employees. In recent years, these accounts have grown with such speed and popularity that they are displacing traditional defined benefit pension plans.¹

Introduced in the United States in 1978, defined contribution plans now have assets exceeding $2 trillion. Assets of 401(k) plans far outweigh the assets in Individual Retirement Accounts (IRAs) and Roth IRAs.

Unlike IRAs, 401(k) plans were not created on purpose. Instead, they grew from a loophole in the Revenue Act of 1978.² Since then, amendments to the Employee Retirement Income Security Act of 1974 (ERISA), regulations promulgated by the Internal Revenue Service and the Department of Labor and court decisions have created a framework for defined contribution plans that works – but not as well as we would like, particularly for lower-wage employees.

"If designed properly, defined contribution retirement plans offer the best hope for a comfortable retirement for most workers."

In 2000, for the first time, the average 401(k) account lost money, shocking many account holders and raising questions about the investment approaches of both participants and plan sponsors. However, if designed properly, defined contribution retirement plans continue to offer the best hope of a comfortable retirement for the most workers. As we will show, design improvements can increase the participation rate of workers, improve returns on investments and reduce market risk.

This paper discusses how to establish more vigorous and secure retirement income for Americans through the free enterprise system. First, we look at recent trends that have had both financial and political consequences for our nation's retirement system. Second, we expose a potentially fatal flaw in the 401(k) market in the United States – one that must be corrected to improve overall retirement security. Finally, we offer common-sense recommendations that are politically and economically achievable. We believe these changes will reinvent retirement income in America.

[page]

Any discussion of retirement income must take into account U.S. and global trends: people are living longer and retiring earlier and thus need more retirement assets. At the same time, fertility is declining and the ratio of workers to pensioners is falling. Retirees will get less help from the next generation either through public or private means.

Longer Lives. Consider: life expectancy has increased 110 days annually for the past 100 years! Life expectancy at birth increased from about 50 years to about 80 years in the 20th century.³ As a result, retirement income has become a major issue around the world. It is not hyperbole to call this the Longevity Revolution.

Until the last century, people over age 65 never made up more than 2 or 3 percent of the population. In today's developed world, they account for nearly 15 percent of the population. They could reach 25 percent by the year 2030, closing in on 30 percent in some countries. According to United Nations projections:⁴

• By 2050, the number of people age 64 to 84 worldwide will grow threefold, from 400 million to 1.3 billion.
• The number of people age 85 and older will grow from 26 million to 175 million (more than a sixfold increase).
• The number age 100 and over will jump from 135,000 to 2.2 million (a 16-fold increase).

The longevity revolution represents a triumph of public health, modern medicine and new technology. The elderly and their families may treasure the extra years of life. Unfortunately, pension plans and other retirement benefit programs were not designed to provide additional years of benefit payments.

"People are living longer and retiring earlier."

Early Retirement. Although the "normal" retirement age enshrined in Social Security is age 65, a growing proportion of workers have elected to retire earlier – at age 62 or even 55. Indeed, from the 1970s to 1990s, many employers seeking to reduce payroll expense while improving productivity offered workers early retirement with fully vested pension benefits as if they had continued working to normal retirement age. Other workers volunteered to retire early with reduced pension benefits.

Social Security benefits are also reduced for workers who retire before the normal age – which under current law is set to gradually increase to age 67 for workers born after 1960. This has reduced the workforce participation of older workers.

Declining Fertility. At the same time that life spans are increasing, fewer babies are being born. This is part of a secular downward trend worldwide that has brought the global fertility rate – the average number of lifetime births per woman of childbearing age – to about 2.7. In the United States, the fertility rate has been declining steadily for more than two centuries, interrupted only by the baby boom after World War II. From a fertility rate of 8.0 in 1800, the U.S. today is fast approaching the replacement rate of 2.1 (the rate required merely to maintain a constant population).⁵ The fertility rate is already below that figure in most developed countries.

Ratio of Workers to Pensioners. Due to population aging and declining fertility, the ratio of working taxpayers to nonworking pensioners in the developed world is around three to one. By 2030, absent reform, this ratio will fall to 1.5:1. In some countries, such as Germany and Italy, it will drop to 1:1 or lower.⁶

What implications do these demographic trends have for retirement income? The primary, obvious one is that with longer life expectancies, people are going to need income for a greater number of years. Further, the prospect of getting less help from the next generation means that individuals will have to save at a higher rate, receive a higher rate of return on their savings or do both if they are to have enough retirement income to maintain a reasonable standard of living.

[page]

Three major elements determine the amount of one's retirement income: the ratio of years worked to years of retirement, the average amount saved during the working years and the average real rate of return on the savings.

"The ratio of years worked to years of retirement has shrunk from 15:1 to less than 2:1."

Working Years vs. Retirement Years. A few generations ago, the males in a typical family started working in their teens. They remained on the job into their 50s, when a final illness often contributed to what we now would call their retirement – which only spanned a few years. After 35 or more years of work, they would have perhaps three years of retirement before death, making the ratio of years worked to years of retirement 12:1.

