This is the accessible text file for GAO report number GAO-10-31 
entitled 'Retirement Savings: Automatic Enrollment Shows Promise for 
Some Workers, but Proposals to Broaden Retirement Savings for Other 
Workers Could Face Challenges' which was released on October 26, 2009. 

This text file was formatted by the U.S. Government Accountability 
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. 

This is a work of the U.S. government and is not subject to copyright 
protection in the United States. It may be reproduced and distributed 
in its entirety without further permission from GAO. Because this work 
may contain copyrighted images or other material, permission from the 
copyright holder may be necessary if you wish to reproduce this 
material separately. 

Report to the Chairman, Special Committee on Aging, U.S. Senate: 

United States Government Accountability Office: 

GAO: 

October 2009: 

Retirement Savings: 

Automatic Enrollment Shows Promise for Some Workers, but Proposals to 
Broaden Retirement Savings for Other Workers Could Face Challenges: 

Retirement Savings: 

GAO-10-31: 

GAO Highlights: 

Highlights of [hyperlink, http://www.gao.gov/cgi-bin/getrpt?GAO-10-31], 
a report to the Chairman, Special Committee on Aging, U.S. Senate. 

Why GAO Did This Study: 

Although employer-sponsored retirement plans can be an important 
component of income security after retirement, only about half of all 
workers participate in such plans. To foster greater participation 
among workers who have access to such plans, Congress included 
provisions that facilitate plan sponsors’ adoption of automatic 
enrollment policies in the Pension Protection Act of 2006. To foster 
greater retirement savings among workers who do not have access to an 
employer-sponsored plan, proposals have been made at the federal level 
for an “automatic IRA” and at the state level for state-based programs. 

Because of questions about the extent of retirement savings and 
prospects for a sound retirement for all Americans, GAO was asked to 
determine (1) what is known about the effect of automatic enrollment 
policies among the nation’s 401(k) plans, and the extent of and future 
prospect for such policies; and (2) the potential benefits and 
limitations of automatic IRA proposals and state-assisted retirement 
savings proposals. To answer these questions, GAO reviewed available 
reports and data, and interviewed plan sponsors, industry groups, 
investment professionals, and relevant federal agencies. 

What GAO Found: 

Automatic enrollment appears to significantly increase participation in 
401(k) plans according to existing studies, but may not be suitable for 
all plan sponsors. Some studies found that participation rates can 
reach as high as 95 percent under automatic enrollment. Available data 
indicate that the percentage of plans with automatic enrollment 
policies increased from about 1 percent in 2004 to more than 16 percent 
in 2009, with higher rates of adoption among larger plan sponsors. In 
most cases, these plans automatically enroll only new employees, rather 
than all employees. We also found that automatic enrollment may not be 
suitable for all plan sponsors, such as those with a high-turnover 
workforce. Further, some data show that while automatic escalation 
policies—which automatically increase saving rates over time—are 
increasingly common, they lag behind adoption of automatic enrollment. 
In combination with low initial contribution rates, this could depress 
savings for some workers. Also, the emergence of target-date funds—
funds that allocate investments among various asset classes and shift 
to lower-risk investments as a “target” retirement date approaches—as 
the typical default investment raises questions in light of the 
substantial losses such funds experienced in the past year. 

Figure: Adoption and Scope of Automatic Enrollment by Plan Size, March 
2009: 

[Refer to PDF for image: illustration] 

Bar graph: 

Plan size: Large; 
Percentage of Plans: All eligible employees: 8.0; 
Percentage of Plans: New or recent hires only: 32.2. 

Plan size: Mid-sized; 
Percentage of Plans: All eligible employees: 7.2; 
Percentage of Plans: New or recent hires only: 24.8. 

Plan size: Small; 
Percentage of Plans: All eligible employees: 6.6; 
Percentage of Plans: New or recent hires only: 7.5. 

Source: GAO presentation of Fidelity Investments data. 

[End of figure] 

Other proposals could expand the portion of the workforce saving for 
retirement, but these proposals could face challenges. Under a 
federally mandated automatic IRA, certain employers could be required 
to enroll eligible employees in payroll-deduction IRAs, unless the 
worker specifically opted out. Such a proposal could broaden the 
population that saves for retirement at minimal cost to employers. 
However, this proposal faces a number of challenges, including 
uncertainty about the extent to which it would help low-income workers 
accumulate significant retirement savings. Proposals for state-assisted 
retirement savings programs could raise coverage and, ultimately, 
savings by involving state governments in facilitating retirement 
savings for workers without access to an employer-sponsored plan. 
However, such programs face uncertainty about employer and worker 
participation levels, as well as legal and regulatory issues. 

What GAO Recommends: 

This report makes no recommendations. 

View [hyperlink, http://www.gao.gov/products/GAO-10-31] or key 
components. For more information, contact Barbara Bovbjerg at (202) 512-
7215 or bovbjergb@gao.gov. 

[End of section] 

Contents: 

Letter: 

Background: 

Automatic Enrollment Raises Plan Participation, but Some Plan Sponsors 
Have Not Adopted Plan Features That Could Further Facilitate Retirement 
Savings: 

Automatic IRA and State-Assisted Retirement Savings Plan Proposals 
Could Improve Access to Retirement Savings Vehicles for Uncovered 
Workers, but Each Proposal Faces Challenges: 

Concluding Observations: 

Agency Comments: 

Appendix I: Scope and Methodology: 

Appendix II: Comments from the Department of Labor: 

Appendix III: Comments from the Department of the Treasury: 

Appendix IV: GAO Contact and Staff Acknowledgments: 

Selected Bibliography on Automatic Enrollment: 

Tables: 

Table 1: Overall Participation Rate Effects of Automatic Enrollment: 

Table 2: Examples of State Roles and Intended Impact under Existing 
Proposals: 

Table 3: Estimated Program Development and Operation Costs in 
Washington State: 

Table 4: Six Studies We Reviewed on the Impact of Automatic Enrollment 
and Related Policies: 

Table 5: Industries and Plan Sponsor Sizes: 

Figures: 

Figure 1: Proportion of Full-Time Workforce Lacking Retirement Plan 
Coverage, 1990 and 2007: 

Figure 2: Proportion of Labor Force That Lacked Retirement Plan 
Coverage in 2007, by Selected Worker Characteristics: 

Figure 3: Adoption and Scope of Automatic Enrollment by Plan Size, 
March 2009: 

Figure 4: Default Contribution Rates for Plans with Automatic 
Enrollment: 

Figure 5: Default Fund Types Used by Plans with Automatic Enrollment 
Policies in 2005 and 2008: 

Figure 6: Steps Involved with Establishing an Automatic IRA: 

Abbreviations: 

ABC: American Benefits Council: 

ASPPA: American Society of Pension Professionals and Actuaries: 

CAP: Center for American Progress: 

CIEBA: The Committee on Investment of Employee Benefit Assets: 

EBRI: Employee Benefits Research Institute: 

ERISA: Employee Retirement Income Security Act of 1974: 

ICI: Investment Company Institute: 

IRA: individual retirement account: 

IRS: Internal Revenue Service: 

PPA: Pension Protection Act of 2006: 

PRC: Pension Rights Center: 

PSCA: Profit Sharing/401(k) Council of America: 

QDIA: Qualified Default Investment Alternative: 

SBCA: Small Business Council of America: 

SIMPLE: Savings Incentive Match Plan for Employees of Small Employers: 

SIPP: Survey of Income and Program Participation: 

TDF: target-date fund: 

[End of section] 

United States Government Accountability Office: 

Washington, DC 20548: 

October 23, 2009: 

The Honorable Herb Kohl: 
Chairman: 
Special Committee on Aging: 
United States Senate: 

Dear Mr. Chairman: 

Although employer-sponsored retirement plans can be an important 
component of retirement security for millions of American workers, only 
about half of all workers participate in such plans. Further, there has 
been little change in the percentage of workers covered by employer- 
sponsored plans in the last quarter century. While some working 
Americans choose not to participate in an employer-sponsored retirement 
savings plan, the large majority of uncovered workers do not have 
access to an employer-sponsored plan. Non-participation in employer- 
sponsored plans may have significant implications for future income, 
particularly for low-income workers. For example, as GAO reported in 
2007, about 63 percent of workers in the lowest income quartile are 
projected to have no defined contribution plan savings at 
retirement.[Footnote 1] Further, workers in this quartile are projected 
to have savings that would replace an average of about 10.3 percent of 
their annualized career earnings after retirement, compared to 
replacement rates of about 34 percent for the highest income quartile. 

To foster retirement saving among the portion of the workforce that has 
been offered an employer-sponsored retirement savings plan but does not 
participate in it, some employers have adopted automatic enrollment 
policies for their defined contribution plans--retirement plans under 
which participants accumulate retirement savings in individual 
accounts, such as a 401(k). Under automatic enrollment, a worker is 
enrolled into the plan automatically, or by default, unless they 
explicitly choose to opt out. Employers who have adopted automatic 
enrollment must also establish default contribution rates and default 
investment vehicles for workers who do not specify these choices on 
their own. Employers may also adopt automatic escalation policies, 
which increase contribution rates automatically over time--typically up 
to a pre-defined maximum contribution rate per participant. 

In addition, federal and state legislators have introduced proposals to 
foster retirement savings among those who work for employers that do 
not sponsor a plan. One group of proposals calls for establishing 
automatic individual retirement accounts (IRA) for workers not covered 
by another retirement plan. Under an automatic IRA, employers would be 
required to offer their employees the opportunity to make contributions 
through automatic payroll deduction, with an opt-out provision for 
participants. Another set of proposals calls for state governments to 
facilitate the establishment of retirement accounts for private-sector 
workers who are not covered by employer-sponsored plans. However, as of 
July 2009, no proposals for a federal automatic IRA or a state-assisted 
retirement savings plan had been passed into law. 

In light of your interest in increasing retirement savings and ensuring 
a sound retirement for all Americans, you asked us to examine certain 
options for expanding retirement plan coverage. Specifically, we 
addressed the following two questions: 

1. What is known about the effect of automatic enrollment policies 
among the nation's 401(k) plans, as well as the extent of and prospects 
for such policies? 

2. What are the potential benefits and limitations of automatic IRA 
proposals and state-assisted retirement savings proposals? 

To answer the first question, we obtained and reviewed data collected 
by large 401(k) plan administrators; reviewed studies on the impact of 
401(k) automatic enrollment polices on participation rates and saving 
patterns; and conducted in-depth interviews with selected plan sponsors 
about their experience with and views of automatic enrollment. We found 
the data from these sources to be sufficiently reliable for our use. To 
answer the second question, we reviewed and analyzed existing 
proposals, as well as feasibility studies of these proposals, and 
interviewed pension industry experts in academia, representatives of 
employee benefit consulting firms, plan administrators, and state and 
federal government officials. Throughout this report, we considered a 
worker to be covered by an employer-sponsored plan if the employer 
offers such a plan, the worker is eligible to participate under the 
plan's rules, and the worker chooses to participate in it. For both 
objectives, we also reviewed relevant federal regulations. 

We conducted this audit from August 2008 to October 2009 in accordance 
with generally accepted government auditing standards. Those standards 
require that we plan and perform the audit to obtain sufficient, 
appropriate evidence to provide a reasonable basis for our findings and 
conclusions based on our audit objectives. We believe that the evidence 
obtained provides a reasonable basis for our findings and conclusions 
based on our audit objectives. Appendix I of this report contains a 
detailed description of the methodology used in this review and its 
limitations. 

Background: 

Traditionally, employers that sponsored retirement plans generally 
established "defined benefit" plans. Under such plans, participation is 
generally automatic for eligible workers, and retirement benefits are 
established by a formula, often based on a worker's salary and years of 
service. Since the 1980s, defined contribution plans--most prominently 
the 401(k) plan--have supplanted defined benefit plans as the dominant 
type of private-sector retirement plan. Under defined contribution 
plans, workers typically must decide whether or not to participate, how 
much to contribute, and how to invest plan assets from a range of 
options provided under the plan. Under a 401(k) plan, if the employee 
does not participate, if contributions made to an employee's account 
are insufficient, or if the investments that an employee chooses yield 
an inadequate return, the employee may have retirement income that is 
insufficient to maintain his or her desired standard of living. 

As defined contribution plans emerged as the dominant form of 
retirement plan, the percentage of the population covered by employer- 
sponsored plans changed very little--remaining at about half of the 
workforce. As figure 1 shows, Current Population Survey data reveal 
that about 48 percent of the total U.S. workforce was not covered by an 
employer-sponsored a plan in 2007.[Footnote 2] About 40 percent worked 
for an employer that did not sponsor a plan, and about 8 percent did 
not participate in the plan that their employer sponsored. 

Figure 1: Proportion of Full-Time Workforce Lacking Retirement Plan 
Coverage, 1990 and 2007: 

[Refer to PDF for image: illustration] 

Bar graph: 

Years: 1990; 
Percentage of workers: Employer sponsors, worker does not participate: 
8; 
Percentage of workers: Employer does not sponsor: 37. 

Years: 2007; 
Percentage of workers: Employer sponsors, worker does not participate:  
8; 
Percentage of workers: Employer does not sponsor: 40. 

Source: GAO presentation of data in Pension Sponsorship and 
Participation: Summary of Recent Trends, Congressional Research 
Service, September 2008. 

[End of figure] 

According to the Current Population Survey, certain segments of the 
working population have consistently had much lower rates of employment 
with employers sponsoring a plan, and lower participation rates than 
the working population overall. As figure 2 illustrates, larger 
portions of certain worker groups, such as lower-income workers, 
younger workers, workers employed by smaller companies, and part-time 
workers lack coverage compared to all full-time workers. 

Figure 2: Proportion of Labor Force That Lacked Retirement Plan 
Coverage in 2007, by Selected Worker Characteristics: 

[Refer to PDF for image: illustration] 

Bar graph: 

Category of worker: All full-time workers; 
Employer sponsors, worker does not participate: 8; 
Employer foes not sponsor: 40. 

