← Browse

Ownership of Retirement Assets: Data in Brief

Published 2022-08-10 · Reports · R47213 · 2,613 words · ~13 min read
Elizabeth A. Myers
Retirement Income Adequacy
View on Congress.gov · PDF
Ownership of Retirement Assets: Data in Brief
August 10, 2022 (R47213)

Contents

Tables

Introduction

Retirement assets are one of several possible income sources in retirement. In 2019, about half of U.S. households had retirement assets, which included savings in defined contribution (DC) pension plans and Individual Retirement Accounts (IRAs).

In DC plans, workers are provided individual accounts funded by their own contributions, contributions from their employers, or both.1 DC plans do not provide guaranteed income. The funds in the account experience investment gains and losses, and the contributions and earnings, if any, are used as a source of income in retirement. Examples of DC plans include 401(k) plans, 403(b) plans, 457(b) plans.2

IRAs are tax-advantaged accounts for individuals (or married couples) to save for retirement, typically outside of the workplace.3 Though individuals with taxable (and certain nontaxable) compensation may contribute directly to IRAs, the majority of inflows to IRAs come from rollovers, which are transfers of savings from a retirement account, such as a 401(k) account, to another retirement account, such as an IRA.4 Most workers with DC plans roll over their savings to IRAs at job change or retirement.5

This report provides data on retirement asset ownership among U.S. households in 2019—specifically, household ownership and account balances of DC accounts, IRAs, and retirement assets (DC accounts plus IRAs)—to provide a snapshot of households' retirement savings based on different socioeconomic and demographic characteristics. As Congress considers various retirement-related bills, a greater understanding of how households save for retirement—either through DC plans, IRAs, or both—could help inform discussions.6

Data in this report is from the 2019 Survey of Consumer Finances (SCF). The SCF is a triennial survey conducted on behalf of the Board of Governors of the Federal Reserve and contains detailed information on U.S. household finances, such as the amount and types of assets owned, the amount and types of debt owed, and detailed demographic information on the reference person and, if applicable, his or her spouse.7 The SCF is designed to be nationally representative of the population of U.S. households, of which there were 128.6 million in 2019.8 Household in the SCF is defined as "the primary economic unit, which consists of an economically dominant single individual or couple (married or living as partners) in a household and all other individuals in the household who are financially interdependent with that individual or couple."

As previously mentioned, retirement assets are not the only source of income in retirement for many households. Most households receive Social Security payments throughout retirement. Social Security is a federal social insurance program that provides monthly benefits to insured retired and disabled workers and their families.9 Because Social Security has a progressive benefit structure, low earners generally receive higher replacement rates as measured by the amount of pre-retirement earnings replaced by monthly benefits.10

Another source of income for many households—particularly among older households and those in the public sector—is defined benefit (DB) plans, which are another type of employer-sponsored pension plan that typically provide workers with monthly benefits throughout retirement. The presence of Social Security and DB plans likely influences households' retirement saving decisions.

Household Ownership of Retirement Assets

Table 1 includes data on DC account ownership and balances, IRA ownership and balances, and retirement asset ownership and balances (DC savings and IRA savings together) among U.S. households in 2019.11 Ownership and balances refer to household ownership and account balances (i.e., a household was counted as owning a DC account if anyone in the household indicated having one; account balances were aggregated for all household members). Because the analysis in this report calculates medians and averages for households with assets, note that the actual median and average balance among all U.S. households would be lower.