The case today is quite different for many people. Those who pursue a college education do not begin full-time work until about age 22. Given current trends, the retirement target for many is age 62. As with previous generations, working years number about 40. But now the retirement years approach 25. In fact, in the case of a healthy employee and spouse both reaching age 65 and retiring in good health, the odds are 50-50 that one will live to age 90. Thus the ratio of years worked to years of retirement has shrunk from 12:1 to less than 2:1.

Average Amount Saved. According to a survey by the Profit Sharing Council of America, in 1999 non-highly compensated employees contributed an average of 5.4 percent of pay to their 401(k)s and the average company contribution was 4.7 percent. Between one-fifth and one-quarter of eligible workers do not participate, and participants who make only minimum contributions fail to take full advantage of the employer match.⁷

"Sufficient retirement income depends on a wise combination of years worked vs. years of retirement, the savings rate and the rate of return."

Average Real Rate of Return. This is an average of the annual rate of return on savings minus the annual rate of inflation. The figure most often used as the annual rate of inflation is the change in the Consumer Price Index, which measures the changes in prices of a fixed amount of selected goods and services. Thus if a person receives a pay increase of 4.25 percent and inflation is 3.25 percent, the real pay increase, in terms of added purchasing power, is 1 percent. After years of high inflation during the late 1970s and early 1980s, the U.S. annual inflation rate has varied from as much as 5.4 percent in 1990 to as little as 1.9 percent in 1986 and 1.6 percent in 1998.

Prospective Retirement Income: An Example. What kind of retirement income will a 401(k), combined with Social Security and private savings, provide?⁸ A successful replacement income model will have a wise combination of the three major elements – the ratio of years worked to years of retirement, the savings rate and the rate of return over inflation.

Let us look at a 25-year-old worker earning $25,000 a year.⁹ Assuming annual pay increases of 4.25 percent (and annual inflation of 3.25 percent), by age 65 he will be earning $126,744 per year. Figure I shows the difference the three major elements make.

• If the worker retires at age 65, he can expect to live 23 years in retirement, and thus will have 1.7 years of future work for each expected retirement year.
• With an average investment return of 5 percent (1.75 after inflation), if he contributes 5 percent to his 401(k) plan, he can expect a pension of $19,136 per year, or 15.1 percent of preretirement income.¹⁰
• If the worker doubles his contribution to 10 percent, it will double his pension to $38,271 per year, or 30.2 percent of preretirement income.
• If the rate of return also doubles to 10 percent (6.75 after inflation), the worker can expect a pension of $165,176 per year, or 30 percent more than his preretirement income.
• If the worker stays on the job until age 70, when his income will be $156,066, he will have 2.5 years of future work for each expected retirement year, and his pension will be $291,640 per year, more than four-fifths greater than his preretirement income.

[page]

"Traditional defined benefit pensions are disappearing."

Traditional defined benefit pensions, company-sponsored and company-paid, are disappearing. They have been replaced by defined contribution plans, funded mostly by employees themselves. A defined benefit plan specifies the monthly payment a retiree will receive. A defined contribution plan specifies the annual contributions to an employee's pension fund.

Defined Benefit Plans. Just a generation ago, the average employee probably participated in a defined benefit pension plan that essentially guaranteed a monthly pension at normal retirement age (nearly always 65) based on two things: (i) final average pay (FAP) and (ii) years of service (YS). Typically:

B = FAP x YS x .015

To illustrate this, assume John Doe was hired at age 30, has had an average annual pay of $80,000 between ages 60 and 65 and intends to retire at age 65. His annual pension would thus be $42,000 [$80,000 x 35 x .015], or $3,500 per month.

Advantages of Defined Benefit Plans. During his 35-year career, Doe will benefit from several very important characteristics of the defined benefit pension:

• He will be protected (both before and after retirement) from any market risk – i.e., whether the stock market soars or crashes, he will receive the same benefit (which is one reason these types of plans are called "defined benefit" plans).
• He also will be protected from preretirement inflation, since his benefit will be based on his average pay over the last five years worked.
• He will not be required to make a personal contribution to the plan and will pay none of the fees and expenses for investing the annual contributions made by the employer to the plan's assets (although economists agree that the employer's contributions are a substitute for wages).
• Finally, he will have no responsibility for managing the investment of the money contributed to the plan.

Disadvantages of Defined Benefit Plans. Defined benefit pensions also have drawbacks and risks. Since pensions traditionally rewarded loyal, practically lifetime service, they imposed vesting requirements. Vesting means that a worker may have to remain with one firm for a number of years, depending on the firm's vesting policy, before becoming entitled to any retirement benefits whatsoever. And workers may not be fully vested until they have worked at that same firm for five to seven years, again depending on the firm's vesting policy.¹¹

"Defined benefit plans penalize workers who change jobs."

As workers have become more mobile, those who change jobs have been serially shut out of pension plans tied to employment with one firm. Even a worker who changes jobs only infrequently and vests in each is affected because the pension benefit is always based on the final average pay of each job rather than the final average pay at retirement. The sidebar illustrates how this affects benefits for a worker who changes jobs.¹²

Although the pensions nominally are paid by the employer, some employers do not back their promises with sufficient invested funds. Thus although the employee is shielded from the direct risk of investing in the market, if the company fails in the marketplace the employee may not get all that was promised. Prior to the passage of the Employee Retirement Income Security Act of 1974 (ERISA), pensioners took their place in line behind other creditors of bankrupt companies. As a result, employees and retirees of some firms found themselves with reduced or nonexistent pension benefits. ERISA requires employers to maintain reserves to pay pension benefits, and many defined benefit plans are insured by the Pension Benefit Guaranty Corporation (PBGC), which collects and invests premiums from covered plans. However, even ERISA insurance does not guarantee that 100 percent of promised benefits will be paid. The maximum guarantee is $3,392.05 per month for someone retiring at age 65 and less for someone retiring earlier.

Defined Contribution Plans. Today the assets in a worker's defined contribution account determine retirement income. Consider John Doe II, now 39 years old and earning $32,750 annually. Assume that his pay will increase 4 percent annually; and that his average pay between ages 60 and 65 will be $80,846, or approximately the same as John Doe I in the old "replacement income" model. John Doe II's company has a 401(k) plan to which John contributes 6 percent of annual pay, with his employer matching 50 cents on the dollar (i.e., 3 percent). If Doe II's account earns an average of 10 percent annually over the 25-year period from age 39 to age 65,¹³ it will give him a benefit of $43,961 annually at retirement.

Thus Doe I will get $42,000 annually from contributions paid 100 percent by the employer, while Doe II, with a 401(k) plan, will get $43,961 annually, with most of that benefit paid by him personally.

Advantages of Defined Contribution Plans. Unlike Doe I, who can draw the maximum benefit only by staying with the same company throughout his career, Doe II can change jobs. And unlike with the defined benefit plan, Doe II's contributions plus the employer's vested match and any interest or dividends reinvested in the plan are Doe II's property and can be passed on to adult children or other heirs, even if there is no surviving spouse or dependent child.

Disadvantages of Defined Contribution Plans. All this comes at a price: John Doe II has to live with far more uncertainties. Some of the uncertainties:

• Doe II is not protected from market risk, so if the stock market plummets, especially during his last working years, Doe II could be in a very tough spot.
• Doe II is not protected from preretirement inflation, and if inflation roars during his last working years as it did in the 1970s, Doe II could see a significant reduction in the purchasing power of his plan account assets. (Of course if all prices rise, the prices of stocks in his account may rise too.)

"Defined contribution plans shift the investment responsibility to employees."

At the same time, Doe II's employer will enjoy a large reduction in the cost of providing Doe II's benefit – from the average defined benefit pension cost of around 7 percent of payroll just a generation ago to a cost of only 3 percent of payroll for Doe II. And instead of directly paying none of the plan's total fees and expenses, which can far exceed the 1.25 percent in our example, Doe II and the other employees will pay all of them. This does not mean that employers gain financially from defined benefit plans. Economists generally agree that wages plus benefits tend to equal the value of each employee to the firm. What it does mean is that the responsibility shifts. Regardless of Doe II's investment skills, Doe II has complete responsibility, albeit with a limited number of investment choices, for managing the money contributed to the plan to provide his benefit – even if Doe II is a novice (or worse).

Most contend that Doe II is being empowered with the opportunity to personally direct the investment of his own funds. Others believe that amateurs are being forced to direct the investment of their own funds.

[page]

Most 401(k) accounts have grown steadily in the past, thanks to a booming economy. But 2000 brought a different story. From 1999 to 2000, even with employee and employer contributions, the average account shrank from $46,740 to $41,919.¹⁴ This focused new attention on the plans and raised questions about the investment choices that plan participants have been making. We expect this will be reinforced by the market decline this year. Does the way workers have been making choices threaten the development of fully funded and portable pensions owned by them? An analysis of how well the plans are faring and of the investing behavior of participants signals the need for some revisions in the plans' operation.

"From 1990 to 1995, defined benefit plans averaged an annual return 1.9 percentage points better than 401(k) plans."

Performance of Company Plans. Each employee benefit plan must file a Form 5500 annually with the Internal Revenue Service, summarizing its results. Using a representative sampling of Form 5500 data tapes for the years 1990 through 1995, the consulting firm Watson Wyatt identified 503 employers that sponsored both a 401(k) plan and a defined benefit plan during all six years. The findings:¹⁵

• The defined benefit plans averaged an annual return 1.9 percentage points better than the 401(k) plans – 10 percent compared to 8.1 percent.
• For 10 percent of plan sponsors, the average annual return of the defined benefit plan exceeded that of the 401(k) plan by 5.1 percent or more.
• For 25 percent of plan sponsors, the average annual return of the defined benefit plan exceeded that of the 401(k) plan by 3.5 percent or more.

To illustrate what a 1.9 percentage point difference means to an account, if two individuals each contribute $4,000 a year to a 401(k) account for 30 years, one at an annual return of 10 percent and the other at an annual return of 8.1 percent, each will have contributed $120,000; however, the 10 percent account will grow to about $690,889 while the other grows to about $480,224 – a difference of $210,665 or 44 percent!

"Plans sponsored by financial service firms underperformed an index of 60 percent stocks and 40 percent bonds."