Category of worker: Forms of <25 employees; 
Employer sponsors, worker does not participate: 4; 
Employer foes not sponsor: 71. 

Category of worker: Employees ages 25-34; 
Employer sponsors, worker does not participate: 11; 
Employer foes not sponsor: 46. 

Category of worker: Lowest earning quartile; 
Employer sponsors, worker does not participate: 11; 
Employer foes not sponsor: 62. 

Category of worker: Part-time workers; 
Employer sponsors, worker does not participate: 15; 
Employer foes not sponsor: 62.  

Source: GAO presentation of data in Pension Sponsorship and 
Participation: Summary of Recent Trends, Congressional Research 
Service, September 2008. 

Source: GAO presentation of data in Pension Sponsorship and 
Participation; Summary of Recent Trends, Congressional Research 
Service, September 2008. 

Note: Except where noted, all data are for full-time workers only. 

[End of figure] 

Workers may choose not to enroll, or delay enrolling, in a retirement 
plan for a number of reasons. For example, according to a recent 
Congressional Research Service presentation of data from the Survey of 
Income and Program Participation, most non-participating workers whose 
employer sponsored a plan said they thought--in some cases, 
incorrectly--they were ineligible.[Footnote 3] The report also found 
that substantial numbers of employees fail to participate because they 
believe they cannot afford to contribute to the plan. For example, 19 
percent of nonparticipating respondents cited this reason, and 10 
percent said they did not want to tie up their money. In recent years, 
exponents of "behavioral economics" have noted that many non- 
participants may not have made a specific decision, but rather fail to 
participate because of a tendency to procrastinate and follow the path 
that does not require an active decision. Further, some workers may not 
participate because of more immediate savings objectives--such as 
saving for education or a home--and, in the case of lower income 
workers, the prospect that Social Security benefits will replace a 
relatively high percentage of income in their retirement years. For 
example, a recent analysis by the Investment Company Institute 
concluded that lower income workers are less likely to save for 
retirement in part because Social Security benefits replace a higher 
proportion of their pre-retirement earnings.[Footnote 4] 

In recent years, automatic enrollment has been advocated as a way to 
encourage greater participation in 401(k) plans among the portion of 
the workforce who have access to such plans but opt not to participate. 
Typically, under 401(k)s and some other types of defined contribution 
plans, workers have been required to decide whether or not to join a 
plan, to specify their saving contribution rates, and to select 
investments from the range of investment options offered by the 
plan.[Footnote 5] Under automatic enrollment, in contrast, a worker 
would be enrolled in a plan unless she or he explicitly opted-out of 
the plan. Plan sponsors that adopt automatic enrollment must specify a 
default contribution rate--the portion of an employee's salary that 
will be deposited in the plan--that applies to employees who do not 
choose a different contribution rate. Also, plan sponsors must select a 
default investment--the fund or other vehicle into which deferred 
savings will be invested--unless the employee specifies an investment 
or investments from those available under the plan. Employers may also 
adopt an automatic escalation policy, under which an employee's 
contribution rates would be automatically increased at periodic 
intervals, such as annually. For example, if the default contribution 
rate is 3 percent of pay, a plan sponsor may choose to increase an 
employee's rate of saving by 1 percent per year, up to some maximum, 
such as 6 percent. While a plan sponsor that adopts an automatic 
enrollment policy must specify a default contribution rate and a 
default investment, plan features such as these and automatic 
escalation may also be adopted in the absence of an automatic 
enrollment policy. 

Automatic enrollment has not been a traditional feature of 401(k) plans 
and, prior to 1998, plan sponsors feared that adopting automatic 
enrollment could lead to plan disqualification.[Footnote 6] However, in 
1998, the Internal Revenue Service (IRS) addressed this issue by 
stating that a plan sponsor could automatically enroll newly hired 
employees and, in 2000, clarified that automatic enrollment is 
permissible for current employees who have not enrolled.[Footnote 7] 
Nonetheless, a number of considerations inhibited widespread adoption 
of automatic enrollment, including remaining concerns such as liability 
in the event that the employee's investments under the plan did not 
perform satisfactorily, and concerns about state laws that prohibit 
withholding employee pay without written employee consent. More 
recently, provisions of the Pension Protection Act of 2006 (PPA) and 
subsequent regulations further facilitated the adoption of automatic 
enrollment by providing incentives for doing so and by protecting plans 
from fiduciary and legal liability if certain conditions are 
met.[Footnote 8] In September 2009, the Department of the Treasury 
announced IRS actions designed to further promote automatic enrollment 
and the use of automatic escalation policies.[Footnote 9] 

PPA Provisions Facilitating Automatic Enrollment: 

Anti-Discrimination Safe Harbor 401(k)(13); 

Plans that adopt automatic enrollment may be exempt from required 
annual testing to ensure that the plan does not discriminate in favor 
of highly compensated employees. To obtain safe harbor protection, 
plans must adopt automatic enrollment as well as other plan features 
and policies. For example, the plan must: 

* Notify affected employees about automatic contributions; 
* Defer at least 3 percent of pay in the first year; 
* Automatically increase contribution by 1 percent each subsequent 
year, to a minimum of 6 percent and a maximum of 10 percent; 
* Invest savings in a type of investment vehicle identified in 
Department of Labor regulations as a Qualified Default Investment 
Alternative (QDIA)[A]; 
* Match 100 percent of the first 1 percent of employee contributions, 
and 50 percent of contributions beyond 1 percent, up to 6 percent of 
wages; 

Protection from Employee Retirement Income Security Act of 1974 (ERISA) 
Fiduciary Liability: 

In the absence of direction from an employee, plans that automatically 
invest contributions in a QDIA are treated as if the employee exercised 
control over management of their savings in the plan. As a result, 
plans that comply with Department of Labor regulations pertaining to 
QDIAs will not be liable for any loss that occurs as a result of such 
investments; 

90-Day Withdrawal Period: 

An automatically enrolled worker has 90 days to opt out and withdraw 
any contributions (including the earnings on those contributions.) 
These amounts will not be subject to the extra tax that normally 
applies to distributions received before age 59½. 414(w)(2)(B); 

Protection from State Wage-Garnishment Laws: 

PPA preempts any state law that would directly or indirectly prohibit 
or restrict the inclusion of an automatic enrollment arrangement in a 
plan. 

Source: GAO analysis of Pension Protection Act and related regulations. 

[A] Final regulations issued by the U.S. Department of Labor specify 
four categories of QDIAs. They are (1) a product with a mix of 
investments that takes into account an individual's age (such as a 
target-date fund), (2) an investment service that allocates assets 
according to an individual's age (such as a managed account), or (3) a 
product with a mix of investments that takes into account the 
characteristics of the group of employees as a whole, rather than each 
individual (such as a balanced fund). A sponsor may also use a capital 
preservation fund as a QDIA for the first 120 days of an individual's 
participation to simplify administration if the worker opts out of the 
plan. 29 C.F.R. §2550.404c-5(c)(4). 

Other proposals have been put forth with the intent of broadening the 
practice of saving for retirement among workers whose employers do not 
sponsor a retirement plan. One such proposal is the automatic IRA, 
which would require employers that do not sponsor a retirement plan to 
facilitate direct deposit or payroll-deduction savings for all 
employees.[Footnote 10] To maximize participation, employees would be 
automatically enrolled, but would be permitted to opt out. Legislative 
proposals to establish an automatic IRA requirement were introduced in 
Congress in 2006 and 2007, and the concept of an automatic IRA was also 
mentioned in the President's 2010 Budget proposal.[Footnote 11] In 
addition, legislative proposals have been introduced in some states' 
legislatures that would involve state governments in facilitating 
payroll-deduction retirement plans or IRAs for employers that do not 
already offer them.[Footnote 12] According to the architect of the 
state plan concept, a state could play an intermediary role to pool 
assets and share expenses among many plan sponsors, thus lowering 
costs. 

Automatic Enrollment Raises Plan Participation, but Some Plan Sponsors 
Have Not Adopted Plan Features That Could Further Facilitate Retirement 
Savings: 

Existing studies show that automatic enrollment significantly increases 
participation rates in 401(k) plans, although beneficial effects of 
automatic enrollment may depend on accompanying policies designed to 
ensure adequate savings and appropriate investment. While defined 
contribution plan sponsors have increasingly adopted automatic 
enrollment in recent years, this approach may not be suitable for all 
plan sponsors, and available data show that some plan sponsors have not 
augmented automatic enrollment policies with other policies designed to 
ensure adequate savings. 

Automatic Enrollment Increases Participation Rates, but Some 
Participants' Acceptance of the Default Contribution Rate and 
Investment Fund May Limit Savings Potential: 

According to analyses we reviewed, automatic enrollment policies result 
in considerably increased 401(k) participation rates for plans adopting 
them, with some participation rates reaching as high as 95 
percent.[Footnote 13] For example, one study followed comparison groups 
hired before and after a company adopted automatic enrollment for new 
employees only and compared the participation rates of the two groups. 
The participation rate for those hired before automatic enrollment was 
adopted was 37 percent at 3 to 15 months of tenure, compared with an 86 
percent participation rate for the group hired after automatic 
enrollment with a similar amount of tenure.[Footnote 14] According to 
the studies we reviewed, automatic enrollment has this effect because 
many people find it easier to delay the decision to enroll in a plan 
out of inertia, procrastination, feelings of intimidation about making 
savings and investment decisions, or other factors. Workers may 
intellectually understand the importance of saving for retirement but 
have trouble overcoming their own inertia to start saving or to save 
effectively.[Footnote 15] A team of researchers found that under 
automatic enrollment, workers' inertia works in favor of saving for 
retirement because workers need to do little or nothing to participate 
in their employers' plan.[Footnote 16] Further, workers may also 
decline to participate, in part, because they believe the decision to 
participate in a 401(k) plan requires time-consuming and complex 
decisions, such as choosing how much to contribute and how to invest 
their contributions. Automatic enrollment, through its default 
contribution rates and default investment vehicles, offers an easier 
way to start saving.[Footnote 17] Table 1 shows overall participation 
rate increases reported in the various studies we reviewed. 

Table 1: Overall Participation Rate Effects of Automatic Enrollment: 

Study: The Power of Suggestion[A]; 
Authors (year of study): Madrian and Shea (2001); 
Overall participation rate: Percent under voluntary enrollment only: 
37; 
Overall participation rate: Percent under automatic enrollment: 86. 

Study: For Better or For Worse[A]; 
Authors (year of study): Choi et al. (2001); 
Overall participation rate: Percent under voluntary enrollment only: 26-
43[B]; 
Overall participation rate: Percent under automatic enrollment: 85 and 
above. 

Study: Nationwide Savings Plan Automatic Enrollment Getting Associates 
PREPared for Retirement[A]; 
Authors (year of study): Swanson and Farnen (2008); 
Overall participation rate: Percent under voluntary enrollment only: 
77; 
Overall participation rate: Percent under automatic enrollment: 95. 

Study: The Importance of Default Options for Retirement Saving 
Outcomes: Evidence from the United States[A]; 
Authors (year of study): Beshears et al. (2008); 
Overall participation rate: Percent under voluntary enrollment only: 
60; 
Overall participation rate: Percent under automatic enrollment: 95. 

Study: Building Futures Volume VIII; 
Authors (year of study): Fidelity Investments (2007); 
Overall participation rate: Percent under voluntary enrollment only: 
53; 
Overall participation rate: Percent under automatic enrollment: 81. 

Study: Measuring the Effectiveness of Automatic Enrollment; 
Authors (year of study): Vanguard Center for Retirement Research 
(2007); 
Overall participation rate: Percent under voluntary enrollment only: 
45; 
Overall participation rate: Percent under automatic enrollment: 86. 

Source: GAO analysis. 

[A] These studies are based on a small number of case studies. 

[B] After 6 months of tenure. This study reports separate results for 
three case-study companies. 

[End of table] 

Three of the studies also found that automatic enrollment has a 
significant effect on subgroups of workers with relatively low 
participation rates, such as lower-income and younger workers. For 
example, the Fidelity Investments study found that 30 percent of 
workers aged 20 to 29 were participating in plans without automatic 
enrollment. In plans with automatic enrollment, the participation rate 
for workers in that age range was 77 percent, an increase of 47 
percentage points.[Footnote 18] In addition, four of the studies found 
that automatic enrollment policies reduced the disparities in 
participation rates between certain groups of workers. For example, the 
Madrian and Shea study examined participation rates by race and 
ethnicity and found that, among workers hired under a voluntary 
enrollment only policy, Hispanic workers had the lowest participation 
rate at 19.0 percent, blacks had a slightly higher participation rate 
of 21.7 percent, and whites had the highest participation rate at 42.7 
percent. Under automatic enrollment, however, the participation rates 
for Hispanic and black employees nearly quadrupled to 75.1 percent and 
81.3 percent, respectively, narrowing the gap with white workers, whose 
participation rate more than doubled, increasing to 88.2 percent. The 
difference in participation rates between white and Hispanic employees 
fell from 23.7 to 13.1 percentage points.[Footnote 19] 

The high level of participation rates under automatic enrollment 
appears to persist after such policies are adopted, according to two of 
the studies in our review.[Footnote 20] Both studies observed 
participation rates at specific companies for about 3 years after an 
automatic enrollment policy was adopted. The studies found that for 
employees hired under automatic enrollment, participation jumped almost 
immediately and then increased slightly over the 3-year period, but 
remained relatively stable. The Vanguard study modeled the 
participation rate over time and found a decline in the participation 
rate over a 3-year period, with the participation rate falling by about 
10 percentage points from its peak of 91 percent but remaining high 
relative to participation rates for plans without automatic 
enrollment.[Footnote 21] Consistent with these results, plan sponsors 
who had adopted automatic enrollment stated that their experience also 
indicated that higher participation rates were sustained. One plan 
sponsor, a large manufacturer, reported that after 2 years of automatic 
enrollment, 95 percent of employees who had been automatically enrolled 
had stayed in the plan.[Footnote 22] 

Although automatic enrollment can increase participation and saving 
rates for most workers, it may have an adverse effect on saving rates 
and investment choices for other workers, depending on the nature of 
default contribution rates and default investment funds used, according 
to some of the studies in our review. Four of the six studies we 
reviewed found that automatically enrolled participants are likely to 
accept the plan's default contribution rate. Three of the studies found 
that some participants would have selected a contribution rate higher 
than the default rate had they not been subject to automatic enrollment 
and had they chosen to enroll in the plan voluntarily. Further, two of 
the three studies also found that some participants were likely to 
accept a default investment fund with relatively low future prospects 
for return on investment, such as money-market or stable-value funds, 
compared to the investment fund they would have selected if they had 
voluntary enrolled.[Footnote 23] Thus, these studies concluded that 
overall savings for these particular participants were lower under 
automatic enrollment.[Footnote 24] Further, the Beshears et al. study 
calculated participation rates for a company that doubled its default 
contribution rate from 3 percent to 6 percent and found that the 
participation rates were virtually identical before and after the 
policy change. In addition, Fidelity Investments reported that 
employees accept the default contribution rate the majority of the 
time, regardless of how high it is. Studies have found that default 
policies have these effects, in part, for the same reasons that 
automatic enrollment increases participation rates--accepting the 
defaults is the path of least resistance and requires no action on the 
part of the worker. In addition, some studies found that some employees 
see default policies as implicit advice from the plan sponsor and that 
they imply optimal saving rates and investment choices. 