Data in Table 1 demonstrate several points regarding household ownership of retirement assets in 2019:

  • Over half (50.5%) of households had retirement assets in 2019. Among households with retirement assets, the median balance was $65,000, and the average balance was $255,469.
  • Ownership rates and balances (among households with accounts) increased with household income, net worth, and education level.12 For example, 85.2% of households with household income of $125,000 or more had retirement assets, compared to 12.5% of households with income of less than $30,000.
  • Ownership and account balances varied based on the race or ethnicity of the household respondent.13 Households in which the respondent identified as Hispanic had the lowest rates of DC account and retirement asset ownership (27.2% and 31.7%, respectively) compared to households in other racial groups (which included those in which the respondent identified as White [non-Hispanic], Black/African-American [non-Hispanic], or another race [non-Hispanic]).14
  • Married households had, on average, more than double the account balance compared to single households. For example, the average retirement asset balance for married households was nearly $310,000, compared to about $140,000 for single households.

Table 1 illustrates the following regarding DC account ownership in 2019:

  • Among U.S. households, 37.5% owned a DC account. Among households with DC accounts, the median balance was $50,000, and the average balance was $173,985.
  • DC account ownership and balances peaked when the reference person was aged 45-54 then began to decline, likely due to households rolling over DC savings into IRAs. In addition, older households were more likely to have retirement income from a DB plan compared to relatively younger households (and so might not have needed to save for retirement to the same extent as younger households).15
  • Households in which respondents identified their race or ethnicity as "Other"—which included non-Hispanic respondents who indicated that they identified as Asian, American Indian/Alaska Native, Native Hawaiian/Pacific Islander, or another race not listed16—had the highest rates of DC account ownership (51.5% owned a DC account), though DC account-owning households in which the respondents identified as White, non-Hispanic, had the highest median and average balances ($55,000, and $196,772, respectively).

Table 1 illustrates the following regarding IRA ownership in 2019:

  • About one-quarter (25.4%) of households owned IRAs. Among IRA-owning households, the median balance was $70,000, and the average balance was $253,088.
  • The median and average IRA balances for all IRA-owning households exceeded those of DC account owners, despite IRA contribution limits being lower than DC plan contribution limits.17 One reason for this is likely due to rollovers from DC accounts as workers change jobs or retire. (The average age of the reference person in an IRA-owning household was 56.5 compared to 47.5 for the reference person in a DC account-owning household.) Another reason could be that IRA-owning households are, at the median, wealthier than non-IRA-owning households and are able to contribute more to both DC plans and IRAs. (In 2019, the median household wealth of IRA-owning households was $580,030, compared to $64,700 for non-IRA-owning households.)
  • Over half (54.2%) of households with household income of $125,000 or greater owned IRAs, compared to 6.7% of households with income of less than $30,000.

Table 1. Retirement Asset Ownership and Account Balances Among U.S. Households in 2019

 

Defined Contribution (DC) Accounts

Individual Retirement Accounts (IRAs)

Retirement Assets
 

Percentage of Households with DC Accounts

Median Account Balance (for Households with Accounts)

Average Account Balance (for Households with Accounts)

Percentage of Households with IRAs

Median Account Balance (for Households with Accounts)

Average Account Balance (for Households with Accounts)

Percentage of Households with Retirement Assets

Median Asset Balance (for Households with Assets)

Average Asset Balance (for Households with Assets)

All Households

37.5%

$50,000

$173,985

25.4%

$70,000

$253,088

50.5%

$65,000

$255,469

Age of the Household Reference Person:

