The survey also considers the extent to which respondents are preparing for the longer-term financial needs that they will face in retirement as well as the current experiences of retirees. In general, the results demonstrate that many individuals are struggling to save for retirement and, even among those who are saving, individuals are uncertain about their ability to successfully manage their self-directed retirement savings. This uncertainty is manifested by over half of savers who have limited or no comfort in their ability to manage these funds. Additionally, among current retirees, there are notable differences in the retirement assets across demographic groups.

Saving for Retirement

Many respondents report that they lack retirement savings. When asked what types of retirement savings or pension they have, 28 percent of non-retired adults indicate that they currently have no retirement savings or pension whatsoever.

Among those who do have savings, the most commonly reported form of retirement savings is a defined contribution plan, such as a 401(k) or 403(b) plan. Half of all non-retirees have money in this type of retirement savings plan (figure 30). This is nearly twice the 25 percent of non-retirees who have a traditional defined benefit pension plan through an employer. Thirty-one percent of non-retirees have an individual retirement account (IRA), and 46 percent indicate that they have retirement savings outside of a formal retirement account.

Figure 30. Forms of retirement savings among non-retirees
Figure 30. Forms of retirement
savings among non-retirees
Accessible Version | Return to text

Note: Among respondents not currently retired. Respondents can select multiple answers.

IRA is individual retirement account.

Perhaps unsurprisingly, the pattern of saving for retirement varies by age, with the likelihood of having savings being lowest among younger respondents. Nearly half of those ages 18 to 29 report that they have no retirement savings or pension, whereas over three-fourths of non-retirees age 30 or older indicate that they have at least some savings.

However, not all individuals of a given age are equally likely to have retirement savings. In particular, whether respondents have retirement savings as they approach retirement is highly dependent on their employment and disability status (figure 31). While almost 90 percent of employed respondents age 45 or older have retirement savings, a much smaller 34 percent of those who are out of work due to a disability have retirement savings. As a result, even though individuals who are out of work due to a disability represent just 12 percent of non-retired individuals age 45 or older, because of their low rate of retirement savings, they represent just over one-third of all non-retirees without retirement savings over that age.

The reported frequency of having retirement savings also increases sharply with income. Ninety-six percent of respondents making at least $100,000 per year report having at least some retirement savings and 87 percent of those making between $40,000 and $100,000 per year have savings. But among respondents making less than $40,000 per year, 44 percent have any retirement savings. This divergence in retirement savings by income is similarly true among individuals within age cohorts (table 33).

Table 33. Non-retirees reporting any retirement savings (by age and family income)


Age Less than $40,000 $40,000-$100,000 Greater than $100,000
18-29 39.0 78.3 96.6
30-39 39.4 84.7 95.5
40-49 52.5 87.4 95.9
50-59 46.4 92.2 95.5
60+ 57.3 92.5 95.8
Overall 43.6 86.7 95.7

Note: Among respondents not currently retired.

The income and employment status of individuals are closely related, meaning that part of the lower rate of retirement savings among lower-income adults results from their lower employment rates. However, even among non-self-employed full-time workers, those in lower-income families are much less likely to have savings (64 percent) than are those in families with higher incomes between $40,000 and $100,000 (93 percent) or over $100,000 (98 percent). This suggests that a portion of the gap is also due to either differences in retirement benefits for the types of jobs held by individuals lower in the income distribution, or the fact that even these lower-income individuals who work full time lack the financial capacity or wherewithal to save and contribute to retirement accounts.

While many adults struggle to save for retirement, among those who have retirement savings, there is also evidence that some may not preserve these savings for retirement and will instead use the savings for other purposes. This is despite the fact that early withdrawals from some retirement accounts may incur a substantial tax penalty. Six percent of those with retirement savings report that they borrowed money from a retirement account during the year before the survey (table 34). Moreover, 6 percent of those with such accounts report that they cashed out (permanently withdrew) some or all of their retirement savings in the prior 12 months, and 1 percent indicate that they both borrowed money from and cashed out retirement accounts in that time. Overall, 13 percent of adults with retirement savings either borrowed from those savings, cashed out those savings, or did both in the prior year. Additionally, 3 percent of non-retirees without retirement savings say that they borrowed from and/or cashed out their retirement savings, reflecting that some individuals previously had savings but have depleted the funds in those accounts.47

Table 34. Borrowing and cashing out of retirement accounts (by presence of current retirement savings)


Response Currently has retirement savings Currently has no retirement savings
Borrowed 6.2 0.5
Cashed out 5.7 2.1
Both 0.6 0.4

Note: Among respondents not currently retired.

Self-Directed Retirement Savings

Recognizing that self-directed retirement savings are the dominant method of preparing for retirement, the survey seeks to better understand how comfortable individuals are at managing these retirement savings. Among those who have at least some self-directed retirement savings (including 401(k)s, IRAs, and savings outside retirement accounts), there are decidedly mixed levels of comfort. Forty-seven percent of these adults are mostly or very comfortable making investment decisions in these accounts. However, the remaining 53 percent of adults with self-directed retirement savings are either not comfortable or are only slightly comfortable making these decisions.

There is also evidence that the level of comfort is lower among men with less education and among women of all education levels (figure 32). Sixty-five percent of men with at least a bachelor's degree report that they are mostly or very comfortable making these investment decisions. In comparison, 45 percent of men with a high school education or less express this level of comfort, and only 38 percent of women with a bachelor's degree and 37 percent of women with a high school degree or less are at least mostly comfortable managing these accounts. Additionally, the patterns of investment comfort across education levels for men and women are noticeably disparate. Men with at least a bachelor's degree are significantly more likely to be very comfortable than those with a high school degree or less. However, increased educational attainment does not significantly improve the likelihood of women being very comfortable when managing self-directed investments. Both 9 percent of women with a high school degree or less, as well as those with at least a bachelor's degree, report that they are very comfortable managing these accounts.

The survey also seeks to understand the factors contributing to individuals not saving for their retirement through defined contribution plans. Among those who are employed for somebody else in their main job (thereby excluding those who are self-employed, contractors, or not working), 33 percent of workers do not have a 401(k), or similar, account.48 This includes 63 percent of individuals who are employed for someone else, but whose family income is less than $40,000, who do not have a 401(k).

Among the one-third of workers employed for someone else who do not have a 401(k) type account, their lack of an account may be due to their decision not to participate in a plan offered by their employer, their employer not offering a plan (or not offering one for which they are eligible), or a combination of a lack of interest in a plan and their employer not offering one. Based on responses to the survey, 38 percent of these workers are offered a 401(k) type plan by their employer that they choose not to participate in (figure 33). An additional 21 percent report that their employer does not offer a plan (or that they don't know if one was offered), but they either would not participate or don't know if they would participate even if offered by their employer. Hence, for 59 percent of workers, their lack of interest or capacity for saving in a 401(k) limits their participation, rather than their employer not providing a plan to invest in.

Figure 33. Reasons for not participating in a 401(k)
Figure 33. Reasons for not
participating in a 401(k)
Accessible Version | Return to text

Note: Among respondents employed for somebody else who do not have a 401(k) type account.

Retirement Decision and Experiences

Respondents who are currently retired are asked about their experiences in retirement and about how they manage their expenses. The most common age to retire is 62, with 20 percent of retirees who recall their retirement age saying that they retired at that age, followed by age 65, when 11 percent of those who recall their retirement age retired. Eighty-five percent of these current retirees who recall their retirement age report that they had retired at or before age 65.

When asked about why they retired at the age that they did, the most common reasons are wanting to spend more time with family, which 59 percent of retirees rate as somewhat or very important to their decision, or wanting to do other things, which 58 percent rate as somewhat or very important. Less frequent are poor health (30 percent rating as at least somewhat important), not liking the work (27 percent), or being forced to retire (24 percent).

The importance of health to one's retirement decision does seem to differ, however, based on the individual's level of education. While 18 percent of retirees with a bachelor's degree indicate that their health factored into their decision to retire when they did, this was at least somewhat important for 35 percent of retirees with a high school degree or less and 33 percent of retirees with some college or an associate degree.

Differences also exist in the factors for when to retire among retirees of different races and ethnicities. Forty percent of black retirees and half of Hispanic retirees indicate that poor health was at least somewhat important to their decision, which exceeds the 26 percent of white retirees for whom this contributed to their decision. White retirees, on the other hand, are more likely to indicate that they retired at least in part because they wanted to do other things (cited by 60 percent of white retirees and 59 percent of Hispanic retirees, compared to 45 percent of black retirees).

When it comes to sources of funds in retirement, 87 percent of those in retirement are drawing Social Security benefits (table 35). Sixty-three percent are drawing a traditional defined benefit pension, 52 percent draw on savings outside a retirement account, 42 percent use savings from an IRA, and 38 percent draw on a defined contribution plan.

There are also notable differences in the sources of retirement income among retirees based on their level of education. The likelihood of having private retirement savings increases with education. In aggregate, 75 percent of retirees with a high school degree or less are drawing from at least some private retirement savings (other than employment during retirement and relying on family), compared to 91 percent of those with at least a bachelor's degree. This can also be seen among the specific private retirement sources in table 35. For example, while 30 percent of retirees with a high school degree have a 401(k), over half of retirees with a bachelor's degree have one.

Table 35. Sources of funds in retirement (by education)


Source High school degree or less Some college Bachelor's degree or more Overall
Social Security 92.1 86.0 79.6 87.2
Defined benefit pension 54.7 64.4 74.9 62.6
Savings outside a retirement account 43.2 50.6 70.0 52.2
Individual retirement account (IRA) 31.7 44.9 59.2 42.5
401(k) or other defined benefit pension 29.8 38.6 53.6 38.4
Income from real estate or the sale of real estate 8.4 15.8 18.8 13.2
My spouse/partner has a job 19.4 25.6 28.0 23.5
I have a job 6.9 11.8 13.6 10.0
Income from a business or the sale of a business 2.5 6.8 6.7 4.8
Relying on children or other family 4.7 4.1 1.3 3.7
Other retirement savings 17.7 21.5 25.1 20.7

Note: Among respondents who are partly or fully retired. Respondents can select multiple answers.

In contrast, less-educated retirees are more likely to be drawing Social Security, which is driven completely by differences in claimant rates before age 70. Among those age 70 or older, over 96 percent of retirees with each level of education are claiming Social Security. But among those ages 62 to 70, 98 percent of retirees with a high school degree or less are claiming Social Security, relative to 82 percent of those with at least a bachelor's degree. This may reflect differences across education level in whether to defer claiming so that annual benefits continue to grow.

Similar differences in sources of retirement funds appear based on the race and ethnicity of the retiree. Black and Hispanic retirees are substantially less likely than whites to have 401(k) savings, IRA savings, or savings outside of a retirement account, and somewhat less likely to have a defined benefit pension or income from real estate (table 36). In aggregate, two-thirds of black retirees and 74 percent of Hispanic retirees have at least some private retirement savings, compared to 86 percent of white retirees with private retirement savings. In contrast, black and Hispanic retirees are somewhat more likely to be relying on other family members for support, although the incidence of relying on family support is relatively low irrespective of one's race and ethnicity.

Table 36. Sources of funds in retirement (by race/ethnicity)


Source White, non-Hispanic Black, non-Hispanic Hispanic
Social Security 88.6 84.1 83.0
Defined benefit pension 64.1 56.0 58.2
Savings outside a retirement account 58.5 21.7 40.2
Individual retirement account (IRA) 47.9 20.3 27.7
401(k) or other defined benefit pension 40.9 27.8 32.9
Income from real estate 14.1 8.7 10.9
My spouse/partner has a job 23.6 33.2 21.9
I have a job 10.0 11.7 12.2
Income from a business 4.5 2.4 9.3
Relying on children or other family 2.9 6.2 7.4
Other retirement savings 22.4 12.6 15.1

Note: Among respondents who are partly or fully retired. Respondents can select multiple answers.




 48. Some of these non-retired respondents may be cashing out from their retirement account to pay for retirement expenses as they near retirement. However, half of those who cashed out a retirement account, 58 percent of those who borrowed money, and 70 percent of those who did both are under age 45, suggesting that many are doing so for other purposes. Return to text

 49. Respondents are specifically asked about their participation in, and access to, a 401(k), 403(b), thrift, or other defined contribution plan from work. References in this section to 401(k) plans include both 401(k) plans as well as these other forms of defined contribution retirement accounts. Return to text

Back to Top
Last Update: June 14, 2017