Adoption of Automatic Enrollment Has Grown Considerably in Recent 
Years, but Many Plan Sponsors Have Not Adopted Policies to Ensure 
Adequate Long-Term Savings: 

Automatic enrollment policies have been increasingly adopted by defined 
contribution plan sponsors in recent years as a result of several 
factors. Nonetheless, a number of considerations may limit adoption of 
automatic enrollment over the long term. Low default contribution rates 
and an apparent lag in the adoption of automatic escalation policies 
raise questions about the adequacy of long-term saving rates under 
automatic enrollment. Further, the widespread adoption of target-date 
funds (TDF)--funds that allocate investments among various asset 
classes and automatically shift to lower-risk, income-producing 
investments as a "target" retirement date approaches--as default 
investments for plans with automatic enrollment has raised concerns 
about investment risk and transparency of investments of TDFs for 
participants nearing retirement. 

Greater Adoption of Automatic Enrollment and Factors Driving Trend: 

Data from two large plan administrators, as well as discussion with 
retirement plan experts, indicate that plan sponsors' adoption of such 
policies has grown considerably in recent years. Data from Fidelity 
Investments show that the percentage of defined contribution plans 
adopting automatic enrollment grew from about 1 percent in December 
2004 to about 16 percent in March of 2009. Comparable data from 
Vanguard show that about 19 percent of plans had adopted automatic 
enrollment as of December 2008, up from 8 percent in June 2006. 

Data from one plan administrator show that large plan sponsors-- 
sponsors of plans with $500 million or more in assets--have adopted 
automatic enrollment policies more often than smaller plans. As figure 
3 illustrates, Fidelity Investments data show that about 40 percent of 
large plans had adopted automatic enrollment by March 2009, while only 
about 14 percent of small plans had done so.[Footnote 25] According to 
Fidelity's data, about 47 percent of all plan participants are included 
in plans that offer automatic enrollment. 

Figure 3: Adoption and Scope of Automatic Enrollment by Plan Size, 
March 2009: 

[Refer to PDF for image: illustration] 

Bar graph: 

Plan size: Large; 
Percentage of Plans: All eligible employees: 8.0; 
Percentage of Plans: New or recent hires only: 32.2. 

Plan size: Mid-sized; 
Percentage of Plans: All eligible employees: 7.2; 
Percentage of Plans: New or recent hires only: 24.8. 

Plan size: Small; 
Percentage of Plans: All eligible employees: 6.6; 
Percentage of Plans: New or recent hires only: 7.5. 

Source: GAO presentation of Fidelity Investments data. 

[End of figure] 

Plan sponsors can choose to apply automatic enrollment "broadly" to all 
employees, or more narrowly to include only newly hired employees. Data 
from Fidelity Investments indicate that automatic enrollment policies 
are typically applied only to new or recently hired employees, but a 
growing percentage of plan sponsors extended automatic enrollment to 
existing employees as well. Among plans with automatic enrollment 
polices, about 58 percent of plans apply such policies to new and 
recently hired employees only, while about 42 percent apply automatic 
enrollment to existing eligible employees as well. [Footnote 26] There 
is considerable variation in this pattern by plan size--as figure 3 
illustrates, the majority of large plans had adopted automatic 
enrollment only for new or recently hired employees, while nearly half 
of small plans applied the policy to all eligible employees. 

According to plan sponsors, retirement plan experts, and others we 
contacted, several considerations have been driving the increase in 
automatic enrollment. These considerations include: (1) plan sponsors' 
desire to increase participation rates, (2) plan administrators' 
marketing of automatic enrollment, and (3) aspects of the Pension 
Protection Act that facilitate and encourage the adoption of automatic 
enrollment. 

Sponsor Desire to Increase Participation Rates: Officials of each of 
the six plan sponsors we contacted that had adopted automatic 
enrollment highlighted the desire to better ensure adequate retirement 
savings for employees. Some plan sponsors noted that automatic 
enrollment was considered necessary because other methods of increasing 
plan participation had not been effective. For example, one plan 
sponsor had sent e-mails and educational materials reminding employees 
of the plan's availability and provided them with analyses of the 
matching funds they had lost by not contributing. While these methods 
raised contribution rates from the 50 percent range to the 65 percent 
range, the company could not further increase contribution rates. 
However, after instituting an automatic enrollment policy, the plan's 
participation rates increased to 93 percent. An official of another 
plan sponsor noted that an automatic enrollment policy was necessary 
because the company had a very young workforce, and the company 
believed that retirement savings was a very low priority and a distant, 
"abstract" benefit for these workers. Some plan sponsors also noted 
that adoption of an automatic enrollment policy may be particularly 
urgent in the case of plans that discontinue other benefits. For 
example, representatives of two plan sponsors stated that automatic 
enrollment was adopted at about the time that an existing defined 
benefit plan was frozen--that is, closed to new entrants. A 
representative of a large plan consulting firm noted that sponsors may 
do this with the long term in view; they want their employees to be 
able to retire at retirement age, partly to ensure that as productivity 
drops off, these workers do not have a reason to stay on indefinitely. 
This plan consultant also said that some plans may adopt automatic 
enrollment to help ensure that the sponsor can pass nondiscrimination 
testing.[Footnote 27] 

Plan Administrator Marketing: 401(k) plan administrators--firms that 
provide and manage retirement plans for plan sponsors--have been 
actively marketing and promoting adoption of automatic enrollment, 
according to plan administrators and others. One large plan 
administrator stated that it encourages automatic enrollment by 
conducting analyses of the effects of automatic enrollment tailored to 
individual plan sponsor clients. The official stated that when plan 
sponsors see the potential benefits to employees of automatic 
enrollment compared to the status quo, about 25 percent of them have 
adopted automatic enrollment. 

The Pension Protection Act: Finally, various aspects of the PPA have 
facilitated the trend toward automatic enrollment. An official of one 
of the nation's largest 401(k) plan administrators noted that the 
criteria and guidelines established in the PPA streamlined and 
simplified the decision-making process about automatic enrollment and 
the related plan design features. One consultant noted that the 
deliberations about the PPA involved considerable industry input and 
had an "announcement effect" that generated considerable publicity 
regarding automatic enrollment. Representatives of two plan sponsors 
said that the PPA safe harbor protection was a consideration in 
adopting automatic enrollment.[Footnote 28] One plan had not adopted 
all of the PPA safe harbor provisions. Instead, it used the safe harbor 
specifications as a guide in setting features such as a matching 
contributions and the 3 percent default contribution rate. The other 
plan sponsor adopted all of the PPA safe harbor provisions. 

Factors Limiting the Adoption of Automatic Enrollment: 

Various factors, including higher costs, management views, concerns 
about legal and regulatory challenges, and lack of awareness may delay 
or prevent the adoption of automatic enrollment policies by some plan 
sponsors. 

Greater Costs: Automatic enrollment implies greater costs for plan 
sponsors, including higher matching costs and greater fees paid to plan 
administrators. One plan administrator stated that automatic enrollment 
could be particularly unattractive to cost-sensitive companies that 
have narrow profit margins. This concern was reflected by a plan 
sponsor who noted that the adoption of automatic enrollment would be 
difficult for one of the company's subsidiaries, which operated in a 
very cost-competitive environment and would therefore have difficulty 
passing on costs to customers. The subsidiary would probably have to 
absorb the additional costs through reduced profit margins. Plan 
consultants and plan administrators noted that the costs of automatic 
enrollment may be a particular concern in the current economic 
environment. One large plan administrator noted that the current state 
of the economy will slow down adoption of automatic enrollment, as many 
companies try to minimize additional costs. While some noted that this 
should be a transient consideration, one expert we contacted said that 
the recession would be a memorable event for some plan sponsors, which 
could have longer-term implications. 

Certain types of plan sponsors may be especially concerned about the 
cost impact of automatic enrollment. For example, plan sponsors that 
employ a low-wage and high-turnover workforce--such as retail 
establishments and restaurant chains--may be especially reluctant to 
adopt automatic enrollment because of the additional cost, 
administrative burden, and prospect of limited benefits for employees. 
One such plan sponsor we contacted explained that adopting automatic 
enrollment would result in a five-fold increase in the number of plan 
participants and associated administrative costs and that the company 
might lower the employer match to mitigate the associated cost 
increases. In addition, the plan would have greater administrative 
burdens related to the need for employee communication and to handle 
inquiries from the much larger pool of participants. A representative 
of this plan sponsor also explained that, even with contribution rates 
of 6 percent, low-wage and short-tenure staff would accumulate very 
small balances and likely abandon them or take a lump-sum distribution 
upon separation. [Footnote 29] 

Management Views: Apart from costs, experts noted that some plan 
sponsors may be reluctant to adopt automatic enrollment due to certain 
management views or out of concern about employee reaction or welfare. 
Some experts told us that some managers view automatic enrollment as 
overly paternalistic or do not wish to reward passivity in employees 
who do not voluntarily join a plan. For example, a representative of 
one plan sponsor told us that the company wanted all employees to 
participate in the plan but wanted active participation and 
conscientious saving. A representative of a small manufacturing company 
stated that management believes that the workforce would be highly 
distressed if the company began summarily taking 401(k) contributions 
from their pay, even with a well-communicated opt-out feature. The 
representative further noted that the sponsor believes the recent 
declines in the equity markets also weigh against adoption of automatic 
enrollment, in light of the firm's fiduciary responsibility for the 
plan. 

Legal and Regulatory Challenges: Some experts noted that some plan 
sponsors may be reluctant to adopt automatic enrollment due to legal 
and regulatory concerns. One expert noted that small plans see the 
legal and regulatory environment surrounding 401(k)s as complex and may 
see a switch to automatic enrollment as overly risky. Another expert 
noted that, unlike larger plan sponsors, small plans without well- 
staffed legal and compliance departments may be risk-averse and slow to 
adopt new policies. 

Lack of Employer Awareness: Some smaller employers may not be aware 
that automatic enrollment is a plan feature available to them. 
Representatives of two small plan sponsors said that they were 
generally unaware of automatic enrollment as a plan option. For 
example, one small plan sponsor stated that neither she nor the 
organization's 401(k) employee advisory committee was familiar with the 
concept of automatic enrollment. Further, the local service provider 
that provides the organization with plan guidance and management 
assistance had not suggested such a policy. Relatedly, representatives 
of one large plan administrator told us that small plan sponsors 
generally lag behind large plan sponsors in adopting innovative 
services, tools, and plan design features, including automatic 
enrollment. 

Impact of Automatic Features on Saving Rates and Investment Risk and 
Transparency: 

Available data show that many plans with automatic enrollment have not 
adopted default automatic escalation policies, which, in combination 
with low default contribution rates, could result in low saving rates 
for participants who do not increase contribution rates over time. 
Also, experts noted that a trend toward TDFs as default investments, 
while potentially beneficial in important respects, also raises 
questions about the investment risks and transparency for those close 
to retirement. 

As figure 4 illustrates, data from two large plan sponsors indicate 
that the majority of plans with automatic enrollment have adopted 
initial default contribution rates of 3 percent. Between 15 to 17 
percent of plans have a default contribution rate of less than 3 
percent, and between 22 and 25 percent of plans have a default 
contribution rate of more than 3 percent.[Footnote 30] Data from 
Fidelity showed that the average default contribution rate grew 
modestly, from about 3.0 percent in March 2005 to about 3.2 percent in 
March 2009. According to a Vanguard official, the average default 
contribution rate was 3.3 percent at the end of 2008.[Footnote 31] 

Figure 4: Default Contribution Rates for Plans with Automatic 
Enrollment: 

[Refer to PDF for image] 

Default deferral percentages: 1 percent; 
Fidelity: 5; 
Vanguard: 2. 

Default deferral percentages: 2 percent; 
Fidelity: 12; 
Vanguard: 13. 

Default deferral percentages: 3 percent; 
Fidelity: 62; 
Vanguard: 60. 

Default deferral percentages: 4 percent; 
Fidelity: 9; 
Vanguard: 10. 

Default deferral percentages: 5 percent; 
Fidelity: 6; 
Vanguard: 7. 

Default deferral percentages: 6 percent; 
Fidelity: 6; 
Vanguard: 7. 