How Financial Services Firms' Plans Fare. Even the 401(k) plans sponsored by firms in the financial services industry have often had below-average returns. While all employees of these firms aren't financially knowledgeable individually, it is reasonable to expect that they would have a higher level of interest, concern and knowledge than employees at non-financial services firms. We looked at the performance of the plans of five well-known financial services firms for the period 1995 through 1998 (the latest year for which data are available). Since two of the five plans included the firms' own stock, our analysis of those two firms excludes that stock and considers only investments chosen by the employees.¹⁶ The firms are:

• Morningstar, the leading provider of mutual fund, stock and variable-insurance investment information and counted on by the 401(k) industry for unbiased data and analysis as well as candid editorial commentary.¹⁷
• Prudential, which offers a broad range of financial products and services for people in the United States and abroad, including mutual funds, annuities, pension and retirement-related services and administration.
• Hewitt Associates, a global management consulting firm specializing in human resource solutions, including helping companies set up employee benefit plans.
• Citigroup, which purports to offer a range of quality financial products and services unmatched in the industry.
• Merrill Lynch, which manages investments for corporate, institutional and governmental clients and claims to be the only firm that ranks among the top three in equity underwriting and secondary trading in all major regions of the world.

"The plans sponsored by financial services firms underperformed a balanced portfolio by 3 to 10 percentage points."

Table I shows that over the four-year period none of the five came close to matching the performance of the stock market or an index of 60 percent stocks and 40 percent bonds. Only in one year, 1996, did any of the 401(k) plans come close to or exceed the 60/40 portfolio. The mix of stocks and bonds returned 32.3 percent in 1995, 12.0 percent in 1996, 24.3 percent in 1997 and 20.8 percent in 1998. A fixed annual rate of return of 21.0 percent would provide the same overall investment return. The corresponding annual rates of return for the financial services firms were 13.1 percent for Morningstar, 10.5 percent for Prudential, 12.6 percent for Hewitt,¹⁸ 17.8 percent for Citigroup and 11.0 percent for Merrill Lynch.

As Figure II shows, using the 60/40 index as a benchmark, Morningstar's plan underperformed by 7.9 percent, Prudential's by 10.5 percent, Hewitt's by 8.4 percent,¹⁹ Citigroup's by 3.2 percent and Merrill Lynch's by 10.0 percent.

While acknowledging that plan participants may not be knowledgeable about investing, a Hewitt spokeswoman said it is difficult to know whether the four years studied are representative of returns over a longer period of time and whether a 60/40 index is the appropriate benchmark for comparison. She noted that while the Standard & Poor's 500 stock index was high over the 1995-98 period, a typical financial planner would have had clients invested in diversified portfolios that also included investments in such areas as small capitalization stocks and international stocks, which produced low returns during that particular period.

The other 401(k) plans did not respond to numerous telephone calls and e-mails.

This is not a criticism of the firms, which operate under the same constraints on giving financial advice to their employees as do other companies. [See "The Threat of Future Liability" below.] Rather, it is reinforcement for concerns about the investment decisions many participants in 401(k) plans are making (or not making).

[page]

The authors have had the opportunity to examine actual plan records over several years for a number of large plans in assorted, geographically dispersed industries. Company names are withheld because of confidentiality, but we can discuss the data. One feature that stands out in most of them is a remarkable difference among individual yields in the same plan. To cite a typical example:

• One plan, with 2,113 participants, had a yield of 18.3 percent for the plan as a whole.²⁰
• However, within the plan, one participant's individual yield was 52.1 percent and another's was a negative 12.8 percent.

"In a typical plan, the higher a participant's average pay, the higher the return on 401(k) investments."

A related feature that stands out in the plans we examined is the pattern of returns by wage income. When the participants are arrayed by income and divided into five groups, the higher a quintile's average pay, the higher the return on its 401(k) investments. An explanation of this pattern is suggested by Figure III, which shows the asset allocation made by each income quintile in a typical plan. Participants in this plan could allocate their contributions among six funds: money market, fixed income, balanced, value equity, growth equity and aggressive equity. The money market and fixed income funds, and 50 percent of the balanced fund, are considered fixed-type investments. The three equity funds and 50 percent of the balanced fund are considered equity-type investments.

• Almost two-thirds of the money in the lowest-income quintile was in the money market fund or other fixed-type investments.
• By contrast, about 85 percent of the money in the highest-income quintile was in equity-type investments.

Why was so much of the money from the lowest-income quintile in fixed-type, low-return investments? Many participants who enroll in 401(k) plans do not choose where their contributions are to be invested, leaving it up to the plan sponsor. This is called a "default election." In the above plan, as in many 401(k) plans, the money market fund – perhaps the least appropriate retirement investment – is automatically designated by the plan sponsor as the default investment. Between one-third and one-half of the employees in the fifth quintile made no choice. Plans that do have money market funds as the default investment reason that these investments will never lose money and therefore relieve the plan of possible liability claims. Of course, this ignores the loss of purchasing power caused by inflation. Meanwhile, those with higher incomes are receiving higher returns.

"Many lower-income employees do not choose where to invest, and their contributions are placed in money market funds, where they tend to stay."

As time goes on, there is a good possibility that the gap between the returns to the highest- and lowest-paid will widen. Research by Hewitt Associates, in conjunction with Harvard University and the University of Chicago, has found that employees who make a default election tend to remain at the same contribution rate and keep the same funds for years, ultimately losing the opportunity to have more money at retirement.²¹

If 401(k) plans sponsored by major financial services firms can manage only below-average returns, is it reasonable to hope that ordinary 401(k) participants will be able to better manage the investment of their retirement funds? In 1998, the index of 60 percent stocks, 40 percent bonds had a return of 20.8 percent. A database we constructed of 401(k) plans with 500,000 participants (approximately 1 percent to 2 percent of the 401(k) market) showed an average return of 8 percent on the plans' assets in 1998 – another indication that most plans performed below the market averages.

[page]

The low returns also reflect a number of inherent failings in 401(k) plans as currently structured, involving participants, plan sponsors and the law.

Problem: Lack of Knowledge. Several studies find that many participants in defined contribution plans have an appalling lack of understanding of basic principles of investing. For example, a recent national survey of participants found:²²

• Respondents generally considered company stock less risky than a diversified domestic equity portfolio.
• 44 percent thought money market funds included stocks and 43 percent thought they also included bonds.
• Nearly 20 percent didn't know they could lose money in equities.
• 65 percent didn't know they could lose money in a bond fund and
• 60 percent didn't know they could lose money in a government bond fund.

"Many participants have an appalling lack of understanding of basic principles of investing."

Small wonder that so many participants in 401(k) plans have little or no grasp of the principles of prudent investing! They may have a limited or extensive list of funds from which to choose, but they base their selection on individual funds rather than investment strategy. The fund offerings may not stress the value of index funds, which invest in the stocks or bonds used to compute a particular index and have low management fees because they are not actively managed. Participants take too little risk, as in the case of those letting most of their assets stay in money market funds or cash, or too much risk, as in the case of those putting the great majority of their assets into high-tech stocks or funds.

Problem: High – and Hidden – Administrative Costs and Management Fees. Administering a 401(k) plan and managing its investments costs money. Many plans have selected low-cost funds, with fees fully disclosed to plan participants and often paid by the employers. However, many other plans have higher fees mostly paid by the participants. Some of the latter contain mutual funds with high retail price structures.²³ Participants often are unaware that they are paying administrative fees for these funds from their accounts. In some cases, especially with smaller employers, plan sponsors choose these funds because in return the mutual funds handle administrative chores such as keeping track of account balances, sending out statements and answering questions.

The New York Times recently reported that some fund companies rebate part of the administrative fees to employers or outside plan administrators.²⁴ The administrative fees, which the Times said usually amount to about 0.25 percent of the assets in an account for large plans, add up. For example:²⁵

• An investment of $5,000 a year for 30 years with a 10 percent annual return amounts to $863,594.
• Annual fees of 0.25 percent will reduce that amount by $40,883.
• If fees are 1 percent, as they often are in smaller plans, the benefit reduction is $151,387 (or 21 percent) – and some plans have costs exceeding 2 percent.

Even if there are no rebates or cost-shifting, companies sponsoring 401(k) plans have little incentive to monitor the fees closely or to negotiate lower fees for plan participants when the costs are paid from the participants' and not the company's funds.

"Thirty-nine percent of those age 18 to 34 cash out their 401(k) accounts when they change jobs."

Problem: Cashing Out. Even modest contributions to one's 401(k) at an early age can grow to a significant sum by the time of retirement. However, almost a third of people with accounts – and 39 percent of those ages 18 to 34 – cash them out when they change jobs. Plan sponsors can (but are not required to) cash out an account balance valued at under $5,000 when the participant terminates employment.²⁶ The employee can roll over the money into an individual retirement account (IRA), move it to a new employer's plan, or take a lump sum payment minus income tax and a penalty for early withdrawal.²⁷ The ERISA Advisory Council reported that only 20 percent of individuals who received lump sum distributions rolled the entire sum into another tax-deferred account.²⁸ A report to the advisory council recommended that all defined contribution plans be required to accept rollovers of cash from other qualified plans. Putnam Investments, a money management firm, estimates:²⁹

• Withdrawals amount to between $33 billion and $39 billion per year.
• Those withdrawing the money pay $7.1 billion to $8.3 billion in unnecessary federal taxes and penalties each year.

Problem: Hardship Distributions and Consumer Loans. Many plans allow an employee to make a hardship withdrawal, usually to purchase a primary residence, pay college tuition, pay unreimbursed medical expenses or prevent eviction from or foreclosure on a principal residence. These withdrawals are subject to income tax and a 10 percent early withdrawal penalty.³⁰ Further, the participant cannot contribute to the account for one year after a hardship withdrawal, thus losing any matching contribution from the employer.³¹

Most 401(k) plans also allow a participant to borrow from his or her account for non-hardship purposes, such as buying a boat or a big-screen television, and to repay the loan to the account with interest. This can be tempting because the interest rate is lower than credit card interest and the interest goes into the participant's own account. However, the interest is paid with aftertax money, which will be taxed again when it is withdrawn in retirement. In addition, the participant loses the return while the funds are out of the 401(k).

[page]

Why aren't the companies that offer 401(k) plans helping their employees make wise investment choices? A major reason is fear of opening themselves to lawsuits charging they are responsible for employees' investment losses. Federal law encourages this silence. Section 404(c) of ERISA, as amended, says that fiduciaries who successfully comply with a maze of complex rules are generally relieved of any fiduciary responsibility for investment losses, provided all plan participants exercise independent control over their accounts.

"Companies fear opening themselves to lawsuits if they give employees investment help."

As is often the case, words written by regulators to guide us seem straightforward enough at first reading. We are told that to be a successful 404(c) plan sponsor one must (i) give participants the opportunity to choose from a broad range of investment alternatives, (ii) allow them to make elections at least quarterly, (iii) assure that investment options are diversified, and (iv) provide sufficient information to enable participants to make informed investment decisions.

But how much information is sufficient? According to the Commerce Clearing House Pension Plan Guide, fiduciaries are never relieved of their constant duty to consider the prudence of the investment alternatives made available to participants under a plan and to maintain oversight over the investment options.³² If the employer points out that investing 401(k) money in equities offers a better opportunity over the long run to build a sizable nest egg for retirement, and equities have one or two down years, is the employer liable because of giving advice?

In this litigious age, lawsuits are likely to be filed alleging that plan sponsors have given participants too much or too little advice or are otherwise derelict in their fiduciary duties.³³ For example, the consulting firm Watson Wyatt has suggested that where 401(k) plans continually underperform, "employees may eventually complain that either the funds or the education offered were inappropriate or insufficient."³⁴

Rep. John Boehner (R-Ohio) has introduced a bill that would allow employers to provide plan participants with access to professional investment advice, and would allow investment firms that manage 401(k) plans to provide advice, with the advisers required to fully disclose their fees and any potential conflicts of interest. The bill is controversial, with opponents contending that the advisers stand to enrich themselves by recommending investment in their products.

[page]

We have devised what we call the "Triple 90 Yardstick" to assess any 401(k) plan. We have found that if plans meet this yardstick, it is almost certain that even low-wage plan participants will receive an adequate replacement income in retirement. A unique characteristic of this tool is that it does not consider plan provisions or exotic features; it simply looks at employee behavior. Briefly, a plan meets our yardstick measure if:

• 90 percent of all eligible employees participate in and contribute to the plan.
• 90 percent of the maximum employer contribution is captured by employees.
• 90 percent of the plan's assets are invested according to professional asset allocation advice. (For example, if professionals would put 70 percent in equities, the plan would put at least 63 percent – 90 percent of 70 – into equities).

"The ideal plan: 90 percent of employees participate; 90 of the employer's match offer is realized; and 90 percent of investments follow prefessional guidlines."

The second item is generally referred to as the plan's match capture ratio. To illustrate, assume that a company contributes 50 cents for each $1 (up to 6 percent of pay) an employee contributes. If all eligible employees participate and all contribute 6 percent of pay, the plan's match capture ratio is 100 percent. If all employees participate, but each only contributes 2 percent of pay, the match capture ratio is 33 percent. In the latter case, participants forfeit 67 percent of the employer's potential annual matching contribution.

The third item measures how closely the asset allocation of the group as a whole compares to the asset allocation that an investment professional would have made. Investment professionals generally recommend that at least 70 percent of funds be invested in equity investments and no more than 30 percent in fixed-income investments. If employees meet 90 percent of the "professional's target" for equity investments, the third measurement is satisfied.

Applying this yardstick, too many plans don't measure up. This means the retirement income for an employee with average income will be inadequate. If a plan measures up to our Triple 90 Yardstick, the reward is almost always a fourth 90 percent measurement; that is, the replacement-income ratio for the employee with average income (with Social Security considered) will also exceed 90 percent.

[page]

The concept of defined contribution plans is a sound one. The assets are owned by the employee, are portable from job to job and a participant's benefits are not affected by job changes. However, as with most new concepts, there are flaws that can be eliminated to the benefit of both employees and employers. To remedy the flaws, we propose a new type of 401(k) plan, which we call the American Freedom 401(k). Employees now in a 401(k) plan should have a choice of remaining there or moving to the American Freedom 401(k) plan. Employers who offer all the features of the American Freedom 401(k) plan should receive a "safe harbor" from litigation, explained below. The plan would have these features:

"The American Freedom 401(k) corrects the faults existing in today's 401(k)s."

Enrollment and Minimum Contributions. The plan would automatically enroll all employees after they satisfy the plan's eligibility period unless they execute a rejection form opting out. The plan would also set an initial minimum contribution rate of about 4 percent to 6 percent of income – an amount that could prudently be expected to provide a reasonable retirement income – unless the employee specifically opts for a smaller amount.³⁵ This minimum contribution requirement would help limit a too-common practice today, where company human resource departments, to make participation rates in 401(k) plans appear to be high, urge employees to "just contribute a dollar or two out of each paycheck."

Premixed Portfolios and Professionally Directed Investments. Since index funds and the managers of defined benefit pension plans have historically produced higher yields on investments, companies adopting the American Freedom 401(k) plan would have to agree to include in participants' options premixed efficient portfolios – ones that give the maximum rate of return at different risk levels – or a professionally directed investment option or both.³⁶ Companies should be encouraged (but not required) to provide employees who choose to manage their own accounts with access to investment advice.

Default Option. The contributions of a participant who made no initial choice of funds should go into a premixed efficient portfolio (e.g., 60 percent stocks and 40 percent bonds) or into the professionally directed investment as a default option.

Fees and Expenses. A plan sponsor either would pay all fees and expenses or would reimburse the plan. This would give employers an incentive to limit fees and expenses (currently there is no required oversight of such spending) and would raise the net returns received by plan participants. As an alternative, fees and expenses might be capped at, say, 1 percent, with the employer required to pay anything above the cap.

Automatic Rollover. The American Freedom 401(k) plan would prohibit benefit cash-outs by the plan or the employee following termination of employment before retirement, death or disability. Instead, the account could be rolled over into a similar qualified plan or could remain in the previous employer's plan if the new place of employment has no qualified plan.

Vesting. Vesting would be 100 percent and immediate.

Hardship Loans. The American Freedom 401(k) plan would have a new feature, the hardship loan, funded from and paid back to the plan's trust fund, not the participant's account. Consumer-type loans and hardship distributions to plan participants from their accounts, now permitted by most plans, would be prohibited. A hardship loan would simply be a loan from plan assets (not the borrower's account), limited to conditions that would meet the legal criteria for a current hardship distribution. This would enable participants to get money from the plan for a true hardship or emergency and pay it back with interest without (i) losing the matching employer contributions for a time, (ii) paying increased taxes due to the prohibition on personal contribution for a time, (iii) being subject to a tax and penalty on the hardship distribution and (iv) affecting the investment return on the account. At the same time, participants seeking a loan for some other purpose could turn to a source of consumer credit and leave retirement funds in the account to grow.

"Employers who establish an American Freedom 401(k) plan should receive 'safe harbor' protection from lawsuits."

Safe Harbor for Employers. Because the American Freedom 401(k) plan would be so beneficial to participants, employers should be given an incentive to establish such a plan. Legislation should provide that, in exchange for providing a plan offering all the features, an employer would have to meet only the basic coverage and nondiscrimination requirements. In addition, the plan would be deemed to comply with technical testing standards now required.³⁷ Finally, the plan sponsor would receive "safe harbor" protection, exempting it from class action civil suits and similar actions alleging breach of fiduciary standards. We would expect industry service providers to respond quickly to such a program.

[page]

In an age when workers often have many jobs, sometimes many careers, defined contribution pension plans such as the 401(k)s allow steady accumulation of retirement funds for a comfortable retirement – but only if the accounts are invested prudently and well. The American Freedom 401(k) plan will make retirement security possible, even for participants with little or no investment knowledge.

NOTE: Nothing written here should be construed as necessarily reflecting the views of the National Center for Policy Analysis or as an attempt to aid or hinder the passage of any bill before Congress.

[page]

1. At the end of 1999, there were 340,000 401(k) plans with 34 million participants. Danny Hakim, "Controlling 401(k) Assets: Fight Brewing over Investment Choices for $1.7 Trillion," New York Times, November 17, 2000.
2. Internal Revenue Code Section 401(k) was quietly inserted into the code by the Revenue Act of 1978, primarily to clear up a dispute over the taxation of profit-sharing plans. It says that an employee savings plan can include a cash or deferred arrangement – a deal in which employees can take their bonuses in cash and pay taxes on them or put them into their savings plans and postpone their tax bill – as long as the plan is designed to benefit low-paid as well as high-paid employees. R. Theodore Benna, a pension consultant, recognized that the law didn't forbid including regular salary and not just bonuses in such a plan and that companies could chip in extra money to encourage employees to save. He set up a plan for Johnson Companies which won Internal Revenue Service approval in November 1981. See Eric Schurenberg, 401(k) Take Charge of Your Future, updated edition (New York: Warner Books, 1996).
3. U.S. Census Bureau, Statistical Abstract of the United States, 1999, 119th ed., Table No. 1421, Washington, D.C.
4. United Nations Population Division, World Population Prospects: The 1998 Revision (New York: United Nations, 1998).
5. Ibid.
6. See Jonathan Gruber and David Wise, "Social Security, Retirement Incentives and Retirement Behavior: An International Perspective," Employee Benefit Research Institute, Issue Brief No. 209, May 1999.
7. For survey http://www.psca.org/.
8. This paper does not examine the income from Social Security. Workers and their employers currently pay a FICA payroll tax of 15.3 percent, split equally between them. (Most economists agree that the entire tax burden falls on the employee, as with employee benefits.) Of this amount, 12.4 percentage points goes to Old-Age, Survivor and Disability Insurance (OASDI) and the remainder to Medicare. The OASDI is levied up to a maximum covered wage base each year. In 2001 the base is $80,400, meaning the maximum tax in 2001 is $9,969.60 ($80,400 x .124). The most recent figures from Commerce Clearing House show that Social Security benefits will replace 53.4 percent to 24.2 percent of income for low- and high-wage base maximum workers. They will replace 39.7 percent of the average worker's income. See Avram Sacks, 2001 Social Security Explained (Chicago: Commerce Clearing House, 2001).
9. Any Social Security benefits at retirement will be in addition to the private pension discussed here.
10. The pension incomes are in nominal rather than real dollars.
11. By law, the vesting period cannot exceed seven years.
12. Another disadvantage of defined benefit plans is that although a worker and his or her spouse have a vested right to a pension, there is no benefit for a minor child or others if the worker dies prior to retirement without a surviving spouse (or a surviving former spouse with a legal right to a portion of the pension).
13. Doe II's investment return is reduced 1.25 percent each year for the plan's annual fees and expenses – which are paid by employees.
14. Report from Cerulli Associates, cited in Danny Hakim, "401(k) Accounts Are Losing Money for the First Time," New York Times, July 9, 2001.
15. "Investment Returns: Defined Benefit vs. 401(k)," Watson Wyatt Insider, September 1998.
16. These calculations are made from annual Form 5500s gathered by http://freeERISA.com.
17. Morningstar does not own, operate or hold any interest in mutual funds, stocks or insurance products.
18. Hewitt's results are for Oct. 1-Sept. 30 fiscal years. The 60 percent stocks-40 percent bonds portfolio had an average return of 19.5 percent for those four fiscal years.
19. See Note 18.
20. Plan assets of $54,353,404 produced total annual investment earnings of $9,926,774.
21. The U.S. Treasury encourages automatic enrollment of employees in companies' 401(k) plans to increase participation rates, but the Hewitt study questioned the desirability of automatic enrollment. "Given the stickiness of defaults, employers may want to consider default funds and contribution rates that are more appropriate for the average participant over the longer run," Lori Lucas, a defined contribution consultant for Hewitt, said. "Time May Not Be on Automatically Enrolled Employees' Side," press release, July 10, 2001, Hewitt Associates.
22. "John Hancock Financial Services Survey: Most Individuals Fundamentally Unprepared to Manage Retirement Assets," John Hancock Financial Services news release, April 24, 2001.
23. Retail mutual funds generally charge higher management fees than institutional funds, which have lower expenses and generally are available only to institutional investors.
24. Virginia Munger Kahn, "When Hidden Fees Erode 401(k)s," New York Times, July 22, 2001.
25. Ibid.
26. Beginning in 2002, this amount is reduced to $1,000.
27. "Retirement Savings in an Unsettled Economy," survey for Putnam Investments, May 22, 2001.
28. "Are We Cashing Out Our Future?" Working Group Report to the ERISA Advisory Council on Employee Welfare and Pension Benefits, November 13, 1998.
29. Ibid.
30. The 10 percent tax penalty may apply if the employee is under age 59 ½; there is no penalty for withdrawal if one becomes disabled as defined by the Internal Revenue Service.
31. Under legislation effective in 2002, the one-year prohibition is reduced to six months.
32. See CCH-EXP, PEN-PLAN-GUIDE, ¶4485, Participant-Directed Accounts, ERISA Reg. §2550.404c-1(b)(2) (i).
33. Experience teaches that predicting the future meaning of seemingly simple words written long ago is far from simple. We will resist further comment, but the reader might note that where the Supreme Court Justices see a fundamental (i.e., common sense) wrong involving a defendant with actual knowledge of odious facts and circumstances who then manifests a deliberate indifference, a judicial s-t-r-e-t-c-h may occur.
34. "Investment Returns: Defined Benefit vs. 401(k)," Watson Wyatt Insider.
35. This would not prevent the employee from later changing the contribution rate or withdrawing from the plan altogether.
36. Premixed portfolios typically would comprise one or more index funds.
37. The complex testing is supposed to ensure that more highly compensated employees do not receive better treatment that less highly compensated employees.

[page]

Brooks Hamilton, a Dallas attorney, has worked in the field of employee benefit consulting for more than 40 years. His practice since the late 1970s has been limited to corporate employee benefits and related executive compensation, financial and estate planning. Prior to that, he had founded and operated a firm providing professional employee benefit computer services to major financial institutions. For several years, Hamilton has been addressing ways to deal with the wide disparity of investment returns among employees with accounts in the same 401(k) plans. He holds a B.S. degree from the University of Houston, did postgraduate work in computer science at the University of North Carolina and received a J.D. degree from Southern Methodist University School of Law.

Scott Burns writes a personal finance column for the Dallas Morning News that is syndicated nationally and distributed by Universal Press Syndicate. The column can also be read on Money Central at msn.com. Burns has his own Web site, www.scottburns.com. Before joining the News in 1985, Burns had been financial editor and a columnist for the Boston Herald since 1977. Burns has been a contributing editor of Worth Magazine, and his articles and columns have appeared in other magazines, including Playboy, Vogue and Harper's Bazaar. He is a graduate of the Massachusetts Institute of Technology with a degree in humanities and biology. He also studied writing with poet Archibald MacLeish at Harvard University.