Younger than 35

40.4%

$13,000

$26,988

12.1%

$8,000

$22,822

45.3%

$13,000

$30,167

35-44

48.3%

$51,000

$107,646

22.1%

$52,000

$97,960

55.8%

$60,000

$131,952

45-54

49.9%

$73,000

$197,633

28.0%

$64,000

$175,452

57.9%

$100,000

$254,720

55-64

42.5%

$99,000

$305,324

30.1%

$100,000

$316,073

54.5%

$134,000

$408,420

65 and older

17.1%

$96,000

$288,795

32.9%

$125,000

$386,662

43.7%

$125,000

$403,384

2018 Household Income (in 2019 Dollars):

Less than $30,000

6.7%

$6,000

$27,553

6.7%

$23,200

$95,306

12.5%

$13,000

$66,024

$30,000-$49,999

25.6%

$12,000

$27,851

14.6%

$35,000

$87,567

37.0%

$18,000

$53,821

$50,000-$74,999

42.5%

$22,000

$59,342

21.8%

$36,000

$108,606

56.3%

$32,000

$86,678

$75,000-$124,999

53.0%

$43,000

$93,208

32.3%

$52,000

$181,937

69.8%

$58,600

$154,903

$125,000 or more

66.6%

$157,000

$365,082

54.2%

$142,800

$405,698

85.2%

$243,000

$539,991

Household Net Worth Quintile

1st (up to $6,370)

18.9%

$5,000

$11,890

4.3%

$2,900

$6,883

21.2%

$4,900

$12,018

2nd ($6,371 to $67,650)

29.2%

$13,000

$20,111

6.5%

$5,000

$11,814

33.4%

$12,500

$19,906

3rd ($67,651 to $200,950)

41.7%

$31,000

$42,012

15.4%

$15,000

$28,237

49.3%

$34,000

$44,308

4th ($200,951 to $557,160)

45.3%

$77,000

$102,078

35.3%

$50,000

$73,746

64.5%

$85,000

$111,955

5th ($557,161 or greater)

52.6%

$300,000

$484,578

65.5%

$210,000

$442,930

84.1%

$404,000

$644,659

Household Marital Status:

Married

46.4%

$60,000

$206,401

32.3%

$85,000

$293,354

61.2%

$85,000

$309,953

Single

26.3%

$27,000

$101,016

16.6%

$47,000

$152,723

36.9%

$40,000

$140,135

Single female

25.4%

$24,000

$69,451

16.7%

$44,300

$135,975

36.4%

$36,000

$110,961

Single male

27.5%

$34,400

$143,795

16.3%

$50,000

$177,885

37.6%

$46,000

$181,616

Race or Ethnicity of the Household Respondent:a

White, non-Hispanic

40.5%

$55,000

$196,772

32.1%

$75,000

$272,638

56.9%

$80,000

$292,423

Black/African-American, non-Hispanic

28.7%

$25,000

$99,451

9.4%

$38,000

$95,639

34.9%

$31,000

$107,410

Hispanic (any race)

27.2%

$28,100

$97,166

9.8%

$23,000

$118,685

31.7%

$31,000

$119,702

Other race, non-Hispanicb

51.5%

$50,000

$170,317

27.3%

$100,000

$241,999

62.9%

$70,000

$244,188

Education Level of the Household Reference Person:

High school graduate or less

26.3%

$25,000

$88,387

12.7%

$40,000

$117,973

34.3%

$32,000

$111,436

Some college or associate's degree

33.4%

$30,000

$104,415

18.9%

$48,000

$137,542

44.4%

$42,000

$136,765

Bachelor's degree

51.6%

$57,000

$184,237

37.7%

$85,000

$292,436

67.4%

$87,300

$303,194

Advanced degree (master's, professional, doctorate)

51.9%

$119,000

$343,855

49.9%

$120,000

$373,130

75.9%

$170,000

$477,360

Source: CRS analysis of the 2019 Survey of Consumer Finances (SCF).

Notes: Retirement assets refer to both DC accounts and IRAs. Median and average account balances are calculated using the aggregated value of all accounts within an account category. (For example, the median DC balance refers to the median balance based on a household's aggregated DC accounts.) Median and average account balances refer to the balances of households with accounts in 2019. DC data includes Keogh accounts because the Internal Revenue Service refers to Keogh plans as qualified plans. Amounts are in 2019 dollars.

a. The SCF's question about race or ethnicity is asked only of the designated respondent. In 79% of sampled households, the designated respondent was the reference person.

b. "Other" includes respondents who indicated that they identified as Asian, American Indian/Alaska Native, Native Hawaiian/Pacific Islander, or another race not listed. The SCF noted that due to small sample sizes, the survey did not have the statistical power to further disaggregate the group of families in which the respondent identified as "other." The SCF allows respondents to indicate more than one race or ethnicity. CRS used the first racial identification reported to classify household respondents. In addition, any respondent that identified as Hispanic or Latino was classified as Hispanic, regardless of race.