Source: GAO presentation of data from Fidelity Investments and Vanguard 
Group. 

Note: Fidelity data are as of March 2009. Vanguard data are as of 
December 2008. Vanguard data indicate that 8 percent of plans had 
default deferral rates of 6 percent or more. 

[End of figure] 

Available data is mixed with regard to the extent to which plan 
sponsors with automatic enrollment have also adopted automatic 
escalation policies. According to one plan administrator's data, about 
45 percent of plans with automatic enrollment had adopted a default 
automatic escalation feature as of March 2009, up from zero in 
2005.[Footnote 32] Further, this administrator's data shows that 
adoption of default automatic escalation policies are much less 
prevalent among large plans than small plans--about 24 percent of large 
plans with automatic enrollment policies have adopted such a policy, 
while about 51 percent of small plans have done so. Data from another 
plan administrator show a much greater rate of adoption of default 
automatic escalation policies--77 percent of all plans with automatic 
enrollment had adopted default automatic escalation policies in 2008, 
up from 31 percent in 2005.[Footnote 33] 

Available data indicate that plans with automatic enrollment policies 
overwhelmingly adopted TDFs as a default investment. TDFs allocate 
their investments among various asset classes and shift that allocation 
from equity investments to fixed income and money market investments as 
a "target" retirement date approaches.[Footnote 34] Eighty-seven 
percent of Vanguard Group plans with automatic enrollment had adopted 
TDFs as a default investment at the end of 2008, compared to 42 percent 
in 2005. Conversely, the use of balanced funds, money market funds, and 
stable value funds as default investments have declined 
significantly.[Footnote 35] This trend toward TDFs as a default 
investment vehicle is corroborated by data from Fidelity Investments, 
which shows that 96 percent of plans with an automatic enrollment 
policy used TDFs as of March 2009, up from 57 percent at the end of 
2005.[Footnote 36] 

Figure 5: Default Fund Types Used by Plans with Automatic Enrollment 
Policies in 2005 and 2008: 

[Refer to PDF for image: chart] 

Pie Graphs: 

Default Investments, 2005: 

Target date funds: 42%; 
Balanced fund: 33%; 
Target data funds: 25%. 

Default Investments, 2008

Target date funds: 87%; 
Balanced fund: 11%; 
Target data funds: 2%. 

Source: GAO presentation of data from Vanguard Group. 

Target-Date Funds: 

Target-date funds (TDF) offer investors certain advantages generally 
not offered by other types of investment vehicles. TDFs allocate their 
investments among various asset classes and shift that allocation to 
more conservative investments as a "target" date approaches. For 
example, a TDF could be designed for workers expecting to retire many 
years in the future and would typically have a much greater allocation 
to equities and a lesser allocation to fixed-income investments. 
Conversely, a fund designed for workers nearing retirement age would 
tend to have a greater allocation to fixed-income investments. TDFs 
thus offer participants a potentially beneficial long-term asset 
allocation strategy while lowering risks as the participant approaches 
retirement age. 

[End of figure] 

This trend is important in part because recent evidence suggests that 
participants who are automatically enrolled in plans with target-date 
fund defaults tend to have a high concentration of their savings in 
these funds. A recent analysis by the Employee Benefit Research 
Institute found that workers who were considered to be automatically 
enrolled in a 401(k) plan were more likely than those who voluntarily 
joined to invest all their assets in a TDF.[Footnote 37] The study 
found that, except for participants in plans with more than 10,000 
participants, more than 90 percent of those automatically enrolled in 
TDFs had all of their allocation in such funds. 

While TDFs may help ensure that workers have a more age-appropriate mix 
of investments, some experts have stated that TDFs may pose certain 
challenges, as recent events in the financial markets have illustrated. 
As a result of the 2008 stock market decline, some TDFs designed for 
those expecting to retire in or around 2010 lost 25 percent or more in 
value. In light of concerns about a number of issues, including how 
plan sponsors evaluate, monitor, and use TDFs, in 2008 the Advisory 
Council on Employee Welfare and Pension Benefit Plans recommended that 
the U.S. Department of Labor provide more specific guidance regarding 
the complex nature of TDFs, and outline the methodology that plan 
fiduciaries should follow in selecting and monitoring them. The 
Advisory Council also stated that additional education materials would 
help plan participants become aware of the value and risks of 
TDFs.[Footnote 38] 

Automatic IRA and State-Assisted Retirement Savings Plan Proposals 
Could Improve Access to Retirement Savings Vehicles for Uncovered 
Workers, but Each Proposal Faces Challenges: 

In order to promote retirement savings among the 40 percent of the 
workforce whose employer does not sponsor a plan, members of Congress, 
in recent years, have introduced bills for federally required 
"automatic IRAs" that would be implemented nationwide. Automatic IRAs 
offer the potential benefit of expanding retirement coverage but some 
have expressed concerns that automatic IRAs may not result in 
significant retirement savings, and raised questions about the costs of 
such a program for employers and the federal government. There have 
also been proposals for state-supported retirement savings programs 
over the past several years. These state proposals also have the 
potential to expand retirement coverage, but on a state-by-state basis. 
Concerns have been expressed regarding the cost of these state 
proposals, as well as employer and employee interest in such plans, and 
potential legal barriers. 

Automatic IRAs Could Provide Benefits to More Workers, but Its Impact 
Is Uncertain: 

A number of existing proposals have described in general terms the 
concept of an automatic IRA, and they contain common elements.[Footnote 
39] Under the 2006 and 2007 congressional bills for automatic IRAs, 
employers that do not sponsor a retirement plan would be required to 
defer a percentage of an employee's pay to an IRA through payroll 
deduction, unless the employee opts out. The requirement would apply to 
all employers with more than 10 employees and who had been in business 
for at least 2 years, or employers with 100 or more employees 
regardless of the length of time the employer has been in business. 
Eligible workers would be those who had worked for an employer for a 
specified period, were at least 18 years old, and were not eligible to 
participate in any other qualified retirement plan the employer 
sponsors. Affected employers could either (1) automatically enroll 
eligible workers, although employees would be offered the option to 
affirmatively opt out or (2) require that employees make an explicit 
yes or no decision on whether to participate; workers not making such a 
decision would be automatically enrolled (see fig. 6). Employers would 
then transfer a portion of employees' pay to either a traditional IRA 
or a Roth IRA through payroll deduction.[Footnote 40] Employers could 
elect to send contributions to IRAs of an employer-designated issuer, 
unless the employee selects his own IRA provider. If neither the 
employer nor the employee designates a specific IRA provider, the 
contributions would be deposited into federally designated default 
accounts. 

Figure 6: Steps Involved with Establishing an Automatic IRA: 

[Refer to PDF for image: flowchart] 

Source: GAO, Adobe systems incorporated. 

[End of figure] 

The automatic IRA is intended to help address the retirement security 
needs of those not already covered by an employer-sponsored retirement 
plan. Further, the automatic IRA is designed to extend the benefits of 
payroll-deduction savings and automatic features of 401(k)s.[Footnote 
41] According to some experts, requiring employers to offer automatic 
IRAs is necessary for a number of reasons. First, employers have had 
the option to establish payroll-deduction IRAs for over 10 years, and 
for a number of reasons, very few employers have done so. As we 
previously reported,[Footnote 42] IRA providers have told us some 
employers may be reluctant to adopt a payroll-deduction IRA because 
they believe that their publicizing the payroll-deduction IRAs may be 
construed as an endorsement of the policy, which could potentially 
violate ERISA.[Footnote 43] Further, employers may not be aware of how 
payroll-deduction IRAs work and some small employers may not be aware 
that this option exists. The automatic enrollment component is 
necessary, according to designers of the automatic IRA, because various 
impediments would prevent many eligible employees from taking advantage 
of an available payroll-deduction IRA. For example, employees would 
have to decide whether to participate, select an IRA provider, select 
investment vehicles, and determine how much to contribute. These 
officials note that many workers may have difficulty overcoming 
inertia, and automatic enrollment would help them overcome this 
difficulty. 

Impact on Employees and Retirement Savings: 

Advocates for automatic IRAs and some pension industry experts reported 
that the automatic IRA could have a positive effect on retirement 
savings. According to the architects of the approach, automatic IRAs 
offer a powerful mechanism for accumulating retirement savings through 
regular payroll deposits that continue automatically. In light of the 
impact of automatic enrollment polices on participation in 401(k)s, an 
automatic IRA program that features default automatic enrollment could 
have a positive impact on participation rates. One study estimated that 
likely automatic IRA participants include younger, part-time, and lower-
or moderate-income workers, as well as workers subject to higher than 
average job turnover. Advocates stated that automatic IRAs could help 
these employees overcome inertia since they would no longer need to 
take the initiative in order to save. Further, two pension industry 
experts told us that the payroll-deduction nature of automatic IRAs 
would ensure that employees of affected companies are saving on a 
regular basis. However, some experts agree that the automatic 
enrollment component of automatic IRAs has the potential to 
significantly increase the number of workers saving for retirement by 
including workers that currently do not have access to an employer- 
sponsored plan. 

Some caution, however, that the benefits resulting from the automatic 
IRA could be relatively small. A 2009 preliminary analysis funded by 
the Department of Labor illustrates potential outcomes of two automatic 
IRA scenarios using specific behavioral assumptions about participation 
and contribution rates, among other things.[Footnote 44] The analysis 
found that the resulting increase in average retirement benefits at age 
70 is small (even when ignoring account fees and offsetting reductions 
in other savings) and is weighted toward the third of the population 
with the highest lifetime earnings. If actual participation rates in 
the automatic IRA are higher than the analysis assumes--and the 
experience of automatic enrollment in 401(k)s indicates this is a 
possibility--the resulting increase in average retirement benefits 
could be higher. The Department of Labor is undertaking additional 
analysis to illustrate the effects of participation rates similar to 
those achieved in 401(k) plans with automatic enrollment. 

Participants may also not fully benefit from the tax incentives 
provided by automatic IRAs. According to analysis sponsored by AARP, 50 
percent of automatic IRA participants would be lower-income. Therefore, 
with the exception of Social Security and Medicare taxes, they would 
pay little, if any, income tax and may not benefit from the tax 
incentives traditional IRAs offer. Further, while some participants 
would be eligible for the Saver's Credit, one analysis estimated that 
the benefit of the credit could be limited. Because the credit is 
nonrefundable, it can only be used to offset income tax liability. If 
the participant does not have income tax liability up to the full 
amount of the credit, the full value of the credit can not be used. For 
example, the study concluded that if participants made the maximum 
permitted contributions to an automatic IRA, then nearly 90 percent 
would not receive the full credit. In part to improve the tax incentive 
of automatic IRAs, the Department of Treasury, in August of 2009, 
proposed making the Saver's Credit fully refundable and depositing it 
automatically into IRAs.[Footnote 45] 

Finally, proper administration of accounts, including record keeping, 
will be important for managing and maintaining the retirement savings 
accumulated through the IRAs, according to some analyses and experts. 
For example, workers would be responsible for ensuring that they do not 
exceed applicable limits on annual contributions to an automatic IRA. 
An analysis sponsored by AARP and pension industry officials noted that 
it might be difficult for some workers to keep track of their accounts 
when they move from job to job or if their employer goes out of 
business.[Footnote 46] A 2007 study noted that automatic IRA proposals 
do not impose record-keeping responsibilities on employers, beyond 
withholding and transmitting IRA contributions. Because of this, the 
report noted that other entities must assume these responsibilities, 
and a companion report recommended a centralized administrator be 
responsible for record keeping. For example, the first report found 
that a central administrator could help prevent lost accounts by 
providing participants with annual reports including their 
contributions and investment earnings for the year, as well as the 
total account balance. In addition, the architects of the automatic IRA 
proposed that the federal government set up a standard default account 
and contract with private financial institutions for record-keeping 
services, among other things, to make managing the IRAs easier. 

Impact on Employers: 

A variety of views exist regarding the cost and administrative burden 
that an automatic IRA would place on affected small employers. As one 
analysis of the automatic IRA concept noted, an automatic IRA has the 
potential to result in some costs and administrative burdens on small 
employers. The analysis noted that employers would need to provide 
employees with election forms and process paperwork with respect to 
employee elections or non-elections, choose an IRA provider or opt for 
the federally designated default investment, and withhold contributions 
from employees' pay. Pension industry officials also stated that some 
small employers do not have payroll deduction systems and send paper or 
spreadsheet files to their record keepers and brokers. They told us 
that these employers may choose to invest in new infrastructure in 
order to remit automatic IRA contributions through payroll deduction. 

Recent legislative proposals for an automatic IRA have recognized the 
potential challenges for small employers and contain provisions to 
mitigate their impacts. For example, the 2007 bills to authorize 
automatic IRAs would have exempted employers with fewer than 10 
employees as well as those that had been in business for less than 2 
years. According to some advocates, the proposals would have therefore 
avoided placing additional requirements on the smallest companies, 
which may not have electronic payroll-deduction systems. It would also 
relieve employers starting a new business from additional costs and 
administrative burdens. In addition, the proposals would establish a 
tax credit in the early years for participating employers with fewer 
than 100 employees to mitigate some administrative and startup costs. 

In light of these automatic IRA features, some experts and a recent 
analysis have found that additional costs may be small for most small 
employers. The architects of the automatic IRA concept have stated that 
because many employers already make deductions for federal income tax 
and payroll tax withholdings, making IRA payroll deductions would 
impose little, if any, new administrative costs. Further, they said 
these employers would not have to bear the costs involved in 
maintaining a retirement plan, such as matching employee contributions. 
These views were supported by a 2009 report sponsored by AARP, which 
found that most small employers would face low costs to implement the 
automatic IRA.[Footnote 47] This study noted, for example, that about 
97 percent of employers with 10 or more employees had automated payroll 
systems, and such employers would face relatively few burdens 
implementing the automatic IRA. The report also found that payroll 
software companies and payroll service providers are likely to adopt 
automatic IRA requirements in their services to small employers. For 
the estimated 3 percent of affected employers that process payroll 
manually, the automatic IRA would also have to be implemented manually. 

Impact on the Federal Government: 

Past proposals have described a federal role that could mitigate some 
of the difficulties and risks of implementing an automatic IRA. For 
example, the proposals would have created a new federal entity that 
would, among other things, establish low-cost default 
investments.[Footnote 48] According to experts we contacted and 
analyses we reviewed, establishing an automatic IRA policy would 
require tax incentives to make it more affordable to some employers and 
federal expenditures to establish and govern the program, among other 
things.[Footnote 49] However, analyses sponsored by AARP found that it 
is not possible to determine what the costs would be to the federal 
government without a more detailed proposal. Further, these studies 
reported that the establishment of an automatic IRA would reduce 
federal tax revenues as a result of the tax credit available to small 
employers, individual tax benefits from deferred employee income, and 
greater use of the Saver's Credit. Two analyses estimated that the 
revenue losses could amount to somewhere between $2 billion and $19 
billion over 10 years. 

Impact on 401(k) Plans: 

Industry officials and some experts we contacted also noted that 
establishment of an automatic IRA could affect the market for 401(k) 
plans. For example, some pension industry experts and representatives 
of two national organizations representing large plan sponsors noted 
that if automatic IRAs are made too attractive, they might displace 
401(k)s and Savings Incentive Match Plan for Employees of Small 
Employers (SIMPLE) plans.[Footnote 50] These officials said that an 
employer might forego adopting a 401(k) plan if they have already been 
required to facilitate an automatic IRA, and that this would be 
unfortunate because an IRA is in many ways an inferior option for 
workers. The officials told us that a 401(k) plan can offer workers 
better benefits, such as an employer match and higher contribution 
rates.[Footnote 51] In addition, some of the officials reported that 
the existence of a 401(k) creates a workplace culture that encourages 
participation in retirement saving. Further, an AARP-sponsored analysis 
noted that small businesses will weigh the costs of establishing or 
maintaining a qualified retirement plan against the costs of complying 
with the automatic IRA requirements. To the extent that the automatic 
IRA approach offers significantly lower costs--including the relative 
costs of fiduciary liability--employers may decide against adopting a 
401(k) plan or may eliminate an existing one.[Footnote 52] In light of 
concerns such as these, officials of one large financial services firm 
said that, if implemented, the automatic IRA program should be 
evaluated to determine if it led to a shift away from 401(k) plans. 

Others, however, have stated that the automatic IRA is not likely to 
erode the popularity of 401(k) plans, and may even promote greater 
adoption of such retirement vehicles. Some experts noted that the 
potential for automatic IRAs to supplant 401(k)s can be minimized by 
careful design of the program. Perhaps most importantly, they said that 
the maximum annual savings for an automatic IRA should be designed so 
that it is less attractive to a small business employer than a 401(k). 
Specifically, they noted that it is important that the automatic IRA 
not permit contributions above the current IRA dollar limits to avoid 
competing with qualified plans. These advocates reason that the 
potential tax advantage of a 401(k) enables a small business owner-- 
who, along with his employees, may choose to participate in the 401(k) 
or IRA plan his business sponsors--to save a much higher percentage of 
his tax-deferred income than does an automatic IRA, making the 401(k) 
plan a more attractive option. Further, some industry experts told us 
that an automatic IRA would likely be a stepping stone toward adopting 
a 401(k) for some small employers. For example, according to an 
official of one large organization representing pension professionals 
stated that payroll deduction automatic IRA arrangements will 
ultimately encourage more employers to sponsor 401(k) plans, and 
contribute on their employees' behalf. 

State-Assisted Retirement Savings Proposals Intend to Increase 
Retirement Plan Savings but Raise Questions Regarding Participation, 
Program Design, and Legal Barriers: 

In recent years, 10 states have considered proposals that would involve 
state governments in facilitating retirement savings plans for workers 
whose employer does not sponsor a plan.[Footnote 53] One study on such 
state proposals reported that employers would like to provide a 
retirement savings vehicle for their employees, but are inhibited by 
the cost, complexity, and time that would be required to do so. 

Under the proposals, state governments could take a number of actions 
to help address some of these issues and facilitate the use of payroll- 
deduction or SIMPLE IRAs or the adoption of 401(k) plans. For example, 
a state would promote private pension coverage and facilitate 
retirement savings for small business, moderate-income, and lower- 
income workers--who are less likely to be covered by an employer- 
sponsored plan--by acting as an intermediary between employers and 
financial institutions. In addition, a state could help small employers 
pool their investments and administrative activities. 

The type of retirement plans that would be established by the programs 
varies. For example, California's proposal would establish a payroll- 
deduction, traditional, or SIMPLE IRA, or a combination of these 
options; Connecticut's proposal would establish a 401(k) or other type 
of defined contribution plan; and proposals in Maryland and Washington 
would establish a defined contribution plan and IRA options.[Footnote 
54] Regardless, the state would initiate and oversee the programs, 
while private-sector companies under contract to the state would manage 
the investment vehicles and day-to-day administration. For example, a 
Washington study outlined an option under which the state would design 
the basic features of a 401(k) plan and market the plan to private- 
sector employers who do not currently offer a plan. The state would 
then contract with a private-sector plan administrator to provide 
access to investment funds, direct customer service, Web-based account 
access, and to distribute account statements and other communications. 
Table 2 shows examples of state roles and the intended impact of 
automatic IRAs under existing proposals. 

Table 2: Examples of State Roles and Intended Impact under Existing 
Proposals: 

Potential state role: Publicize program and provide information to 
employers; 
Intended impact: Through outreach and marketing, the state could inform 
smaller employers abut the availability and benefits of 401(k) plans or 
payroll-deduction IRAs. As a result, small employers would become aware 
of the state program and the benefits their employees would derive from 
participation. 

Potential state role: Develop, approve, and help market a prototype 
401(k) plan; 
Intended impact: A prototype plan document would facilitate employer 
participation by eliminating a potentially high up- front cost. Such a 
document could be developed using in-house expertise in coordination 
with private-sector financial firms. 

Potential state role: Negotiate on behalf of small employers and 
contract with service providers; 
Intended impact: The state would have bargaining power with financial 
service providers in light of the prospect that the program could grow 
to large scale. For example, through competitive bidding the state 
could negotiate for services that are low in cost and incorporate best 
practices. 

Potential state role: Pool savings of employees of small employers; 
Intended impact: By facilitating the pooled savings of numerous small 
accounts, the state would obtain lower-cost retirement plans. Small 
employers are less likely to sponsor retirement plans in part because 
of higher per capita costs and because financial service providers are 
less likely to be interested in serving small accounts. 

Potential state role: Assist employers with plan administration and 
compliance; 
Intended impact: By providing participating small employers with 
convenient and professional assistance with plan investments, 
administrative tasks, and compliance, the state would further encourage 
small employers to adopt a plan or facilitate a payroll-deduction IRA. 

Source: GAO analysis. 

[End of table] 

Challenges States Would Face: 

Program Participation: Little is known about the extent to which 
employers and employees would participate in a state-assisted 
retirement savings program. According to representatives of two 
organizations in favor of state-assisted retirement savings plans, 
employer and employee interest in such a program could be considerable. 
Moreover, one of the representatives noted that proposals in Washington 
and other states have been specifically designed to address small 
companies' concerns that the cost of setting up payroll deduction 
prevents them from adopting a 401(k) plan or a payroll-deduction IRA. 
However, no known rigorous assessment exists of the extent to which 
small employers or employees would opt to participate in such a 
program. We obtained state-sponsored studies from three of the four 
states we examined, and while none of the studies included an analysis 
measuring the magnitude of the market demand for such a program, two 
noted concerns regarding the potential demand for a state program. For 
example, a study prepared for the Maryland legislature compared a 
Maryland proposal to a number of other state financial programs, such 
as a state-sponsored college savings program.[Footnote 55] In comparing 
the two, the report noted that the college savings program offers 
distinctive tax and pre-payment guarantee advantages that are not 
otherwise available in the private market. However, the study noted 
that this distinction, which helps ensure a market demand for the 
college savings plans, would not exist for the state-assisted 
retirement program. The report noted that there will be no additional 
employee tax benefit for participants in the proposed state-assisted 
retirement savings program and, for that reason, the program would have 
to compete on an equal footing with plans in the marketplace. The 
Maryland analysis concluded that the program might be difficult to 
establish or market in the absence of a federal requirement that all 
employers have a retirement savings plan. 

Representatives of the financial services industry also indicated that 
state-assisted programs to some extent could have a "zero-sum" effect. 
In commentary on the Connecticut proposal in March of 2008, the Small 
Business Council of America (SBCA) and American Society of Pension 
Professionals and Actuaries (ASPPA) stated that low-cost retirement 
options exist now, and the state-assisted program would result in 
little difference in cost. The SBCA and ASPPA added that state 
government should not compete with small private businesses unless 
there is a clear market failure or some inherent unfairness that 
disadvantages its citizens. Representatives of the Connecticut Bankers 
Association were concerned that rather than expand retirement savings 
vehicles to new employees, the state program would attract the "already 
served market" with initial low costs. Similarly, an SBCA 
representative stated that providing retirement services to small 
employers is not very profitable for financial services firms, and if 
such a proposal resulted in the state obtaining half of the small 
business market, for example, it is conceivable that large plan 
administrators would exit the business since they only profit in this 
sector though large volume. Further, a fiscal analysis of a Washington 
state proposal noted that the program would have no initial plan assets 
and uncertain levels of participation and, as a result, vendors may 
have difficulty estimating the total cost of record-keeping services. 
This, in turn, could affect vendor interest in providing services for 
the program. 

Program Design and Costs to State: Studies from California, Maryland, 
and Washington about the feasibility of state-assisted retirement 
savings programs identified important questions about program design 
and related costs. For example, state governments will need to 
determine the extent to which administrative and management 
responsibilities will be borne by the state, and how much will be 
contracted out to financial services providers. States would also need 
to determine what types of investment funds will be used, including any 
default funds. Further, analyses from the three states showed they 
would face initial and ongoing costs. In addition, a fiscal analysis of 
the Washington proposal identified three major costs categories-- 
including program development and administration, communications and 
marketing, and record keeping. As table 3 illustrates, the Washington 
analysis estimated a total cost of about $4.4 million to implement and 
operate the program in the first 2 years. 

Table 3: Estimated Program Development and Operation Costs in 
Washington State: 

Cost category: Program development and ongoing administration; 
Examples: * Working with a contracted defined contribution consultant: 
- Develop program design; 
- Conduct procurement process for obtaining record keeper and vendors; 
- Create communication plans; 
- Test systems, processes, and communications materials; 
- Provide oversight of program, including legal support and analysis, 
contract management, and financial reporting; 
Estimated costs[A]: $2,224,000. 

Cost category: Communication and marketing; 
Examples: * Develop statewide marketing and communication program 
targeted toward employers and employees; 
* Publish and disseminate participant and employer communications 
through mailings and other media such as TV, radio, and Web; 
Estimated costs[A]: $1,606,000. 

Cost category: Record keeping; 
Examples: * Contract with third-party record keeper to provide record-
keeping service, including: 
- Daily valuations of participant accounts; 
- Quarterly statements; 
- Customized Web site; 
- Internet account access; Estimated costs[A]: $575,000. 

Cost category: Total; Examples: [Empty]; Estimated costs[A]: 
$4,405,500. 

Source: Fiscal Note to Washington Senate Bill 5791. 

[A] Cost estimates are for 2009-2011. The estimate indicated that 
comparable total costs would also be incurred in future years. 

[End of table] 

Lack of startup funding may be a significant barrier in some states. 
While the Maryland and Connecticut proposals allow for state budget 
appropriations (which may later be recovered through participant fees), 
the California proposal stated that initial funding could come from a 
state budget appropriation or a non-profit or private entity. However, 
state budget appropriations may be difficult to obtain given these 
states' current budget shortfalls. The Washington proposal stated that 
its program could not be started until the state had obtained federal 
and/or philanthropic funds sufficient to support the first 3 years of 
the program. According to an advocate for the Washington proposal, his 
organization recognized that, due to the current economy, the states 
are not in a position to self-fund state-assisted retirement savings 
programs. He reported that his organization has been working on 
obtaining federal funds to cover the startup costs of such programs. 

Legal and Regulatory Challenges: State-assisted savings programs would 
also have to comply with both federal and state law that, according to 
the state analyses, could provide additional challenges. One analysis 
noted that states would need to obtain requisite federal approvals to 
ensure that the programs adhere to all federal requirements governing 
the operation of retirement plans. For example, if the program 
establishes 401(k) accounts, the state would need to submit plan 
documents to the IRS and Department of Labor for approval. Further, the 
Maryland and Washington analyses found that if a state sponsored the 
establishment of a 401(k) plan, such plans would be subject to ERISA's 
fiduciary requirements and expose either the state or the employer to 
potential liability in the event that participants suffer financial 
losses. According to the Maryland analysis, participants may try to 
hold the state or the employer liable if they incur investment losses 
after being sold an investment unsuitable for their needs or if they 
received misleading communications about investments. However, both 
analyses noted that there are steps the states can take to minimize 
their liability. For example, according to the Maryland study, the 
state could limit investment options to reduce the possibility of 
unsuitable choices or miscommunication. 

The Washington and Maryland analyses discussed a number of other 
potential liabilities that states could face. For example, the Maryland 
analysis noted that a failure to file required forms and conduct 
transactions under applicable standards could subject the state or 
employers to significant penalties imposed by IRS or the Department of 
Labor. Officials from Washington noted that a 401(k) option could 
create a new and complex compliance and monitoring role for the state 
retirement agency and that it could be administratively difficult for 
the state to assume this role. They added that the state would face a 
steep learning curve in addressing ERISA and liability issues, and 
might have to contract with outside expertise to deal with compliance 
and oversight issues. 

The state role envisioned in the proposals may be precluded by some 
states' constitutions. Analyses of proposals in California and 
Washington specifically cited aspects of the state constitutions that 
could affect the states' ability to operate a plan. For example, 
California's constitution prohibits the gift or loan of state credit to 
associations, companies or corporations, and prohibits the state from 
loaning its credit, subscribing to, or being interested in the stock of 
any company, association, or corporation.[Footnote 56] According to the 
California study, although California's constitution specifically 
exempts the retirement board of a public retirement system from this 
prohibition, it is not clear whether the exemption would extend to a 
program for private-sector employees. The California study noted that 
if the program is structured in such a way that the state internally 
manages funds associated with the program, this could be seen as the 
state having a financial interest outside the limits of the public 
employees' retirement fund. The California study also observed that a 
constitutional amendment may be needed to address this issue. 
Washington's constitution has a similar prohibition.[Footnote 57] The 
Washington report noted that the proposed retirement savings program 
may be permissible because it serves a public purpose by helping 
individuals save for retirement and reducing the risk that individuals 
will rely on state assistance in the absence of adequate retirement 
savings. However, the report also noted that no Washington case has 
considered proposals similar to those discussed in the report. The 
report concluded that it was not possible to predict how a court would 
rule should the program be challenged. In addition, because no other 
states have enacted such programs, there is no guidance available from 
other courts. 

Concluding Observations: 

Automatic enrollment of workers in 401(k) plans has proven to be an 
effective means of increasing plan participation rates. Because such 
policies are being increasingly adopted by defined contribution plan 
sponsors in the wake of the Pension Protection Act of 2006, many 
additional workers will be brought into plans who might not otherwise 
have participated. Nonetheless, a number of considerations could 
potentially limit the extent or impact of such policies. First, the 
benefits of automatic enrollment are inherently limited to workers that 
have access to an employer-sponsored plan but do not participate. 
Second, some types of employers, such as employers with high-turnover 
workforces and small employers, may find automatic enrollment too 
costly or inappropriate for their workforce. Third, initially low 
default contribution rates and the absence of default automatic 
escalation policies at some plans may result in inadequate long-term 
savings for some workers. 

Automatic IRAs may hold promise for workers who do not have access to 
an employer-sponsored plan. The proposal has potential in that it could 
foster retirement savings among the roughly 40 percent of the workforce 
whose employers do not sponsor a plan. As such a policy is designed, 
however, a number of important issues remain to be considered. For 
example, it is not clear that an automatic IRA will offer low-income 
workers a significant benefit. Further, in order to ensure that the 
intended beneficiaries accumulate and retain savings, some central 
administration--possibly by the federal government--may be required to 
assume significant and long-term involvement for record keeping and 
administration. The nature and costs of such a role have not yet been 
publicly assessed or compared against the potential benefits and 
limitations of an automatic IRA. In addition, while state-assisted 
retirement savings plans may also hold some promise for expanding 
retirement coverage for workers, none of these proposals has been 
enacted and they could face significant legal barriers to 
implementation. 

Both the growth of automatic enrollment and the introduction of 
automatic IRA proposals have brought renewed attention to the question 
of how to extend retirement coverage to the half of the workforce not 
covered by an employer-sponsored plan. This is an important step 
forward, as past debate over retirement security has largely focused on 
increasing retirement savings for those already participating in 
retirement plans. As plan sponsors and participants gain more 
experience with automatic enrollment, it will be helpful to learn from 
these experiences, especially in light of the recent recession. The 
lessons learned may have important implications for related 401(k) plan 
features, such as automatic escalation, and for the potential 
feasibility and usefulness of automatic IRA and state-assisted 
retirement savings proposals. Further, it would also be helpful to 
carefully consider the various concerns raised in the automatic IRA 
debate to increase the likelihood that, if such a proposal becomes law, 
it is administered in an efficient and effective way. Finally, while 
state efforts could be helpful in increasing the number of workers 
saving for retirement, these efforts may not be necessary depending on 
the potential implementation of automatic IRAs. Further, fiscal 
difficulties in some states may make such proposals difficult to 
implement in the near future. 

Agency Comments: 

We provided a draft of this report to the Department of Labor and the 
Department of the Treasury for review and comment. The Department of 
Labor generally agreed with our findings. With regard to the potential 
impacts of an automatic IRA on retirement benefits, Labor said that the 
Employee Benefits Security Administration is undertaking additional 
analysis to illustrate the effect of higher participation rates, 
similar to those achieved in 401(k) plans. The Department of Labor 
provided technical comments, which we incorporated as appropriate. The 
Department of Labor's formal comments are reproduced in appendix II. 

The Department of the Treasury also generally agreed with our findings, 
and provided technical comments, which we incorporated as appropriate. 
The Department of the Treasury's formal comments are reproduced in 
appendix III. 

As agreed with your office, unless you publicly announce its contents 
earlier, we plan no further distribution until 30 days after the date 
of this letter. At that time, we will send copies of this report to the 
Secretary of Labor, the Secretary of the Treasury, appropriate 
congressional committees, and other interested parties. In addition, 
the report will be available at no charge on GAO's Web site at 
[hyperlink, http://www.gao.gov]. 

If you have any questions concerning this report, please contact 
Barbara Bovbjerg at (202) 512-7215. Contact points for our Office of 
Congressional Relations and Public Affairs may be found on the last 
page of this report. GAO staff who made major contributions to this 
report are listed in appendix II. 

Sincerely yours, 

Signed by: 

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

[End of section] 

Appendix I: Scope and Methodology: 

To determine what is known about the effect of automatic enrollment 
policies among the nation's defined contribution plans, as well as the 
extent of and prospects for such policies, we first reviewed reports 
examining the impact of automatic enrollment, default contribution 
rates, and default investment funds on participation rates and saving 
patterns. Table 4 shows the six reports presenting original research 
that we reviewed. 

Table 4: Six Studies We Reviewed on the Impact of Automatic Enrollment 
and Related Policies: 

Study: Measuring the Effectiveness of Automatic Enrollment; 
Authors (year of study): Vanguard Center for Retirement Research 
(2007). 

Study: For Better or For Worse; 
Authors (year of study): Choi et al. (2001). 

Study: Nationwide Savings Plan Automatic Enrollment Getting Associates 
PREPared for Retirement; 
Authors (year of study): Swanson and Farnen (2008). 

Study: The Power of Suggestion; 
Authors (year of study): Madrian and Shea (2001). 

Study: Building Futures Volume VIII; 
Authors (year of study): Fidelity Investments (2007). 

Study: The Importance of Default Options for Retirement Saving 
Outcomes: Evidence from the United States; 
Authors (year of study): Beshears et al. (2008). 

Source: GAO. 

[End of table] 

Officials at the Department of Labor, as well as other pension industry 
experts, verified our selections. The six reports include two conducted 
by large plan administrators that analyzed the records of their 
respective defined contribution plan sponsors and participants. The 
remaining four reports conducted case studies of companies that adopted 
automatic enrollment and analyzed participation rates, contribution 
rates, and investment fund allocations before and after the policy was 
implemented. For each study, we analyzed available evidence on: (1) the 
impact of automatic enrollment on participation rates and the 
durability of any increases in participation, (2) the characteristics 
of the workers whose participation rates are affected by automatic 
enrollment,(3) the impact of automatic enrollment on contribution 
rates, and (4) the impact of automatic enrollment on selection of 
investment funds. However, the findings of these case studies may not 
be generalizable for three reasons. First, each study examined the 
experience of only one to three companies. Second, many of the 
companies in the four case studies seem to have been facing difficulty 
meeting nondiscrimination testing requirements. Third, some offered 
matching contributions to their employees and it is unclear whether the 
presence of a match affects automatic enrollment participation rates. 
Therefore, the experiences of these companies may not be representative 
of all 401(k) sponsors. 

To determine the extent to which plans had adopted automatic enrollment 
policies, we obtained data from two large plan administrators--Fidelity 
Investment and Vanguard Investments. Data from Fidelity represents the 
18,100 qualified defined contribution plans Fidelity administers, 
encompassing about 14 million plan participants and over $600 billion 
in assets. Data from Vanguard is drawn from Vanguard's universe of 
defined contribution plans--more than 2,200 qualified plans that 
encompass more than 3 million participants and almost $200 billion in 
assets. We determined that these data accurately reflect the experience 
of Fidelity and Vanguard, but are not necessarily representative of the 
universe of defined contribution plans. 

We conducted in-depth interviews with 12 plan sponsors to obtain their 
perspectives on their experiences with automatic enrollment and related 
policies, as well as prospects for the policies. We used Form 5500 data 
from the Department of Labor to select a sample that included plan 
sponsors from a variety of industries including those that may be 
considered to have low wages and high turnover, and vice-versa, and 
small, medium, and large plans, as measured by number of participants. 
Six of the plan sponsors had adopted automatic enrollment, including 
one that had significantly narrowed the scope of its automatic 
enrollment policy, two that had adopted a 401(k) plan within the past 5 
years, and 10 that have sponsored a 401(k) plan for more than 5 years. 
The remaining six plan sponsors have not adopted automatic enrollment. 
Table 5 shows the industries and plan sponsor sizes for the 12 sponsors 
that we contacted. 

Table 5: Industries and Plan Sponsor Sizes: 

Industry: Professional and business services; Number of plan sponsors: 
1. 

Industry: Utilities; 
Number of plan sponsors: 1. 

Industry: Manufacturing; 
Number of plan sponsors: 2. 

Industry: Accommodation and food services; 
Number of plan sponsors: 1. 

Industry: Arts, entertainment, and recreation; 
Number of plan sponsors: 1. 

Industry: Transportation and warehousing; 
Number of plan sponsors: 1. 

Industry: Information; 
Number of plan sponsors: 1. 

Industry: Finance and insurance; 
Number of plan sponsors: 2. 

Industry: Retail trade; 
Number of plan sponsors: 2. 

Plan Size: Small (<150 employees); 
Number of plan sponsors: 3. 

Plan Size: Medium (150-4,999 employees); 
Number of plan sponsors: 3. 

Plan Size: Large (>5,000 employees); 
Number of plan sponsors: 6. 

Source: GAO analysis of Form 5500 data from the Department of Labor. 

[End of table] 

In addition, we conducted interviews with officials at the Departments 
of the Treasury and Labor, as well as academic experts from the 
Employee Benefits Research Institute (EBRI), Brookings, The Heritage 
Foundation, Harvard University, and the New School of Social Research. 
We also interviewed 401(k) plan administrators, providers, and 
consultants including Deloitte, Fidelity Investments, Vanguard, Mercer, 
Watson Wyatt, T. Rowe Price, ADP, State Street Global Advisors, and 
Renaissance Institutional Management. Finally, we interviewed industry 
and research organizations such as the Investment Company Institute 
(ICI), the Pension Rights Center (PRC), AARP, the Profit Sharing/401(k) 
Council of America (PSCA), the American Benefits Council (ABC), 
American Society for Pension Professionals and Actuaries (ASPPA), the 
Center for American Progress (CAP), the AFL-CIO, the Small Business 
Council of America (SBCA), and the Committee on Investment of Employee 
Benefit Assets (CIEBA). 

To determine the potential benefits and limitations of automatic IRA 
proposals and state-assisted retirement savings plan proposals, we 
analyzed the Automatic IRA Acts of 2006 and 2007 as well as state- 
assisted retirement savings proposals from four states. The Economic 
Opportunity Institute identified nine states that have introduced state-
assisted retirement savings proposals: California, Connecticut, 
Maryland, Massachusetts, Michigan, Pennsylvania, Rhode Island, Vermont, 
Virginia, and Washington.[Footnote 58] In addition, the architect of 
the state-assisted retirement savings concept identified Vermont as 
having introduced a proposal. We selected four--California, 
Connecticut, Maryland, and Washington--for an in-depth review because 
these states covered a range of plan type offerings and we were able to 
obtain feasibility studies or testimony prepared for state 
congressional hearings on their proposals. We did not conduct an 
independent legal review of these proposals. We analyzed the work of 
two researchers from Brookings and The Heritage Foundation that have 
developed proposals for the automatic IRA and state-assisted retirement 
savings plans. We then reviewed four studies sponsored by AARP 
examining the feasibility of automatic IRAs; survey reports by AARP and 
Prudential on employee and employer attitudes toward automatic IRAs; a 
microsimulation analysis of the impact of automatic IRAs on workers' 
savings accumulations and retirement security; and three feasibility 
studies on California, Maryland, and Washington states' proposals. In 
addition, we reviewed testimony and written materials from hearings 
held in Connecticut and Washington to obtain the perspectives of state 
officials, small business representatives, and pension industry 
representatives on state-assisted retirement savings proposals. We also 
reviewed relevant federal laws and regulations. 

We interviewed researchers who have focused on the topic, including 
those from Brookings, The Heritage Foundation, EBRI, the Economic 
Opportunity Institute, Harvard University, and the New School for 
Social Research as well as officials from AARP, PSCA, ASPPA, CIEBA, 
ABC, ICI, CAP, and PRC. In addition, we interviewed state officials 
from Washington, Maryland, and California as well as officials from 
pension plan administrators and consultants, including Mercer, Watson 
Wyatt, T. Rowe Price, ADP, State Street Global Advisors, and 
Renaissance Institutional Management. Finally, we interviewed a 401(k) 
consultant for small businesses and an official from SBCA to obtain the 
perspective of representatives of the small business community. 

