The Role of Pensions in Retirement Income: Trends and ...

The Role of Pensions in Retirement Income: Trends and Questions

by Virginia P. Rena*

Pensions are an important and increasingly common supplement to Social Security benefits for persons aged 65 or older-particularly for those in the the middle and upper income quintiles. By 1990, pension income was reported by 44 percent of all elderly units-57 percent of the couples and 34 percent of unmarried persons.

This article discusses the role of pensions in the income of the elderly; private pension coverage, vesting, and types of plans among active workers; how the shift toward defined contribution plans poses new problems in assessing the role of private pensions in providing retirement income security; and expected pension receipt rates for the future elderly.

Pension receipt among the elderly is expected to continue to grow over the next 20-30 years because of past growth in coverage and vesting. Microsimulation models are a relatively new tool for forecasting the future distribution of pension income. The models offer a framework for considering the research questions that, if answered, would help improve our understanding of the impact of the pension system on future income security.

* Office of Policy and External Affairs, Social Security Administration. The author is currently on assignment with the National Academy of Social Insurance. Any views expressed in this article are those of the author and do not represent the views of the Social Security Administration or the National Academy of Social Insurance. This article was adapted from a paper prepared for the National Academy on Aging, Executive Seminar on Pensions in a Changing Economy, October27-28, 1992, Greenberg House, Washington, DC. Full conference proceedings,Pensions in a Chan@ng Economy, edited by Richard V. Burkhauser andDallas L. Salisbury, are published by the Employee Benefit Research Institute, Washington,DC, 1993.

Pensions from public or private plans are an important and increasingly common supplement to Social Security benefits for the elderly. By 1990,

pension income was reported by 44 percent of all elderly units-57

percent of couples and 34 percent of unmarried persons.

Social Security is the most evenly distributed source of income of the elderly, and it is the major share of

income of those in the lowest three income quintiles and is the largest single source of income for all but those persons in the highest income quintile. Among the elderly, receipt of pensions, assetincome, and earnings rises with income. The vast majority of pension, asset,and earnings income is received by those in the middle and upper income quintiles. Among the middle and upper income elderly, pensions are

more evenly distributed than are earnings or assetincome, according to data from the Current Population

Survey (CPS). Among active workers, the

proportion covered by a private pension plan grew rapidly in the 1940's and 1950's and more slowly in the 1960's and 1970's. The coverage rate declined slightly between 1979 and 1988. The decline in private plan coverage affected men more than women, thereby narrowing the gender gap in the private pension coverage rate. Among Mltime workers, the decline was greater

for young men (under age 35) and among both men and women with less than a high school education,

Participation in defined contribu-

tion plans is growing, both as a source of primary pension coverage and as supplemental coverage to a primary defined benefit or defined contribution plan. The expansion in private plan coverage has been in multiple plan

coverage-that is, workers with any private pension coverage are increasingly likely to be included in more than one plan. Often the supplements are 40 1(k) plans, which are more often offered to and used by higher-earning workers. The shift toward defined contribution private plans, and the lump-sum distributions they typically

Social Security Bulletin l Vol. 56, No. 1 l Spring 1993

29

pay, poseproblems in assessingIhe total role of private pensions in the retirement income s~stcm

Pcnsim rcccipt among the elderly- is espected to continue to grow over the nest 20 or 30 years becauseof past growth in coverage and vesting. Microsimulntion models are a relatively new tool for forecasting the distribution of pension income for the elderly. The models offer a framework for considering the research questions that. if answered. would help improve our understanding of the impact of the pension s! stem on future income security.

This section examines the role of pensions in the income of the elderly in 1990, rcvicws the changes in the role of pensions since the mid-l 970's, and examines the role of pensions by total income levels of the elderly in 1990. All the estimates are based 011 tabulations of the March Current Population Surveys that have been published biennially by the Oflice of Research and Statistics of the Social Security Administration.

Sources of Income, 1990

Of the lotal population of married couples and nonmarricd persons aged 65 or older in I990. 44 percent reported someincome from a public or private pension They include 30 percent with a private pension and 17 percent with a public pension, and 3 percent who received both public and private pensions (table 1). A couple is counted as rccciving a pension if either the husband or bj,ife has a pension. Nearly 6 in 10 (57 percent) elderly couples had pensions. while among the unmarried. 41 percent of the men and 32 percent of the women received a pension.

Social Security is the most commonly received source of retirement income. with about 9 in 10 elderly couples and unmarried persons receiving benefits. Public and private pensions differ somewhat in the roles they have filled relative lo Social Security. Becausenearly-all private sectorjobs have

been covered by Social Securit); for

Security, the median public pension was

several decades. nearly all private

$S.100. while the median for the smaller

pensions are designed to supplement

group who did not receive Social

Social Security be&its. In 1988. the

Security was $14,660, reflecting the

median annual private pension income likely casethat the pension is a substitute

received by elderly individuals was

for both Social Security and a pension.

$3.590. Because private pensions are

Public pensions that seme dual roles are

designed lo supplement Social Security. likely to be the recipient's major source

such pensions rarely are the major source of income. Of all elderly who received

of income for the elderly. In 1990, for government employee pensions in 1990.

example. private pensions accounted for the pensions were more than 50 percent

50 percent or more of total income for of total income for 36 percent of the

just 8 percent of the elderly units who

recipients.

received such pensions (or 2 percent of

Income from public and private

all the elderljz).

pensions combined accounted for about

Public pensions from Federal, State, 18 percent of the reported aggregate

or local government employment have income of the elderly in 1990 (chart 1).

traditionally had varied roles in relation Because public pension amounts are

to Social Security. Federal civilian

larger. they are a slightly larger share

employees were not covered by Social

of aggregate income than private

Security until Januaq 1, 1984, when

pensions-despite the fact that private

coverage was extended on a mandatory pensions are nearly twice as likely to be

basis to those hired after that date. Since received.

the 1950's. State and local government employees have been covered under

Trends in Income, 1976-90

Social Security at the option of the

The proportion of the elderly receiv-

governmental entity for which they work. ing pensions has increased steadily since

In recent years. about 7 in 10 State and the mid- 1970's. The proportion receiv-

local government employees have been ing private pensions grew from 20 to

covered by Social Security J Those not 30 percent, while the public pension

covered by Social Security generally

receipt rate grew from 13 to 17 percent

have pension plans that are meant to

(table 2).

substitute for both Social Security and a

Although the proportion of the

supplemental pension, while the others elderly receiving pensions has grown

have supplemental pensions akin to

steadily since the mid- 1970's. the share

private pensions. This mix of roles of of aggregate income the elderly receive

public employee pensions is reflected in from pensions has grown only modestly.

the pension amounts. In 1988, for

As the proportion of the elderly receiving

individuals who also received Social

public or private pensions grew from 3 1

30

Social Security Bulletin l Vol. 56, No. 1 l Spring 1993

to 44 percent, the share of aggregate income from public and private pensions combined grew from 16 to 18 percent (tables 2 and 3, respectively). The relative role of pensions in aggregate income is influenced not only by the rate of receipt and amount of pensions but also by the size of other income sources.

From 1976 to 1990, the largest changesin the composition of the income of the elderly were a decline in the share of earnings from work (from 23 to 18 percent) and an increase in the shareof income from assets(from 18 to 24 percent)-largely in the form of interest, dividends, and rental income. During this period, assetincome asa shareof the aggregate income of the elderly was highest in 1984, when interest rates were unusually high.

The growth in the share of income the elderly receive from assetsmay also reflect changes in the private pension system. Defined contribution plans accountfor a growing share of private pension payouts. These payments typically are paid as lump-sum distributions. If the retiree annuitizes the lumpsumpayment, it should be counted as "private pensions or annuities" in the CPS. If, however, the lump sum is held asan income-producing asset,it would becounted as assetincome. We have no information to quantify the extent to which lump-sum pensions may be reflected in asset income.

Pensions and Other Sources

by Income Quintiles, I990

When we divide the elderly into quintiles based on total money income asreported in the CPS, we find that the likelihood of receiving pensions rises with income-from 8 percent for those in the lowest income quintile, to 26 percent in the second quintile, to 50 percent in the middle quintile, and to 67 percent in the top two quintiles (table 4). The likelihood of receiving earnings from work or asset income also risessharply with income.

Social Security is received by the large majority of the elderly in all income levels; about 95-96 percent of thosein the middle three quintiles

Chart 1. -Shares of aggregate income from major sources, 1990

Couples and unmarried persons aged 65 or older

Asset income

Public pensions 9.4%

Other 3.1%

Earnings 17.7%

Social Security 36.4%

Source: Susan Grad, lit~omc of the Populatim 55 01 Older, 1990, Office of Research and Statistics, Social Security Administration, April 1992.

Table 2.-Percent of elderly units' receiving various sources of income.

1976-90

Source of income

1976 I

1980 I

19x4 I

1988 1

I990

Social Security

x9

90

91

92

92

Pensions, total z

31

34

38

42

44

Public

13

14

16

16

17

Private

20

22

24

29

.30

Earnings

2s

23

21

22

22

Asset income

56

66

68

68

69

I

I Couples and untnarrted person5 aged 65 or older. zIncludec some who receive both public and private pensions

Source: Susan Grad, Iwor~~r c!f I/V Pq~~lution 55 or O/dr,-. /YYO, and comparable publications by Grad in earlier years, Office of Research and Statistics, Social Security Administration.

receive it. Social Security receipt rates are lower among those in the bottom quintile (85 percent). Of those elderly persons in the lowest income group, some may not have qualified for Social Security benefits based on their own or a family member's earnings and therefore rely on public assistance.primarily Supplemental Security Income. The Social Security receipt rate is also below average in the top income quintile

(88 percent), which may include some high-earning persons aged 65-69 who are not receiving Social Security because they are not yet retired. At the age of 70, Social Security benefits are paid regardless of the beneficiary's earnings.

When we look at sharesof aggregate income of the elderly by total income quintiles, we seethat pensions are an important supplement to Social Security for the middle and upper income elderly.

SocialSecurityBulletin l Vol. 56, No. I l Spring 1993

31

Pensions account for 16 percent of aggregate income for those in the middle income quintile and about 20 percent of

the income of those in higher income quintiles (table 5). Social Security accounts for over 75 percent of aggregate income for those in the lowest two income quintiles, over 50 percent of total income for those in the middle income quintile, and is the largest single source of income for those in the fourth income quintile. In the highest income quintile,

income from assetsis the largest single source of income, earnings are the second largest source, followed by pensions.

The relative size of mean total income by income quintile and the respective roles of Social Security, pensions, earnings, and asset income are illustrated in chart 2.

Of the four main sourcesof income of the elderly, Social Security is the most

evenly distributed across income groups. This is to be expected becauseit is the most widely received source of income, the benefits are designed to replace a higher proportion of lifetime earnings for

low earners than for high earners, and it provides continuity in benefit income to elderly widows or widowers of deceased workers.

Pension income is more concentrated in the upper end of the income distribution than is Social Security, but it is more evenly distributed between the middle and upper income elderly than are reported assetincome or earnings.

Of pension income reported by the elderly in the CPS, more than 50 percent was received by those in the top income quintile, while nearly 40 percent was

received by those in the third and fourth income quintiles (table 6). In contrast, about 70 percent of assetincome and nearly 80 percent of earnings were received by those in the top income

quintile. While pension income is highly

concentrated in the top income quintile, it is less concentrated at that level than is assetincome or earnings. As will be

discussed later, the growth in private pension plans that pay lump-sum distributions make it increasingly difficult to distinguish pensions from

Table 3.-Percent of aggregate income of the elderly 1 from various sources, 1976-90

Source of income

Total

Social Security ......

Pensions

......

Earnings

......

Asset income

......

Other

......

1976

100 39 16 23 18

4

1980

100 39 16 19 22 4

1984

100 38 IS 16 28

3

1988

100 38 17 17 2s

3

1990

100 36 18 18 24

3

I Couples and unmarried persons aged 65 or older.

Source: Susan Grad, Income of the Po~pulcrrion 55 or Older. 1990. April 1992, and comparable reports by Grad in earlier years, Office of Research and Statistics, Social Security Administration.

Chart 2.-Composition of mean income by total income quintiles, 1990

Couples and unmnrried

personsaged 65 or older

Mean income

$60,00(

$50,00(

$53,25

F

$40,00(

$30,00(

$20,00( . . . . . . . . . . . . . . . . . $13,670

$lO,OO(

mother 0 Asset income. BE amings KII Pensions EIS ocial Security

$t

2nd

3rd

4th

5th

Income quintiles

32

Social Security Bulletin l Vol. 56, No. 1 l Spring 1993

asset income as reported by the elderly in the CPS.

Pension Coverage and Vesting for Workers

This section describes pension coverage of active workers in 1988, trends in coverage since 1940, recent trends by age. gender, and educational attainment, trends in vesting, and in the types of private plans covering workers today. Most of the data are from four special supplements to the CPS, which were conducted in April or May of 1972, 1979, 1983, and 1988.

Pension Coverage, I988

Estimates of the proportion of workers covered by pension plans vary depending on how coverage is defined and measured. This article uses measures of pension availability and coverage for wage and salary workers that are based on a series of questions in the May 1988 CPS pension supplement. First, workers were asked if their employer or union had a "pension or retirement plan" for any employees in their organization. Those who did not answer "yes" were asked if their employer offered a particular kind of retirement plan-a deferred profit-sharing plan or a stock plan. Later in the interview, it was determined whether or not the employer offered a 40 1(k) type plan that included employer contributions. A positive response to any of these questions is counted here as pension "availability."

Workers are counted as "covered" if they indicated that they were actually included or participating in any of these plans.' Self-employed workers are counted as covered if they reported that they contributed to a Keogh plan or an Individual Retirement Account (IRA) in the preceding year.

Of all workers, 44 percent met these criteria for being covered by a pension or retirement plan (table 7). The coverage rate for private wage and salary workers was 40 percent, while a significantly larger group (56 percent) said that their employer had a plan for at least some of its employees. Among government employees, the coverage rate was

Table 4.-Percent of the elderly 1 receiving various sources of income by total income quintiles2 1990

Source of income

Quintiles

Total

1St

2nd

3rd

4th

5th

Number (in millions)

Social Security

Pensions

._...,....

Earnings

Income from assets

Public assistance

23.1

4.6

4.6

4.6

4.6

4.6

92

85

96

95

95

88

44

R

26

50

61

67

22

5

9

19

29

46

69

31

56

75

X7

96

I

22

8

3

1

0

I Couples and unmarried persons aged 65 or older. `Quintile limits are $6,570. $10,752, $17,208, and $28,714

Source: Susm Grad, lr~ome of the Poprrlurion 55 or Older, 1990, Office of Research and Statistics, Social Security Administration, April 1992.

Table 5.-Percent of aggregate income of the elderly I from various sources, by total income quintiles.2 1990

Unit source of income

Number (in millions) Total percent Social Security Any pension Public pension\ Private pensions or annuities Earnings Asset income Public assistance Other .._......._.........._

Quintiles

Total

1st

2nd

3rd

4th

5th

23.1

4.6

4.6

4.6

4.6

4.6

100

100

100

100

100

100

36

79

76

59

41

18

18

3

8

16

23

20

9

2

4

8

11

11

9

2

4

8

11

10

18

1

3

7

12

21

2.5

4

9

1.5

21

33

8

11

3

1

(3)

0

2

2

3

3

3

2

I Couples and unmarried persons aged 65 or older. ZQuintile limits are $6,570, $10,752, $17.208, and $28,714 3Less than 0.5 percent.

Source: Susan Grad, Iwome of /he Pop~lution 55 01' Older, 1990, Office of Research and Statistics, Social Security Administration, April 1992.

Table 6.-Percent of aggregate income from major sources received by low, middle, and high income elderly,' 1990

Source

Social Security Pensions .._..._..,,_.... Asset income Earnings

I

Total

Income quintiles

from

"LOW"

"Middle"

source

(1st & 2nd)

(3rd & 4th)

100

27

41

100

4

39

100

4

27

100

1

20

"High" (5th)

26 57 70 78

1Coupleq and unmarried persons aged 65 or older Source: Author'5 calculations hased on data in chart 2 and tahle 5.

75 percent; and 21 percent of the selfemployed were contributing to either an IRA or Keogh plan.

Private pension coverage grew rapidly during the 1940's and 1950's

and more slowly in the 1960's and 1970's (chart 3). The proportion of private employees covered by a pension plan grew from about 12 percent in 1940 to about 37 percent in 1960.

Social Security Bulletin l Vol. 56, No. 1 l Spring 1993

33

 ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download