Footnotes

1.

For more background on pension plans, see CRS Report R47119, Pensions and Individual Retirement Accounts (IRAs): An Overview.
2.

For more information on different types of DC plans, see CRS Report R47152, Private-Sector Defined Contribution Pension Plans: An Introduction.
3.

Congress has authorized two types of IRAs: traditional and Roth. Contributions to traditional IRAs may be deductible from taxable income (depending on household adjusted gross income and workplace pension coverage), and withdrawals are included in taxable income. Contributions to Roth IRAs are not tax-deductible, but qualified distributions (those made after age 59½, death, or disability from an account that is at least five years old) are tax free. An individual can move traditional IRA savings to a Roth IRA in what is referred to as a conversion. Congress has also authorized several types of IRA-based retirement plans that employers can offer, such as SIMPLE IRAs and SEP-IRAs.
4.

Examples of compensation include wages, salaries, tips, commissions, self-employment income, nontaxable combat pay, and alimony (which is treated as compensation for IRA purposes).
5.

Lump sum payments from defined benefit plans may also be rolled over to IRAs. In 2018 (the latest year for which data are available), about 97% of inflows to traditional IRAs were from rollovers, while about 52% of Roth IRA inflows came from rollovers and conversions.
6.

See, for example, H.R. 2954 and S. 4353 in the 117th Congress, and the Enhancing American Retirement Now section-by-section summary, at https://www.finance.senate.gov/imo/media/doc/EARN%20Act%20section%20by%20section%20summary1.pdf.

7.

The reference person is the single individual in a single-person household, the male in a mixed-sex couple household, and the older individual in a same-sex couple household. The SCF codebook states that no judgment about the internal organization of the household is implied by this organization of the data. In some cases, the respondent for a household was different than the reference person. Data referring to the household respondent rather than the reference person is noted throughout this report when applicable. In 79% of sampled households, the designated respondent was the reference person. More information about the SCF, including the data and codebook, is available at https://www.federalreserve.gov/econres/scfindex.htm. Because household wealth is highly concentrated, the SCF includes an oversample of relatively wealthy households.
8.

Estimates in this report are adjusted using population weights provided in the SCF dataset.
9.

For more information on Social Security, see CRS Report R42035, Social Security Primer.

10.

For more information on Social Security benefits, see CRS Report R46658, Social Security: Benefit Calculation.
11.

A Keogh plan is a retirement plan for self-employed individuals. The Internal Revenue Service (IRS) indicates that the term Keogh is seldom used because the law no longer distinguishes between corporate and other plan sponsors. CRS classified Keogh accounts as DC accounts in this report because IRS refers to Keogh plans as qualified plans.
12.

Education level refers to the highest education level completed by the reference person at the time of the survey.

13.

The SCF's question about race or ethnicity is asked only of the designated respondent.
14.

Hispanic household respondents may be of any race.

15.

Over the past five decades, private-sector employees have become less likely to be covered by DB pension plans and more likely to be covered by DC pension plans. See CRS In Focus IF12007, A Visual Depiction of the Shift from Defined Benefit (DB) to Defined Contribution (DC) Pension Plans in the Private Sector.
16.

The SCF combined these categories in the public dataset. The SCF noted that due to small sample sizes, the survey did not have the statistical power to further disaggregate this group of families. The SCF allows respondents to indicate more than one race or ethnicity. CRS used the first racial identification reported to classify household respondents. In addition, any respondent who identified as Hispanic or Latino was classified as Hispanic, regardless of race.

17.

In 2019, for workers under age 50, the IRA contribution limit was $6,000 compared to a $19,000 elective deferral limit for DC plans.