Skip to content

Older Workers' Pension Plan and IRA Coverage


Coverage by pension and retirement plans is especially important for older workers, defined here as those aged 55 and older. These workers are very close to retirement and will soon be counting on income from their pensions and savings to supplement their Social Security benefits.

Age 55 is also significant because it is a common age for early retirement under many employer-sponsored pension plans and a point at which many individuals begin leaving the labor force. Evidence indicates that eligibility to receive a pension benefit significantly affects the decision to leave the labor force. In addition, the type of pension coverage a worker has is important to this decision: workers with defined benefit (DB) plans are more likely to exit the labor force when they are eligible for benefits than are workers with defined contribution (DC) plans. Older workers are more likely to be covered by a DB plan than by a DC plan.

The Survey of Income and Program Participation (SIPP)

This paper examines the pension and retirement plan coverage of workers approaching retirement; it uses data from the "Retirement Expectations and Pension Coverage" topical module (Wave 7) of the 1996 Survey of Income and Program Participation (SIPP) released by the U.S. Census Bureau in February 2002 (the data collected are for the period from March to June 1998). This SIPP dataset includes the most recent and detailed data available on pension and retirement plans for a large representative sample of the U.S. population.

The SIPP data permit a broad-gauged examination of retirement plan coverage, i.e., they capture any retirement plan coverage obtained by workers during their careers. Most analyses of retirement plan coverage look at current primary job coverage only. Consequently, they produce estimates of aggregate pension coverage for older workers that are biased downward. Retirement plan coverage is defined here as (1) employer pension plan coverage obtained at any time during a worker's career (i.e., current primary job, current secondary job, and any previous jobs), or (2) having an Individual Retirement Account (IRA), or (3) both. Inclusion of IRAs acknowledges the blurring of the distinction between employer-provided DC pension benefits and personal retirement savings.

In 1998, 17.6 million or 13.3 percent of all workers who had a job or owned a business were older workers (Table 1). As expected, there is a dramatic drop-off in the working population after age 64. In 1998, there were only 4.4 million workers aged 65 and older. This means that by age 65, three-fourths of all older workers (i.e., 13.2 million) have left the labor force (see Table 1).

Retirement Plan Coverage of Older Workers

A total of 12.4 million, or 70.3 percent of all older workers had some type of retirement plan coverage in 1998 (Table 2). Of those covered, 7.1 million (39.3 percent) of all older workers had coverage from a current primary job, and 4.6 million of all older workers (25 percent) had coverage from previous jobs. Allowing for overlap, 9.8 million (55.6 percent) had some kind of career pension coverage. This rate was more than 15 percentage points higher than the standard measure (i.e., pension on current primary job) of pension coverage. Another 2.6 million workers without pension coverage participated in IRAs (overall, 37.4 percent of all older workers had IRAs, see Figure 1). This underscores the importance, especially for older age groups, of using a career-long measure of pension coverage as well as expanding the definition of coverage to include IRA participation (Figure 1)

(Click here to see Table 1: Number of Workers Age 16 and Older Who Had a Job or Owned a Business in 1998..)

(Click here to see Table 2: Retirement Plan Coverage of Workers Age 55 and Older by Socio-economic Characteristics in 1998 (in percentages).)

Of all older workers, almost 30 percent of men and 19 percent of women had pension coverage from previous jobs. Similarly, almost 26 percent of older white workers and 19 percent of their non-white counterparts had one from previous jobs (for details by other characteristics, see Table 2).

Older workers had the highest rate of IRA coverage among all age groups. This is expected since the bulk of IRA assets are rolled over from previous pensions.

Older female workers were less likely than their male counterparts to have retirement plan—66.7 percent compared to 73.4 percent. This difference is mainly due to higher pension coverage from previous jobs for men than women.

Older workers with longer job tenure, i.e., five years or more, were more likely to have retirement plan coverage than were shorter-tenured older workers. Older workers with less than five years' tenure on their current job were much less likely than their counterparts with five or more years' tenure to have any pension coverage. This is due to the minimum five-year vesting requirement for DB plans; the minimum vesting requirement for employer contributions in DC plans is three years.

Perhaps it is not surprising that older nonwhite workers were more likely than their white counterparts to have pension coverage on their current primary job. However, older nonwhites were much less likely than their white counterparts to have an IRA. These findings may be the result of older nonwhite concentration in unionized industries which tend to have DB plans and the fact that lump sum distributions are less available in DB plans than in DC plans. Such lump sums comprise the bulk of IRA assets.

Higher-income older workers were much more likely than their lower-income counterparts to have retirement plan coverage—either pension or IRA coverage. About 85 percent of workers with incomes of $40,000 to $59,999 had retirement plan coverage, compared to only 54 percent of workers with incomes less than $20,000, and 77 percent with incomes between $20,000 and $39,999. Workers with $80,000 to $99,999 in annual income had the most retirement plan coverage, the result of having the highest level of pension coverage on their current primary job. However, workers with annual incomes of $100,000 and over had the highest levels of pension coverage from previous jobs and during their careers as well as IRA participation.

Older workers with more education were more likely to have retirement plan coverage than their counterparts with less education. Highly educated older workers were especially likely to have high IRA participation. This is probably a result of both pension rollovers and these workers' higher propensity to save.

Older workers in the public sector were more likely to have pension coverage than their counterparts working for private employers. Federal employees had the highest level of pension coverage. Surprisingly, there was little difference by sector in IRA participation. Such participation by public sector workers was expected to be lower than that for private sector workers because of the high level of DB coverage in the public sector, which translates into the relative availability of lump sum distributions that can be rolled over into IRAs.

Pension/Retirement Plan Type

From the 1998 SIPP data it is possible to identify two general pension plan types, but only for a worker's current job: (1) DB plans—which base retirement benefits on a formula, and (2) DC or 401(k)-type plans—broadly defined as plans that allow workers to make voluntary tax-deferred contributions to an individual account, usually with some matching contribution by the employer.

Older workers were more likely to be covered by a DB only than by a DC plan only. In 1998, 19.4 percent had a DB plan only, while 13.6 percent had a DC plan only (Table 3). Only 6.3 percent of older workers had both types of plans. There was no difference by sex in type of single plan coverage. However, older male workers were somewhat more likely than their female counterparts to have dual coverage. Older nonwhites were much more likely than their white counterparts to have DB coverage only, while older whites were slightly more likely to have DC coverage only. Higher-income older workers were much more likely than their lower-income counterparts to have either single or dual coverage.


More than 70 percent of older workers have some type of retirement plan coverage. However, this means that a sizable proportion of the population rapidly approaching retirement lacks any type of pension coverage or an IRA in their own name and will have to rely solely on Social Security and possibly other savings for their retirement income security. Lower-income older workers are especially likely to have inadequate resources when they retire. These results argue for broadened retirement plan coverage early in careers to ensure income supplements to Social Security benefits and other savings. Ways to enhance coverage could include universal availability of pension payroll deductions, reducing administrative costs for starting pension plans, and expanding 401(k) automatic enrollments. Increasing coverage is a necessary but not a sufficient condition for attaining retirement income security. Ways of enhancing benefit amounts, e.g., reducing cash outs of lump sum distributions should also be examined.


  1. Joseph F. Quinn, "Retirement Patterns and Bridge Jobs in the 1990s," EBRI, Issue Brief, No. 206 (February 1999). A DB plan provides a specified benefit at retirement using a formula; a DC plan defines benefits as contributions to an individual account by employers, employees or both and investment returns.
  2. Satyendra K. Verma and Jules H. Lichtenstein, "Retirement Plan Coverage of Baby Boomers and Retired Workers: Analysis of 1998 SIPP Data," Issue Paper #2003-10, Public Policy Institute, AARP, (July 2003).
  3. This was a sizable group representing more than one-quarter (26 percent) of the working boomer population (aged 33 to 52) in 1998.
  4. Note that the sum of each coverage source in Table 2 does not equal the total for either "[pension coverage] during career" or "any covrage." This is because some workers have more than one type of coverage and the "[pension coverage] during career" and "any coverage" categories include each worker only once if that worker had coverage from any one source.
  5. Verma and Lichtenstein, 2003.
  6. Annika Sunden and Alicia Munnell, "Private Pensions: Coverage and Benefit Trends," prepared for "Conversation on Coverage," Washington, DC (July 24-25, 2001)

Written by Jules H. Lichtenstein and Satyendra Verma, AARP Public Policy Institute
October 2003
©2003 AARP
All rights are reserved and content may be reproduced, downloaded, disseminated, or transferred, for single use, or by nonprofit organizations for educational purposes, if correct attribution is made to AARP.
Public Policy Institute, AARP, 601 E Street, NW, Washington, DC 20049