This is the accessible text file for GAO report number GAO-02-225 
entitled 'Private Pensions: Improving Worker Coverage and Benefits' 
which was released on April 9, 2002. 

This text file was formatted by the U.S. General Accounting Office 
(GAO) to be accessible to users with visual impairments, as part of a 
longer term project to improve GAO products’ accessibility. Every 
attempt has been made to maintain the structural and data integrity of 
the original printed product. Accessibility features, such as text 
descriptions of tables, consecutively numbered footnotes placed at the 
end of the file, and the text of agency comment letters, are provided 
but may not exactly duplicate the presentation or format of the 
printed version. The portable document format (PDF) file is an exact 
electronic replica of the printed version. We welcome your feedback. 
Please E-mail your comments regarding the contents or accessibility 
features of this document to Webmaster@gao.gov. 

United States General Accounting Office: 
GAO: 

Report to Congressional Requesters: 

April 2002: 

Private Pensions: 

Improving Worker Coverage and Benefits: 

GAO-02-225: 
		
Contents: 

Letter: 

Results in Brief: 

Background: 

Ability of Reforms to Expand Pension Coverage and Benefits May Be 
Limited: 
Improving Retirement Income Outside the Single-Employer Voluntary 
Pension System Involves Tradeoffs and Challenges: 

Concluding Observations: 

Agency Comments: 

Appendix I: Advantages of Tax-Qualified Retirement Savings: 

Appendix II: Perspectives on Adequacy of Retirement Income: 

Related GAO Products: 

Tables: 

Table 1: Plan Sponsorship Varies with Firm Size: 

Table 2: Non-Single-Employer-Based Approaches to Increasing Retirement 
Income: 

Table 3: Tax Treatment of Saving in a Defined Contribution Plan versus 
a Regular Savings Account (constant tax rates): 

Table 4: Comparison of Tax Treatments of Saving in a Defined 
Contribution Plan and Saving in a Roth IRA: 

Figures: 

Figure 1: Pension Plan Participation, by Plan Type: 

Figure 2: Percentage of Employees Aged 25-64 Years Who Are Covered by 
a Pension Plan, by Finn Size, 1999: 

Figure 3: Participation in Retirement Plans by Earnings, 2000 
(Private, Full-Time Workers Aged 25-64 Years): 

Abbreviations: 

EGTRRA: The Economic Growth and Tax Relief Reconciliation Act of 2001: 

ERISA: Employee Retirement Income Security Act: 

EITC: Earned Income Tax Credit: 

ESOP: employee stock ownership plans: 

HRS: Health and Retirement Study: 

IRA: individual retirement arrangement: 

MUPS: minimum universal pension system: 

SEP SIMPLE: Simplified Employee Pension: 

SSI: Savings Incentive match Plan for Employees Supplemental Security 
Income: 

TIAA-CREF: Teachers Insurance Annuity Association and College 
Retirement Equities Fund: 

[End of section] 

United States General Accounting Office: 
Washington, DC 20548: 

April 9, 2002: 

The Honorable Robert E. Andrews: 
Ranking Minority Member: 
Subcommittee on Employer-Employee Relations: 
Committee on Education and the Workforce: 
House of Representatives: 

The Honorable Major R. Owens: 
Ranking Minority Member: 
Subcommittee on Workforce Protections: 
Committee on Education and the Workforce: 
House of Representatives: 

Private pensions, as a key supplement to Social Security, can help 
assure that workers receive adequate incomes in retirement. Although 
private pensions are an important source of retirement income for many 
workers, millions of other workers have no individual pension 
coverage, which places them at risk of inadequate income during their 
retirement years.[Footnote 1] Since the 1970s, only about half of the 
nation's workers have been covered by private employer-sponsored 
pensions. Although it is difficult to predict whether any particular 
worker currently in the labor force will ultimately earn a pension 
benefit, at present only about 52 percent of retirees receive pension 
income. Over the past 25 years, considerable attention has been 
focused on modifying pension law, in part to improve coverage and 
ultimately retirement income adequacy, yet a significant portion of 
the workforce remains without pension coverage and the opportunity to 
earn pension income. 

A wide variety of further reforms have been suggested to improve 
pension coverage and benefits. You asked us to examine the issues 
surrounding pension coverage and benefit adequacy and what measures 
might improve the income prospects of future retirees. Accordingly, 
this report discusses the potential for reform of the private pension 
system to improve workers' pension coverage and benefits, and it 
reviews approaches other than the voluntary, single-employer-based 
pension system that might expand pension coverage and improve the 
retirement income of those workers likely to lack pensions. 

In conducting this study, we surveyed an array of literature relevant 
to the topic. We reviewed numerous academic and policy studies, 
supplemented by information obtained from benefit professionals, 
academic researchers, and employer and employee groups. Our work was 
conducted between January and December 2001 in accordance with 
generally accepted government auditing standards. 

Results in Brief: 

Traditional reforms to the voluntary, single-employer-based pension 
system may have limited potential to significantly expand pension 
coverage and improve benefits for workers who traditionally lack 
pensions. Reforms aimed at encouraging plan sponsorship have focused 
primarily on improving tax incentives and reducing the burden of 
pension regulation on small employers. But the effect of reforms aimed 
at increasing sponsorship and coverage may be offset by other policy 
actions. For example, increasing the annual limits on the pension 
benefits that can be earned or on the contributions that can be made 
may improve the tax incentives for sponsorship, but lowering marginal 
income tax rates may offset some of the impact of raising the limits. 
There are also numerous proposals that attempt to affect pension 
coverage and benefits by further modifying the framework of rules 
governing pensions. Past reforms to these rules, such as improved 
vesting,[Footnote 2] and trends in the pension field, such as the 
enhanced portability of defined contribution plans and increased 
worker knowledge about retirement planning, suggest that more workers 
and their spouses could receive pension income in the future. But the 
intended effects of changing some pension rules may be counteracted by 
the responses of employers and workers. For example, requiring 
employers to broaden pension coverage or provide higher benefits to 
certain workers could lead to decreased plan sponsorship. Also, 
efforts to improve retirement saving by restricting workers' ability 
to receive and use lump sum distributions from their plans could make 
it less likely that they would participate in and contribute to their 
plans. Some analysts question whether additional pension reforms will 
have significant results for the types of workers who traditionally 
lack pensions, particularly those with low incomes, because many 
reforms offer only incremental changes. As a result, some reformers 
suggest proposals that move outside the voluntary, single-employer 
private pension system. 

Outside the voluntary, single-employer private pension system, there 
are three broad categories of reform approaches—pooled employer 
reforms, universal access reforms, and universal participation 
reforms. Pooled employer reforms[Footnote 3] focus on increasing the 
number of firms offering pension coverage through centralized third-
party administration and aim to increase employee pension portability. 
However, the employer's loss of control of plan design and concern 
about cost and administrative requirements may limit employer interest 
in such plans. Universal-access reforms aim to increase retirement 
savings by providing all workers with an opportunity to save through a 
payroll-based account without mandating an employer contribution. 
However, these reforms raise concerns about the difficulty faced by 
workers, particularly low-income workers, in setting aside money for 
retirement and about the administrative burden placed on employers. 
Universal-participation reforms aim to ensure coverage and retirement 
income for all workers by mandating pension availability and 
participation, similar to the existing Social Security system. Such 
reforms raise concerns about the increase in employers' administrative 
burden, as well as potential adjustments to other forms of 
compensation to offset higher pension costs. Several pension-related 
proposals aimed at improving the availability and level of retirement 
income for lower-earning workers are similar in many respects to 
current proposals to introduce an individual account-based option into 
Social Security. Such approaches entail cost and design challenges, 
but it is important to recognize the relationship between concerns 
about private pension coverage and benefits and the Social Security 
policy debate in any retirement policy reforms that emerge. 

Background: 

The standard of living of the elderly depends on total retirement 
income,[Footnote 4] which includes Social Security, pensions, income 
from assets, and earnings from employment. In addition, benefits from 
public assistance programs, such as Supplemental Security Income 
(SSI), and health insurance programs, such as Medicare, may also be 
relevant in assessing the standard of living of the elderly. Pensions 
generally supplement Social Security, which has a progressive benefit 
structure that provides higher relative benefits to lower earners. 
[Footnote 5] As a result, although private pensions account for only 
about one-tenth of the aggregate income of the elderly, they are an 
important source of retirement income for many households, 
particularly those in the middle to higher ranges of the income 
distribution. Recent research suggests that about two-thirds of 
households nearing retirement have rights to some pension income, but 
these amounts can vary widely. 

The ability to earn and receive retirement income under a voluntary, 
private pension system is the result of decisions made by both the 
employer and the worker within a legal and regulatory framework that 
has developed over time. The Internal Revenue Code and the Employee 
Retirement Income Security Act (ERISA) of 1974, as amended, are the 
basis of pension law today.[Footnote 6] Employers make the decision to 
sponsor a plan and choose the features that it will include, taking 
into account that workers may have different preferences for pensions 
in comparison with other forms of compensation such as cash wages and 
health insurance. Workers also make numerous employment-related 
decisions over the course of their career, such as where to work, how 
much to work, and whether to change jobs, that can affect their 
ability to earn pension income. They also make decisions about how 
much to save for retirement and whether to preserve funds distributed 
from their plans. The result of employer and worker interactions in 
the marketplace is that not all workers will earn pension income and 
receive it in retirement. 

Employers Sponsor Pensions to Provide Tax-Deferred Income to Workers: 

Among the most important reasons that employers sponsor pensions are 
the need to attract and retain a productive workforce and the tax 
advantages associated with pensions.[Footnote 7] Pensions can be a 
means of providing deferred compensation that may encourage workers to 
make a long-term commitment to the employer, thus reducing turnover 
and making for a more stable, productive workforce. But in deciding 
whether to offer a pension, companies must assess the nature of their 
particular workforce to determine if offering pensions is a necessary 
employment inducement. For example, some workers may view pensions as 
less important than cash wages or other benefits, particularly health 
insurance. For such workers, the employer may have little incentive to 
offer a pension. Employers also choose to sponsor pension plans 
because of the favorable federal tax treatment of pension 
contributions and investment returns. This tax treatment, particularly 
the deferral of taxation on invested income, is especially attractive 
to those facing higher marginal tax rates,[Footnote 8] such as some 
business owners and higher-paid employees, and can be an important 
incentive to sponsor a plan. 

Employers also consider the benefits of offering a pension plan in 
comparison with its overall cost. The major cost of the pension to the 
employer will depend on the contributions necessary to finance or fund 
the pension. Other costs involve the administration of the plan, such 
as record keeping, calculation of benefits, outside administrative 
help and advisers, communication with employees, investment management 
fees, and compliance with government rules and regulations. The result 
of weighing the benefits and costs of offering a plan is that not all 
employers will find it desirable to sponsor a pension plan. For 
example, compared with medium and large employers, small employers are 
less likely to sponsor a pension plan. Small businesses may face 
greater uncertainty, especially with regard to profitability, and may 
face cost pressures that can affect their ability to offer 
compensation packages that compare favorably with those offered by 
larger, more stable firms. While small businesses often cite the cost 
of pensions as an obstacle to sponsorship, surveys suggest that the 
firm's lack of profitability and employee preferences are also 
important obstacles. 

Employers Can Limit Plan Coverage to a Portion of the Workforce: 

An employer has discretion to determine which workers will be covered 
by its pension plan, and the employer's plan design decision may 
result in certain types of workers' not having the opportunity to 
participate.[Footnote 9] In designing the plan, the employer may cover 
employee groups on the basis of objective business criteria, such as 
pay (hourly or salaried), job location, or job categories. An employer 
may have one plan to cover a wide range of categories of workers, or 
it may have separate plans for different groups depending on business 
objectives. The employer is also bound by a federal rule on 
eligibility that covers all pension plans. Under this rule, a pension 
plan may exclude employees younger than age 21 or those who have less 
than one year of service from participating in the plan.[Footnote 10] 

Plans that seek tax-qualified status must also satisfy a set of 
"nondiscrimination" rules that seek to ensure that the plan design 
does not exceed certain limits in the extent to which it favors highly 
compensated employees in participation and benefits. Even so, in 
addition to age and service requirements, the nondiscrimination rules 
may permit a firm to exclude between 30 to 80 percent of the non-
highly compensated workers from the plan. 

Employers' Plan Design Decisions Determine How Pension Benefits Are 
Earned: 

A worker can be offered either a defined benefit plan or a defined 
contribution plan.[Footnote 11] Some workers may participate in both 
types of plans if their employer offers more than one type of plan. 
[Footnote 12] Figure 1 shows that defined contribution plans account 
for most of the growth in pension plan participation since the mid-
1970s. Under the typical formula used for defined benefit plans, the 
annual (or periodic) increment in benefits earned (benefit accrual) 
tends to increase over the worker's career with the employer, which 
makes this type of plan advantageous for workers who stay with one 
employer over their working careers. Under a defined contribution 
plan, the benefit accumulation each period may fluctuate over the 
course of the worker's career; frequently, however, such accounts are 
depicted in terms of an average or steady return over the worker's 
tenure with the employer, making the accumulation pattern more even in 
comparison with a defined benefit plan. This means that younger or 
shorter-tenured workers may have higher benefit accumulations compared 
with the benefits they would accrue under a traditional defined 
benefit plan. 

Figure 1: Pension Plan Participation, by Plan Type: 

[Refer to PDF for image: stacked line graph] 

The graph depicts total participants in defined benefit and defined 
contribution plans for the years 1975 through 1997. 

Source: Employee Benefit Research Institute and Department of Labor. 

[End of figure] 

Employers make other decisions about how pension benefits will accrue 
and be distributed. These decisions are subject to legal requirements. 

Limits on Contributions and Benefits: 

ERISA sets limits on annual contributions and benefits that qualified 
retirement plans may provide for each participant.[Footnote 13] These 
requirements are generally intended to limit the tax benefits provided 
through pensions, particularly to highly compensated individuals. 
Separate limits exist for defined benefit and defined contribution 
plans.[Footnote 14] In addition, employers must ensure that their 
plans comply with nondiscrimination rules that seek to balance benefit 
accruals of highly paid participants with those of non-highly paid 
participants by specifying the extent to which the benefit accruals 
of, or contributions made for, highly paid workers can exceed those of 
non-highly paid workers.[Footnote 15] 

Vesting Rules: 

Vesting provisions specify when workers acquire the irrevocable right 
to pension benefits. ERISA requires a plan to adopt vesting standards 
at least as liberal as one of the following schedules: full (or 
"cliff") vesting after five years or gradual vesting over seven years, 
except that matching contributions must fully vest within three years 
or gradually vest over five years.[Footnote 16] 

These rules affect how and when pension benefits will be paid out to 
workers. Pension plans provide for distribution of accrued benefits in 
the event of the worker's retirement, death, disability, or other 
severance of employment. Present law limits the circumstances under 
which plan participants may obtain preretirement distributions. 
[Footnote 17] Defined benefit plans typically provide benefits in the 
form of an annuity,[Footnote 18] which provides benefits throughout 
the period of retirement, and generally have age and service 
provisions that determine when an employee becomes eligible for 
receipt of benefits.[Footnote 19] Employers may also allow their 
workers to elect to receive pension payments as a lump sum. Because 
defined contribution plans are not required to offer annuities, lump 
sum distributions are typical and raise concerns about whether pension 
benefits will be preserved throughout retirement. 

Pension Benefits Depend on Workers' Decisions about Work and Saving 
for Retirement: 

The decisions that workers make also play an important role in 
determining how much pension income they will earn and receive in 
retirement. When a worker accepts employment, he or she accepts a 
compensation package that may or may not include a pension. Many 
workers may prefer cash wages or other benefits, such as health 
insurance, to pension benefits. The extent to which the worker values 
the pension component of compensation depends on many individual 
factors, including how aware he or she is about the need for future 
retirement income. 

Some workers also decide how long to remain employed by the plan's 
sponsor, and this decision determines whether they will earn pension 
income. Workers who stay with a plan sponsor for a number of years are 
more likely to meet the vesting requirements and to accrue benefits. 
For some plans, such as 401(k) plans, workers must also decide to 
participate, how much to contribute, and how to invest the assets in 
the plan.[Footnote 20] Workers who exhibit less attachment to the 
workforce may be less likely to become covered and participate in the 
plan. 

Even if a worker earns pension benefits, he or she must make decisions 
that determine whether these savings will contribute to their standard 
of living in retirement. When workers become eligible to receive 
distributions from a plan—either preretirement or upon retirement—they 
are faced with a choice of whether to preserve the distribution in a 
form that could provide income over their remaining lifetime, such as 
by transferring the funds to an Individual Retirement Arrangement 
(IRA)[Footnote 21] or choosing an annuity. The option of cashing out a 
lump sum distribution from a pension plan without rollover to an IRA 
raises concerns about future retirement income. Lump sum distributions 
can have advantages, because they allow flexibility for workers who 
have high-priority needs such as medical treatment, purchasing a home, 
or investing in a business. Lump sums distributions may also make 
sense when the amount is small and can be invested more profitably 
elsewhere. The potential disadvantage of lump sums distributions is 
that the assets may not be preserved for retirement income, as would 
be the case with a rollover to an IRA or purchase of an annuity. 
However, the importance of the lump sum issue to retirement income 
adequacy is the subject of debate and continuing research. Some see a 
problem given the number of workers taking preretirement lump sums 
without rollover to an IRA. However, some research has concluded that 
the impact of this practice on retirement income is very small, since 
these workers tend to have small account balances. Other research 
shows that larger sums generally are preserved through rollover into 
an investment account and that the proportion of workers cashing out 
lump sums is declining.[Footnote 22] 

Some Workers Unlikely to Benefit under a Voluntary System
Under a voluntary private pension system, the linkage between work, 
pension coverage, and the receipt and level of pension income in 
retirement is complex and depends on an array of factors, such as 
employer plan sponsorship and plan design, the framework of government 
rules, and worker decisions and choices over a lifetime. The result of 
employer and worker interactions in the marketplace is that not all 
workers will earn and receive pension income in retirement. Research 
suggests that some of the demographic characteristics of those who 
lack pension income in retirement are similar to the characteristics 
of workers who lack pension coverage during their working years. 
[Footnote 23] For example, those without pension income in retirement 
are more likely to be single, to be women, and to have low levels of 
education. But data on pension coverage are only a partial indicator 
of future pension receipt. The receipt of pension income involves 
factors that span a worker's career, and it is difficult to predict 
whether any particular worker currently in the labor force will 
ultimately receive a pension benefit. However, available research 
suggests that those who accumulate no pension income, or relatively 
low pension income, are more likely to include the following: 

* Workers employed by small firms. Compared with medium and large 
employers, small employers are less likely to sponsor a pension plan. 
As table 1 shows, the pension sponsorship rate drops dramatically as 
firm size gets smaller-86 percent of firms employing more than 1000 
workers offer pensions, while only 13 percent of firms with fewer than 
10 employees offer pensions. Figure 2 illustrates that worker 
participation in pension plans is lower for those employed by small 
firms. 

Table 1: Plan Sponsorship Varies with Firm Size: 
	
Firm size (no. of workers): Less than 10; 
Percentage of firms sponsoring plan: 12.9. 

Firm size (no. of workers): 10-24; 
Percentage of firms sponsoring plan: 28.6. 

Firm size (no. of workers): 25-49; 
Percentage of firms sponsoring plan: 39.7. 

Firm size (no. of workers): 50-99; 
Percentage of firms sponsoring plan: 53.5. 

Firm size (no. of workers): 100-249; 
Percentage of firms sponsoring plan: 67.5. 

Firm size (no. of workers): 250-499; 
Percentage of firms sponsoring plan: 76.7. 

Firm size (no. of workers): 500-999; 
Percentage of firms sponsoring plan: 79.3. 

Firm size (no. of workers): More than 1000; 
Percentage of firms sponsoring plan: 86.3. 

Source: Employee Benefit Research Institute. 

[End of table] 

Figure 2: Percentage of Employees Aged 25-64 Years Who Are Covered by 
a Pension Plan, by Firm Size, 1999: 

[Refer to PDF for image: vertical bar graph] 

Less than 25 employees: 23%; 
25-29 employees: 43%; 
100-999 employees: 57%; 
More than 1000 employees: 70%. 

Source: Munnell and Sunden, "Private Pensions," 40. 

[End of figure] 

* Workers employed part time or part year. Employers are less likely 
to provide pension coverage to part-time, seasonal, and contingent 
workers. For example, recent data show that about 60 percent of 
workers employed full time and year round have some form of pension 
coverage, but only 21 percent of part-time workers have pension 
coverage. 

* Workers with low earnings. Low earners are less likely than middle 
and high earners to be offered a pension plan and participate when a 
plan is offered. As figure 3 shows, pension participation varies by 
earnings levels ranging from over 70 percent participation for the top 
earning group to about 30 percent for the lowest earners.[Footnote 24] 
For those who are participants, some plans that are integrated with 
Social Security permit a reduction in pension benefits for the lowest 
earners to offset their proportionally higher Social Security benefits. 

Figure 3: Participation in Retirement Plans by Earnings, 2000 
(Private, Full-Time Workers Aged 25-64 Years): 

Employee annual earnings: Less than $20,000; 
Employer sponsors plan: 42%; 
Employee participating: 28%. 

Employee annual earnings: $20,000—$39,999; 
Employer sponsors plan: 65%; 
Employee participating: 55%. 

Employee annual earnings: $40,000—$59,999; 
Employer sponsors plan: 80%; 
Employee participating: 73%. 

Employee annual earnings: More than $60,000; 
Employer sponsors plan: 80%; 
Employee participating: 75%. 

Source: Congressional Research Service analysis of Current Population 
Survey (CPS) data. 

[End of figure] 

* Workers who frequently change jobs over the course of a career. Even 
"covered" workers who frequently change jobs can fail to accrue 
pension wealth for a significant fraction of their working lives owing 
to eligibility rules or to vesting rules and the resulting forfeiture 
of nonvested contributions or accruals. In addition, under defined 
benefit plans, the annual benefit accrual may be small relative to 
that for longer-service workers because of the age- and service-
weighted features used in these plans. Finally, many plans provide for 
lump sum cash-outs of accounts or accruals, which often are not rolled 
into other retirement savings vehicles. 

* Workers who place little value on saving. Some workers, either by 
preference or from lack of knowledge, may not be predisposed to saving 
or to committing to saving over the long term for retirement. The 
determinants of saving behavior are not completely understood, 
[Footnote 25] but it appears that inadequate retirement saving occurs 
at all income levels (see appendix 2). 

Concerns remain about the ability of workers with these 
characteristics to earn pension income and receive it in retirement. 
The federal government has several policy tools to provide incentives 
for expanding pension coverage, and various reforms to pension rules 
have been enacted with the aim of protecting and improving pension 
benefits for workers. Efforts to further improve coverage and benefits 
generally involve incremental reforms within the existing framework of 
the voluntary pension system. 

Ability of Reforms to Expand Pension Coverage and Benefits May Be 
Limited: 

Traditional reforms to the voluntary, single-employer-based pension 
system may have limited potential to significantly expand pension 
coverage and improve benefits for workers who traditionally lack 
pensions. Reforms aimed at encouraging plan sponsorship have focused 
on improving tax incentives and reducing the burden of pension 
regulation on small employers, but the effect of reforms aimed at 
increasing pension sponsorship and coverage may be offset by other 
policy actions. Also, numerous proposals attempt to directly affect 
pension coverage and benefits by revising the framework of rules 
governing pensions. Past reforms to these rules, such as improved 
vesting, and trends in plan design, such as the enhanced portability 
and accrual patterns associated with defined contribution plans, 
suggest that more workers and their spouses could receive pension 
income in the future. But the responses of employers and workers to 
further rule revisions may offset some of the revisions' intended 
effect. Some analysts question whether additional reforms to the 
voluntary, employer-based pension system can significantly expand 
pension sponsorship and increase coverage for workers traditionally 
lacking pensions and improve benefits for workers with pensions. 

Tax and Regulatory Reforms Aim to Provide Incentives for Plan 
Sponsorship: 

Much of the pension policy debate is concerned with the issue of how 
to increase pension plan sponsorship, particularly among small 
employers, as a basis for fostering increases in worker coverage and 
participation and for providing opportunities to earn pension 
benefits. The major policy tools to encourage pension sponsorship 
include increasing the tax preferences for pensions and simplifying 
pension regulations, and these tools are aimed at making it easier for 
employers to decide to sponsor plans. 

Reforms to Improve Tax Incentives: 

Tax incentives are an important tool to encourage employers to provide 
pensions. The success of tax incentives to encourage pension 
sponsorship has been questioned, however, in part because data show 
that only about half of the workforce is covered by a pension. At 
least two important factors may limit the effect that tax incentives 
provide for pension sponsorship. First, tax regulations limit 
employers' ability to direct tax preferences to the higher-paid 
employees who likely most value pensions. As a result, recent pension 
reform efforts typically have been aimed at relaxing these limits on 
pension tax preferences. Second, marginal tax rates have been lower in 
recent decades, which may have reduced the value of pensions to 
workers and thus the incentive for employers to sponsor or expand 
pensions. 

The progressive structure of income tax rates, that is, levying higher 
marginal tax rates as income increases, makes the benefits of the tax 
preference for pensions relatively greater for higher-income workers 
who pay higher marginal tax rates than for lower-income workers. Thus, 
this tax preference provides an incentive for owners and officers of 
firms to sponsor a pension plan for themselves and their higher-income 
employees. In turn, because sponsors may also want to provide pension 
benefits for other workers in the firm, and because pension law 
encourages plan sponsors to extend pensions broadly to their work 
force, these tax incentives may result in increased worker coverage. 
Some pension regulations, such as contribution limits and 
nondiscrimination rules, are designed to limit the use of tax 
preferences and to ensure that they do not benefit specific groups of 
workers, typically the higher paid, disproportionately; however, these 
regulations may reduce the incentive for employers to offer pensions. 
As these pension rules are made more stringent, the incentive may be 
further reduced.[Footnote 26] Relaxing limits and nondiscrimination 
rules is viewed by many employers as improving incentives to sponsor 
and expand plans. While such changes may lead to increased retirement 
savings by some workers, it is not clear whether they can 
significantly improve pension coverage and benefits for workers who 
traditionally lack pensions. Workers advocates may also view such 
changes as reducing the equity with which pension benefits are 
provided among workers. 

In addition, during the last two decades, marginal income tax rates 
have been lowered, which may have reduced the tax incentive to sponsor 
pensions. Reagan and Turner studied the pattern of marginal rates 
during the 1980s to determine whether decreases in marginal tax rates 
have reduced pension coverage.[Footnote 27] They found that, on 
average, a decrease of one percentage point in the marginal tax rate 
is consistent with a decline of 0.4 percentage points in the worker 
coverage rate. Thus, they conclude that declines in marginal tax rates 
appear to have lessened the incentives for plan sponsorship. 

Reforms to Simplify Pensions: 

Some reforms have sought to simplify the regulations imposed on 
qualified pension plans, so that business owners will be more likely 
to sponsor plans. Government involvement in pensions generally seeks 
to promote protection of employee benefit rights. Over time, however, 
with the enactment of new legislation and subsequent regulations, 
pensions have become more complex and costly to administer. Employers 
often argue that the burden of complying with pension regulations is 
excessive to the point of discouraging plan sponsorship, thus limiting 
the opportunity to increase coverage. 

The cost of sponsoring a pension plan can be an important deterrent to 
sponsorship in the small business sector.[Footnote 28] As a result, 
there have been calls for "pension simplification" to reduce the 
administrative complexity and cost of pensions while retaining the 
flexibility to design pensions that meet employers' needs. Proposed 
solutions generally involve reducing or eliminating various 
requirements with which sponsors must comply.[Footnote 29] Worker and 
public policy advocates, however, seek plan designs that improve 
worker coverage and benefits. Policymakers have sought to balance 
these competing demands by adopting reforms that reduce the legal and 
regulatory requirements on plan sponsors if they adopt specific plan 
designs that expand coverage to more workers and specify employer 
contributions. 

Two examples of pension simplification reforms are the creation of the 
Simplified Employee Pension (SEP) and the Savings Incentive Match Plan 
for Employees (SIMPLE).[Footnote 30] Created in 1978, a SEP is 
essentially an IRA that an employer provides to each eligible 
employee. The employer is subject to minimal reporting requirements 
and is not subject to nondiscrimination rules. Although employers are 
not required to contribute to an employee's SEP, when employer 
contributions are made they must be distributed as a uniform 
percentage of pay to all employees. 

In 1996, Congress also instituted a new plan design, SIMPLE, that 
allows workers to defer a portion of their salary. While SIMPLES are 
also exempt from certain nondiscrimination rules and reporting 
requirements, the employer must match the employee's elective 
contributions according to a specified formula or provide a 2 percent 
contribution for all eligible employees.[Footnote 31] 

Although plan designs such as SEP and SIMPLE offer some potential for 
increasing small business plan sponsorship, it is not clear that this 
general approach to pension simplification can make significant 
strides toward increasing plan sponsorship further among small 
employers or increasing worker coverage in that sector. Surveys 
indicate that some small employers remain unfamiliar with the 
availability of simplified plan designs. Moreover, the relief from 
many requirements and the benefits offered by such alternatives may 
not be sufficient to offset the cost or burden of offering them, and 
small employers may still be unwilling to sponsor plans given business 
conditions or worker preferences. 

Reforms to Pension Rules Attempt to Increase Coverage and Benefits: 

In addition to reforms aimed at increasing pension plan sponsorship, 
various reforms attempt to improve pension coverage and benefits by 
modifying the framework of rules governing pensions and the process 
that workers must navigate in earning pension income. Past and 
proposed reforms to eligibility, coverage, and participation 
provisions attempt to increase the number of workers who have the 
opportunity to participate in a pension plan, particularly workers who 
tend to have lower earnings. Reforms to vesting provisions could 
provide another means of helping workers gain the opportunity to earn 
pension income and possibly increase the total amounts that they 
accrue. Similarly, reforms to the regulatory provisions that set 
conditions on plan benefit designs, such as limits and 
nondiscrimination rules, as well as more direct specification of 
allowable plan designs, could affect how much workers accrue through 
their pension plans. Reform proposals that affect the distribution of 
accrued pension benefits tend to revolve around the issue of 
preretirement lump sums and whether they will contribute to workers' 
retirement income. Another issue arising from the trend toward defined 
contribution plans concerns the choices that workers will make 
regarding their investments and whether they will preserve their 
accumulations to provide lifetime income in retirement. But it is not 
clear whether most of these reforms can significantly affect coverage 
or benefits because of offsetting factors associated with employer or 
worker behavior. 

Eligibility Reforms: 

Plan eligibility provisions allow the employer to limit participation 
among younger workers or among those who do not work full time; 
further restrictions on these provisions could provide these workers 
with the opportunity to participate in a plan. However, employers may 
have little incentive to extend eligibility to workers with generally 
higher turnover, and changing these provisions could raise 
compensation costs or conflict with worker preferences in compensation. 

Coverage Reforms: 

Because pension plans are defined for specific employee groups, job 
locations, or job categories, requiring employers to expand coverage 
and give greater numbers of workers the opportunity to participate in 
a plan may be difficult. As a result, direct efforts to improve 
coverage may focus on the level of a worker's compensation by 
requiring that plans cover more workers who are not highly 
compensated. This is typically accomplished by modifying 
nondiscrimination rules (minimum coverage rules) or nondiscrimination 
testing rules. But improving coverage in this manner could conflict 
with the desire of the employer to design its plan to meet business 
needs and to direct compensation to its most valued employees. 

Participation Reforms: 

Participation reforms seek to ensure that workers who historically 
have had low participation rates, such as low-income workers, 
participate in pension plans. Some proposals to encourage 
participation in 401(k) plans would automatically enroll workers at 
the time of employment and would require them to choose to opt out of 
the plan if they so desire. Some plans have instituted such 
provisions, and research suggests that automatic enrollment does 
increase participation. Research has also shown that individuals 
enrolled in this way tend to exhibit inertia with regard to the 
amounts that they contribute, staying with the default contribution 
rates and, in their investment choices, staying with conservative 
investments such as money market funds.[Footnote 32] One automatic 
enrollment plan design, where workers agree to save a portion of their 
future salary increases, has shown promising results.[Footnote 33] 

Vesting Reforms: 

Vesting reforms seek to give workers rights to their pension accruals 
more quickly by making vesting periods shorter or even immediate. 
Previous reforms to vesting requirements appear to have substantially 
improved the percentage of plan participants who are vested. Also, the 
movement toward 401(k) plans, which have immediate vesting of employee 
contributions, helps address concerns about younger and higher-
turnover workers. Also, EGTRRA provides for faster vesting of matching 
employer contributions. From an employer's perspective, shorter or 
immediate vesting can increase the cost of providing pensions. As a 
result, the scope for further improvements in vesting may be limited, 
because employers might prefer to retain or simplify the existing 
rules and the flexibility that these rules provide to design pensions 
to meet business objectives and limit compensation costs. 

Accrual Reforms: 

Some workers, such as those with lower earnings or who change jobs 
frequently, are less likely to earn pension benefit accruals. 
Improving accruals for mobile workers generally means smoothing out 
the accrual pattern across the factors that are important in a defined 
benefit plan, namely, age, length of service, and salary. For example, 
granting higher accruals for early years of service and smaller 
accruals for higher tenure could foster the goal of providing higher 
accruals to the lower-paid, shorter-service workers. To some degree, 
the movement toward defined contribution and cash balance plans has 
alleviated concerns about greater accruals for these types of workers. 
Other means of inducing more even accrual patterns could include 
strengthening nondiscrimination rules by altering the tests to 
encourage greater accruals for individuals who are not highly 
compensated. Consistent with the theme of pension simplification, some 
reformers suggest that the pension system should allow fewer plan 
designs.[Footnote 34] However, the goal of providing more even 
accruals for all workers can conflict with the desire of employers for 
flexibility in benefit design and their ability to direct compensation 
to their most valued employees. 

Preservation Reforms: 

Preservation reforms address the issues of preretirement lump sum 
distributions and spousal rights in defined contribution plans. 
Workers who roll over a lump sum distribution into an IRA or another 
defined contribution plan can preserve the funds in a tax-deferred 
arrangement; this may provide more assurance that the pension saving 
will be preserved for retirement. As a result of concerns that lump 
sums may be consumed rather than saved, proposals have been made to 
place more restrictions on them. One option is to increase the penalty 
for not rolling the funds over into an IRA or another qualified 
retirement plan. Another option is simply to require that the funds be 
rolled over. EGTRRA generally requires a rollover to be automatic 
unless a participant elects a lump sum. This provision will go into 
effect when regulations are finalized by the Department of Labor. Such 
measures could improve benefit preservation, but some research 
suggests that greater restrictions on the use of lump sums may 
decrease workers' willingness to participate in 401(k) plans.[Footnote 
35] 

Another important issue concerns the rights of spouses regarding 
distribution from defined contribution plans. While defined benefit 
plans are required to offer an annuity with a provision that the 
spouse be able to approve the form of distribution, defined 
contribution plans are not generally required to offer an annuity 
option. Providing such an option could affect the cost of 
administering the defined contribution plan. 

Benefit Security Reforms: 

Key factors that affect workers' benefit security during the 
preretirement period involve the prudent investment of pension assets 
and workers' decisions about distributions from their plans. Pension 
plans are protected by ERISA fiduciary rules, and most defined benefit 
plan participants' benefits are protected by PBGC pension insurance. 
Although defined benefit plans are subject to a rule that no more than 
10 percent of plan assets can be invested in the securities of the 
employer, this rule does not apply generally to defined contribution 
plans. In the past and more recently, proposals have been made to 
apply restrictions on employer stock to all defined contribution plans 
or specifically to 401(k) plans, with the aim of reducing the risk 
that participants may bear. However, restrictions on investment in 
employer securities could reduce opportunities for workers to earn 
retirement income and make it less attractive for employers to 
contribute matching funds to 401(k)s.[Footnote 36] 

Participant Education Reforms: 

The trend toward defined contribution plans and increasing individual 
responsibility for retirement raises a general concern with regard to 
whether workers have sufficient knowledge and information regarding 
retirement planning and such matters as the investment of plan assets, 
preserving distributions prior to retirement, and assuring that income 
will be available throughout the retirement period. Some proposals 
would allow employers to provide plan participants with investment 
advice regarding the participant-directed assets in their 401(k) plans 
from financial service firms that administer such plans. However, 
concerns have been raised that such proposals would not adequately 
protect plan participants from potential conflicts of interest by 
investment advisors who also provide other services to their plan. 
Some pension plans are already acting to ensure that their 
participants have access to necessary information. The growth of 
401(k) plans, increased amounts of information provided through 
financial and insurance entities, and general economic and social 
trends may be encouraging workers to increase their knowledge about 
saving, investment, and retirement. Also, new strategies for improving 
worker knowledge about retirement planning are being examined. 
[Footnote 37] 

Although a variety of reforms attempt to encourage plan sponsorship 
and improve pension coverage and benefits, several analysts note that 
the ability of the voluntary, employer-based pension system to 
significantly expand pension sponsorship and extend coverage to 
workers may be limited.[Footnote 38] In particular, one study 
concluded that, at best, legislative changes are capable of extending 
coverage to a quarter to a third of uncovered workers, with actual 
results likely to be considerably lower.[Footnote 39] Consistent with 
such results, some question whether additional reforms will have 
significant results for workers who traditionally lack pensions, 
particularly those with low incomes, since these reforms offer only 
incremental changes to the voluntary, single-employer pension system. 
As a result, some reformers suggest proposals that move beyond the 
voluntary, single-employer private pension system. 

Improving Retirement Income Outside the Single-Employer Voluntary 
Pension System Involves Tradeoffs and Challenges: 

Three broad categories of reform approaches outside the single-
employer, voluntary pension system have been advanced to improve 
worker coverage and retirement income. These categories are (1) pooled 
employer reforms, (2) universal access reforms, and (3) universal 
participation reforms. Pooled employer reforms[Footnote 40] focus on 
increasing the number of firms offering pension coverage through 
centralized third-party administration. Pooled employer plans aim to 
increase worker coverage and improve pension portability, but there 
are limits to the receptiveness of employers to pooled employer plans 
given the employer's loss of control of plan design and concern with 
cost and administrative requirements. Universal access reforms attempt 
to increase retirement savings by making payroll retirement saving 
accounts available to all workers without mandating an employer 
contribution. However, these reforms raise concerns about the 
administrative burden placed on employers and, because the reforms 
rely on employee contributions, about the difficulty faced by workers, 
particularly low-income workers, in setting aside money for 
retirement. Universal participation reforms are intended to ensure 
coverage and retirement income for all workers by mandating pension 
availability and participation, similar to the existing Social 
Security system. Reforms based on universal participation raise 
concerns about increases in employer administrative burden and because 
of their broad potential economic effects on labor cost. Table 2 
provides examples of these three approaches. 

Table 2: Non-Single-Employer-Based Approaches to Increasing Retirement 
Income: 

Plan type: Existing plans; 
Pooled employer plans: Multiemployer collectively bargained plans, 
Multiple-employer trade and
association plans; 
Universal access plans: IRAs; 
Universal participation plans: Social Security (SS). 

Plan type: Proposed reforms; 
Pooled employer plans: Model pooled plans; Tax credits; Treat 
professional and trade associations as employers; 
Universal access plans: Universal voluntary individual accounts; Tax 
credits; Matching contributions; 
Universal participation plans: Mandatory private pension; Raise SS 
base benefits; Mandatory SS individual accounts. 

[End of table] 

Reforms to Advance Pooled Employer Plans Offer Advantages but Rely on 
Voluntary Employer Action: 

Existing pooled employer plans, which include multiemployer and 
multiple-employer plans, cover about 12 percent of all pension plan 
participants. Proposals advancing the pooled employer model promote 
establishing these plans in more industries and encourage small 
employer membership. Advocates of pooled employer plans maintain that 
the advantages of the plans' portability, their industry or trade 
focus, and their low administrative cost make them a viable approach 
for increasing pension coverage, particularly to employees of small 
businesses. Others contend that little incentive exists for employers 
to join a pooled employer plan, as they must sacrifice control of plan 
design and costs. In the view of these critics, existing alternatives 
such as 401(k) plans offer portability and low administrative cost and 
are even easier to administer. 

Collectively bargained pooled employer plans exist already in many 
industries and trades. These multiemployer plans, in which 
participants can negotiate the plan characteristics, must be jointly 
governed by management and labor representatives. Since their 
inception in 1929, these plans have been advanced by labor unions and 
have developed a variety of benefit structures. Usually, multiemployer 
plans provide pension coverage to labor union workers from the same 
industry or trade. Although most are defined benefit plans, 
multiemployer defined-contribution plans do exist, and hybrid models 
have developed where the employer's contribution and the worker's 
benefit are both specified. 

Non-collectively-bargained pooled employer plans, or multiple-employer 
plans, also exist and are normally administered by a professional or 
trade association. For example, the Teachers Insurance Annuity 
Association and College Retirement Equities Fund (TIAA-CREF) offers a 
multiple-employer plan organized around education and research 
professions. Employers, such as member colleges and universities, make 
contributions for their employees. TIAA-CREF offers a defined 
contribution plan, in which contributions are accumulated over a 
career and paid out at retirement, often as an annuity. 

Proposals advancing pooled employer plans would include both proposals 
that would facilitate collectively bargained plans and proposals that 
would advance development of professional and trade association plans. 
One proposal would create a model small-employer group pension plan 
with minimal administrative responsibilities. Other proposals would 
provide tax incentives to employers to encourage participation in 
pooled employer arrangements. Another proposal would make changes in 
income tax law to allow professional and trade associations to be 
treated as employers for purposes of sponsoring pooled employer 
pensions or health plans for their members. 

Advocates of pooled employer plans reason that both employers and 
employees benefit from the portability and trade focus of this 
arrangement. The portability of the plans improves worker pension 
receipt by allowing short-service workers to accumulate pension 
benefits with different employers. This portability diminishes the 
effects on pension accruals of company ownership changes and failures, 
because workers can continue to participate with new or reorganized 
employers. The trade focus enhances the advantages of portability, 
because even though workers may change employers, many stay in the 
same industry or trade. Similarly, employers benefit by having a pool 
of workers with previous work and training in their industry or trade, 
and pooled employer plans are likely to have pension features, such as 
early retirement provisions, to meet the needs of a common industry or 
trade. Advocates also note that workers in small business, in 
particular, could benefit from the pooled employer model because small 
employers generally have high rates of employee turnover and high 
business termination rates. 

By lowering the cost of administering a pension plan, pooled employer 
plans also offer employers a more cost-effective way of providing 
pensions to their employees. Because they provide economies of scale 
and reduce employer costs, such plans are easier for some employers to 
offer. Advocates note that pension administrative costs per employee 
are normally higher for small employers who have smaller numbers of 
workers over which to spread implementation and administrative costs. 
Pooled employer pension plans spread these costs over a larger number 
of workers. 

Despite these possible benefits, some pension experts have expressed 
doubt that pooled employer models can be widely expanded beyond 
current levels, because pooled employer plans are still dependent on 
voluntary employer action. They note that pooled employer plans have 
been available for many years, yet small businesses have shown little 
interest in them. Employers may be less likely to adopt pooled 
employer plans, because they have little control over plan design and 
are less able to assure that the plan meets their needs. Further, 
little evidence exists that proposals such as employer tax credits 
will lead to adoption of pooled employer plans by businesses without 
pension plans. Moreover, employers may have little incentive to choose 
a pooled-employer defined-benefit plan instead of a single-employer 
401(k) plan, which also is portable and offers low administrative 
costs. 

Reforms to Create Universal Access to Retirement Saving Rely on 
Voluntary Worker Election: 

Recognizing that many employers do not provide pension plans to 
workers and that some employees with coverage need additional 
retirement savings, some analysts and policymakers embrace reforms to 
assure universal access to tax-favored retirement savings accounts 
such as IRAs.[Footnote 41] Although legislation has created different 
IRA types and provisions, workers generally establish IRAs outside the 
workplace. Proposals that would expand universal access accounts 
beyond IRAs vary in coverage and in incentive features such as tax 
credits to encourage employer or employee participation. Many of these 
proposals seek to provide employees with a payroll-based opportunity 
for retirement saving.	 

Some form of IRA is currently available to all workers. ERISA 
introduced the IRA in 1974 as a means of promoting retirement savings 
for workers without employer-sponsored pensions. Since then, 
legislation has modified provisions and created new types of tax-
advantaged IRAs.[Footnote 42] Today, traditional IRAs can be purchased 
with pretax dollars if a person is not covered by a pension plan or if 
his or her income is less than specified amounts. IRAs can also be 
purchased with after-tax dollars, regardless of income. For these 
traditional IRAs, earnings are taxed as income at retirement. 

Reforms advancing universal access accounts aim to facilitate 
increased retirement saving. To increase the likelihood of worker 
participation, most proposals call for payroll-based accounts. Some 
would offer universal access accounts to employees regardless of other 
pension coverage; others would apply only to employees without pension 
coverage. Some proposals would require employers to establish the 
accounts, while other proposals would make the accounts available at 
an employer's or employee's election. Also included in proposals is 
the option of a government-managed payroll account as an alternative 
for employers, particularly small employers, who want to minimize 
their administrative involvement with employee accounts. To encourage 
employee saving, some proposals include incentives such as tax credits 
and matching of employer contributions.[Footnote 43] 

Advocates of universal access accounts reason that requiring such 
accounts would facilitate employee and employer contributions even 
without a required employer contribution. They reason that workers are 
more likely to routinely set aside retirement savings when they have a 
payroll deduction account and when they receive employer contributions 
to that account. Further, employers may be more likely to make 
contributions when there is an existing account. 

IRA experience may be useful in predicting the effects of universal 
access accounts. Although an estimated 42 percent of households owned 
some type of IRA as of May 2001, evidence suggests that IRAs serve 
more as a parking place for distributions from other tax-qualified 
retirement savings plans than as accounts for active retirement 
saving. Rollover contributions from other tax-qualified retirement 
accounts are estimated to represent more than 90 percent of current 
IRA contributions. A study of a large sample of individual tax returns 
found that only about 5 percent of individuals reporting income made a 
contribution to an IRA in 1995.[Footnote 44] 

Studies show that low-income workers have the lowest rate of IRA 
saving and that the rate of contributions to IRA accounts rises as 
incomes rise.[Footnote 45] In 1995, only one percent of those with 
income of less than $10,000, compared with 17 percent of those with 
income more than $100,000, contributed to an IRA. Observers note that 
the low rate of IRA saving by low-income workers is not surprising in 
that low-income workers have the smallest amount of disposable income 
for saving. Further, low-income workers obtain the least tax savings 
from tax-deferred treatment, because they pay the lowest marginal tax 
rates. However, universal access account proposals that include tax 
credits or matches by the government or employer, based on the 
contributions of the worker, attempt to improve saving incentives for 
lower earners.[Footnote 46] 

Critics of universal access reform proposals argue that universal 
access accounts are not the best way of increasing retirement saving. 
They suggest that such proposals may increase the administrative 
burden on employers, particularly small employers, and create numerous 
small accounts with relatively high administrative expenses Experts 
disagree about whether 401(k) plan accounts or IRA accounts have 
increased personal saving. They note that lower-income workers face 
lower tax rates and therefore benefit less from the tax-deferred 
nature of the accounts. In addition, these critics note that such 
plans shift investment risk to the individual and that lower-income 
workers have little investment management experience. Some are 
concerned that individual accounts could supplant existing private 
pensions, resulting in employers' feeling less need to offer 
traditional pension benefits and leading to a possible drop in 
national saving. The proposals also entail substantial design 
challenges to ensure that universal accounts are effectively 
implemented and administered. These challenges include determining how 
records would be kept, what investment options and controls would be 
offered, and when workers would gain access to savings in the accounts. 

Universal Participation Involves Mandates; Broader Effects Are 
Uncertain: 

Although reforms requiring universal participation in a pension system 
are aimed at improving workers' retirement income, concerns exist 
about the broad economic effects of such reforms. Three primary types 
of reform employ universal participation: (1) reforms mandating 
private pension coverage in addition to Social Security, (2) reforms 
increasing base-level Social Security benefits, and (3) reforms 
establishing mandatory Social Security individual savings accounts. 
[Footnote 47] 

Mandatory Pensions as a Second Tier to Social Security: 

Mandatory pension proposals differ in specific provisions but 
generally require pension coverage and employer contributions for all 
employees. Under mandatory pension proposals, employers would be 
required to establish pension accounts and make contributions for 
workers. Proponents of these reforms suggest that mandatory pensions 
would increase private retirement saving, particularly for low-income 
workers, and would take advantage of the existing private pension 
infrastructure. Proposals mandating employer pensions aim to provide 
retirement income as a second tier to Social Security, but critics 
suggest that if these proposals are implemented, they may have adverse 
impacts on the national economy because of the increased cost of labor 
and potentially increased layoffs. 

Several mandatory pension proposals have been suggested. For example, 
the 1981 report from the President's Commission on Pension Policy 
recommended an advance-funded minimum universal pension system (MUPS). 
The commission recommended that employers establish pension accounts 
for all employees[Footnote 48] and contribute a minimum of 3 percent 
of pay annually. The MUPS proposal required immediate vesting and 
prohibited integration with Social Security. Under MUPS and other 
mandatory pension proposals, employers would be required to establish 
pension accounts and make contributions for workers. Another, more 
recent proposal required employers to provide uniform pension coverage 
for all employees in a given line of business but allowed for workers 
with income below a certain threshold to be excluded from employer-
sponsored coverage and to instead receive their retirement income from 
the government. To help ensure employer participation, this proposal 
offered increased employer flexibility in benefit and contribution 
limits.[Footnote 49] 

Proponents of a mandatory pension system reason that mandatory 
pensions can take advantage of the existing private pension 
infrastructure and increase national saving by providing a retirement 
saving mechanism to more workers. Low- and moderate-income workers 
represent a disproportionate share of those without pensions, so 
mandating pension coverage would increase the retirement incomes of 
these workers, who generally lack retirement income other than Social 
Security. Because of the low rate of retirement and other savings, 
particularly for lower-income workers, some proponents of a mandatory 
pension system believe that mandating pensions would increase personal 
retirement savings. Mandating pensions would increase pension coverage 
provided by small employers, where it has been difficult to increase 
coverage. In addition, a mandatory pension system could take advantage 
of the existing private sector pension system infrastructure. 

However, critics of mandatory universal pension proposals suggest that 
such plans may adversely affect both employees and employers. 
Mandatory pensions may require workers to receive compensation in the 
form of pension benefits when they might prefer cash wages, which may 
be a particular concern of low-income workers. Mandatory pensions 
would reduce workers' ability to allocate earnings to other valuable 
uses, such as health insurance, housing, and education. Employees with 
current pension coverage could be adversely affected if employers 
chose to reduce benefits to the mandatory minimum. In addition, 
mandatory pensions could have negative consequences for employers, 
increasing employers' costs for pension implementation, 
administration, and contributions. Mandatory pensions could also 
restrict employers' ability to design pensions to meet their business 
objectives. Such reforms raise concerns about the increase in 
employers' administrative burden, as well as potential adjustments to 
other forms of compensation to offset higher pension costs. 

Increase Social Security Base-Level Benefits: 

Some analysts acknowledge that extending pension coverage and benefits 
to workers by making the voluntary system mandatory is a difficult 
option and that it may make more sense to simply modify the existing 
mandatory Social Security system. One proposed reform involves raising 
the base level of Social Security retirement benefits. Such a proposal 
attempts to increase Social Security benefits for low-earning workers, 
recognizing that they generally lack pension income, have very little 
retirement savings, and are therefore dependent on Social Security. 
Proponents of such a proposal cite the simplicity and low 
administrative cost of increasing the base level benefits, but 
concerns remain about the potential impact of this approach when a 
Social Security financing shortfall already exists. 

Proposals to raise the base level of Social Security benefits try to 
offset the effect on retirement income of low wage, part-time, or 
seasonal employment as well as periods of unemployment. These 
proposals would raise Social Security benefits so that low earners 
would receive higher replacement of preretirement income.[Footnote 50] 
Proposals have different ways of providing the higher benefits for low 
earners. One option is to revise Social Security's minimum benefit 
provision.[Footnote 51] Other options would change the benefit formula 
for specific workers, and others would count unemployment insurance 
payments and the Earned Income Tax Credit (EITC)[Footnote 52] as 
earnings in computing Social Security benefits. 

Proponents of increasing base-level Social Security benefits cite the 
simplicity of using the existing, relatively efficient Social Security 
system to compensate for the lack of pensions and retirement savings 
of many low earners. They reason that the workers who would benefit 
most from this change are those with the least retirement savings and 
the greatest dependence on Social Security. 

Critics of these proposals suggest that raising the base benefit level 
may detract from Social Security's financial integrity and popular 
support. Increasing Social Security benefits, even for a limited 
segment of retirees, would further compound the existing shortfall in 
Social Security financing. Restoring solvency in light of these 
benefit increases may require reducing benefits to workers with higher 
earnings or increasing worker and employer contributions. Some fear 
that such adjustments might cost the program the support of these 
higher-income workers, if Social Security came to be viewed as a 
welfare program. Moreover, increasing Social Security benefits may 
have implications for private pensions, making employers less likely 
to want to provide pension benefits for their lower-earning workers. 

Social Security Individual Accounts: 

Some current efforts to reform Social Security financing call for the 
establishment of individual Social Security savings accounts.[Footnote 
53] These proposals seek to partially replace the current pay-as-you-
go financing of Social Security in which current contributions are 
generally used to pay current retiree benefits. Advocates of these 
proposals suggest that such accounts would increase overall worker 
retirement income with higher market investment returns and would 
provide greater worker control of retirement savings. However, critics 
question whether individual accounts can increase retirement income, 
and they counter that low-income workers would benefit the least from 
such accounts because they have relatively little to contribute and 
modest investment experience. 

Individual account reform proposals vary, but they generally allow 
workers to own and, to varying degrees, manage their own accounts. The 
proposals would create individual accounts in different ways. Some 
would finance individual accounts with new contributions, while others 
would allocate some portion of the current Social Security taxes to 
fund the accounts. Still others would allow supplementary voluntary 
contributions to mandatory individual accounts or be based completely 
on voluntary contributions. Most proposals retain some features of the 
current Social Security system. One hybrid proposal would completely 
redesign the Social Security program into a two-tier program, with the 
second tier consisting of an individual account.[Footnote 54] 

Proponents of Social Security individual accounts maintain that such 
accounts allow workers to invest a portion of their contributions and, 
with the returns, to fund future retirement benefits. Advocates of 
Social Security individual accounts point to the potential for 
increased returns for participants that could result from allowing 
investment in stocks and bonds. Some advocates indicate that in 
addition to offsetting the need to raise payroll taxes or cut benefits 
to restore financial solvency to Social Security, individual accounts 
could eventually increase the overall retirement income of future 
retirees. Furthermore, Social Security individual accounts could 
provide an administrative infrastructure for other retirement savings 
plans, such as plans based solely on employee payroll deductions. 
Workers might also become more inclined to contribute an increased 
portion of their wages to retirement savings if such plans were 
available. Advocates therefore reason that Social Security individual 
accounts could increase private and national saving and lead to more 
capital formation. 

Individual Social Security accounts also have critics. Critics of 
individual accounts point out that investing in stocks and bonds 
introduces investment risk that could, in certain cases, result in 
lower retirement income. Moreover, they argue that individual accounts 
are unlikely to restore Social Security's solvency without the need 
for additional financing through tax revenues, benefit reductions, or 
government borrowing. Concerns have also been raised about the impact 
on benefits, in that lower-income workers would have fewer funds going 
to their individual accounts and would have the least investment 
experience. Finally, concerns have been raised that employers may 
redesign their pensions or drop pension coverage if they feel that 
Social Security individual accounts allow workers to accumulate 
adequate retirement income. 

Concluding Observations: 

The concern about the low rate of private pension coverage among 
certain segments of the workforce and the desire to improve pension 
and retirement income, particularly for lower earners, has led to 
various proposals to reform the existing voluntary employer-based 
system, as well as some proposals that move outside that system. 
However, each type of reform introduces issues that make the likely 
effects of reform difficult to determine. For example, under the 
existing system, the effect of policies aimed at improving incentives 
for plan sponsorship through the tax system or by simplifying pension 
rules may be limited by other policy actions. The intended effects of 
changing pension rules may be counteracted by the responses of 
employers and workers. As a result, additional reforms to the 
voluntary, single-employer-based system have only a limited ability to 
significantly expand pension sponsorship and extend coverage and 
benefits to workers who traditionally lack pensions. 

In considering proposals that move outside the voluntary, single-
employer system, employers may find long-standing proposals, such as 
those that would expand pooled employer arrangements and mandate 
private pensions, unattractive in part because they may increase 
compensation costs. While raising the base level of Social Security 
benefits might be an effective means of addressing some of the 
concerns about lower-earning workers, such a reform would need to be 
considered as part of the broader Social Security financing reform 
discussion. Several pension-related proposals aimed at improving the 
availability and level of retirement income for lower-earning workers 
are similar in many respects to current proposals to introduce an 
individual account-based option into Social Security. The 
infrastructures of private pensions or Social Security could be 
modified to provide a universal, payroll-based opportunity to
save for retirement. While many lower-earning workers may have 
difficulty saving out of current income, supplementing a worker's 
account through tax credits and contribution matches might increase 
saving incentives among those with low levels of income and retirement 
wealth. Such approaches entail cost and design challenges, but it is 
important to recognize the relationship between concerns about private 
pension coverage and benefits, and the Social Security policy debate, 
in any retirement policy reforms that emerge. The outcome of reform 
efforts will define a new balance between voluntary and mandatory 
approaches to providing retirement income. 

Agency Comments: 

We provided draft copies of this report to the Department of Labor and 
the Department of the Treasury for their review. The Department of 
Labor had no comment on the report. The Department of the Treasury 
provided us with technical comments, which we incorporated as 
appropriate. 

We are providing copies of this report to Secretary of Labor Elaine L. 
Chao, Secretary of the Treasury Paul H. O'Neill, and appropriate 
congressional committees. We will make copies available to others on 
request. The report is also available on GAO's home page at 
[hyperlink, http://www.gao.gov]. Please call me on (202) 512-7215 or 
George A. Scott on (202) 512-5932 if you or your staff have questions. 
Other major contributors to this report include Kenneth J. Bombara, 
Timothy Fairbanks, Edward Nannenhorn, Corinna Nicolaou, Roger J. 
Thomas, and Charles Walter III. 

Signed by: 

Barbara D. Bovbjerg, Director: 
Education, Workforce, and Income Security Issues: 

[End of section] 

Appendix I: Advantages of Tax-Qualified Retirement Savings: 

The purpose of this appendix is to show (1) how the tax treatment of 
saving through a qualified pension plan differs from the tax treatment 
of saving in a regular bank savings account, (2) how the magnitude of 
the difference depends on the tax rates individuals face, and (3) that 
the tax treatment of pension saving can be equivalent to exempting the 
earnings on pension contributions. 

If a person's employment compensation is paid as wages, those wages 
would be taxable income. If he or she then saves some of these wages 
in a regular bank savings account, the income earned in the account 
would be taxable each year as it is earned. When funds are withdrawn 
from the account, no further tax would be owed.[Footnote 55] 

If the same employee receives compensation in the form of a 
contribution to a qualified pension plan, that pension contribution 
would not be counted as income to the employee at the time of the 
contribution. In addition, earnings on the contribution would 
accumulate tax deferred. When the contributions and earnings are 
withdrawn or distributed, they would be subject to tax at the regular 
income tax rates applicable at that time.[Footnote 56] 

Table 3 shows a hypothetical example of how the tax treatment afforded 
to pensions can benefit savers. It also shows how the tax benefit from 
saving in a pension depends on a person's income tax rate. The example 
in this table supposes that two people are subject to different tax 
rates, one to a 15-percent tax rate and the other to a 28-percent 
rate, throughout their lives. Both receive a higher after-tax return 
from saving through a pension than they would have received in a 
regular taxable account. In both cases, the value of their pension 
accounts at retirement is greater than the value of their regular 
savings account at the time funds are withdrawn. This reflects the 
effect of taxes not paid at the time of the initial deposit in the 
pension account and taxes not paid on the earnings in the pension 
account over time. Despite the fact that both individuals have to pay 
tax on the value of the pension account when the funds are 
distributed, while no additional tax is owed on the funds in the 
regular saving account, both individuals gain by saving through the 
pension instead of the regular account. 

Table 3 also shows that the person with the higher, 28-percent tax 
rate benefits more from saving through a pension, compared with a 
regular savings account, than the person with the lower, 15-percent 
rate. 

Table 3: Tax Treatment of Saving in a Defined Contribution Plan versus 
a Regular Savings Account (constant tax rates): 

Contribution: 
15% tax rate: Regular: $1,000; 
15% tax rate: Pension: $1,000; 
28% tax rate: Regular: $1,000; 
28% tax rate: Pension: $1,000. 

Tax on contribution: 
15% tax rate: Regular: $150; 
15% tax rate: Pension: $0; 
28% tax rate: Regular: $280; 
28% tax rate: Pension: $0. 

Net deposit in account: 
15% tax rate: Regular: $850; 
15% tax rate: Pension: $1,000; 
28% tax rate: Regular: $720; 
28% tax rate: Pension: $1,000. 

Value at withdrawal: 
15% tax rate: Regular: $2,280; 
15% tax rate: Pension: $3,172; 
28% tax rate: Regular: $1,668; 
28% tax rate: Pension: $3,172. 

Tax on withdrawal: 
15% tax rate: Regular: $0; 
15% tax rate: Pension: $476; 
28% tax rate: Regular: $0; 
28% tax rate: Pension: $888. 

Net withdrawal: 
15% tax rate: Regular: $2,280; 
15% tax rate: Pension: $2,696; 
28% tax rate: Regular: $1,668; 
28% tax rate: Pension: $2,284. 

Gain over regular account: 
15% tax rate: Pension: $416; 
28% tax rate: Pension: $616. 

Percentage gain: 
15% tax rate: Pension: 18%; 
28% tax rate: Pension: 37%. 

Note: Funds in both accounts accumulate for 15 years at an interest 
rate of 8 percent. For any given tax rate, the tax benefit of a 
pension will also vary with the rate of return and period of 
accumulation of funds in the account. 

Source: Congressional Budget Office. 

[End of table] 

The example in table 3 assumed that the lifetime tax rate—when 
contributions are made, as earnings accrue, and when funds are 
withdrawn or distributed—remains constant. When tax rates vary over 
time, the tax benefits from saving through a pension are greater if 
the rates that are applicable when contributions are made and as 
earnings accrue exceed the rates applicable when the funds are 
withdrawn. In other words, if the tax rate during a person's working 
life is higher than the tax rate during retirement, the tax benefits 
from pension saving will be greater. Conversely, if tax rates are 
higher during retirement than during a person's working life, the 
relative tax benefits are smaller. When tax rates are low during a 
person's working life and much higher during retirement, the person 
might be better off saving in a regular taxable account.
Another way to look at the tax treatment of pension savings is to 
compare it with that of an account in which contributions are taxable 
but no further tax is owed on earnings. In a Roth IRA, for example, 
wages are subject to tax when they are earned, but any account 
earnings can be permanently exempt from tax.[Footnote 57] Table 4 
shows that if tax rates remain constant over time as in the example 
underlying table 3, the after-tax return from saving through a pension 
can be equivalent to saving through a Roth IRA. 

Table 4: Comparison of Tax Treatments of Saving in a Defined 
Contribution Plan and Saving in a Roth IRA: 

Contribution: 
15% tax rate: Regular: $1,000; 
15% tax rate: Pension: $1,000; 
28% tax rate: Regular: $1,000; 
28% tax rate: Pension: $1,000. 

Tax on contribution: 
15% tax rate: Regular: $150; 
15% tax rate: Pension: $0; 
28% tax rate: Regular: $280; 
28% tax rate: Pension: $0. 

Net deposit in account: 
15% tax rate: Regular: $850; 
15% tax rate: Pension: $1,000; 
28% tax rate: Regular: $720; 
28% tax rate: Pension: $1,000. 

Value at withdrawal: 
15% tax rate: Regular: $2,696; 
15% tax rate: Pension: $3,172; 
28% tax rate: Regular: $2,284; 
28% tax rate: Pension: $3,172. 

Tax on withdrawal: 
15% tax rate: Regular: $0; 
15% tax rate: Pension: $476; 
28% tax rate: Regular: $0; 
28% tax rate: Pension: $888. 

Net withdrawal: 
15% tax rate: Regular: $2,696; 
15% tax rate: Pension: $2,696; 
28% tax rate: Regular: $2,284; 
28% tax rate: Pension: $2,284. 

Source: Congressional Budget Office. 

[End of table] 

[End of section] 

Appendix II: Perspectives on Adequacy of Retirement Income: 

Currently, an active research debate is addressing the questions of 
whether workers and households will achieve adequate retirement income 
and the role that pensions play in retirement income. Data are 
generated for the current retired population, and estimates are made 
for those who will retire in the future. The current status of 
retirees is typically examined through comparisons with the poverty 
line or with replacement rates, which relate actual or expected 
retirement income to the income level at a period of time during the 
worker's career. The status of future retirees also can be assessed 
through estimates of such measures but is increasingly examined in the 
context of whether workers are accumulating sufficient assets while 
working (i.e., saving) to assure themselves of a stream of retirement 
income adequate to meet certain standards or targets. 

Income Adequacy for Existing Retirees: 

Data on existing retirees recently presented by GAO[Footnote 58] 
suggests that those without pension income in retirement are more 
likely to be in poverty. In 1998, about 4.2 million of 36.6 million 
retired persons, or 11.5 percent, had total retirement incomes below 
the poverty line. In addition, about half of those retired (17.6 of 
36.6 million) reported that they did not receive income from a pension 
of their own or from that of a spouse. Of those not receiving pension 
income, about 21 percent had retirement incomes below the federal 
poverty line; of those who did receive some pension income, only 3 
percent had incomes below the poverty line. Furthermore, the study 
noted that some of the characteristics of those who lack pension 
income in retirement are similar to the characteristics of workers who 
lack pension coverage during their working years. For example, those 
without pension income in retirement are more likely to be single, to 
be women, and to have low levels of education. However, it is not 
possible to predict whether any particular worker currently in the 
labor force will ultimately receive a pension benefit. That is, the 
linkage between work, pension coverage, and the receipt and level of 
pension income in retirement is complex and depends on an array of 
factors, such as employer plan sponsorship and benefit design, the 
framework of government rules, and worker decisions and choices over a 
lifetime. 

Data on the status of current retirees also focuses on the replacement 
rates that are provided via Social Security and pensions. Typically, 
pension professionals suggest that a worker or family needs 
approximately 65 to 85 percent of preretirement income to maintain the 
preretirement living standard.[Footnote 59] The achievement of this 
level of income replacement depends significantly on Social Security 
and pension income and may require income from other sources, such as 
earnings from employment, home equity, and nonpension saving.[Footnote 
60] Studies show that many workers need to save for retirement beyond 
the income they can expect from Social Security and pensions. Owing to 
the tilt of Social Security benefits toward lower earners, it follows 
that those in lower earnings categories generally need to save 
proportionately less than those in higher earnings groups to reach an 
adequate replacement rate. At the same time, workers in lower earnings 
categories are less likely than higher earners to have pension income 
in retirement. 

Income Adequacy for Future Retirees: 

Research has also focused on the question of whether future retirees 
will have adequate retirement income. In the early to mid-1990s, a 
number of research studies engaged the retirement income adequacy 
question and reached different conclusions. Studies by 
Andrews[Footnote 61] and the Congressional Budget Office[Footnote 62] 
(CBO) reached generally positive conclusions concerning the retirement 
income status of future retirees. Research by Bernheim reached less 
optimistic conclusions, finding that a broader range of workers were 
not saving sufficiently more than the amounts they could receive from 
Social Security and pensions to assure themselves of an adequate 
retirement income.[Footnote 63] More recently, data from the Health 
and Retirement Study (HRS)[Footnote 64] has been applied in several 
studies of retirement income adequacy. 

In general, the adequacy debate continues, with researchers 
interpreting the data in different ways. These studies tend to focus 
on measuring asset (wealth) accumulation in a present value context in 
which retirement income sources such as Social Security and pensions 
are represented as asset values. The studies estimate the likely total 
asset accumulation at retirement by workers in their sample, and some 
studies may incorporate a target saving rate approach that is 
analogous to the replacement rate concept. Using HRS data, Gustman and 
Steinmeier reached positive conclusions about the retirement saving of 
future retirees and found pensions to be widely distributed among 
households.[Footnote 65] However, Mitchell and Moore, also using HRS 
data, concluded that the majority of households nearing retirement age 
will not be able to maintain current levels of consumption in 
retirement without additional saving.[Footnote 66] They found 
considerable variation in wealth across the income distribution but 
also wide variation in wealth among households within a given earnings 
level. 

They also found a rather low correlation of wealth to earnings. This 
means that low retirement saving is not strictly a low earnings 
phenomenon: there are high earners with low retirement wealth and low 
earners with relatively high retirement wealth. Mitchell and Moore's 
results also suggest that although the need to save increases with 
higher earnings, when households are arrayed according to retirement 
wealth, those with the lowest wealth face significant risk of 
inadequate retirement income.[Footnote 67] Recent research by Engen, 
Gale, and Uccello provides a different interpretation of Mitchell and 
Moore's results, but their findings are consistent with the conclusion 
that there appear to be different preferences or propensities in the 
population for accumulating retirement wealth and that inadequate 
retirement income appears to be associated with low retirement saving 
by a segment of the workforce.[Footnote 68] 

[End of section] 

Related GAO Products: 

Private Pensions: Key Issues to Consider Following the Enron Collapse, 
[hyperlink, http://www.gao.gov/products/GAO-02-480T]. Washington, 
D.C.: February 27, 2002. 

Social Security: Program's Role in Helping Ensure Income Adequacy. 
[hyperlink, http://www.gao.gov/products/GAO-02-62]. Washington, D.C.: 
Nov. 30, 2001. 

Private Pensions: Issues of Coverage and Increasing Contribution 
Limits for Defined Contribution Plans. [hyperlink, 
http://www.gao.gov/products/GAO-01-846]. Washington, D.C.: Sept. 17, 
2001. 

Retirement Savings: Opportunities to Improve DOL's SAVER Act Campaign. 
[hyperlink, http://www.gao.gov/products/GAO-01-634]. Washington, D.C.: 
June 26, 2001. 

National Saving: Answers to Key Questions. [hyperlink, 
http://www.gao.gov/products/GAO-01-591SP]. Washington, D.C.: June 1, 
2001. 

Cash Balance Plans: Implications for Retirement Income. [hyperlink, 
http://www.gao.gov/products/GAO/HEHS-00-207]. Washington, D.C.: Sept. 
29, 2000. 

Private Pensions: Implications of Conversions to Cash Balance Plans. 
[hyperlink, http://www.gao.gov/products/GAO/HEHS-00-185]. Washington, 
D.C.: Sept. 29, 2000. 

Social Security Reform: Implications for Private Pensions. [hyperlink, 
http://www.gao.gov/products/GAO/HEHS-00-187]. Washington, D.C.: Sept. 
14, 2000. 

Private Pensions: "Top-Heavy" Rules for Owner-Dominated Plans. 
[hyperlink, http://www.gao.gov/products/GAO/HEHS-00-141]. Washington, 
D.C.: Aug. 31, 2000. 

Pension Plans: Characteristics of Persons in the Labor Force Without
Pension Coverage. [hyperlink, 
http://www.gao.gov/products/GAO/HEHS-00-131]. Washington, D.C.: Aug. 
22, 2000. 

Social Security: Evaluating Reform Proposals. [hyperlink, 
http://www.gao.gov/products/GAO/AIMID/HEHS-00-29]. Washington, D.C.: 
Nov. 4, 1999. 

Integrating Pensions and Social Security: Trends Since 1986 Tax Law 
Changes. [hyperlink, http://www.gao.gov/products/GAO/HEHS-98-191R]. 
Washington, D.C.: July 6, 1998. 

Social Security: Different Approaches for Addressing Program Solvency. 
[hyperlink, http://www.gao.gov/products/GAO/HEHS-98-33]. Washington, 
D.C.: July 22, 1998. 

401(k) Pension Plans: Loan Provisions Enhance Participation But May 
Affect Income Security for Some. [hyperlink, 
http://www.gao.gov/products/GAO/HEHS-98-5]. Washington, D.C.: Oct. 1, 
1997. 

Retirement Income: Implications of Demographic Trends for Social 
Security and Pension Reform. [hyperlink, 
http://www.gao.gov/products/GAO/HEHS-97-81]. Washington, D.C.: July 
11, 1997. 

[End of section] 

Footnotes: 

[1] U.S. General Accounting Office, Pension Plans: Characteristics of 
Persons in the Labor Force Without Pension Coverage, [hyperlink, 
http://www.gao.gov/products/GAO/HEHS-00-131] (Washington, D.C.: Aug. 
22, 2000). 

[2] Vesting provisions specify when workers acquire the irrevocable 
right to pension benefits. 

[3] For simplicity, we use "pooled employer plans" when referring to 
plans covering two or more employers. These plans include 
"multiemployer plans," defined as plans maintained pursuant to 
collective bargaining agreements to which more than one employer 
contributes, and "multiple-employer plans," in which employers are 
often members of, or otherwise related to, a professional or trade 
association. 

[4] Researchers note that defining retirement income is difficult 
because the concept of retirement is elusive and the definition 
depends on which sources of income are considered. Pension income is 
particularly difficult to define owing to changes in plan types and 
the increase in the use of lump sum distributions. 

[5] Social Security's progressive benefit structure is illustrated by 
means of the replacement ratio, which is the benefit earned by the 
retiree divided by a measure of preretirement earnings. Typical 
replacement rates for a worker retiring at 65 years of age in January 
2001 are as follows: low earner, 53 percent; average earner, 39 
percent; high earner, 32 percent. Assuming that the total percentage 
of preretirement income replaced by Social Security and employer-
sponsored pensions combined is reasonably constant, private pensions 
would tend to play a larger role in the retirement income of higher 
earning workers. 

[6] If a plan meets the Internal Revenue Code requirements and becomes 
qualified, contributions to the plan and plan earnings are afforded 
favorable income tax treatment. Contributions made to qualified plans 
and plan earnings are not included in the taxable income of the 
employee until the contribution and earnings are distributed. In 
contrast, if an employee saves on his own, contributions to a saving 
account are taxed, and earnings are generally taxed as they accrue in 
the account (see appendix I). 

[7] For a more detailed discussion, see Alan L. Gustman, Olivia S. 
Mitchell, and Thomas L. Steinmeier, "The Role of Pensions in the Labor 
Market: A Survey of the Literature," Industrial and Labor Relations 
Review 47, no. 3 (1994): 417-438. 

[8] Also, the tax advantages of pensions have traditionally played a 
role in the financial management of the corporation, allowing firms 
some flexibility in minimizing their tax liability and funding plans 
less expensively. For example, subject to certain conditions and 
limitations, a firm may contribute more to the plan during profitable 
years, thus lowering its tax liability, and contribute less during 
times when profitability is poor. Thus, funding methods and rules play 
a role in the employer's decision to sponsor a plan. 

[9] In defining the term "coverage," it is important to understand the 
different ways that a worker can be associated with a pension plan. 
First, a worker who is employed by a firm that sponsors a plan is 
considered to have a plan "available." Second, a worker may be 
"covered" by the plan but not eligible for benefits as a participant. 
Third, a worker may be covered by the plan and actually participate in 
it. Thus, while coverage and participation are not strictly the same 
concepts, they are often used interchangeably in discussions of 
pensions. One exception is in 401(k) arrangements where participation 
is voluntary. For further discussion see, Alicia Munnell and Annika 
Sunden, "Private Pensions: Coverage and Benefit Trends" (paper 
presented at "Conversation on Coverage," a conference at the Pension 
Rights Center, Washington, D.C., July 2001). 

[10] Typically defined as at least 1,000 hours of work in a 12-month 
period. 

[11] In a defined benefit plan, the employer promises a worker set 
payments, often calculated according to the worker's earnings and 
tenure with the firm, paid out for the duration of his or her 
retirement. Because the employer agrees to these future payments, the 
firm bears the risk associated with funding the plan. In a defined 
contribution plan, an employer contributes a specific amount to an 
account for each worker; the worker's benefit in retirement is based 
on the cumulative account balance. Under defined contribution plans, 
workers bear the investment risk associated with the account, because 
there is no promise made by the employer that money will be available 
during retirement. There are various types of defined contribution 
plans that involve employer contributions, including money purchase 
plans, profit sharing plans, stock bonus plans and employee stock 
ownership plans (ESOP). This last plan is typically invested in 
company stock. Alternatively, in 401(k) plans (based on section 401(k) 
of the Internal Revenue Code), workers are allowed to make tax-
deferred contributions to the plan, which may also include employer 
contributions. 

[12] Some pension plans are "hybrid plans"—-for example, cash balance 
plans are defined benefit plans with features resembling those of 
defined contribution plans. 

[13] Section 415 of the Internal Revenue Code. 

[14] The Economic Growth and Tax Relief Reconciliation Act of 2001 
(EGTRRA) revised the limits applicable to various plans in 2002. For 
example, the maximum amount of annual compensation for an individual 
in a defined benefit plan was increased from $170,000 to $200,000. The 
maximum annual benefit payable from the plan was increased from 
$140,000 to $160,000. For defined contribution plans, the maximum 
annual contribution to the plan (employer and employee combined) for a 
worker was increased from $35,000 to $40,000. The maximum amount of 
salary that an employee can defer under 401(k) plans was increased 
from $10,500 to $11,000 (gradually increasing to $15,000 for 2006). 
Additional deferrals (up to $5,000 for 2006) are permitted after 50 
years of age. 

[15] The "nondiscrimination" limits on coverage and benefits require 
(a) that the proportion of "non-highly compensated employees" (making 
less than $90,000 annually) covered by the plan be at least some 
minimum fraction of the "highly compensated employees" (making $90,000 
or more); and (b) that the contributions or "benefits" going to the 
lower paid be "equivalent" to those going to the highly paid. In 
making this comparison, as provided under section 401(a)(5)(C) of the 
Internal Revenue Code, a plan can be "integrated" with Social Security 
under specific permitted disparity rules. Under integration, a portion 
of the Social Security (OASD1) benefits or FICA taxes for an employee 
are considered paid by the employer. Because Social Security benefits 
and FICA taxes tend to be disproportionately larger for low- and 
moderate-income workers, integration permits a qualified retirement 
plan to provide disproportionately larger benefits to highly 
compensated employees. If an employer designs a plan to the regulatory 
limit, the proportion of lower-paid workers covered can be as little 
as one-fifth of the proportion of highly paid who are covered, but the 
plan must provide higher average benefits to the lower paid than to 
the higher paid to reach this coverage limit. By using complex benefit 
testing rules to the limit, an employer can give highly paid, older 
workers a contribution 36 times greater than that for lower-paid, 
younger, and shorter-service workers in a defined contribution plan. 
In a defined benefit plan, the current value of the accrual for older, 
highly paid workers can exceed by 100 times that of lower-paid, 
younger, and shorter-service workers. The regulations, however, also 
provide "safe harbor" coverage and benefit design options with 
simpler, more uniform concepts of equity in coverage and benefits, so 
that a plan need not undergo expensive and complex nondiscrimination 
tests. These options generally result in broader coverage and higher 
benefits for the lower-paid workers than does a plan designed to 
operate at the regulatory limits. 

[16] These rules apply to benefits attributable to employer 
contributions to a single-employer pension plan. Benefits attributable 
to employee contributions to either defined contribution or defined 
benefit plans, and investment income earned on employee contributions 
to defined contribution plans, are immediately vested. Multiemployer 
plans generally had 10-year cliff vesting, but this has been recently 
changed to conform with other qualified plans. 

[17] Most 401(k) plans permit participants to take a loan from 
accumulated assets. See U.S. General Accounting Office, 401(k) Pension 
Plans: Loan Provisions Enhance Participation But May Affect Income 
Security for Some, [hyperlink, 
http://www.gao.gov/products/GAO/HEHS-98-5] (Washington, D.C.: Oct. 1, 
1997). 

[18] Providing annuities helps ensure that plan participants will 
actually receive income from the pension to supplement their Social 
Security benefits and other sources of retirement income throughout 
the period of retirement. When an annuity is provided from an employer 
plan, ERISA requires a qualified joint and survivor annuity as the 
normal method of payment for a married participant. 

[19] The provisions for retirement age are implicitly linked to the 
retirement ages in the Social Security program. Plans must generally 
allow postretirement benefits to begin at the Social Security normal 
retirement age (NRA). Age and service provisions allow employers to 
structure plans in ways that allow eligible workers to retire earlier 
than at the NRA. Security. Employers can also offer their workers 
further incentives to retire early if this meets the goals of the 
firm. In general, employers with defined contribution plans can begin 
to make distributions to workers at age 59 1/2. Conversely, provisions 
of the Internal Revenue Code (minimum distribution rules) specify when 
distributions from any type of tax-favored retirement vehicle must 
begin, generally by the age of 70 1/2 or at severance from employment, 
if later. 

[20] A broader issue concerns whether increased individual retirement 
saving through, for example, 401(k) pensions or individual retirement 
arrangements, represents an increase in the aggregate level of 
personal saving, and hence national saving. Contributions to 
retirement income vehicles may represent saving that would have 
occurred even without the tax incentives in pensions, or amounts 
merely shifted from taxable assets or financed by borrowing. For 
further discussion see, U.S. General Accounting Office, National 
Saving: Answers to Key Questions, [hyperlink, 
http://www.gao.gov/products/GAO-01-591SP] (Washington, D.C.: June 1, 
2001), 98-99; Eric M. Engen and William G. Gale, The Effects of 401(k) 
Plans on Household Wealth: Differences Across Earnings Groups, 
National Bureau of Economic Research (Cambridge, Mass.: 2000). 

[21] An IRA is a personal, tax-deferred retirement arrangement that an 
employed person or spouse can set up with a tax-deductible deposit 
limited in 2002 to $3,000 or $3,500 if aged 50 or older ($6,000 for a 
couple or $7,000 if both are aged 50 or older). Failure to roll over 
most tax-deferred preretirement distributions directly into another 
qualified retirement plan could result in taxation of the lump sum as 
ordinary income and a 10-percent early-withdrawal penalty. 

[22] John Sabelhaus and David Weiner, "Disposition of Lump-Sum 
Distributions: Evidence from Tax Returns," National Tax Journal 52, 
no. 3 (September 1999): 593-614; Leonard E. Burman, Norma B. Coe and 
William G. Gale, "Lump Sum Distributions from Pension Plans: Recent 
Evidence and Issues for Policy and Research," National Tax Journal 52, 
no. 3 (September 1999): 553-562; James M. Poterba, Steven F. Venti, 
and David A. Wise, Pre-Retirement Cashouts and Forgone Retirement 
Saving: Implications for 401(k) Asset Accumulation (Cambridge, Mass.: 
National Bureau of Economic Research, 1999); John R. Woods, "Pension 
Vesting and Pre-Retirement Lump Sums among Full Time Private Sector 
Employees," Social Security Bulletin 56, no. 3 (1993): 3-21. 

[23] U.S. General Accounting Office, Pension Plans: Characteristics of 
Persons in the Labor Force Without Pension Coverage. 

[24] Patrick J. Purcell, Pension Sponsorship and Participation: 
Summary of Recent Trends (Washington, D.C.: Congressional Research 
Service, 2001). 

[25] For background on issues related to saving, see U.S. General 
Accounting Office, National Saving: Answers to Key Questions, GAO-01-
591SP (Washington, D.C.: June 1, 2001). 

[26] Tax incentives may be particularly important for the small 
business sector, where the low level of sponsorship and worker 
coverage makes it a focus of policy measures aimed at encouraging 
business owners to establish pension plans. However, the tax incentive 
in pensions may accrue to the benefit mainly of one owner or a small 
number of key employees, with the lower-paid workers of the small 
business benefiting minimally. In the past, the potential for small 
business owners to use pensions essentially as a tax shelter led to 
the adoption of special nondiscrimination rules termed "top-heavy 
rules." While these rules may have limited the use of pensions as a 
tax shelter, some employer groups have concluded that such rules limit 
incentives for small businesses to sponsor pensions. See U.S. General 
Accounting Office, Private Pensions: "Top-Heavy" Rules for Owner-
Dominated Plans, [hyperlink, 
http://www.gao.gov/products/GAO/HEHS-00-141] (Washington, D.C.: Aug. 
31, 2000). 

[27] Patricia B. Reagan and John A. Turner, "Did the Decline in 
Marginal Tax Rates During the 1980s Reduce Pension Coverage?" in 
Employee Benefits and Labor Markets in Canada and the United States, 
ed. William T. Alpert and Stephen A. Woodury (Kalamazoo, Mich.: W.E. 
Upjohn Institute for Employment Research, 2000), 475-495. 

[28] A provision of EGTRRA introduces a tax credit for small employer 
pension plan start-up costs. The provision allows small employers 
(defined as employers with no more than 100 employees earning more 
than $5,000 per year) a credit of up to $500 per year for three years 
for "qualified plan startup costs" for plans established and costs 
incurred after December 31, 2001. 

[29] The Joint Committee on Taxation recently issued a major study of 
the tax code and made recommendations for simplifying the code. The 
study includes a number of recommendations regarding pension 
regulations. See Joint Committee on Taxation, Study of the Overall 
State of the Federal Tax System and Recommendations for 
Simplification, Pursuant to Section 8022(3)(B) of the Internal Revenue 
Code of 1986, JCS-3-01 (Washington, D.C.: 2001). 

[30] The Revenue Act of 1978 created SEP IRAs. In 1996, the Small 
Business Job Protection Act created SIMPLE IRAs for small employers 
with less than 100 employees. Simplified small business pension plan 
designs based on a defined benefit model have also been proposed. Two 
such proposals during the 106th Congress included the Secure Assets 
For Employees (SAFE) plan, introduced as H.R. 2190, and the Secure 
Money Annuity or Retirement Trust (SMART) plan, which was introduced 
as H.R. 1213. These plans also involve a tradeoff of simplified 
requirements for providing specified benefit levels. For a fuller 
description, see Purcell, Pension Sponsorship, 7-8. 

[31] Among the plan design options for employers, particularly small 
and medium-sized firms, are IRS's master and prototype (M&P) plan 
program and volume submitter plan program. Under M&P plans, employers 
adopt simple plan design features that are preapproved by IRS for the 
institution offering the plan, such as a bank or an insurance company. 
Volume submitter plans offer employers greater flexibility in plan 
design options than M&P plans and a streamlined plan approval process. 
A principle distinction between plans maintained under these programs 
and individually designed plans is that the institution offering the 
M&P plan or the volume submitter plan, rather than the employer, is 
responsible for maintaining and updating the plan. As a result, such 
plans reduce the administrative burden and costs on individual 
employers. According to the Department of the Treasury, in 1998, there 
were over 3,000 sponsors of approximately 720,000 M&P and volume 
submitter plans with about 26 million participants. 

[32] Vanguard Center for Retirement Research, Automatic Enrollment: 
Benefits and Costs of Adoption (Valley Forge, Pa.: Vanguard Center for 
Retirement Research, 2001); John J. Choi, David Laibson, Brigitte C. 
Madrian, and Andrew Metrick, For Better or Worse: Default Effects and 
401(k) Savings Behavior (Cambridge, Mass.: National Bureau of Economic 
Research, 2001). 

[33] Shlomo Benartzi and Richard H. Thaler, Save More Tomorrow: Using 
Behavioral Economics to Increase Employee Saving (Los Angeles: 
University of California at Los Angeles, 2001). 

[34] Pamela Perun and Eugene Steuerle, ERISA at 50: A New Model for 
the Private Pension System, The Retirement Project, no. 4 (Washington 
D.C.: The Urban Institute, March 2000). 

[35] Poterba, Venti, and Wise, Pre-Retirement Cashouts, 34. 

[36] For further discussion see, U.S. General Accounting Office, 
Private Pensions: Key Issues to Consider Following the Enron Collapse, 
[hyperlink, http://www.gao.gov/products/GAO-02-480T] (Washington, 
D.C., February 27, 2002) and Patrick J. Purcell, The Enron Bankruptcy 
and Employer Stock in Retirement Plans (Washington, D.C.: 
Congressional Research Service, 2002). 

[37] For more information on the Department of Labor's activities 
under the Savings Are Vital to Everyone's Retirement (SAVER) Act of 
1997, see U.S. General Accounting Office, Retirement Savings: 
Opportunities to Improve DOL's SAVER Act Campaign, [hyperlink, 
http://www.gao.gov/products/GAO-01-634] (Washington, D.C.: June 26, 
2001). 

[38] For example, this point was noted by the Technical Panel on 
Trends and Issues in Retirement Savings as part of the 1994-1995 
Advisory Council on Social Security; see Report of the 1994-1995 
Advisory Council on Social Security, Volume H (Washington D.C.: 
Advisory Council on Social Security, 1995), 52-53. See also Richard P. 
Hinz and John A. Turner, "Pension Coverage Initiatives: Why Don't 
Workers Participate?" in Living With Defined Contribution Pensions, 
ed. Olivia S. Mitchell and Sylvester J. Schieber (Philadelphia: 
Pension Research Council, 1998), 19-37. 

[39] Hinz and Turner, "Pension Coverage Initiatives," 35. 

[40] For simplicity, we use "pooled employer plans" when referring to 
plans covering two or more employers. A "multiemployer plan" is 
defined as a plan maintained pursuant to collective bargaining 
agreements to which more than one employer contributes. In a "multiple-
employer plan," employers are often members of, or otherwise related 
to, a professional or trade association. 

[41] Variable annuities offered by the insurance industry also provide 
tax-favored retirement savings. 

[42] SEPs and SIMPLE IRAs are not universally available accounts, 
because they are available at the employers' discretion and are thus 
not available to all workers. 

[43] A provision of EGTRRA introduces a "savers tax credit" for 
eligible taxpayers who contribute to a retirement plan or an IRA. The 
savers credit is a nonrefundable income tax credit for taxpayers with 
adjusted gross incomes that do not exceed $50,000. It is equal to a 
specified percentage of certain employee contributions made to an 
employer-sponsored retirement plan or to a specified percentage of 
certain individual or spousal contributions to an IRA beginning in 
2002. 

[44] Paul A. Smith, A Longer Term Perspective on IRA Participation: 
Evidence from a Panel of Tax Returns (Washington, D.C.: U.S. 
Department of Treasury, Office of Tax Analysis, 2001). 

[45] Paul Yakoboski, "IRA Ownership," EBRI Notes, vol. 20 (December 
1999): 1-3. 

[46] A recent paper by Kotlikoff and Gokhale suggests that low and 
moderate income workers may not have much incentive to participate in 
401(k)s and other tax-deferred savings vehicles. For example, if 
workers face low marginal tax rates earlier in their career but higher 
marginal rates later in their career and when withdrawals occur, it is 
possible that their lifetime taxes may be higher as a result of 
contributions to a 401(k) early in their career than if they had not 
contributed or had contributed to a Roth IRA. Such results may call 
into question certain policies aimed at increasing tax-deferred saving 
for low- and moderate-income individuals. See Jagadeesh Gokhale, 
Laurence J. Kotlikoff, and Todd Neumann, Does Participating in a 
401(k) Raise Your Lifetime Taxes? Federal Reserve Bank of Cleveland 
Working Paper 01-08 (Cleveland, Ohio: 2001). 

[47] Most Social Security individual account proposals call for 
mandatory participation and are therefore discussed here under 
universal participation reforms. However, some individual account 
proposals call for voluntary Social Security individual accounts that 
would provide universal access and would involve issues discussed in 
this report under Reforms to Create Universal Access to Retirement 
Saving Rely on Voluntary Worker Election (pages 28-31). 

[48] Employers were not required to make pension contributions for 
employees younger than 25 or with less than 1,000 hours of employment. 

[49] Daniel I. Halperin and Alicia H. Munnel, Assuring Retirement 
Income for AU Workers (Chestnut Hill, Mass.: Center for Retirement 
Research at Boston College, 2000). 

[50] Increasing the base level of Social Security benefits could also 
increase benefits for the disabled, spouses, dependents, and survivors. 

[51] Kilolo Kijakazi, Low-Wage Earners: Options for Improving Their 
Retirement Income (Washington, D.C.: Center on Budget and Policy 
Priorities, 2000). 

[52] The E1TC is a refundable tax credit established by Congress in 
1975. The E1TC offsets much of the impact of Social Security taxes 
paid by low-income workers and is intended to encourage persons with 
low-incomes to seek work rather than welfare. 

[53] This option was the focus of the recently completed report by the 
President's Commission to Strengthen Social Security, Strengthening 
Social Security and Creating Personal Wealth for Americans: The Final 
Report of the President's Commission to Strengthen Social Security 
(Washington, D.C.: President's Commission to Strengthen Social 
Security, 2001). A number of GAO reports have also addressed the topic 
of Social Security reform (see Related GAO Products on page 46 of this 
report). 

[54] Under this proposal, everyone with a full 35 years of work would 
receive the first-tier standard defined benefit, regardless of income. 
The first-tier benefit would give all workers with a full work history 
the standard benefit; however, the standard benefit would be reduced 
for workers with less than a full 35-year work history. The second-
tier benefit would be based on an individual account financed by a 
combination of employee and employer contributions and by a limited 
matching contribution from Social Security. 

[55] The income tax treatment of a funded, vested nonqualified pension 
plan is similar to that of a regular savings account. Contributions 
are included in the employee's income. Income earned by the plan is 
taxable to the employee annually. No additional tax is owed when 
contributions and earnings are withdrawn. 

[56] The tax treatment afforded a contribution made to a deductible 
IRA is equivalent. 

[57] Distributions from a Roth IRA are not taxed if the distribution 
(1) is made at least 5 tax years after the first contribution to the 
account and (2) is made after the taxpayer has attained 59 1/2 years 
of age, on account of death or disability, or (3) is for a first-time 
home purchase. 

[58] U.S. General Accounting Office, Pension Plans: Characteristics of 
Persons in the Labor Force Without Pension Coverage, [hyperlink, 
http://www.gao.gov/products/GAO/HEHS-00-131] (Washington, D.C.: Aug. 
22, 2000). 

[69] Recommended replacement rates are less than 100 percent for the 
following reasons: (1) The need to save for retirement may diminish, 
(2) taxes may decline (e.g., payroll taxes may not be due and income 
tax rates may be lower), (3) work-related expenses may decline, (4) 
family size may decline, (5) some households may pay off home 
mortgages, and (5) households may consume assets, not just income. 

[60] For example, a 1990 study by Grad using data from the early 1980s 
showed that Social Security and pension income replace a substantial 
portion of preretirement earnings. However, these two sources alone 
did not provide sufficient retirement income for most workers to 
achieve replacement rate targets. Grad reported that only about one-
fourth of retired workers were able to achieve replacement rates equal 
to two-thirds (67 percent) of their preretirement income. See Susan 
Grad, "Earnings Replacement Rates of New Retired Workers," Social 
Security Bulletin 53, no. 10 (1990): 2-19. 

[61] Andrews reported on simulations of two models. The results 
indicated that retirement income adequacy could be expected to improve 
in the future, with the possible exception that the percentage of 
unmarried women near the poverty line would not decline appreciably. 
These models predicted that for workers and households, (1) overall 
median real retirement income would increase, (2) the percentage 
receiving pension income would increase, and (3) the share of pension 
income as a share of total retirement income would increase. See Emily 
S. Andrews, "Gaps in Retirement Income Adequacy," in The Future of 
Pensions in the United States, ed. Ray Schmitt (Philadelphia: 
University of Pennsylvania Press, 1993), 1-31. 

[62] CBO found that most of the baby-boom generation would likely 
achieve a higher standard of living in retirement than its parents' 
generation. At the same time, it was noted that single, poorly 
educated baby boomers and those who did not have other assets such as 
equity in a home would have difficulty achieving adequate retirement 
income levels. See Joyce Manchester, Baby Boomers in Retirement: An 
Early Perspective (Washington, D.C.: Congressional Budget Office, 
1993). 

[63] B. Douglas Bernheim, "Is the Baby Boom Generation Preparing 
Adequately for Retirement?" Technical report prepared for Merrill 
Lynch & Co., Inc., August 1992. 

[64] The HRS provides information on the income and wealth holdings 
for a nationally representative sample of 7,607 families who had at 
least one member born between 1931 and 1941. That is, during the first 
wave of interviews in 1992, respondents were between the ages of 51 
and 61. 

[65] Gustman and Steinmeier used HRS data to study wealth accumulation 
and found that the households in the study had accumulated, on 
average, 86 percent of final earnings in nominal terms and 60 percent 
of final earnings in real terms. Moreover, they found that contrary to 
general impressions, pensions are distributed widely among households. 
Although only half of employed individuals have a pension at any point 
in time, three-fourths of HRS households were covered by a pension at 
one time, and two-thirds of HRS households own the rights to a pension 
or pension income. See Alan L. Gustman and Thomas L. Steinmeier, 
"Effects of Pensions on Savings: Analysis with Data from the Health 
and Retirement Study," Carnegie-Rochester Conference Series 50 (July 
1999): 271-326. 

[66] James F. Moore and Olivia S. Mitchell, "Projected Retirement 
Wealth and Saving Adequacy," in Forecasting Retirement Needs and 
Retirement Wealth, ed. Olivia S. Mitchell, P. Brett Hammond, and Anna 
Rappaport (Philadelphia: University of Pennsylvania Press, 2000), 68-
94. 

[67] Related work by Venti and Wise suggests that inadequate wealth 
accumulation seems more related to the decision to save or consume 
income while working and less related to (random) factors such as luck 
or chance. See Steven F. Venti and David A. Wise, "The Cause of Wealth 
Dispersion at Retirement: Choice or Chance?" American Economic 
Association Papers and Proceedings 88, no. 2 (1998): 185-191. 

[68] Engen, Gale, and Uccello point out that estimates of wealth-to-
earnings ratios for a population or a given earnings category should 
not merely be compared with a given saving target to determine savings 
adequacy. Rather, since households earnings vary from period to 
period, there will be a distribution of wealth-to-earnings ratios for 
any given group, and some may experience "good" years with high wealth-
to-earnings ratios while others in the same group may experience "bad" 
years with low wealth-to-earnings ratios. This means that a given 
wealth-to-earnings target represents only a median for the group and, 
taking into account the variation around this median, changes previous 
interpretations of savings inadequacy. Even those with low measured 
wealth-to-earnings ratios in a given period could be saving 
adequately. Those who are saving adequately tend to more consistently 
exceed the wealth-to-earnings targets, and those who are not saving 
adequately tend to more consistently fall below the targets. Using 
this approach, the researchers conclude that most households have 
adequate retirement savings but that there is perhaps a specific 
group, who are more likely to have low wealth in relation to earnings. 
See Cori E. Uccello, Are Americans Saving Enough for Retirement? Issue 
in Brief 7 (Boston: Center for Retirement Research at Boston College, 
2001). Also see Eric Engen, William G. Gale, and Cori E. Uccello, The 
Adequacy of Household Saving, Brookings Papers on Economic Activity, 
no. 2 (Washington, D.C.: Brookings Institution, 1999), 65-187. 

[End of section] 

GAO’s Mission: 

The General Accounting Office, the investigative arm of Congress, 
exists to support Congress in meeting its constitutional 
responsibilities and to help improve the performance and accountability 
of the federal government for the American people. GAO examines the use 
of public funds; evaluates federal programs and policies; and provides 
analyses, recommendations, and other assistance to help Congress make 
informed oversight, policy, and funding decisions. GAO’s commitment to 
good government is reflected in its core values of accountability, 
integrity, and reliability. 

Obtaining Copies of GAO Reports and Testimony: 

The fastest and easiest way to obtain copies of GAO documents at no 
cost is through the Internet. GAO’s Web site (www.gao.gov) contains 
abstracts and full-text files of current reports and testimony and an 
expanding archive of older products. The Web site features a search 
engine to help you locate documents using key words and phrases. You 
can print these documents in their entirety, including charts and other 
graphics. 

Each day, GAO issues a list of newly released reports, testimony, and 
correspondence. GAO posts this list, known as “Today’s Reports,” on its 
Web site daily. The list contains links to the full-text document 
files. To have GAO e-mail this list to you every afternoon, go to 
www.gao.gov and select “Subscribe to daily E-mail alert for newly 
released products” under the GAO Reports heading. 

Order by Mail or Phone: 

The first copy of each printed report is free. Additional copies are $2 
each. A check or money order should be made out to the Superintendent 
of Documents. GAO also accepts VISA and Mastercard. Orders for 100 or 
more copies mailed to a single address are discounted 25 percent. 
Orders should be sent to: 

U.S. General Accounting Office: 
441 G Street NW, Room LM: 
Washington, D.C. 20548: 

To order by Phone: 

Voice: (202) 512-6000 

TDD: (202) 512-2537 

Fax: (202) 512-6061 

To Report Fraud, Waste, and Abuse in Federal Programs: 
Contact: 

Web site: www.gao.gov/fraudnet/fraudnet.htm 

E-mail: fraudnet@gao.gov 

Automated answering system: (800) 424-5454 or (202) 512-7470: 

Public Affairs: 

Jeff Nelligan, managing director, NelliganJ@gao.gov: 
(202) 512-4800: 
U.S. General Accounting Office: 
441 G Street NW, Room 7149: 
Washington, D.C. 20548: