ȱ
ȱȱȱ ȱȱȱ
ŘŖŖŝȱ
ȱȱ
ȱȱȱ¢ȱ
ȱŞǰȱŘŖŖşȱ
ȱȱȱ
ŝȬśŝŖŖȱ
ǯǯȱ
řŖşŘŘȱ
ȱȱȱ
Pr
epared for Members and Committees of Congress

ȱȱȱ ȱȱȱŘŖŖŝȱ
ȱ
¢ȱ
About half of all workers in the United States participate in an employer-sponsored retirement
plan of some kind, a proportion that has remained relatively stable over the past thirty years.
Beginning in the early 1980s, however, employers began to move away from traditional pension
plans – also known as defined benefit (DB) plans – to defined contribution (DC) plans, like those
authorized under section 401(k) of the Internal Revenue Code. Unlike DB plans, which are
required by federal law to offer a benefit in the form of a life annuity, DC plans are individual
accounts that typically pay the employee a lump sum at retirement. In 2007, approximately 21
million workers in the private sector participated in defined benefit plans, while more than 40
million workers participated in defined contribution plans.
One of the key distinctions between a defined benefit plan and a defined contribution plan is that
in a DB plan, it is the employer who bears the investment risk. The employer must ensure that the
pension plan has sufficient assets to pay the benefits promised to workers and their surviving
dependents. In a DC plan, the worker bears the risk of investment losses. The worker’s account
balance at retirement will depend on how much has been contributed to the plan over the years
and on the performance of the assets in which the plan is invested. Because DC plans and
Individual Retirement Accounts (IRAs) represent a large share of the assets available to
households to pay their expenses during retirement, Congress needs current, detailed information
on amounts that workers have accumulated in these plans to assess both workers’ preparedness
for retirement and the effectiveness of the tax incentives created for retirement savings plans.
Once every three years, the Board of Governors of the Federal Reserve System collects data on
household assets and liabilities through the Survey of Consumer Finances (SCF). The most recent
such survey was conducted in 2007, and the survey results were released to the public in February
2009. This CRS report presents data from the 2007 SCF with respect to household ownership of,
and balances in, retirement savings accounts.
Because the majority of assets held in retirement accounts are invested in stocks, trends in stock
prices have a significant impact on households’ retirement account balances. As a result of the
broad decline in stock prices in 2008, the retirement account balances that households reported on
the 2007 SCF may be greater than many of those households would report in 2009. The effect of
the current recession on household finances will be reflected in the next SCF, which will be
fielded in 2010. Nevertheless, the 2007 SCF provides the most comprehensive and current data
available on the amount and type of retirement assets owned by American households.
In 2007, 53% of U.S. households owned at least one retirement account, whether an individual
retirement account, a 401(k) plan, or other employment-based retirement account. The median
combined balance of all retirement accounts owned by households with at least one account was
$45,000. Twenty-five percent of households had total retirement account balances of $140,000 or
more, and 25% of households had total retirement account balances of $11,000 or less.
The median value in 2007 of all retirement accounts owned by households headed by persons
between the ages of 55 and 64 was $100,000. For a 65 year-old man retiring in April 2009,
$100,000 would be sufficient to purchase a level, single-life annuity that would pay income of
$700 per month for life, based on current interest rates. Because women have longer average life
expectancies than men of the same age, $100,000 would purchase a level, single-life annuity that
would pay income of $650 per month for life to a 65 year-old woman retiring in April 2009.
ȱȱȱ

ȱȱȱ ȱȱȱŘŖŖŝȱ
ȱ
ȱ
Trends in Retirement Plan Design................................................................................................... 1
The Survey of Consumer Finances.................................................................................................. 1
Participation in Employer-Sponsored Retirement Plans of All Kinds............................................. 2
Participation in Defined Contribution Plans.................................................................................... 3
Retirement Savings of American Households ................................................................................. 7
Retirement Account Ownership in 2007 ................................................................................... 7
Retirement Account Balances in 2007 ...................................................................................... 8
All households .................................................................................................................... 8
Households with One or More Workers under Age 65 ....................................................... 9
Retirement Account Balances by Age of Household Head..................................................... 10
Retirement Plan Contributions in 2007................................................................................... 12
Plan Loans............................................................................................................................... 13
Household Net Worth .................................................................................................................... 14
Conclusion..................................................................................................................................... 15
References ..................................................................................................................................... 16

ȱ
Table 1. Participation in Employer-Sponsored Retirement Plans of Any Kind............................... 3
Table 2. Participation in Defined Contribution Plans: 2001, 2004, 2007........................................ 6
Table 3. Household Retirement Account Ownership: 2001, 2004, and 2007.................................. 8
Table 4. Household Retirement Account Balances: 2001, 2004, and 2007................................... 10
Table 5. Household Retirement Account Balances by Age of Householder ..................................11
Table 6. Monthly Contributions to Defined Contribution Plans in 2007 ...................................... 13
Table 7. Median Household Net Worth in 2001, 2004, and 2007 ................................................. 14

ȱ
Author Contact Information .......................................................................................................... 16

ȱȱȱ

ȱȱȱ ȱȱȱŘŖŖŝȱ
ȱ
ȱȱȱȱȱ
Since the 1970s, the proportion of workers who participate in employer-sponsored retirement
plans has remained relatively stable at approximately half of the workforce. Since the early
1980s, however, employers have moved away from defined benefit (DB) plans to defined
contribution (DC) plans. Defined benefit plans – what most people think of as traditional
pensions – are required by federal law to offer plan participants a retirement benefit in the form of
a lifelong annuity. The amount of the annuity typically is based on the employee’s length of
service and average salary. In the private sector, DB plans usually are funded solely by employer
contributions and investment earnings on those contributions. Defined contribution plans, in
contrast, are more like savings accounts maintained by employers on behalf of each participating
employee. In the most common type of DC plan—those established under section 401(k) of the
tax code—the employee defers a portion of his or her salary, which is invested in stocks, bonds,
or other assets. The employer often matches some or all of the employee’s contribution to the
plan. At retirement, the balance in the account is the sum of past contributions plus interest,
dividends, and capital gains—or losses. The account balance is often distributed to the departing
employee as a single lump sum.
One of the key distinctions between a defined benefit plan and a defined contribution plan is that
in a DB plan, the employer bears the investment risk. The employer must ensure that the plan has
sufficient assets to pay the benefits promised to workers and their surviving dependents. In a DC
plan, the worker bears the risk of investment losses. The worker’s account balance depends on
how much has been contributed to the plan over the years and how the plan’s underlying
investments have performed. At year-end 2007, 78% of the $3.7 trillion in assets held by DC
plans was invested in stocks and stock mutual funds.1 The high percentage of DC plan assets
invested in equities makes these plans sensitive to trends in stock prices. Due to the sharp decline
in the major stock market indices during 2008, the total assets held by DC plans fell by almost
$1.1 trillion (28%) between December 31, 2007 and December 31, 2008. By year-end 2008, the
proportion of DC plan assets invested in stocks and stock mutual funds had fallen to 70%.
ȱ¢ȱȱȱȱ
This Congressional Research Service (CRS) report presents data on retirement savings account
ownership and retirement account balance collected through the Survey of Consumer Finances
(SCF) in 2001, 2004, and 2007. The SCF is a survey of households sponsored by the Board of
Governors of the Federal Reserve System in cooperation with the Department of the Treasury. It
is conducted once every three years to collect information on the assets and liabilities of U.S.
households, the sources and amounts of their income, their demographic characteristics,
employment, and participation in employer-sponsored health and retirement plans. Data from the
SCF are widely used by economists at the Federal Reserve, other government agencies, and by
private-sector research organizations and academic institutions to study trends in the amount and
distribution of assets and liabilities among U.S. households. Since 1992, SCF data have been
collected by the National Organization for Research at the University of Chicago (NORC). In

1 Board of Governors of the Federal Reserve System, Flow of Funds Accounts of the United States: Flows and
Outstandings, Fourth Quarter 2008, March 12, 2009, p. 113.
ȱȱȱ
ŗȱ

ȱȱȱ ȱȱȱŘŖŖŝȱ
ȱ
2001, members of 4,449 households were interviewed for the SCF. In 2004, members of 4,522
households were interviewed.2 For the 2007 SCF, members of 4,422 households were
interviewed. With the appropriate sample weights applied, the 2001, 2004, and 2007 SCF
interview samples were representative of all U.S. households.
Most of the information collected in the Survey of Consumer Finances—such as total assets and
liabilities—is reported at the household level. The only data that are reported separately for the
householder and his or her spouse or partner describe these individuals’ employment, pension
coverage, and demographic characteristics. In this report, retirement plan participation is shown
for employed household heads and employed spouses of household heads. 3 Contributions to
retirement accounts are shown for both employed household heads and employed spouses. Their
combined contributions are shown as the household’s total retirement plan contributions. Account
balances are shown only at the household level, i.e. as the sum of the account balances owned by
the members of the household. Net worth – the sum of household assets minus the sum of
household liabilities – is reported at the household level.
Data from a survey of employers – the Department of Labor’s National Compensation Survey
(NCS) – indicate that 61% of workers in the private sector worked for employers that sponsored
retirement plans in 2007, and that 51% of private-sector workers participated in employer-
sponsored retirement plans.4 The Department of Labor’s employer survey also indicates that 55%
of employees in the private sector worked for employers that sponsored defined contribution
plans, and that 43% of private-sector employees participated in DC plans in 2007. The rates of
DC plan sponsorship and participation reported by employers on the National Compensation
Survey are slightly higher than those reported by households on the SCF. Surveys of households
typically find lower rates of participation in employer-sponsored retirement plans than are found
in surveys of employers. In some cases, the person responding for the household is unsure what
kind of retirement plan he or she participates in. Under-reporting is also common with respect to
the respondent’s spouse. Although the SCF reports lower participation in employer-sponsored
DC plans than the NCS, it collects information that is not available from the employer survey,
such as workers’ demographic characteristics, their retirement account balances, and the amount
of their contributions to retirement accounts.
ȱȱ¢Ȭȱȱ
ȱȱȱ ȱ
Data from the Survey of Consumer Finances indicate that 62% of workers under age 65 were
employed at jobs that offered some form of employer-sponsored retirement plan in 2007. This

2 For more information, see http://www.federalreserve.gov/pubs/oss/oss2/scfindex.html.
3 This report refers to households rather than to families because the unit of analysis in the SCF is more comparable to
the Census Bureau’s definition of a household than to its definition of a family. In the survey, the household head is
designated as the male in a mixed-sex couple and the older person in a same-sex couple. The SCF staff note that this
designation is not intended to convey a judgment about how an individual family is structured. It is merely a means of
organizing the data consistently. For more information, see Bucks, Kennickell, Mach, and Moore, Federal Reserve
Bulletin, 2009 or refer to the survey documentation on the Federal Reserve Board’s web site at
http://www.federalreserve.gov/Pubs/OSS/oss2/2007/scf2007home.html.
4 U.S. Department of Labor, Bureau of Labor Statistics, National Compensation Survey: Employee Benefits in Private
Industry in the United States, March 2007, Summary 07-05, August 2007.
ȱȱȱ
Řȱ

ȱȱȱ ȱȱȱŘŖŖŝȱ
ȱ
figure includes both defined benefit plans and defined contribution plans. Forty-nine percent of
workers under age 65 participated in employer-sponsored retirement plans in 2007, little changed
from the 48% who participated in employer-sponsored plans in 2004 and the 50% who
participated in 2001.5 (See Table 1.). Workers at large firms were much more likely to have been
offered a retirement plan than those employed at small firms. In 2007, 77% of workers employed
at firms with 100 to 499 employees, and 88% of workers employed at firms with 500 or more
employees, worked for employers that offered a retirement plan of some kind. In contrast, just
15% of employees at firms with fewer than 20 employees worked for employers that offered a
retirement plan in 2007. This represents a substantial drop from 2001, when 22% of employees at
firms with fewer than 20 employees worked at firms that offered a retirement plan.
Full-time workers were much more likely than part-time workers to have been offered the
opportunity to participate in an employer-sponsored retirement plan. In 2007, 66% of full-time
workers were offered a retirement plan at work, compared to 36% of part-time workers. More
than half of full-time workers (53%) participated in employer-sponsored retirement plans in 2007,
compared to just 19% of part-time workers.
Table 1. Participation in Employer-Sponsored Retirement Plans of Any Kind
(Working Household Heads and Spouses Under Age 65, in percent)

Offered Any Type of Plan
Participated in Any Plan

2001 2004 2007 2001 2004 2007
Size of Firm

500 or more workers
88.5
86.5
87.5
69.8
69.3
70.2
100 to 499 workers
80.1
77.9
76.5
60.0
56.2
59.1
21 to 99 workers
56.0
57.8
56.6
41.9
43.0
42.5
Under 20 workers
22.4
19.5
15.4
16.9
15.1
10.9
Employment

Full-time
68.8 67.0 66.2 54.6 53.1 53.4
Part-time
38.1 38.2 36.4 21.2 23.1 19.4
Total
64.3 61.9 62.0 49.6 47.9 48.6
Source: CRS analysis of the Federal Reserve Board’s 2001, 2004, and 2007 Survey of Consumer Finances.
Notes: Data represent 109.9 million workers in 2001, 112.0 million workers in 2004, and 115.3 million workers
in 2007. Data include workers in both the public sector and the private sector.
ȱȱȱȱȱ
The data displayed in Table 1 show the percentage of workers participating in employer-
sponsored retirement plans of any kind, whether defined benefit, defined contribution, or both
types of plan. To participate in a defined contribution plan, an employee usually must to elect to
defer some of his or her salary into the plan. As a result, participation rates in DC plans depend on
both the percentage of employers who offer a plan and the percentage of employees who elect to
contribute to the plan.

5 The data presented in Table 1 reflect the plan participation of workers who were identified as either the household
head or the spouse or partner of the household head.
ȱȱȱ
řȱ

ȱȱȱ ȱȱȱŘŖŖŝȱ
ȱ
Table 2 shows the percentage of private-sector workers whose employer sponsored a defined
contribution plan, the percentage of workers who participated in DC plans, and the percentage of
workers who were offered a plan who participated in it. This last percentage – the “take-up rate”
– differs among workers with different economic and demographic characteristics.
The first three columns of Table 2 show the percentage of workers whose employer offered a DC
plan in 2001, 2004, and 2007. The middle three columns show the percentage of workers who
participated in DC plans in those years.6 The last three columns show the take-up rate among
employees who were offered the opportunity to participate in a DC plan. The percentage of
workers whose employer offered a defined contribution plan fell from 51% in 2001 to 47% in
2004 and then rose to 49% in 2007. The percentage of workers who participated in employer-
sponsored defined contribution plan also fell between 2001 and 2004 and rose from 2004 to 2007.
In 2001, 38% of private-sector wage and salary workers participated in employer-sponsored DC
plans. This proportion fell to 37% in 2004 and rose to 39% in 2007. The take-up rate rose
throughout the period. In 2001, 74% of workers who were offered a DC plan participated in the
plan. The take-up rate in DC plans rose to 78% in 2004 and to 79% in 2007.
The data presented in Table 2 show that workers under age 35, those who did not attend college,
those with income in the lowest quartile, and those who worked at small firms were relatively less
likely than other workers to have worked for an employer that sponsored a defined contribution
retirement plan. In 2007, 46% of workers under age 35 worked for an employer that sponsored a
defined contribution plan, compared to 49% of workers between the ages of 35 and 44, and 55%
of those aged 45 to 54. Thirty-nine percent of workers with a high school diploma or less
education worked for employers that sponsored defined contribution plans, compared to 51% of
workers with some college and 60% of workers with a college degree. Only 30% of workers
whose household was in the lowest income quartile worked for firms that sponsored defined
contribution plans in 2007, compared to 62% of workers in the top income quartile and 55% of
workers in the second-highest income quartile. Just 12% of workers employed at firms with fewer
than 20 employees worked for firms that sponsored defined contribution plans in 2007, compared
to 46% of workers at firms with 20 to 99 employees, 59% of workers employed at firms with 100
to 499 employees, and 70% of workers employed at firms with 500 or more employees.
Participation in defined contribution plans usually requires the employee to elect to participate in
the plan. Although some plans have adopted automatic enrollment for eligible employees, almost
two-thirds of DC plans continue to require employees to elect to participate in the plan. The
percentage of employees offered a plan who elect to participate is called the “take-up rate.” The
participation rate in DC plans depends on both the percentage of workers offered a plan and the
take-up rate among those whose employers sponsor a plan. Low participation rates can result
from either a low percentage of employers offering a plan, a low percentage of employees who
are offered a plan electing to participate, or both.
The distinction between low participation rates that result mainly from low take-up rates and
those that mainly result from relatively few workers being offered a plan can have important
implications for public policy. For example, only 32% of workers under the age of 35 participated
in defined contribution plans in 2007. Although this was due in part to the lower percentage of
these workers working for employers who sponsored plans, another important factor was the low

6 ERISA allows employers to exclude workers who have completed less than one year of service or who work fewer
than 1,000 hours during the year.
ȱȱȱ
Śȱ

ȱȱȱ ȱȱȱŘŖŖŝȱ
ȱ
take-up rate among younger workers who were offered a plan. Only 70% of workers under age 35
whose employers sponsored DC plans participated in those plans in 2007. In contrast, the take-up
rate among workers aged 35 to 44 was 82%, and the take-up rate among workers aged 45 to 54
was 83%.7 On the other hand, the low participation rate among employees of small firms was
attributable mainly to the much lower proportion of these workers who were employed at firms
that sponsored plans for their employees rather than to low take-up rates.
In 2007, only 9% of workers employed at firms with fewer than 20 employees participated in a
defined contribution plan, compared to 35% of workers at firms with 20 to 99 employees, 46% of
employees at firms with 100 to 499 employees, and 57% of employees at firms with 500 or more
employees. It is important to note, however, that only 12% of employees who worked at firms
with fewer than 20 employees worked for firms that offered DC plans to their employees. Among
workers employed at firms with 20 to 99 employees, 46% worked at firms that sponsored DC
plans, and among workers employed at firms with 100 to 499 employees, 59% worked for firms
that sponsored DC plans. Seventy percent of workers at firms with 500 or more employees were
employed by firms that sponsored a DC plan in 2007. The take-up rate in 2007 among employees
at firms with fewer than 20 employees (77%) was similar to the take-up rates among workers at
firms with 20 to 99 employees (75%) and only slightly lower than the take-up rates at firms with
100 to 499 employees (78%) and firms with 500 or more employees (81%). These results imply
that efforts to boost plan participation among younger workers should be targeted at raising take-
up rates (e.g., through automatic enrollment or more education for workers about the importance
of saving for retirement), while boosting participation among workers at small firms will require
policymakers to find ways to make offering a retirement plan less burdensome and costly to the
employer.

7 The take-up rate is the percentage of workers who participated divided by the percentage offered a plan. For workers
under age 35, the take-up rate in 2007 was .318/.456 = .697.
ȱȱȱ
śȱ

ȱȱȱ ȱȱȱŘŖŖŝȱ
ȱ
Table 2. Participation in Defined Contribution Plans: 2001, 2004, 2007
(Working Householders and Spouses Under Age 65, in percent)

Offered a DC Plan
Participated in a Plan
Take-up Rate

2001 2004 2007 2001 2004 2007 2001 2004 2007
Relationship

Householder
53.6 48.6 50.8 40.5 37.8 40.5 75.5 77.7 79.6
Spouse/partner 47.3 44.8 46.7 34.1 34.5 36.8 72.1 77.0 78.8
Age

Under
35
47.5 42.4 45.6 33.1 28.6 31.8 69.6 67.5 69.7
35
to
44
56.1 48.4 49.0 43.9 38.2 40.2 78.3 79.1 82.1
45
to
54
54.5 50.1 54.6 40.2 40.0 45.3 73.7 79.9 83.0
55
to
64
43.0 49.9 48.2 32.9 43.1 40.2 76.4 86.5 83.5
Race/ethnicity

Caucasian
52.4 50.0 51.4 39.4 39.7 41.6 75.1 79.5 80.9
Other
48.0 40.4 45.0 34.5 28.8 33.9 72.0 71.2 75.2
Sex

Male
51.3 49.5 50.4 42.1 38.7 40.9 82.2 78.1 81.1
Female
47.6 45.1 48.4 34.1 34.6 37.4 71.7 76.8 77.3
Marital status

Married
52.1 49.1 49.9 39.9 39.3 41.2 76.6 79.9 82.5
Not
married 50.0 44.0 48.5 35.0 31.9 35.4 70.0 72.4 72.9
Education

College
grad. 61.1 58.6 59.5 50.4 49.1 50.6 82.5 83.8 85.0
Some
college 50.5 46.3 50.5 36.2 35.1 39.3 71.7 75.7 77.9
High
school
less 43.9 37.8 39.4 29.3 26.5 28.6 66.9 70.2 72.7
HH Income

Top
quartile 61.4 60.9 61.5 50.4 53.0 53.9 82.2 87.1 87.5
Second
quartile 61.8 52.6 54.5 46.2 41.4 45.6 74.7 78.7 83.8
Third
quartile 46.2 43.5 46.3 31.8 31.1 33.6 68.7 71.4 72.5
Bottom
quartile 29.0 25.8 29.7 17.4 13.9 17.0 60.0 54.1 57.4
Firm size

500
or
more 72.3 67.7 70.4 52.5 52.8 57.0 72.7 78.0 80.9
100
to
499
62.8 56.1 58.7 45.8 41.4 45.8 72.9 73.7 78.0
20
to
99
44.1 43.6 46.3 34.3 34.3 34.9 77.7 78.8 75.4
Under
20
17.2 14.9 11.9 14.1 11.9 9.2 82.3 80.1 76.9
Employment

Full-time
56.5 52.6 53.9 42.4 40.7 43.3 75.0 77.3 80.3
Part-time
22.3 22.9 22.0 14.7 17.6 14.3 65.7 76.7 65.3
Union status

Union

67.2 58.4 62.3 46.1 45.4 49.5 68.6 77.6 79.5
Non-union
48.1 45.3 47.1 36.6 35.1 37.4 76.1 77.5 79.3
Total
51.4 47.3 49.4 38.2 36.7 39.2 74.4 77.5 79.3
Source: CRS analysis of the Federal Reserve Board’s 2001, 2004, and 2007 Survey of Consumer Finances.
Notes: Data represent 109.9 million workers in 2001, 112.0 million workers in 2004, and 115.3 million in 2007.
ȱȱȱ
Ŝȱ

ȱȱȱ ȱȱȱŘŖŖŝȱ
ȱ
ȱȱȱȱ ȱ
With the trend away from defined benefit plans to defined contribution plans, workers now bear
much of the responsibility of preparing for retirement. Workers who contribute to DC plans, such
as those authorized under sections 401(k), 403(b), and 457 of the Internal Revenue Code, or to an
IRA can accumulate retirement savings on a tax-deferred basis while they are working.8
The following tables show the retirement savings of all households and of households in which
there was at least one worker under age 65. According to the SCF, of the 116.1 million U.S.
households in 2007, there were 81 million households in which either the household head, the
household head’s spouse, or both, were employed adults under age 65.
ȱȱ ȱȱŘŖŖŝȱ
Data from the SCF show that the percentage of households that owned a retirement account of
some kind—whether an individual retirement account or a 401(k) plan or other DC plan—fell
from 2001 to 2004 and rose from 2004 to 2007. (See the top panel of Table 3.) Fifty-three percent
of all households owned at least one retirement account in 2007, compared to 51% in 2004 and
53% in 2001. Thirty-eight percent of households owned a defined contribution plan from current
or past employment in 2007, compared to 36% in 2004 and 35% in 2001. Thirty-one percent of
households owned an IRA or Keogh plan in 2007, compared to 29% in 2004 and 31% in 2001. In
2007, 15% of households owned both a defined contribution account from current or past
employment and an IRA or Keogh plan. This was up slightly from 14% in 2004 and 13% in 2001.
Sixty-three percent of households in which either the household head, the household head’s
spouse, or both, were workers under age 65 owned a retirement account of some kind in 2007.
(See the bottom panel Table 3.) This was higher than in 2004, but roughly the same percentage as
in 2001. Fifty percent of households with at least one worker under age 65 owned a defined
contribution plan from current or past employment in 2007, compared to 47% in both 2004 and
2001. Thirty-three percent of households with one or more workers under age 65 owned an IRA
or Keogh plan in 2007. This was three percentage points higher than in 2004 and the same
percentage as in 2001. Twenty percent of households with one or more workers under age 65
owned both a defined contribution account from current or past employment and also owned an
IRA or Keogh plan in 2007. This was up slightly from 18% in both 2004 and 2001.

8 In a traditional IRA, pre-tax contributions can be made only if the worker is not covered by an employer-sponsored
retirement plan or has income below amounts specified in law. Taxes are deferred on contributions and investment
earnings until money is withdrawn from the account. Roth IRAs accept only after-tax contributions; however, qualified
withdrawals from a Roth IRA are tax-free.
ȱȱȱ
ŝȱ

ȱȱȱ ȱȱȱŘŖŖŝȱ
ȱ
Table 3. Household Retirement Account Ownership: 2001, 2004, and 2007
All Households
2001
2004
2007
Number of households , in thousands
106,496 112,109 116,122
Percent with either an IRA/Keogh or a defined contribution plan
52.8
51.1
53.4
Percent with a defined contribution plana
34.7 35.8 37.7
Percent with an IRA or Keogh planb 31.3
29.0
30.6
Percent with both an IRA/Keogh plan and a defined
contribution plan
13.2
13.7
14.7
Percent with neither IRA/Keogh nor a defined contribution plan
47.2
48.9
46.4
Households with a worker under age 65
2001
2004
2007
Number of households, in thousands
75,693 79,622 80,997
Percent with either an IRA/Keogh or a defined contribution plan
62.5
59.4
63.2
Percent with a defined contribution plana
47.3 47.3 50.2
Percent with an IRA or Keogh planb 33.0
29.9
32.5
Percent with both an IRA/Keogh plan and a defined
contribution plan
17.8
17.8
19.5
Percent with neither IRA/Keogh nor a defined contribution plan
37.5
40.6
36.8
Source: CRS analysis of the Federal Reserve Board’s 2001, 2004, and 2007 Survey of Consumer Finances.
a. Includes DC accounts from current and past jobs. Household may also have owned an IRA or Keogh plan.
b. Household may also have owned a defined contribution plan.
ȱȱȱȱŘŖŖŝȱ
ȱȱ
As shown in Table 3, 53% of all households owned at least one retirement account in 2007.
Table 4 shows the balances in these accounts at he 50th percentile (the median), as well as at the
75th and 25th percentiles. In 2007, among all households that owned one or more retirement
accounts, the median balance of all accounts owned by those households was $45,000. This was
$5,430 more than the median balance of all household retirement accounts in 2004, and $9,940
higher than the median balance of household retirement accounts in 2001. (All amounts shown in
Table 4 have been adjusted to 2007 dollars.) Twenty-five percent of all households with at least
one retirement account in 2007 had retirement savings of $140,000 or more and 25% of
households with one or more accounts had retirement savings of $11,000 or less.
Among households that owned one or more defined contribution accounts from current or past
employment of the householder or the householder’s spouse, the median combined balance of all
DC accounts owned by the household in 2007 was $35,000. Twenty-five percent of households
with at least one defined contribution account had account balances of $110,000 or more in 2007,
and 25% of households with one or more DC accounts had balances of $8,900 or less.
Among households that owned one or more individual retirement accounts or Keogh accounts for
the self-employed, the median combined balance of all IRAs and Keogh accounts owned by those
households in 2007 was $34,000. Twenty-five percent of all households with at least one IRA or
ȱȱȱ
Şȱ

ȱȱȱ ȱȱȱŘŖŖŝȱ
ȱ
Keogh account in 2007 had IRA/Keogh balances of $100,000 or more and 25% of households
with one or more IRAs or Keogh accounts had balances of $10,000 or less.
ȱ ȱȱȱȱȱȱȱŜśȱ
The median value in 2007 of all retirement accounts owned by households with at least one
worker under age 65 was $45,000. Twenty-five percent of all households with one or more
workers had combined retirement account balances of $137,700 or more, and 25% of households
with at least one worker under 65 had total retirement account balances of $10,220 or less.
Among households with one or more workers under age 65 that owned one or more defined
contribution accounts from current or past employment of the householder or the householder’s
spouse, the median combined balance of all defined contribution accounts owned by the
household in 2007 was $32,500. Twenty-five percent of all worker households with at least one
defined contribution account in 2007 had DC plan balances of $106,000 or more and 25% of
households with one or more DC accounts had combined DC account balances of $8,700 or less.
Among the households with one or more workers under age 65 that owned one or more individual
retirement accounts or Keogh accounts for the self-employed, the median combined balance of all
IRAs and Keogh accounts owned by the household in 2007 was $30,000. Twenty-five percent of
all households with at least one IRA or Keogh account in 2007 had account balances of $89,500
or more, and 25% of households with one or more IRAs or Keogh accounts had account balances
of $9,600 or less.
ȱȱȱ
şȱ

ȱȱȱ ȱȱȱŘŖŖŝȱ
ȱ
Table 4. Household Retirement Account Balances: 2001, 2004, and 2007
(Amounts in 2007 dollars)
All households with retirement accounts
2001
2004
2007
Households with any type of retirement account

Combined balance of all retirement accounts:

75th percentile
$111,250 $131,900 $140,000
50th percentile (median)
35,060
39,570
45,000
25th percentile
8,180
10,990
11,000
Households with one or more defined contribution accountsa

Combined balance of all defined contribution accounts:

75th percentile
75,960
100,130
110,000
50th percentile (median)
23,370
29,680
35,000
25th percentile
5,840
8,240
8,900
Household with one or more IRA or Keogh accountsb

Combined balance of all IRA/Keogh accounts:

75th percentile
93,490
109,910
100,000
50th percentile (median)
31,550
32,970
34,000
25th percentile
9,000
9,000
10,000
Households with a worker under age 65
2001
2004
2007
Households with any type of retirement account

Combined balance of all household retirement accounts:

75th percentile
$99,330 $131,670 $137,700
50th percentile (median)
31,550
37,370
45,000
25th percentile
7,010
10,880
10,220
Households with one or more defined contribution accountsa

Combined balance of all defined contribution accounts:

75th percentile
72,450
98,920
106,000
50th percentile (median)
23,370
29,680
32,500
25th percentile
5,840
8,570
8,700
Households with one or more IRA or Keogh accountsb

Combined balance of all IRA/Keogh accounts:

75th percentile
75,960
94,520
89,500
50th percentile (median)
24,540
30,550
30,000
25th percentile
7,010
7,690
9,600
Source: CRS analysis of the Federal Reserve Board’s 2001, 2004, and 2007 Survey of Consumer Finances.
a. Includes defined contribution plans from current and past jobs. May also have owned an IRA or Keogh plan.
b. May also have owned a defined contribution plan.
ȱȱȱ¢ȱȱȱ ȱ ȱ
An individual’s age is an important consideration when evaluating the adequacy of his or her
retirement savings. The more time that a person has until reaching retirement, the greater the
opportunity to make additional contributions and for investment earnings to build up his or her
retirement account balance. Table 5 shows rates of retirement account ownership and average
retirement account balances, categorized by the age of the household head.
ȱȱȱ
ŗŖȱ

ȱȱȱ ȱȱȱŘŖŖŝȱ
ȱ
Between 2004 and 2007, the percentage of households that owned a retirement account of any
kind rose among households headed by persons aged 45 to 54 and among households headed by
persons aged 65 and older. It remained relatively stable among other age groups. In 2007, 43% of
households in which the householder was under age 35 owned one or more retirement accounts.
Among households in which the householder was 35 to 44 years old, 59% owned at least one
retirement account, as did 66% of households headed by persons aged 45 to 54 and 62% of
households headed by persons aged 55 to 64. Forty-one percent of households headed by persons
aged 65 and older owned a retirement account in 2007.
The median combined balance of all retirement accounts owned by households with at least one
such account was $45,000 in 2007. The median balance of household retirement accounts was
just $9,600 among households headed by persons under age 35. Median balances were
progressively higher for the next three age groups. Households headed by persons aged 35 to 44
had median retirement savings of $37,000 in 2007. Households headed by persons aged 45 to 54
had median retirement savings of $63,000, and households headed by persons aged 55 to 64 had
median retirement savings of $100,000 in 2007. Among households with at least one retirement
account that were headed by persons aged 65 and older, the median combined balance of all
household retirement accounts was $60,800 in 2007.
Table 5. Household Retirement Account Balances by Age of Householder
(Number of households in thousands; Account balances in 2007 dollars)
Households Percent Mean
Median
Age of
Number of
with
with
Value, all Value, all
householder Households Accounts Accounts Accounts Accounts
2001
Under 35
24,211
11,067
45.7%
$22,262
$ 8,180
35 to 44
23,751
14,837
62.5
76,573
33,890
45 to 54
21,941
14,063
64.1
152,920
56,090
55 to 64
14,107
8,395
59.5
230,540
64,270
65 or older
22,486
7,913
35.2
178,180
64,270
All households
106,496
56,275
52.8%
$122,227
$35,060
2004

Under 35
24,874
10,254
41.2%
$29,786
$12,090
35 to 44
23,115
13,550
58.6
78,008
30,775
45 to 54
23,279
13,892
59.7
163,656
65,946
55 to 64
17,086
10,938
64.0
253,048
93,424
65 or older
23,755
8,707
36.7
188,822
57,153
All households
112,109
57,340
51.1%
$140,350
$39,570

2007

Under 35
25,148
10,708
42.6%
$25,279
$ 9,600
35 to 44
22,745
13,306
58.5
81,308
37,000
45 to 54
24,120
15,968
66.2
156,124
63,000
55 to 64
19,564
12,199
62.4
271,920
100,000
65 or older
24,545
10,050
40.9
207,321
60,800
All households
116,122
62,231
53.6%
$148,579
$45,000
Source: CRS analysis of the Federal Reserve Board’s 2001, 2004, and 2007 Survey of Consumer Finances.
Note: Includes IRAs, Keoghs, and DC plan account balances from both current and past employment.
ȱȱȱ
ŗŗȱ

ȱȱȱ ȱȱȱŘŖŖŝȱ
ȱ
ȱȱȱȱŘŖŖŝȱ
In defined contribution plans, the benefit available to the worker is the amount in his or her
account. The account balance depends on the amount that the employer and employee have
contributed to the plan, the investment gains or losses on those contributions, and the fees
charged to participants. The maximum permissible annual contribution is limited by federal law,
but very few workers contribute amounts near the legal maximum.9 Table 6 shows workers’
monthly contributions to defined contribution plans in 2007.
The top panel of Table 6 shows the distribution of employee salary deferrals, employer
contributions, and total contributions from the employee and employer. Contributions are shown
separately for householders and for spouses that participated in DC plans, and also at the
household level. Household contributions are the combined contributions made by householders
and their spouses
The median monthly salary deferral by household heads in 2007 was $280. This is equivalent to
$3,360 on an annual basis.10 Twenty-five percent of household heads that contributed to a DC
plan in 2007 contributed $600 or more per month, and 25% contributed $130 or less. The median
monthly salary deferral by spouses of household heads in 2007 was $200. This is equivalent to
$2,400 on an annual basis. Twenty-five percent of spouses of household heads that contributed to
a DC plan in 2007 contributed $430 or more per month and 25% contributed $100 or less.
The median monthly contribution in 2007 by employers of household heads was $180 and the
median monthly contribution by employers of spouses was $130. The median monthly total
contribution — including both employee deferrals and employer contributions — was $420 for
working householders who participated in DC plans and $300 for working spouses.
Just as retirement account balances may be considered a financial resource of the entire
household, so too are contributions to retirement plans generally made for the future retirement
security of both the householder and his or her spouse. Table 6 shows the combined contributions
of the householder and spouse and their employers. The median monthly household contribution
in 2007 among households in which either the householder, the householder’s spouse or both
contributed to a DC plans was $290. Twenty-five percent of households contributed $656 per
month or more, and 25% of households contributed $130 per month or less. The median monthly
employer contribution in 2007 among households in which employers made contributions on
behalf of either the householder, the householder’s spouse, or both was $190. Twenty-five percent
of households received monthly employer contributions of $360 or more, and 25% of households
received employer contributions of $92 per month or less. The median monthly total contribution
to DC plans in 2007 from all sources — the householder, the householder’s spouse, and/or an
employer — was $440. In 25% of households, monthly contributions from all sources were $920
or more, and in 25% of households total monthly contributions were $208 or less.

9 The maximum annual employee salary deferral into a defined contribution plan is limited by I.R.C. §402(g). As
amended by P.L. 107-16, this limit was set at $11,000 in 2002 and increased in $1,000 dollar increments to $15,000 in
2006. Since 2007, the limit has been indexed to the Consumer Price Index in $500 increments. The limit in 2009 is
$16,500. I.R.C. §415(c) limits the total annual addition to defined contribution plans — comprising the sum of
employer and employee contributions. The §415(c) limit in 2009 is $49,000.
10 The maximum permissible employee salary deferral under I.R.C. §402(g) in 2007 was $15,000.
ȱȱȱ
ŗŘȱ

ȱȱȱ ȱȱȱŘŖŖŝȱ
ȱ
The bottom panel of Table 6 shows employee contributions, employer contributions, and total
contributions to DC plans in 2007 as a percentage of earnings. The median employee salary
deferral by household heads who contributed to a DC plan in 2007 was 6.0% of the household
head’s earnings. The median salary deferral by working spouses who contributed to a DC plan in
2007 was 6.0% of the spouse’s earnings. The median salary deferral by households in which
either the householder or the householder’s spouse contributed to a DC plan in 2007 was 5.1% of
total household earnings.
Table 6. Monthly Contributions to Defined Contribution Plans in 2007
(Monthly Contributions in dollars and as a percentage of earnings)
Relationship to
Employee
Employer
Total
Householder
Contributions
Contributions
Contributions
Householder

75th percentile
$600
$330
$845
50th percentile (median)
280
180
420
25th percentile
130
90
200
Spouse/partner

75th percentile $430
$242
$630
50th percentile (median)
200
130
300
25th percentile
100
60
150
Householder and
Spouse/partner

75th percentile $656
$360
$920
50th percentile (median)
290
190
440
25th percentile
130
92
208
Householder

75th percentile 9.9%
5.8%
13.0%
50th percentile (median)
6.0
3.8
9.4
25th percentile
3.9
2.7
6.0
Spouse/partner

75th percentile 10.0%
5.5%
1.9%
50th percentile (median)
6.0
3.5
8.9
25th percentile
3.5
2.7
5.3
Householder and
Spouse/partner

75th percentile 8.3%
5.0%
11.9%
50th percentile (median)
5.1
3.1
8.0
25th percentile
2.9
1.8
4.5
Source: CRS analysis of the Federal Reserve Board’s 2007 Survey of Consumer Finances.
Notes: Householder’s contributions are reported as a percentage of the householder’s earnings. Spouse’s
contributions are reported as a percentage of the spouse’s earnings. Household’s contributions are reported as a
percentage of the household’s earnings.
ȱȱ
Ten percent of households in which either the householder, the householder’s spouse, or both
participated in a defined contribution plan in 2007 reported that they had a loan currently
ȱȱȱ
ŗřȱ

ȱȱȱ ȱȱȱŘŖŖŝȱ
ȱ
outstanding against at least one plan. The mean outstanding balance of all household loans
against retirement plans was $6,672 and the median household loan balance was $5,000. Twenty-
five percent of all households with loans against retirement plans had outstanding balances of less
than $1,800, and 25% of households had outstanding loan balances of more than $9,000. Five
percent of households with loans against retirement plans had outstanding loan balances of more
than $20,000.
ȱȱȱ
Many households have wealth other than retirement accounts on which they will be able to draw
during retirement. More than 96% of workers in the United States are covered by Social Security,
and about one-fifth of workers in the private sector participate in defined-benefit pension plans.11
In addition, many households have other assets that could be used to pay expenses during
retirement. For example, the most valuable asset owned by many families is their home, and some
people may find when they are older that they prefer to live in a smaller house or apartment, or
they may choose to move to an area where property taxes and other living expenses are lower
than where they lived during their working years. In addition to equity in their homes, many
individuals have financial assets, equity in businesses, real estate, or other valuables that can
either provide a stream of income through interest, dividends, or rents, or that can be fully or
partially liquidated to finance their consumption needs during retirement. The broadest measure
of net household wealth—the difference between a household’s total assets and total liabilities—
is called “net worth.” The median net worth of all households in the United States in 2001, 2004,
and 2007 is shown in Table 7 in 2007 dollars.
Table 7. Median Household Net Worth in 2001, 2004, and 2007
(Amounts in 2007 dollars)
Age of
Householder
Median Net Worth
Percentage Change

2001
2004
2007 2001-2004 2004-2007
Under 35

$13,700
$15,600
$11,800
13.9%
-24.4%
35 to 44

90,700
76,200
86,600
-16.0
13.6
45 to 54

155,400
158,900
182,500
2.3
14.9
55 to 64

216,800
273,100
253,700
26.0
-7.1
65 to 74

207,900
208,800
329,400
0.4
57.8
75 or older

181,600
179,100
213,500
-1.4
19.2
All households

101,200
102,200
120,300
1.0
17.7
Source: Bucks, Kennickell, Mach, and Moore, Federal Reserve Bulletin, Feb. 2009.
Between 2001 and 2004, median net worth among all U.S. households (in 2007 dollars) rose by
just 1%, increasing from $101,200 to $102,200. From 2004 to 2007, median household net worth
rose by almost 18% to $120,300. Changes in median net worth between 2004 and 2007 differed
substantially by the age of the household head. Median net worth fell by 24% among households

11 All permanent federal employees and about four-fifths of employees of state and local governments participate in
defined benefit pension plans.
ȱȱȱ
ŗŚȱ

ȱȱȱ ȱȱȱŘŖŖŝȱ
ȱ
headed by persons under age 35, although the percentage change in net worth from 2004 to 2007
among households in this age group must be viewed in light of the small base ($15,600 in 2004)
against which it is measured. Among households headed by individuals between 35 and 44 years
old, median net worth rose by 14% between 2004 and 2007, increasing from $76,200 to $86,600.
The median net worth of households headed by persons between the ages of 45 and 54 rose by
15% between 2004 and 2007, while median net worth fell by 7% among households headed by
persons aged 55 to 64. The greatest increase in median net worth between 2004 and 2007
occurred among households headed by persons aged 65 to 74. Among these households, net
worth rose from $208,800 to $329,400, an increase of nearly 58%. Households headed by persons
aged 75 or older experienced an increase in median net worth of 18% between 2004 and 2007.
ȱ
Are Americans saving adequately – and wisely – for retirement? The median retirement account
balances reported on the 2007 Survey of Consumer Finances would not by themselves provide an
income in retirement that most people in the United States would find adequate. Among the 12.2
million households headed by persons aged 55 to 64 in 2007 that owned at least one retirement
account, the median balance of all their accounts was $100,000. This is less than twice the
median annual income of households in this age group. Moreover, almost 7.4 million households
headed by persons 55 to 64 years old had no retirement savings accounts in 2007. Including the
households with zero balances, a total of 13.5 million households headed by persons aged 55 to
64—69% of all households in this age group—had retirement account balances of less than
$100,000 in 2007. Among the 81 million households that included at least one worker under age
65 in 2007, 29.8 million—or 37%—did not own a retirement savings account of any kind. The
median retirement account balance among households with one or more workers under age 65
that owned a retirement account in 2007 was just $45,000.
For workers who are not included in a defined-benefit pension where they are employed—which
is about 80% of all workers in the private sector—saving from current income is essential to
preparing for retirement. Whether workers save by putting money into accounts that are
earmarked for retirement or by accumulating other assets on which they can draw after they have
retired is not necessarily important. The act of saving is of greater consequence to retirement
security than the manner in which it is accomplished. Nevertheless, the fact that 30 million
households that included a worker under age 65 had no retirement savings accounts in 2007
indicates that many households are not taking advantage of the tax deferrals available to most
workers through IRAs and to many workers through employer-sponsored retirement plans.
On the one hand, the widespread adoption of tax-favored retirement savings plans over the past
30 years indicates that many workers are taking seriously their responsibility to save for
retirement. On the other hand, the balances in these accounts—even among those who are near
retirement—are generally low. For example, if the median retirement account balance of
$100,000 among households headed by persons 55 to 64 years old in 2007 were converted to an
annuity, it would provide a monthly income of $700 per month ($8,400 annually) to a man
retiring at age 65 in 2009. This amount would replace just 15% of the median income of
households headed by individuals who were 55 to 64 years old in 2007.12 Another issue that many
households – particularly those nearing retirement age – must consider in the wake of the sharp

12 In 2007, the median income of households headed by persons aged 55 to 64 was $54,600.
ȱȱȱ
ŗśȱ

ȱȱȱ ȱȱȱŘŖŖŝȱ
ȱ
decline the major stock market indices that occurred in 2008 is whether their retirement savings
are adequately diversified. Although stocks have historically earned a higher average annual rate
of return on investment than bonds, the annual rates of return on stocks are more volatile than the
rates of return on bonds. A sharp decline in retirement account balances in the years just before –
or after – retirement could throw a household’s finances into disarray.
The uncertain future of Social Security and the declining prevalence of defined-benefit pensions
that provide a guaranteed lifelong income have put much of the responsibility for preparing for
retirement directly on workers. The low rate of personal saving in the United States and the lack
of any retirement savings accounts among millions of American households indicate that there is
a need for greater awareness among the public about the importance of setting aside funds to
prepare for life after they have stopped working. Most workers in the United States will need to
begin saving more of their income if they wish to maintain a standard of living in retirement
comparable to that which they enjoyed while working. The alternatives would be to work longer
or to greatly reduce their standard of living when they retire.
ȱ
Brian K. Bucks, Arthur B. Kennickell, Traci L. Mach, and Kevin B. Moore. “Changes in U.S.
Family Finances from 2004 to 2007: Evidence from the Survey of Consumer Finances.” The
Federal Reserve Bulletin, Feb. 2009. http://www.federalreserve.gov/pubs/bulletin/2009/pdf/
scf09.pdf

ȱȱȱ

Patrick Purcell

Specialist in Income Security
ppurcell@crs.loc.gov, 7-7571

ȱȱȱ
ŗŜȱ