[End of section] 

Appendix II: Comments from the Department of Labor: 

U.S. Department of Labor: 
Employee Benefits Security Administration: 
Washington, D.C. 20210: 

October 5. 2009: 

Ms. Barbara D. Bovbjerg: 
Director, Education, Workforce, and Income Security Issues: 
United States Government Accountability Office: 
Washington, DC 20548: 

Dear Ms. Bovbjerg: 

We have reviewed the Government Accountability Office's (GAO) draft 
report entitled "Retirement Savings: Automatic Enrollment Shows Promise 
for Some Workers, but Proposals to Broaden Retirement Savings for Other 
Workers Face Challenges" (GAO-10-31). 

GAO is to be commended for this thoughtful report. Automatic enrollment 
has proven to be an important policy and market tool for promoting 
retirement saving. Its positive impact in 401(k) plans has been well 
documented. Its promise for individual retirement accounts (IRAs) is 
considerable but not yet demonstrated. With respect to the latter, the 
Employee Benefits Security Administration (EBSA) would like to 
highlight two open questions that GAO's report considers but does not 
fully explore. 

The first such question relates to the potential impact of automatic 
IRA proposals on retirement savings of lower-paid workers. According to 
the report, "it is not clear that an automatic IRA would help low-
income workers" save. GAO cites a "preliminary analysis" from the 
Policy Simulation Group (PSG) which finds little positive impact. This 
analysis, which EBSA funded, illustrates the effect on future 
retirement income of automatic IRA participation rates similar to those 
posited by some of the proposals' advocates. EBSA is undertaking 
additional analysis that is intended to illustrate the effect of higher 
participation rates similar to those achieved in automatic 401(k) 
plans. 

The second such question relates to government oversight of an 
automatic IRA program. GAO's report notes that EBSA's "enforcement 
activities currently include pursuing 401(k) plan payroll deductions 
that are not timely remitted to employees' accounts. The government 
would have to undertake similar activities in instances where automatic 
IRA payroll deductions are not timely remitted." Given the large number 
of affected small employers, the large volume of small transactions, 
and the risk that workers saving by default might be inattentive, EBSA 
believes that the level of effort required to detect and correct such 
lapses, as well as to keep accounts linked to their owners as they move 
from job to job, could be large. 

EBSA is committed to protecting the employer-sponsored benefits of 
American workers. retirees, and their families. We appreciate having 
had the opportunity to review and comment on the draft report. Please 
do not hesitate to contact us if you have questions concerning this 
response or if we can be of further assistance. 

Sincerely, 

Signed by: 

Michael L. Davis: 
Deputy Assistant Secretary: 

[End of section] 

Appendix III: Comments from the Department of the Treasury: 

Department Of The Treasury:  

Washington, D.C. 20220: 

October 15, 2009: 

Ms. Barbara D. Bovbjerg: 
Director, Education, Workforce, and Income Security Issues: 
United States Government Accountability Office: 
441 G Street, NW: 
Washington, DC 20548: 

Dear Ms. Bovbjerg: 

We appreciate the excellent work that you and your colleagues at GAO do 
to help improve the performance and accountability of the federal 
government for the benefit of the American people. We particularly 
appreciate your work in the area of retirement security and retirement 
savings, and are writing to comment briefly on your draft report GAO-10-
31 to the Chairman, Special Committee on Aging, U.S. Senate, regarding 
the opportunities and challenges of expanding the benefits of 
retirement savings to additional American workers through automatic 
enrollment in 401(k) plans and automatic IRAs. 

GAO's report issued in November 2007 titled "Low Defined Contribution 
Plan Savings May Pose Challenges to Retirement Security, Especially for 
Many Low-Income Workers" decried low participation rates in defined 
contribution plans, particularly for low-income workers who have less 
opportunity to participate in such plans than the average worker. The 
automatic IRA proposal (first introduced as proposed legislation in 
2006) was one of only a few options discussed in the report that seek 
to increase retirement savings coverage and participation. 

The President proposed the automatic IRA in his FY 2010 budget to 
expand retirement savings and help address our nation's low personal 
saving rate. Since then, many stakeholders representing the interests 
of workers, employers, and service providers have expressed an interest 
in this important effort. We appreciate the enthusiasm shown by various 
stakeholders and the fact that they have suggested a number of ideas in 
discussions with us and the congressional sponsors of the legislation. 
Our and the congressional sponsors' outreach to stakeholders has helped 
inform the thinking on how best to structure an automatic IRA program 
in order to make this expansion of retirement savings as effective and 
efficient as possible. Your work will also be helpful to us as we work 
to flesh out the proposal with a view to maximizing participation in 
automatic IRAs; maximizing the formation of new 401(k) and other plans 
as employers "graduate" from automatic IRAs to sponsorship of 
retirement plans; and minimizing any costs to employers and employees. 

Thank you for the opportunity to review the draft of your report. We 
are optimistic that automatic IRAs will go far toward making the 
benefits of tax-favored retirement saving easily available to the tens 
of millions of working families who do not yet participate in those 
benefits. 

Yours sincerely, 

Signed by: 

J. Mark Iwry: 
Senior Advisor to the Secretary of the Treasury Deputy Assistant 
Secretary for Retirement and Health Policy: 

[End of section] 

Appendix IV: GAO Contact and Staff Acknowledgments: 

GAO Contact: 

Barbara D. Bovbjerg (202) 512-7215 or bovbjergb@gao.gov: 

Staff Acknowledgments: 

David Lehrer, Assistant Director, and Michael Hartnett, Analyst-in- 
Charge, managed this review. Jennifer Gregory also led portions of the 
research and made significant contributions to this report in all 
aspects of the work. 

Edward Nannenhorn and Jay Smale provided methodological assistance. 
Kate van Gelder provided assistance with report preparation. Roger 
Thomas provided legal assistance. Ashley McCall assisted in identifying 
relevant literature. Cheron Brooks developed the report's graphics. 
Charlene Johnson, Michaela Monaghan, and Bryan Rogowski verified our 
findings. 

[End of section] 

Selected Bibliography on Automatic Enrollment: 

Beshears, John, James J. Choi, David Laibson, and Brigitte C. Madrian. 
"The Importance of Default Options for Retirement Saving Outcomes: 
Evidence from the USA." Chapter 3 in Lessons from Pension Reform in the 
Americas, edited by Stephen J. Kay and Tapen Sinha. New York: Oxford 
University Press Inc., 2008. 

Copeland, Craig. Use of Target-Date Funds in 401(k) Plans, 2007. Issue 
Brief 327. Washington, D.C.: Employee Benefits Research Institute, 
March 2009. 

Fidelity Investments. Building Futures: Auto Solutions Data. Boston, 
Mass.: September 2008. 

--. Building Futures Volume VIII: A Report on Corporate Defined 
Contribution Plans. Boston, Mass.: 2007. 

Choi, James J., David Laibson, Brigitte C. Madrian, and Andrew Metrick. 
"For Better or For Worse: Default Effects and 401(k) Savings Behavior." 
Chapter 2 in Perspectives on the Economics of Aging, edited by David A. 
Wise. Chicago: University of Chicago Press, June 2004. 

--. "Saving for Retirement on the Path of Least Resistance." Chapter 11 
in Behavioral Public Finance, edited by Edward J. McCaffery and Joel 
Slemrod. New York, N.Y.: Russell Sage Foundation, 2006. 

Madrian, Brigitte C. and Dennis F. Shea. "The Power of Suggestion: 
Inertia in 401(k) Participation and Savings Behavior." The Quarterly 
Journal of Economics, vol. CXVI, no. 4 (November 2001): 1149-1187. 

Mitchell, Olivia S. and Stephen P. Utkus. "Lessons from Behavioral 
Finance for Retirement Plan Design." Chapter 1 in Pension Design and 
Structure New Lessons from Behavioral Finance, edited by Olivia S. 
Mitchell and Stephen P. Utkus. New York, N.Y.: Oxford University Press 
Inc., 2004. 

Nessmith, William E., Stephen P. Utkus, and Jean A. Young. Measuring 
the Effectiveness of Automatic Enrollment. Valley Forge, Pa.: Vanguard 
Center for Retirement Research, December 2007. 

Swanson, Mark D. and D. Bryan Farnen. "Nationwide Savings Plan 
Automatic Enrollment Getting Associates PREPared for Retirement." 
Benefits Quarterly, vol. 24, no. 3 (Third Quarter 2008): 13-19. 

[End of section] 

Footnotes: 

[1] See GAO, Private Pensions: Low Defined Contribution Plan Savings 
May Pose Challenges to Retirement Security, Especially for Many Low- 
Income Workers, [hyperlink, http://www.gao.gov/cgi-bin/getrpt?GAO-08-8] 
(Washington, D.C.: November 2007), p. 3. 

[2] The Current Population Survey is a monthly survey conducted by the 
U.S. Census Bureau among a nationally representative sample of 
approximately 100,000 households, primarily in order to estimate the 
rates of employment and unemployment. During March of each year, the 
survey includes supplemental questions about retirement plan 
participation and other financial matters. 

[3] Patrick Purcell, Retirement Plan Participation and Contributions: 
Trends from 1998 to 2006. (Washington, D.C.: Congressional Research 
Service, Jan. 30, 2009). This report is based on an analysis of the 
Survey of Income and Program Participation (SIPP), which is conducted 
by the U.S. Census Bureau and encompasses a nationally representative 
sample of the civilian, noninstitutionalized population of the United 
States. The SIPP is a longitudinal survey, which means that it measures 
changes pertaining to the same individuals and households over time. 

[4] Brady, Peter, and Stephen Sigrist, Who Gets Retirement Plans and 
Why, Investment Company Institute Perspective 14, no. 2 (Washington 
D.C.: September 2008). 

[5] In some other types of defined contribution plans, such as profit- 
sharing and money-purchase plans, all workers are typically included in 
the plan, and the employer makes contributions and often invests the 
money in the employees' account. 

[6] A plan must be considered "qualified" to obtain favorable tax 
treatment under federal law. One requirement for a qualified 401(k) 
plan is that participants must elect to have the employer provide an 
amount of salary to the employee in cash, or to defer the amount of 
salary and deposit the amount in the employee's plan account. 

[7] The 1998 IRS ruling--IRS Revenue Ruling 98-30--held that employer 
contributions made to a plan on an employee's behalf, in lieu of cash 
payment, are considered elective contributions, so long as the employee 
has an opportunity to receive cash instead, and has not affirmatively 
expressed a desire to do so. In a subsequent ruling--IRS Revenue Ruling 
2000-8--the IRS determined that contributions made on behalf of either 
a newly hired or current employee in lieu of cash compensation were 
valid elective contributions. 

[8] Pub. L. No. 109-280 (2006) §902 of PPA added Code sections 
401(k)(13), 401(m)(12) and 414(w). See final regulations of the 
Internal Revenue Service regarding "Automatic Contribution 
Arrangements" at 74 Fed. Reg. 8200 (Feb. 24, 2009). 

[9] These IRS actions include Revenue Ruling 2009-30, which 
demonstrates ways a 401(k) plan sponsor can include automatic 
contribution increases in its plan, and Notice 2009-65, which includes 
sample automatic enrollment plan language that a 401(k) plan sponsor 
can adopt with automatic IRS approval. 

[10] IRAs were established under the Employees Retirement Income 
Security Act of 1974. (Pub. L. 93-406, codified at 29 U.S.C. 1001 et 
seq.) IRAs are intended to provide individuals not covered by employer- 
sponsored retirement plans an opportunity to save for retirement in 
their own tax-deferred accounts, and to give retiring workers or those 
changing jobs a way to preserve assets in employer-sponsored plans by 
allowing them to transfer plan balances to an IRA. Over the past 30 
years, Congress has created several types of IRAs with different 
features for individuals and small businesses. 

[11] The two legislative proposals were the Automatic IRA Act of 2007, 
S. 1141 and H.R. 2167, 110th Cong., 1st sess; and the Automatic IRA Act 
of 2006, S. 3952 and H.R. 6210, 109th Cong., 2nd sess. 

[12] For example, see Wash. Second Substitute Senate Bill 5791, 61ST 
Legislature, 2009 Regular Session, and Md. Senate Bill 728, 2008 
Regular Session. 

[13] Four of the six analyses we reviewed were case studies that 
compared participation rates at companies before and after automatic 
enrollment was adopted, while two of the analyses were conducted by 
large plan administrators and analyzed the records of their respective 
defined contribution plan sponsors and participants. Because these 
analyses do not represent a random sample of 401(k) plan sponsors, 
their findings are not generalizable. For more information about the 
studies, see appendix I. 

[14] Brigitte C. Madrian and Dennis F. Shea, "The Power of Suggestion: 
Inertia in 401(k) Participation and Savings Behavior," The Quarterly 
Journal of Economics, vol. CXVI, no. 4: 1149-1187. 

[15] Olivia S. Mitchell and Stephen P. Utkus, "Lessons from Behavioral 
Finance for Retirement Plan Design." Chapter 1 in Pension Design and 
Structure: New Lessons from Behavioral Finance, eds. Olivia S. Mitchell 
and Stephen P. Utkus. (New York, N.Y.: Oxford University Press Inc., 
2004). 

[16] James J. Choi, David Laibson, Brigitte C. Madrian, and Andrew 
Metrick, "Saving for Retirement on the Path of Least Resistance." 
Chapter 11 in Behavioral Public Finance, eds. Edward J. McCaffery and 
Joel Slemrod. New York: Russell Sage Foundation, 2006. 

[17] Madrian and Shea. 

[18] Fidelity Investments, Building Futures Volume VIII: A Report on 
Corporate Defined Contribution Plans (Boston, Mass.: 2007). Fidelity 
compared the participation rate for automatic enrollment-eligible 
employees in plans with automatic enrollment with the participation 
rate for eligible employees in plans without automatic enrollment. The 
study analyzed the data of more than 10 million participants in nearly 
13,000 Fidelity-administered plans, representing $674 billion in 
assets. 

[19] Madrian and Shea. 

[20] James J. Choi, David Laibson, Brigitte C. Madrian, and Andrew 
Metrick, "For Better or For Worse: Default Effects and 401(k) Savings 
Behavior." Chapter 2 in Perspectives on the Economics of Aging, ed. 
David A. Wise. Chicago: University of Chicago Press, June 2004. 

John Beshears, James J. Choi, David Laibson, and Brigitte C. Madrian, 
"The Importance of Default Options for Retirement Saving Outcomes: 
Evidence from the USA." Chapter 3 in Lessons from Pension Reform in the 
Americas, eds. Stephen J. Kay and Tapen Sinha. (New York: Oxford 
University Press Inc., 2008) 

[21] William E. Nessmith, Stephen P. Utkus, and Jean A. Young, 
Measuring the Effectiveness of Automatic Enrollment (Valley Forge, Pa.: 
Vanguard Center for Retirement Research, December 2007). Vanguard 
reported predicted effects from a logistic regression modeling 
comparing 527 plans with voluntary enrollment and 48 plans that 
implemented automatic enrollment for new hires only. Plans introducing 
automatic enrollment during the analysis period have employees hired 
under both voluntary and automatic enrollment. As of December 31, 2007, 
Vanguard administered over 2,200 defined contribution plans with more 
than 3 million participants. A Vanguard official reported that total 
plan assets at the end of 2008 were approximately $194 billion. 

[22] It should be noted that rates of participation in employer- 
sponsored plans without automatic enrollment tend to increase with the 
length of employee tenure. For this reason, the impact of automatic 
enrollment on participation rates may be significantly less pronounced 
as length of tenure increases. 

[23] A money market fund is a type of mutual fund that is required by 
law to invest in low-risk securities. These funds have relatively low 
risks compared to other mutual funds. A stable-value fund typically 
invests in bonds and achieves rates of return 2 to 3 percentage points 
higher than money market funds. The fund manager guarantees the 
investors a certain rate of return. 

[24] Choi et al. and Madrian and Shea. While particular types of 
assets, such as money-market and stable-value funds, may result in 
lower asset accumulations for some participants, they must also 
consider market risk when assessing their asset allocations. Market 
risk is the possibility of an overall decline in a broad class of 
assets, such as stocks. In 2008, stock values fell across the board and 
even well-diversified portfolios could not protect participants' 
portfolios from depreciating. Most investment advisors recommend 
diversifying across asset classes as well as within a particular asset 
class. For example, bond prices have historically been less volatile 
than stock prices, and there have been long periods when returns on 
bonds and stocks have not been closely correlated. 

[25] For purposes of this data, Fidelity defines large plans as those 
with more than $500 million in assets, mid-sized plans as those with 
between $35 million and $500 million, and small plans as those with 
less than $35 million in assets. 

[26] Fidelity Investments, Building Futures: Auto Solutions Data 
(Boston, Mass.: September 2008). 

[27] To maintain their 401(k) plans' qualified status, employers must 
perform tests to ensure that their plans comply with federal tax 
nondiscrimination rules (26 U.S.C. §401). These rules seek to balance 
benefit accruals of highly paid participants with those of non-highly 
paid participants. 

[28] The plan sponsor may utilize a "safe harbor" to avoid otherwise 
required ADP and ACP antidiscrimination testing of these plans. 
Specifically, in exchange for utilizing the safe harbor, the employer 
is required to make certain contributions to plan participants and use 
an accelerated vesting schedule. 

[29] A recent GAO report discusses the impact of lump sum 
distributions--that is, cashouts of account balances at job separation 
that are not rolled over into another retirement plan--and other forms 
of 401(k) account "leakage" 401(k) Plans: Policy Changes Could Reduce 
the Long-term Effects of Leakage on Workers' Retirement Savings 
[hyperlink, http://www.gao.gov/cgi-bin/getrpt?GAO-09-715] (Washington, 
D.C.: Aug. 28, 2009). 

[30] While there is no consensus about what constitutes an adequate 
contribution rate, some pension experts have cited combined 
contribution rates--including employer and employee contributions--of 
between 12 and 20 percent of pay. 

[31] Interviews with plans sponsors and others revealed a number of 
reasons for the predominance of the 3 percent default rate. For 
example, a representative of a firm that consults with many retirement 
plans stated that plan sponsors do not want a high default contribution 
rate that would risk causing automatically enrolled participants to 
drop out of the plan. A large plan administrator noted that the Pension 
Protection Act's specification of a 3 percent minimum default rate for 
safe harbor plans was seen by many as implied advice to start at 3 
percent. 

[32] According to data from Fidelity, another 46 percent of automatic 
enrollment plans have adopted optional automatic escalation, which 
means that participants have the option of increasing their 
contribution rate automatically at periodic intervals, such as once per 
year. 

[33] According to a representative of a plan administrator, a number of 
factors could explain this difference. For example, one plan 
administrator may have actively promoted default automatic escalation.

[34] This shift in TDF asset allocation is commonly referred to as the 
fund's "glide path." Sometimes TDFs have significant equity components 
at the retirement date and even at the endpoint of the glide path. 

[35] Both money market and stable value funds were often used as 
default investments before the PPA because employers were concerned 
about legal liability if investments they had chosen declined in value 
as a result of market fluctuations. As a result, they were led to 
invest workers contributions in such low-risk, low-return default 
investments. 

[36] Target-date funds are also sometimes referred to as "lifecycle 
funds." The two are generally synonymous in that they both adjust the 
proportion of different asset classes over time so that the investor 
has lower risk as they approach retirement. 

[37] Craig Copeland, Use of Target-Date Funds in 401(k) Plans, 2007, 
Issue Brief 327 (Washington, D.C.: Employee Benefit Research Institute, 
March 2009). This analysis was based on the Employee Benefit Research 
Institute/Investment Company Institute database, which contains data on 
nearly 22 million participants in participant-directed retirement 
plans. Although this database does not include information on plans' 
use of automatic enrollment, the author of the study developed a proxy 
to identify participants who were considered to have been automatically 
enrolled. 

[38] The Department of Labor and the Securities and Exchange Commission 
held a joint public hearing in June 2009 on TDFs and are currently 
reviewing the testimony presented at the hearing and supplemental 
materials submitted for the public record. (Hearing on Target Date 
Funds and Similar Investment Options, 74 Fed. Register 24052.) 

[39] S. 1141 and H.R. 2167, 110th Cong., 1st sess; S. 3952 and H.R. 
6210, 109th Cong., 2nd sess; David C. John and J. Mark Iwry, Pursuing 
Universal Retirement Security Through Automatic Iras, Joint Written 
Statement before Subcommittee on Long-Term Growth and Debt Reduction 
Committee on Finance, United States Senate. Washington, D.C.: June 29, 
2006. The White House's FY 2010 Budget Proposal also called for an 
automatic workplace pension program. Further, in its August 2009 
proposals for financial regulatory reform, the Department of the 
Treasury called for the enactment of the automatic IRA (The Department 
of the Treasury, Financial Regulatory Reform: A New Foundation: 
Rebuilding Financial Supervision and Regulation (Washington, D.C.: Aug. 
11, 2009)). At this writing, no bill for an automatic IRA had been 
introduced before the 111th (current) Congress. 

[40] A traditional IRA allows individuals to defer taxes on investment 
earnings accumulated in these accounts until distribution at 
retirement. A Roth IRA, in contrast, allows employees to make after-tax 
contributions to an IRA and, after age 59½, take tax-free distributions 
of their investment earnings. Account owners can also make tax-free 
withdrawals of their contributions after age 59½, as long as they have 
held the account for 5 years. 

[41] We first reported on the potential benefits of automatic IRAs in 
our 2007 report Private Pensions: Low Defined Contribution Plan Savings 
May Pose Challenges to Retirement Security, Especially for Many Low- 
Income Workers, [hyperlink, http://www.gao.gov/cgi-bin/getrpt?GAO-08-8] 
(Washington, D.C.: Nov. 29, 2007). We projected DC plan savings 
assuming that all employees participate immediately, rather than 
waiting for eligibility or opting not to participate. We found that 
instant and universal participation not only raised average savings for 
the sample as a whole, but had a relatively strong impact on workers in 
the lowest income quartile. Although our projections were for DC plans, 
they suggested that automatic features intended to increase 
participation can have a strong impact on retirement savings levels. 

[42] See GAO, Individual Retirement Accounts: Government Actions Could 
Encourage More Employers to Offer IRAs to Employees, [hyperlink, 
http://www.gao.gov/cgi-bin/getrpt?GAO-08-590] (Washington, D.C.: June 
4, 2008). 

[43] Employers may establish payroll-deduction IRAs and are not subject 
to ERISA fiduciary requirements so long as they follow regulations set 
by the Department of Labor. Further, the Small Business Job Protection 
Act of 1996, Pub. L. No. 104-188, established SIMPLE IRAs specifically 
for the purpose of encouraging more employers to sponsor IRAs. SIMPLE 
IRAs were designed to encourage employers with 100 or fewer employees 
to assist workers with retirement savings. Under a SIMPLE IRA, an 
employer can both direct a portion of an employee's salary to an IRA 
and offer a matching contribution. 

[44] This analysis was conducted by the Policy Simulation Group at the 
request of the Department of Labor. The analysis considers the impact 
of an automatic IRA fully implemented in 2010 on the annual retirement 
benefits of individuals aged 70 in the year 2065. The analysis creates 
a hypothetical birth cohort and models the cohort's lives from birth to 
death, including all life events such as marriages, births, education 
and job decisions, pension coverage and behavior, and retirement. 

[45] The Department of the Treasury, Financial Regulatory Reform: A New 
Foundation: Rebuilding Financial Supervision and Regulation 
(Washington, D.C.: Aug. 11, 2009). 

[46] Mary M. Schmitt and Judy Xanthopoulos, Automatic IRAs: Are They 
Administratively Feasible, What Are the Costs to Employers and the 
Federal Government, and Will They Increase Retirement Savings?, 
Preliminary Report Prepared for AARP (Optimal Benefit Strategies, LLC, 
Mar. 8, 2007). 

[47] Most Small Employers Face Low Costs to Implement Automatic IRAs, a 
report prepared for AARP (Optimal Benefit Strategies, LLC, Aug. 19, 
2009). 

[48] Although the 2007 House and Senate bills for an automatic IRA 
included establishment of a governing board modeled on the Federal 
Thrift Investment Board--a so-called TSP II Board--the bill did not, as 
the AARP-sponsored analysis noted, clearly delineate the amount of 
reporting and record keeping that would be required. 

[49] For example, the Department of Labor's enforcement activities 
currently include pursuing 401(k) plan payroll deductions that are not 
remitted to employees' accounts in a timely manner. Department of Labor 
officials stated that the government would have to undertake similar 
activities in instances where automatic IRA payroll deductions are not 
timely remitted. 

[50] An employer with 100 or fewer employees may establish a SIMPLE 
plan. A SIMPLE plan is a simplified retirement plan for small employers 
that is not subject to some of the same requirements that the Internal 
Revenue Code imposes on qualified pension plans. 

[51] The 2009 maximum annual tax-favored contribution levels are as 
follows: $5,000 for a traditional or Roth IRA, $11,500 for a SIMPLE 
IRA, and $16,500 for a 401(k). 

[52] Schmitt and Xanthopoulos 2007. 

[53] The 10 states are California, Connecticut, Maryland, 
Massachusetts, Michigan, Pennsylvania, Rhode Island, Vermont, Virginia, 
and Washington. We focused our analysis on proposals in four states--
California, Connecticut, Maryland, and Washington--because these 
proposals cover a range of retirement plan options, among other 
criteria (for more on selection criteria, see app. I). 

[54] See Cal. Assembly Bill No. 125, 2009-10 Regular Session; Conn. 
General Assembly Raised Bill No. 971, January Session, 2009; Md. Senate 
Bill 728 and House Bill 1228, 2008 Regular Session; Wash. Second 
Substitute Senate Bill 5791 and House Bill 1893, 61ST Legislature, 2009 
Regular Session. 

[55] Michael Rubenstein, Maryland Voluntary Employee Accounts Program 
Study (Annapolis: January 2008). 

[56] CA Const. art. XVI, § 17. 

[57] WA Const. art. VIII, § 5 and art. XII, § 9. 

[58] The Economic Opportunity Institute is a non-profit public policy 
organization located in Washington state that focuses on retirement 
security and several other issues. 

GAO's Mission: 

The Government Accountability Office, the audit, evaluation and 
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 GAO's Web site [hyperlink, http://www.gao.gov]. Each 
weekday, GAO posts newly released reports, testimony, and 
correspondence on its Web site. To have GAO e-mail you a list of newly 
posted products every afternoon, go to [hyperlink, http://www.gao.gov] 
and select "E-mail Updates." 

Order by Phone: 

The price of each GAO publication reflects GAO’s actual cost of
production and distribution and depends on the number of pages in the
publication and whether the publication is printed in color or black and
white. Pricing and ordering information is posted on GAO’s Web site, 
[hyperlink, http://www.gao.gov/ordering.htm]. 

Place orders by calling (202) 512-6000, toll free (866) 801-7077, or
TDD (202) 512-2537. 

Orders may be paid for using American Express, Discover Card,
MasterCard, Visa, check, or money order. Call for additional 
information. 

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

Contact: 

Web site: [hyperlink, http://www.gao.gov/fraudnet/fraudnet.htm]: 
E-mail: fraudnet@gao.gov: 
Automated answering system: (800) 424-5454 or (202) 512-7470: 

Congressional Relations: 

Ralph Dawn, Managing Director, dawnr@gao.gov: 
(202) 512-4400: 
U.S. Government Accountability Office: 
441 G Street NW, Room 7125: 
Washington, D.C. 20548: 

Public Affairs: 

Chuck Young, Managing Director, youngc1@gao.gov: 
(202) 512-4800: 
U.S. Government Accountability Office: 
441 G Street NW, Room 7149: 
Washington, D.C. 20548: