Determining How Current and Future Social Security Beneficiaries Make Retirement Decisions

Transcription

1 Determining How Current and Future Social Security Beneficiaries Make Retirement Decisions July 2011

2 Determining How Current and Future Social Security Beneficiaries Make Retirement Decisions Report Prepared By: S. Kathi Brown and Rebecca Perron Copyright 2011 AARP Research & Strategic Analysis 501 E Street, NW Washington, DC Reprinting with permission

3 AARP is a nonprofit, nonpartisan organization with a membership that helps people 50+ have independence, choice and control in ways that are beneficial and affordable to them and society as a whole. AARP does not endorse candidates for public office or make contributions to either political campaigns or candidates. We produce AARP The Magazine, the definitive voice for 50+ Americans and the world's largest-circulation magazine with over 35.1 million readers; AARP Bulletin, the go-to news source for AARP's millions of members and Americans 50+; AARP VIVA, the only bilingual U.S. publication dedicated exclusively to the 50+ Hispanic community; and our website, AARP.org. AARP Foundation is an affiliated charity that provides security, protection, and empowerment to older persons in need with support from thousands of volunteers, donors, and sponsors. We have staffed offices in all 50 states, the District of Columbia, Puerto Rico, and the U.S. Virgin Islands. Acknowledgements The authors would like to thank the following AARP staff for their valuable input: Jean Setzfand, Gary Koenig, Kelley Coates-Carter, Beau Ballinger, Jeff Love, Jennifer Leslie, and Sarah Conroy. All media inquiries about this report should be directed to AARP s Media Relations at (202) Inquiries from others should be directed to S. Kathi Brown at (202) or skbrown@aarp.org.

4 EXECUTIVE SUMMARY In 2009, with Social Security s 75 th birthday approximately one year away, AARP decided to conduct a survey to examine current beneficiaries and near-beneficiaries knowledge, experience with, and intentions regarding the program. The survey queried 3,045 Americans ages to examine how people make decisions regarding when to collect Social Security retirement benefits and to assess knowledge of how benefits are determined. This report focuses on the factors that influence the decision to collect benefits and includes related findings such as how people decide when to retire, expectations of working in retirement, attitudes about financial planning, and sources of information used to learn about Social Security. 1 KEY FINDINGS Key findings include: Pre-retirees average expected retirement age was 66 while retirees indicated an average actual retirement age of Approximately 10 percent of current retirees were working at the time of the survey and more than one in four (28%) said they had worked for pay at some point since retiring. The desire to retire while still healthy enough to enjoy retirement was the statement that resonated with the most pre-retirees when considering factors that may influence their retirement age. Many also indicated that how much they enjoy their work, how physically demanding their work is, and when they become eligible for full Social Security payments may influence their retirement age. Similarly, among retirees, the desire to retire while still healthy enough to enjoy retirement had been the most common influence on when they retired. Approximately one quarter (26%) of respondents who had not yet claimed Social Security retirement benefits ( non-claimers ) expected to claim at age 62, while 22 percent expected to collect benefits after age 62 but before age 66. Of all non-claimers, men, those with bachelor s or post-graduate degrees, those in excellent health, and those with less than $100,000 in savings expected to claim later. More than three quarters (78%) of respondents who had already claimed Social Security retirement benefits ( claimers ) indicated that they had claimed Social Security benefits at age 62 or younger. 3 Of these claimers, most (64%) indicated that they had started collecting benefits about when they expected they would. 1 A separate, complementary report related to the understanding and knowledge of the relationship between one s claiming age and the amount of one s Social Security retirement benefits is available. 2 Because the sample was limited to respondents ages 55 to 66, all retirees in the survey would have had to have retired prior to age 67. This most likely contributed to the relatively low average retirement age among the retired survey respondents. 3 The sample s focus on respondents ages 55 to 66 would also mean that all claimers in the survey would have had to have claimed Social Security retirement benefits prior to age P age

5 When considering issues that may influence their Social Security claiming age, nonclaimers most frequently pointed to the need for money. Specifically, more than three in four (78%) non-claimers strongly or somewhat agreed that their Social Security claiming age will be influenced by how much they need the money. Additionally, just over seven in ten non-claimers (73%) agreed that their claiming age will be influenced by when they decide to retire. Claimers also most commonly identified the need for money as having influenced their claiming age. For claimers, the second most cited influence was advice from others followed closely by the decision about when to retire. When asked how confident they were that they and their spouse will have enough money to live comfortably throughout their retirement years, about seven in ten respondents indicated that they were very or somewhat confident. Older respondents, those with higher educational attainment, and respondents with higher household incomes were more confident. Almost six in ten reported that they were either very or somewhat confident that the Social Security program will be able to provide the benefits that are owed to them. Older, more educated, and wealthier participants were more confident. In regard to financial literacy, over eight in ten appeared to have a basic understanding of the benefits of asset diversification, two thirds of respondents appeared to understand how compound interest works, and only half of the sample understood how inflation affects their buying power. When asked which organizations or sources they had consulted for information about Social Security benefits, by a large margin, respondents were most likely to have consulted the Social Security Administration (62%). Friends and family were consulted by almost half of the sample (47%), followed by a professional financial advisor (30%), financial magazines or books (27%), and AARP (25%). Each of these sources was then deemed very or somewhat useful by at least 90 percent of the respondents who had used the source. CONCLUSIONS Results from the survey suggest that the decision to claim Social Security benefits is strongly linked to the need for money. Additionally, not surprisingly, the decision to claim Social Security benefits and the decision to retire are connected, as many pre-retirees indicate that eligibility for full Social Security benefits will influence to some degree the age at which they retire, and many nonclaimers and claimers indicate that their retirement age will influence (or did influence) when they claim (or claimed) Social Security benefits. Furthermore, a key driver of the retirement decision is the desire to retire while still healthy enough to enjoy retirement. While the survey did reveal that other factors also influence the decision to claim Social Security benefits and the decision to retire, these top influences the need for money and the desire to enjoy retirement have the potential to cause individuals to claim benefits too early thereby exposing them to the risk of receiving monthly payments that will be inadequate in their later years. Notably, very few respondents reported having considered the impact of their own claiming age on their spouse s benefits. With education and advice that explains how their claiming age affects their monthly benefit amount as well as the monthly benefits available to their spouse, some 2 P age

6 individuals who may be inclined to claim prior to their full retirement age might decide to wait a year or more in order to ensure larger monthly payments to themselves and/or their loved ones. For example, among non-claimers in our survey who expected to claim benefits later than they did two years ago, the most common reason for the change in their expected claiming age was that they had learned something new about how benefits are calculated. Education and advice offered about when to claim Social Security benefits should be offered in a manner that helps future beneficiaries weigh the pros and cons of claiming benefits early to address any immediate financial needs versus postponing claiming in order to maximize (or receive larger) monthly benefits in the future when one is older. The survey suggests that both of these issues the immediate need for money and the risk of inadequate payments in later years are concerns that resonate with many respondents and may make the claiming decision challenging. For example, the survey revealed that the most vulnerable population those with low savings tend to be among the most likely groups to be swayed by messages that address both of these issues even when the messages seemingly contradict one another. For example, while non-claimers with low savings were among the most likely to agree that the age at which they claim benefits will be influenced by when they are eligible for the highest monthly benefit (which implies waiting until age 70), they were also among the most likely to agree that the age at which they claim benefits will be influenced by how much they need the money (which could presumably mean claiming early). INTRODUCTION In the summer of 2009, with the 75th anniversary of Social Security approximately one year away, AARP embarked on a two-part study to examine knowledge of and attitudes toward Social Security retirement benefits among current and future beneficiaries. We included individuals ages 55 to 66 who were already receiving Social Security retirement benefits, those who were not yet receiving them but were already eligible to receive them, and those who would be eligible to receive them within the next few years. While Social Security provides various types of benefits (including retirement benefits, disability benefits, spousal benefits, and survivor benefits for widows/widowers and other dependents), this study focused specifically on Social Security retirement benefits. The study, which consisted of focus groups followed by a survey, had the following key objectives: To assess current and future beneficiaries knowledge of how their Social Security retirement benefits are determined as well as how the age at which they claim benefits may affect benefits available to their spouse or widow To determine the factors that influence people s decisions regarding when to claim benefits To determine whether different methods of presenting the relationship between one s claiming age and the amount of the benefit may lead to different decisions regarding when to claim benefits The findings from the study will be published in two reports. This report includes the findings related to only the second objective determining the factors that influence people s decisions about when to claim benefits. A separate report presents the findings related to the first objective assessing knowledge of Social Security retirement benefits. 3 P age

7 This report presents overall findings from the entire sample and, for certain key questions, notes any significant differences by demographics (gender, education, age, household income, current health status, spouse s health status), whether Social Security benefits have already been claimed, financial literacy, sources of information used to learn about Social Security, and how likely respondents think they (or their spouse) are to live until ages 78 and METHODOLOGY This project began in September 2009 with focus groups, which were conducted with individuals ages in Baltimore, Maryland and Schaumburg, Illinois, a suburb of Chicago. Focus group participants were screened to recruit only individuals who either were already receiving Social Security retirement benefits or were likely to be eligible for Social Security retirement benefits in the future. All participants also said that they were healthy enough to work full time. The purpose of the focus groups was to explore how people decide when to claim Social Security benefits and to use these findings to design survey questions for the second part of the project. The second part of the project consisted of an online survey administered by Knowledge Networks of Menlo Park, California, to its national Knowledge Panel. Invitations to participate in the survey were sent only to individuals ages 55 to 66. In an effort to boost the number of respondents who had already claimed Social Security benefits, the sample for the survey was designed so that approximately half of the respondents would be ages 62 to 66. All prospective survey respondents were screened in an effort to accept only respondents who were either already receiving Social Security retirement benefits based on their own work history or would be eligible to receive Social Security retirement benefits based on their own work history in the future. Respondents were also screened to accept only those who reported that they were physically able to work on at least a part-time basis. The sample was also restricted to persons who had not participated in an earlier similar study conducted for the National Bureau of Economic Research. A total of 3,045 respondents ages 55 to 66 met these screening criteria and completed the survey from December 23, 2009 through January 18, Of these 3,045 respondents, 173 were obtained from a panel other than the Knowledge Panel in order to meet our oversample quota for respondents ages 62 to 66. Results were weighted by gender, age, race/ethnicity, education, employment status, internet access, census region, and metropolitan area to be nationally representative of adults ages 55 to 66 who were either already receiving Social Security retirement benefits or would be eligible to receive benefits in the future. Of the 3,045 respondents, 81 percent had not yet claimed Social Security retirement benefits and 19 percent had already claimed Social Security retirement benefits. Throughout this report, respondents who had already claimed benefits are referred to as claimers, while those who had not yet claimed benefits are referred to as non-claimers. 4 All differences presented in the report are statistically significant at p< P age

8 DETAILED FINDINGS THE RETIREMENT DECISION Because we believed that the decision to retire and the decision to collect Social Security retirement benefits may be related, we asked respondents a series of questions related to their retirement decision. Specifically, we were interested in learning the age at which they retired (or expected to retire), whether they worked (or expect to work) in retirement, and the issues that influenced (or will influence) their retirement age. Pre-Retirees Expected Retirement Approximately two in three (68%) respondents were not yet retired ( pre-retirees ). When asked to estimate the age at which they expect to retire, respondents reported an average expected retirement age of 66. While more than four in ten (42%) expected to retire at age 65 or earlier, just as many (42%) expected to retire at age 66 or later, and another 16 percent reported that they never expect to retire. (See Table 1) Table 1: At What Do You Expect to Retire? Base: Respondents who have never been retired regardless of current employment status (n=1,566) Younger than 62 5% 62 12% % % % Never expect to retire 16% Average 65.8 Source: Question R4. Differences in Expected Retirement by Demographics Expected retirement ages varied by gender, current age, educational attainment, household income, and total savings, as well as self-assessment of current health status and expectations of living to age 85. For example, on average, men expected to retire later than women (age 66.1 vs. 65.6). Additionally, respondents ages expected to retire later than those ages (age 67.2 vs. 65.6), which is not surprising since the respondents ages were by definition already older than respondents ages The relationship between educational attainment and expected retirement age worked in an opposite direction from the relationship between household income and expected retirement age and that between savings and expected retirement age. Specifically, those with low educational attainment (high school degrees or less) expected to retire at an average age of 65.2, whereas respondents with some college education or more expected to retire later - at an average age of In contrast, relatively low household incomes and relatively low savings levels were associated with higher average expected retirement ages. For example, respondents with household incomes of less than $60,000 expected to retire an average of one year later than those with household incomes of $100,000 or more (age 66.2 vs. age 65.3), 5 P age

9 a difference that most likely reflects a greater sense of financial security among those with the highest income levels. Similarly, respondents with total savings of less than $100,000 also expected to retire an average of one year later than those with savings of $100,000 or more (age 66.3 vs. 65.2). Differences in Expected Retirement by Current Health Status and Expected Longevity Additionally, respondents who described their health status in relatively favorable terms as well as those who thought that they had a relatively good chance of living to age 85 expected to retire later than other respondents. Specifically, respondents who described their current health as excellent were more likely than those who described their health as good, fair, or poor to expect to retire at 70 or later (21% of those in excellent health vs. 15% of those in good health and 12% of those in fair or poor health). Respondents in excellent health were also more likely than their counterparts in fair or poor health to expect to never retire (20% vs. 10%). Moreover, those who thought that they had at least a 75-percent chance of living to age 85 expected to retire just over one year later on average than those who thought that they had less than a 50-percent chance of living to age 85 (expected retirement age of 66.5 vs. 65.3). When those who never expect to retire were asked for their reasons, the majority (63%) cited the need for money while just one in three (34%) cited the enjoyment derived from working. Respondents with higher education and income levels were more likely than their less-educated and lower-income counterparts to cite enjoyment. Specifically, among those who never expect to retire, more than six in ten (61%) of those with post-graduate degrees attributed this to the enjoyment derived from working compared to just 31% of those with high school degrees or less and 20% of those with only some college education. Similarly, nearly six in ten (58%) of those with household incomes of $100,000 or more cited enjoyment from working compared to just 29% of those with household incomes of less than $30,000 and 21% of those with household incomes of $30,000 - $60,000. Of all respondents who estimated an age at which they expect to retire, the majority (62%) said that they expect to work for pay in retirement. Expectations of working for pay in retirement rose with educational attainment and were particularly widespread among those with the lowest household incomes and the lowest savings. Specifically, respondents with bachelor s degrees (70%) and post-graduate degrees (67%) were more likely than those with only high school degrees or less (57%) and those with only some college education (57%) to expect to work in retirement. Additionally, respondents with household incomes of less than $30,000 (71%) were more likely than those with household incomes of $100,000 or more (58%) to expect to work while retired as were those with savings of less than $25,000 (71%) compared to those with savings of at least $100,000 (54%). 6 P age

10 Actual Retirement of Retirees Approximately one in three (32%) respondents considered themselves to be retired, including some who were retired and working and others who were completely retired. When asked for the age at which they had retired, retirees reported an average retirement age of Nearly four in ten (38%) indicated that they had retired at age 55 or younger, one in three (33%) retired between ages 56 and 61, nearly one in five (18%) retired at age 62, and one in ten (10%) retired between ages 63 and 66. (See Table 2) Differences in Actual Retirement by Demographics Retirees with lower educational attainment, lower household incomes, and lower savings had retired later on average than their counterparts with higher educational attainment, higher household incomes, and higher savings. Interestingly, actual retirement age among retirees decreased as educational attainment rose even though the relationship between expected retirement age and educational attainment was the opposite for pre-retirees. Specifically, retirees with a high school degree or less reported an average retirement age of 57.5, which was a year and a half later than the average retirement age (56.0) of retirees with a post-graduate degree. Similarly, retirees with household incomes of less than $30,000 and $30,000-$60,000 reported average retirement ages (57.6 and 57.1, respectively) that were approximately two years later than the average retirement age reported by those with household incomes of $100,000 or more (55.3). Additionally, retirees with less than $25,000 in savings retired almost two years later than those with savings of $250,000 or more (average retirement age of 57.9 vs. 56.0). Differences in Actual Retirement by Current Health Status and Expected Longevity Retirees actual retirement ages also varied by current health status and expected chance of living to age 85. However, as with the relationship between retirees educational attainment and their actual retirement age, the relationship between retirees current health status and actual retirement age was not the same as the relationship between pre-retirees expected retirement age and current health status. Specifically, for retirees, actual retirement age was higher among those who described their current health in relatively unfavorable terms than among those who described their health more favorably. For example, only eight percent of retirees who described their current health as either excellent or good retired at age 63 or later compared with 18 percent of retirees who described their current health as fair or poor. However, similar to pre-retirees, retirees who thought that they had a relatively good chance of living to age 85 had retired later on average than retirees who thought that they had a lower chance of living to age 85. For example, retirees who thought that they had at least a 50-percent chance of living to age 85 had retired an average of one year later than those who thought that they had less than a 50-percent chance of living to age 85 (actual retirement age of 57.1 vs. 56.1). 5 Because the sample was limited to respondents ages 55 to 66, all retirees in the survey would have had to have retired prior to age 67. This most likely contributed to the relatively low average retirement age among the retired survey respondents. 7 P age

11 Differences in Actual Retirement by Financial Literacy Actual retirement age declined with financial literacy. 6 Specifically, the survey revealed that retirees who correctly answered all of the survey s financial literacy questions retired an average of a year and a half earlier (age 55.9) than those who correctly answered fewer financial literacy questions (age 57.4) This finding shows how higher financial literacy is associated with better savings habits, which allow workers to retire earlier. For example, retirees who answered all financial literacy questions correctly were more likely to have $250,000 or more in savings than were those who answered fewer questions correctly. Additionally, the retirees who answered all financial literacy questions correctly were the most likely to have a post-graduate degree, and actual retirement age was lower among retirees with a post-graduate degree than among those with a high school degree or less. These findings are not surprising as one would expect relatively high financial literacy among well-educated individuals as well as among those who have accumulated more savings. Table 2: At What Did You Retire? Base: Respondents who have retired regardless of current employment status (n=1,473) 55 or younger 38% % 62 18% % Average 56.9 Source: Question R9. Working in Retirement Previous research has suggested that many people expect to work in some capacity during their retirement years. 7 8 For this reason, we asked respondents whether they had worked in retirement, are currently working in retirement, or expect to work in retirement, and also explored their motivations for doing so. Previous Work in Retirement Of retirees who were not working at the time of the survey, more than one in four (28%) said that they had worked for pay at some point since retiring. Men (34%) were more likely than women (22%) to have worked as were retirees with post-graduate degrees (46%) compared to those with less education (21% of those with high school degrees, 25% of those with some college education, and 33% of those with bachelor s degrees). 6 Financial literacy was measured by the number of financial literacy-related survey questions that a respondent answered correctly. The survey contained three questions designed to gauge financial literacy. One dealt with compound interest, another with inflation, and another with asset diversification. See the annotated questionnaire in the appendix (survey questions F7, F8, and F9) for more information. 7 Staying Ahead of the Curve 2007: The AARP Work and Career Study. AARP, Staying Ahead of the Curve 2003: The AARP Working in Retirement Study. AARP, P age

12 Retirees who had previously worked in retirement were most likely to have worked part-time for someone else (61%) while about one in five had worked full-time for someone else (18%) and another one in five had worked for themselves (19%). Men who had worked in retirement (23%) were about twice as likely as women (10%) to have worked full-time. When asked to describe the main reason that they had worked in retirement, approximately half of retirees who had worked since retiring attributed this to their need for money (49%) while just as many attributed it to the enjoyment that they derive from working (50%). As might be expected, likelihood to cite the need for money was considerably lower among those with relatively high household incomes. Specifically, only about three in ten (29%) of those with household incomes of $100,000 or more cited the need for money compared to approximately half (53%) of those with household incomes under $100,000. Current Work in Retirement As mentioned previously, approximately one in ten respondents (10%) were working in retirement at the time of the survey, including 8 percent who were working part time and 2 percent who were working full time. The majority of the working retirees (59%) reported that they were working mainly for the money while about four in ten (41%) indicated that they were working mainly for enjoyment. The need for money was particularly likely to be cited by women, those with relatively low household incomes, and those with relatively low levels of savings. For example, nearly seven in ten (69%) women who were working in retirement attributed the fact that they were working to the need for money whereas just over half (53%) of males working in retirement pointed to the need for money. Moreover, two in three (67%) working retirees with household incomes under $60,000 attributed their employment to the need for money compared to just over four in ten (43%) working retirees with household incomes of $100,000 or more; and eight in ten (81%) working retirees with total savings less than $25,000 said that they were working mainly for the money compared to fewer than half (45%) of working retirees with $100,000 or more in savings. Expected Work in Retirement Of retirees who were completely retired at the time of the survey and were not looking for work, nearly three in ten (29%) said that they expect to work for pay at some point in the future. While likelihood to expect to work at some point in the future did not vary by household income, it did vary by reported level of savings. Specifically, retirees who reported total savings of less than $25,000 (43%) were considerably more likely than those with savings of at least $100,000 (24%) to expect to work at some point in the future. Of pre-retirees who expected to retire one day, the majority (62%) expected to work for pay during their retirement years. 9 Pre-retiree s likelihood to expect to work for pay in retirement increased as educational attainment increased and also rose among those with low household incomes and savings. Specifically, pre-retirees with a bachelor s degree or higher were more likely to expect to 9 When asked at what age they expected to retire, 16% of pre-retirees indicated that they expected to never retire. The respondents who expected to never retire were not asked whether they expected to work in retirement. 9 P age

13 work in retirement than were those without a bachelor s degree (68% vs. 57%, respectively). Expectations of working in retirement were also more common among pre-retirees with household incomes of less than $30,000 (71%) than among those with incomes of at least $100,000 (58%) and also more common among those with savings of less than $25,000 (71%) than among those with savings of $250,000 or more (51%). The majority of retirees and pre-retirees who expected to work in retirement expected to work parttime for someone else (72%), while 20 percent expected to work for themselves and less than one in ten (7%) expected to work full-time for someone else. Men were more likely than women to expect to work for themselves (25% vs. 16%), and women were more likely than men to expect to work part-time for someone else (76% vs. 68%). When describing the main reason that they expected to work in retirement, the majority of retirees and pre-retirees with expectations of working in retirement cited the need for money (60%) and four in ten cited enjoyment (40%). Not surprisingly, retirees and pre-retirees with post-graduate degrees, household incomes of $100,000 or more, and savings of $250,000 or more were considerably less likely than their counterparts with less education, lower incomes, and lower savings levels to attribute their work expectations to the need for money. In fact, of these respondents with the highest educational attainment, highest household incomes, and highest savings levels, the majority attributed their expectations of working in retirement to the fact that they enjoy working. Expected at Which You Still Stop Working Completely On average, respondents who were either working at the time of the survey or expected to work in the future reported that they expected to stop working completely at age 72. More than one in three (37%) expected to stop working between ages 70 and 74, while another one in three (34%) expected to continue working until age 75 or later. (See Table 3) On average, among those who were either working or who expected to work in the future, men expected to stop working completely nearly two years later than women (age 72.9 vs. 71.1). Additionally, respondents with post-graduate degrees expected to stop working 1 ½ or more years later than those with less education (age 71.1 for those with a high school degree or less, 71.8 for those with some college education, 72.2 for those with a bachelor s degree, and 73.7 for those with a post-graduate degree). This may reflect the fact that less-educated workers often hold jobs that are more physically demanding and, therefore, more difficult to perform at older ages than the jobs held by better-educated workers. Table 3: At What Do You Expect to Stop Working Completely? Base: Respondents who were working, looking for work, or who expect to work in the future (n=1,721) Younger than age 62 1% 62 3% % % % % Average 72.1 Source: Question R P age

14 Factors Influencing Retirement Many issues may affect the age at which a person retires. We asked respondents to indicate the degree to which a variety of issues, including eligibility for Social Security benefits, affected (or would be likely to affect) their retirement age. Pre-retirees: Factors Influencing Expected Retirement When pre-retirees were asked to rate their agreement with a series of 12 statements about issues that may influence their retirement age, the issue that generated the most agreement was the desire to retire while still healthy enough to enjoy retirement (82% strongly or somewhat agreed that this would influence their retirement age). Other issues that the majority of pre-retirees agreed would influence their retirement age included how much they enjoy their work (74%); how stressful or physically demanding their work is (70%); becoming eligible for full Social Security payments (62%); reaching their retirement savings goals (59%); and paying off debt, mortgages, or loans (51%). (See Chart 1) Among the issues that are likely to affect the fewest pre-retirees retirement ages are the accomplishment of career goals (29%), the retirement age of one s spouse or partner (32% of married and partnered pre-retirees), and becoming eligible for early Social Security payments (33%). Retirees: Factors Influencing Actual Retirement When retirees were asked to rate the degree to which the same 12 issues had influenced their retirement age, the issue that the largest number of retirees agreed had influenced their retirement age was the same issue that generated the most agreement among pre-retirees: the desire to retire while still healthy enough to enjoy retirement (77% strongly or somewhat agreed that this had influenced their retirement age). (See Chart 2) Only two other issues were viewed by the majority of retirees as having influenced their retirement age: becoming eligible to receive payments from an employer pension or retirement savings plan (55%) and how stressful or physically demanding their work was (53%). However, enjoyment of work had also been a fairly important factor as nearly half (46%) of retirees agreed that how much they enjoyed their work had influenced their retirement age. Among the issues that the fewest retirees felt had influenced their retirement age were becoming eligible for Medicare (6%); becoming eligible to receive full payments from Social Security (9%); and paying off their debt, mortgage, or loans (21%). 11 P age

15 Chart 1: The age at which I retire will be influenced by.* (Base: Never been retired, n=1566**) my desire to retire while I'm still healthy enough to enjoy retirement 34% 48% how much I enjoy my work 28% 46% how stressful or physically demanding my work is 18% 51% when I become eligible to receive full payments from Social Security 25% 37% when I reach my retirement savings goals 15% 44% when I pay off my debt, mortgage, or loans 17% 33% Strongly agree when I become eligible for Medicare 15% 33% Somewhat agree...when I become eligible to receive payments from an employer pension or retirement savings plan. 17% 29% when I become eligible for employer sponsored retiree health benefits 12% 26%...when I become eligible to receive early payments from Social Security. 10% 23% when my spouse/partner retires 8% 23% 0% 20% 40% 60% 80% 100% *Question R19. For each of the following statements concerning your decision about when to retire, please indicate whether you strongly agree, somewhat agree, somewhat disagree, or strongly disagree. **The when my spouse/partner retires item was only asked of respondents who were married or had a domestic partner (n=1,051). 12 P age

16 Chart 2: The age at which I retired was influenced by.* (Base: Retired, n=1473**) my desire to retire while I'm still healthy enough to enjoy it 36% 41%...when I became eligible to receive payments from an employer pension or retirement savings plan. 33% 22% how stressful or physically demanding my work was 20% 34% how much I enjoyed my work 15% 31% when I became eligible for employer sponsored retiree health benefits 16% 19% when I reached my retirement savings goals 10% 23% when I had accomplished certain career or jobrelated goals...when I became eligible to receive early payments from Social Security. when my spouse/partner retired 9% 24% 11% 17% 9% 14% Strongly agree Somewhat agree when I paid off my debt, mortgage, or loans 5% 17% when I became eligible to receive full payments from Social Security 4% 6% when I became eligible for Medicare 2% 4% 0% 20% 40% 60% 80% 100% *Question R20. For each of the following statements concerning your decision about when to retire, please indicate whether you strongly agree, somewhat agree, somewhat disagree, or strongly disagree. **The when my spouse/partner retired item was asked only of respondents who were married or had a domestic partner (n=1,137). 13 P age

17 Pre-Retirees vs. Retirees Compared to pre-retirees expectations of the factors that would influence their retirement age, retirees responses were more narrowly focused. For example, the majority of pre-retirees expected that six of the 12 issues would influence their retirement age while only three of the 12 issues were identified by the majority of retirees as having influenced their retirement age. As a result, pre-retirees were more likely than retirees to identify most of these issues as an influence on their retirement age. The only issues that pre-retirees were not more likely than retirees to identify as an influence on retirement age included becoming eligible for an employer pension plan or retirement savings plan, reaching career or job-related goals, and becoming eligible for employer-sponsored retiree health benefits. Furthermore, eligibility for an employer pension or retirement savings plan was the only issue that retirees were more likely than pre-retirees to cite as an influence. In fact, not only was eligibility for an employer pension or retirement savings plan more likely to be cited by retirees than by pre- retirees, but it was also the second most frequently identified influence among retirees whereas it was only the eighth most frequently identified influence among pre-retirees. The greater influence of eligibility for pensions or retirement savings plans among retirees is not surprising given the fact that retirees (68%) were more likely than pre-retirees (47%) to indicate that they or their spouse have an employer-provided defined benefit pension plan (either a traditional pension or a cash balance plan), with the benefit based on salary and years of service. Retirees and pre-retirees were equally likely to identify eligibility for employer-provided retiree health benefits and achievement of career or job-related goals as an influence on their retirement age. THE SOCIAL SECURITY CLAIMING AGE DECISION Non-claimer s Expected Claiming When asked whether they had previously thought about when they themselves might start collecting Social Security retirement benefits, the majority (81%) of non-claimers indicated that they had thought about this. Regardless of whether they had previously thought about when they would start collecting benefits, all non-claimers were asked at what age they expected to start collecting benefits. In response, approximately one quarter (26%) of non-claimers indicated that they expected to start collecting benefits at the earliest age possible (age 62) and another one in five (22%) expected to start collecting benefits after age 62 but before age 66. Another quarter (24%) expected to start collecting benefits at their full retirement age (age 66), while another 12 percent expected to start between ages 67 and 69 and another 15 percent expected to start at age 70 or later. (See Table 4) This means that approximately half (49%) of non-claimers expected to begin collecting benefits before they reach the age at which they qualify for full Social Security retirement benefits (age 66), while approximately another half (51%) expect to wait until they become eligible for full benefits or later. The average age at which non-claimers expect to claim benefits is age P age

18 Table 4: At What Do You Expect to Start Collecting Social Security Retirement Benefits? Base: Respondents who have not yet claimed benefits (n=1,988) Younger than 62 * 62 26% % 66 (Full Retirement ) 24% % 70 or higher 15% Average 65.5 Source: Question SS2 and SS2a. * Less than 0.5%. Differences in Expected Claiming by Demographics The average age at which non-claimers expected to claim Social Security benefits varied by gender, age, educational attainment, and savings. For example, on average, men expected to claim about half a year later than women (age 65.9 vs. 65.2, respectively). Additionally, nonclaimers with a bachelor s degree and those with a post-graduate degree expected to claim about one year later than those with only a high school degree or less (ages 65.9 and 66.2 vs. 65.0, respectively). This relationship between educational attainment and expected claiming age is similar to that found between educational attainment and expected retirement age and may reflect the fact workers with less education tend to hold jobs that are relatively physically demanding while better-educated workers tend to hold jobs that rely more on knowledge than physical stamina. Although average expected claiming age did not vary by household income, it did vary by savings level. Specifically, non-claimers with less than $100,000 in total savings expected to claim about half a year later than non-claimers with at least $100,000 in savings (age 65.8 and 65.2, respectively). Differences in Expected Claiming by Current Health Status and Expected Longevity Non-claimers who described their health status in relatively favorable terms as well as those who thought that they had a relatively good chance of living to age 85 or who expected that their spouse had a relatively good chance of living to age 85 expected to claim later than other nonclaimers. For example, non-claimers who described their health as excellent expected to claim later than those who described their health as only fair or poor (ages 65.8 and 65.2, respectively). This is similar to the relationship uncovered between health status and expected retirement age and is not surprising as individuals in excellent health may expect to be able to work longer (and postpone claiming benefits longer) than individuals in relatively poor health. Additionally, non-claimers who felt that they have at least a 50-percent chance of living to age 85 expected to claim an average of nearly one year later (age 65.9) than non-claimers who felt that they have less than a 50-percent chance of living to age 85 (age 65.0). 10 Moreover, expected 10 A similar relationship was uncovered between expected chance of living to age 85 and expected retirement age, with individuals who felt that they had at least a 75-percent chance of living to age 85 more likely to expect to retire at age 70 or later. 15 P age

19 claiming age also varied by expectations of spouse longevity. For example, non-claimers who expected their spouses to have at least a 75-percent chance of living to age 85 expected to claim later than those who thought that their spouses had less than a 50-percent chance of reaching age 85 (age 65.8 vs. 65.0, respectively). Differences in Expected Claiming by Financial Literacy Expected claiming age increased with financial literacy. Specifically, non-claimers who answered at least two of the financial literacy questions correctly expected to claim at an average age of 65.7 whereas those who answered only one or none of the questions correctly expected to claim at an average age of This may at least partially reflect educational attainment as non-claimers with more education were more likely than their less-educated counterparts to answer the financial literacy questions correctly and they also expected to claim later than those with less education. Changes in Expected Claiming Over Past Two Years When asked whether their expected claiming age had changed within the past two years, nearly three in four (74%) non-claimers indicated that it had not changed while just under one in five (18%) indicated that it had increased and just a few (8%) indicated that it had declined. Among those who indicated that their expected claiming age had increased over the past two years, the most common reasons included something that they had learned about how benefits are calculated (41%), health care costs (36%), and investment losses (32%). (See Table 5) Among those who indicated that their expected claiming age had declined over the past two years, the most common reasons included job loss (their own or a family member s, 30%), investment losses (24%), and health care costs (23%). Non-claimers who now expect to claim later than they once did were more likely than those who now expect to claim earlier to attribute their changing expectations to health care costs (36% vs. 23%) and to something that they had learned about how Social Security benefits are calculated (41% vs. 13%). In contrast, non-claimers whose expected claiming age had decreased were more likely than those whose expected claiming age had increased to cite job loss (30% vs. 14%) when naming the reasons for their changing expectations. 16 P age

20 Table 5: Why Have Your Expectations Changed? Base: Non-claimers who now expect to claim benefits earlier or later than they expected two years ago Base: Now expect to claim earlier (n= 155) Base: Now expect to claim later (n=359) (A) (B) Financial (net) 44% 57% A*** Health care costs 23% 36% A Investment losses 24% 32% Credit card debt 5% 11% Need the money* 6% 4% Education costs ** 3% Economy/recession* 1% 3% Other financial issues* 1% 2% Advice/Learning (net) 26% 47% A Something that I learned about 13% 41% A how Social Security benefits are calculated Advice that I received 17% 15% Job/Work (net) 34% B 16% Job loss my own or a family 30% B 14% member s Decided to keep working* ** 2% Job stress/dissatisfaction* 2% -- Other job/work issues* 2% 1% Health (net) 15% 9% Health problems my own or a 14% 7% family member s Health improvements my 1% 3% own or a family member s Family (net) 7% 4% Change in marital status 6% 2% Caregiving/family 1% 2% responsibilities* Other family issues* 1% 1% Other (net) 11% 7% Government/Congress actions* 3% 2% Concerns about Social Security 4% 2% solvency/fear of benefits reduction* Other 5% 3% * On the survey, this question was displayed with pre-printed response options and an other space in which respondents were allowed to write in a response. Responses displayed above with asterisks were written in by respondents. Multiple responses were allowed per respondent. Table lists answers to all pre-printed response options as well as all written-in responses that were provided by at least 2 percent of respondents in either column. ** Less than 0.5% *** Statistically significant differences between those who now expect to claim later and those who now expect to claim earlier are marked with the letters A and B. For example, percentages marked with a superscript A reflect percentages that are significantly larger than the corresponding percentage in column A. Similarly, percentages marked with a superscript B reflect percentages that are significantly larger than the corresponding percentage in column B. Source: Question SS4 17 P age

21 Claimers Actual Claiming Just over three in four (78%) claimers indicated that they had started collecting Social Security retirement benefits at age 62 or younger, including 1 percent who believed that they had started collecting benefits before age 62 and 77 percent who reported that they had started collecting benefits at age Approximately one in five (22%) had started collecting benefits between ages 63 and Among all claimers, the average claiming age was (See Table 6) Table 6: At What Did You Start Receiving Social Security Retirement Benefits? Base: Respondents who have claimed benefits (n=1,057) Younger than 62 1% 62 77% 63 8% 64 4% 65 8% 66 2% Average 62.5 Source: Questions SS6 and SS6a. The average age at which claimers claimed Social Security benefits did not vary by demographics such as gender, age, educational attainment, household income, savings, or current health status. Nor did it vary by other factors that might be expected to influence claiming age such as expectations of one s own chances of living to age 85, expectations of one s spouse living to age 85, or financial literacy (as measured by number of financial literacy questions answered correctly). 13 Actual Claiming vs. Claimers Expectations When asked whether they had started collecting Social Security benefits earlier than they expected, later than they expected, or about when they expected, the majority of claimers (64%) indicated that they had started collecting benefits about when they expected while just over one in three (35%) said that they had started collecting benefits earlier than expected. Virtually no claimers (1%) started collecting benefits later than they had expected. Among those claimers who had started collecting benefits earlier than expected, the most common reason was job loss (their own or a family member s, 45%) followed by advice that they received (26%), something that they had learned about how benefits are calculated (15%), and health problems of their own or a family member (13%). Given that claimers indicated that job loss led to earlier claiming, the Social Security system could be put in jeopardy during lean economic times when job loss is high, payroll taxes are lower, and claims may be higher. (See Table 7) 11 Although respondents were informed that age 62 is the earliest age at which one can start collecting Social Security retirement benefits, some respondents (0.6%) insisted that they had started collecting Social Security retirement benefits before age 62. Their responses are included in the 78 percent figure noted above even though they may either have incorrectly recalled their claiming age or have been referring to another type of Social Security benefit, such as disability benefits. 12 It is important to reiterate that the survey was limited to respondents ages 55 to 66. Therefore, none of the respondents were older than age 66 and none could have claimed benefits beyond age While we might have expected to see differences in claiming age by some of these variables, it may be that the relatively small size of the claimer sample precluded the detection of statistically significant differences that would have emerged if the survey had included more claimers. 18 P age

22 Table 7: What Caused You to Start Collecting Benefits Earlier Than You Had Expected?* Base: Respondents who claimed benefits earlier than they had expected to (n=357) Job / Work Issues (Net) 52% Job loss my own or a family member s 45% Job stress/dissatisfaction* 4% Early retirement* 3% Other job/work issues* 1% Advice / Learning (Net) 32% Advice that I received 26% Something that I learned about how Social 15% Security benefits are calculated Health (Net) 16% Health problems my own or a family 13% member s Health improvements my own or family 3% member s Financial (Net) 16% Investment losses 8% Health care costs 6% Needed the money* 5% Credit card debt 5% Education costs 1% Other financial issues* 1% Family (Net) 4% Change in marital status 2% Other family issues* 2% Other 4% * This question was displayed with pre-printed response options and an other space in which respondents were allowed to write in a response. Responses displayed above with asterisks were written in by respondents. Multiple responses were allowed per respondent. Table lists answers to all pre-printed response options as well as all written-in responses that were provided by at least 2 percent of respondents. Source: Question SS8. Working After Collecting Social Security Some people stop working once they begin collecting Social Security benefits, while others continue working while receiving benefits. We asked respondents whether they worked (or expect to work) while collecting benefits in order to understand how many respondents had supplemented (or expect to supplement) their Social Security benefits with earned income from work. Non-claimers Expectations Six in ten (60%) non-claimers reported that they expect to work for pay after they begin collecting Social Security retirement benefits, while nearly all other non-claimers (40%) reported that they do not expect to work after collecting benefits. 19 P age

23 Claimers Actual Experiences While the majority (51%) of claimers were completely retired and neither working nor looking for work, a considerable portion of claimers were either employed or looking for work at the time of the survey or had worked at one time since claiming benefits. Specifically, approximately one in four (27%) claimers were employed part time, another six percent were employed full time, and another eight percent were unemployed and looking for work. Moreover, of claimers who were completely retired, approximately one in five (21%) reported that they had worked for pay at some point after they began collecting Social Security benefits. Factors Influencing Social Security Claiming In an effort to understand the degree to which certain issues influence the age at which people claim Social Security benefits, non-claimers were asked to rate their agreement with a series of seven statements about issues that may influence their claiming age while claimers were asked to rate their agreement with a similar set of statements about issues that did influence their claiming age. Non-claimers: Factors Influencing Expected Claiming When non-claimers were asked to rate their agreement with a series of statements about issues that may influence their Social Security claiming age, the need for money and their expected retirement age rose to the top of the list. Specifically, more than seven in ten non-claimers strongly or somewhat agreed that their claiming age will be influenced by how much they need the money (78%) or by when they decide to retire (73%). Additionally, approximately two in three agreed that their claiming age will be influenced by whether they are able to collect benefits while working without significantly reducing their benefits (66%) or when they become eligible for the highest possible monthly benefit (65%). (See Chart 3) Relatively few respondents (43%) felt that their claiming age would be affected by advice received from others. Furthermore, among respondents who are married, only a minority (38%) agreed that their claiming age would be affected by the impact on their spouse s benefits and only about one in three married or partnered respondents felt that their claiming age would be affected by when their spouse or partner collects benefits. Differences by Demographics, Health Status, Expected Longevity, and Financial Literacy 14 The age at which I start collecting Social Security benefits will be influenced by whether I can collect benefits and still work for pay without significantly reducing my benefits. : This statement resonated the most with non-claimers with relatively high education levels, those with low household incomes, and those with low savings levels. For example, non-claimers who were more likely than others to strongly or somewhat agree with this statement included respondents with post-graduate degrees (71%, vs. 63% of those with high school degrees or less) and respondents with household incomes of less than $30,000 (73%, vs. 60% of those with incomes of $100,000 or more). For example, respondents with less than $25,000 in total savings (77%) were more likely to agree with this statement than were those with more than $25,000 in savings (61%). 14 This section lists only those statements for which agreement varied by demographics such as gender, education, household income, and total savings; by health status; by expected longevity; or by financial literacy. Each respondent s financial literacy was measured by how many of the three financial literacy survey questions the respondent answered correctly. Statements for which responses did not vary by any of these factors are not listed. 20 P age

24 Agreement with the statement was also relatively common among non-claimers whose survey responses suggested low-to-moderate financial literacy levels, which may reflect the statement s greater popularity among those with low incomes and low savings as respondents with the lowest incomes and those with the lowest savings were more likely to exhibit lower financial literacy. Specifically, while nearly seven in ten (69%) respondents who answered two or fewer of the financial literacy questions correctly agreed with this statement, just six in ten (60%) respondents who answered all three questions correctly agreed. The age at which I start collecting Social Security benefits will be influenced by when I am eligible for the highest possible monthly Social Security benefit. : Likelihood to agree with this statement was higher among non-claimers with relatively low savings as well as those who expect that they have at least a 50-percent chance of living to age 85. For example, while nearly three in four (73%) non-claimers with less than $25,000 in savings agreed with this statement, just over half (56%) of those with savings of $250,000 or more agreed. Additionally, while nearly seven in ten (68%) non-claimers who expect to have at least a 50-percent chance of living to age 85 agreed with this statement, fewer than six in ten (58%) non-claimers who believe that they have less than a 50-percent chance of living this long agreed. Non-claimers who said there is a 75-percent chance or more that they will live until age 78 or older (68%) were more likely to agree with this statement than those who said that there was less than a 50-percent chance (56%) or a 50-to-75 percent chance (62%) that they will live until age 78 or older. The age at which I start collecting Social Security benefits will be influenced by when my spouse/partner decides to collect Social Security benefits. : Respondents with the lowest educational attainment (high school degrees or less, 37%) and those with the highest educational attainment (post-graduate degrees, 36%) were more likely than those with only some college education or bachelor s degrees (27%) to agree with this statement. Agreement with this statement was also more common among respondents with $100,000 to $250,000 in savings (40%) than among those with $25,000 to $100,000 in savings (29%) and those with more than $250,000 in savings (30%). The age at which I start collecting Social Security will be influenced by how much I need the money. : As might be expected, agreement with this statement was higher among those with lower household incomes, those with lower savings levels, and those who believe that their spouses have a relatively low chance of living to age 85. For example, while about eight in ten (81%) non-claimers with household incomes of $60,000 or less agreed with this statement, only about seven in ten (71%) non-claimers with household incomes of $100,000 or more agreed. Moreover, non-claimers with total savings of less than $250,000 (81%) were more likely than those with savings of at least $250,000 (71%) to agree. Additionally, non-claimers who expect that their spouses have less than a 50-percent chance of living to age 85 are more likely to agree than are those who believe that their spouses have at least a 50-percent chance of living this long (83% vs. 72%). 21 P age

25 Claimers: Factors Influencing Actual Claiming When claimers were asked to indicate the extent to which they agreed with a series of similar statements about issues that had influenced their actual Social Security claiming age, the need for money rose to the top of the list just as it did when non-claimers were asked to anticipate which issues would influence their claiming age. In fact, the need for money was the only issue that the majority of claimers identified as having been an influence on their claiming age (58% strongly or somewhat agreed). Other issues that claimers were relatively likely to identify as influences on their claiming age included advice received from others (46%) and the age at which they expected to retire (44%). Relatively few claimers agreed that their claiming age had been influenced by the age at which they became eligible for the highest possible monthly benefit (13%) or by when their spouse/partner decided to collect benefits (14% of married/partnered claimers). Moreover, fewer than one in ten (7%) married claimers agreed that their claiming age had been influenced by the impact on their spouse s benefits. Differences by Demographics, Health Status, Expected Longevity, and Financial Literacy 15 The age at which I began collecting Social Security was influenced by advice received from others. : Women were more likely than men to strongly or somewhat agree with this statement (57% vs. 38%). Respondents who estimated that they have at least a 75-percent chance of living to age 78 (54%) were more likely to agree with this statement than those who estimated themselves to have less than a 50-percent chance (29%) or a 50-to-75 percent chance (33%). Additionally, respondents who believed that they had at least a 50-percent chance of living to age 85 were more likely to agree than were those who believed that they had less than a 50-percent chance of living this long (50% vs. 36%). The age at which I began collecting Social Security was influenced by whether I could collect benefits and still work for pay without significantly reducing my benefits. : Claimers with less than $25,000 in savings were more likely than those with savings of $100,000 or more to agree with this statement (44% vs. 25%). Claimers who believe they have at least a 75-percent chance of living to age 78 or older (35%) were more likely to agree with this statement than those who believe they only have a 50-to-75 percent chance (21%). The age at which I began collecting Social Security was influenced by when I became eligible for the highest possible monthly Social Security benefit. : As might be expected, this statement also elicited more agreement from claimers with lower savings. Specifically, claimers with less than $25,000 in savings were twice as likely as those with savings of $250,000 or more (17% vs. 8%) to agree with this sentiment. 15 This section lists only those statements for which agreement varied by demographics such as gender, education, household income, and total savings; by health status; by expected longevity, or by financial literacy. Each respondent s financial literacy was measured by how many of the three financial literacy survey questions the respondent answered correctly. Statements for which responses did not vary by any of these factors are not listed. 22 P age

26 The age at which I began collecting Social Security was influenced by when I expected to retire. : Claimers who described their current health status as excellent were more likely than those who described their current health status as good, fair, or poor to agree with this statement (54% vs. 41%). The age at which I began collecting Social Security was influenced by when my spouse/partner decided to collect Social Security benefits. : Women were about four times more likely than men to strongly or somewhat agree with this statement (25% vs. 6%). This may reflect the fact that married women were considerably more likely than married men to report that their spouse was the primary breadwinner (73% of women vs. 13% of men) as well as the fact that women were more likely than men to be younger than their spouse/partner (61% of women vs. 18% of men). The age at which I began collecting Social Security was influenced by how much I needed the money. : Not surprisingly, likelihood of agreeing with this statement was higher among those with relatively low household incomes. For example, more than six in ten (63%) claimers with household incomes less than $60,000 strongly or somewhat agreed with this statement compared with just half (50%) of claimers with household incomes of $60,000 or more. 23 P age

27 Chart 3: The age at which I will start (started) collecting Social Security will be (was) influenced by.* (Non-Claimers n=1,988**) (Claimers n=1,057**)...how much I need the money. 28% 30% 35% 43%...when I decide to retire. 16% 28% 31% 42%...whether I can collect benefits & still work for pay without significantly reducing my benefits. 10% 28% 20% 38%...when I am eligible for the highest possible monthly Social Security benefits....the impact on my spouse's Social Security benefits. 1% 6 8% 5 10% 26% 28% 39% Strongly agree (Non Claimers) Somewhat agree (Non Claimers) Strongly agree (Claimers)...advice received from others. 6 17% 37% 29% Somewhat agree (Claimers)...when my spouse/partner decides to collect Social Security benefits. 4 10% 6 26% 0% 20% 40% 60% 80% 100% *Question SS10 (Non-Claimers). We would like to know what factors may influence your decision about when to start collecting Social Security retirement benefits. *Question SS12 (Claimers). We would like to know what factors influenced your decision about when to start collecting Social Security retirement benefits. **The impact on my spouse s Social Security benefits item was asked only of respondents who were married (n=1,299 non-claimers and 787 claimers). The when my spouse/partner decides to collect Social Security benefits item was asked only of respondents who were married or had a domestic partner (n=1,380 non-claimers and 814 claimers). 24 P age

28 Non-claimers vs. Claimers Compared to non-claimers expectations of the factors that will influence their claiming age, claimers responses were more narrowly focused. For example, the majority of non-claimers expected that four of the seven issues would influence their claiming age while only one of the seven issues was identified by the majority of claimers as having influenced their claiming age (See Chart 3). As a result, non-claimers were more likely than claimers to identify six of the seven issues as an influence on their claiming age. Advice received from others was the only issue that non-claimers and claimers were equally likely to identify as an influence. However, advice appeared to have been relatively more important to claimers than to non-claimers as it was the influence named second most frequently by claimers and only the fifth most frequently named influence among non-claimers. The fact that advice was the second most often cited influence for claimers and only the fifth most often cited influence by non-claimers may suggest that non-claimers will ultimately rely on advice to a greater extent than they anticipate. Reactions to Statements About Social Security Benefits Non-claimers and claimers were also asked to rate their agreement with each of five different statements about their reliance on Social Security benefits and the connection between claiming age and benefits. Non-claimers were asked to indicate whether they agreed with each statement now, while claimers were instructed to indicate whether they agreed that each statement reflected a thought that they had had at the time that they decided to claim benefits. The purpose of this section of the survey was to gauge the extent to which different statements about Social Security benefits and claiming age resonated with respondents. Non-claimers Reactions All five statements elicited ratings of strongly agree or somewhat agree from the majority of nonclaimers. However, the statement with which the most non-claimers agreed (77% strongly or somewhat agreed) related to the advantages of delaying claiming - If I delay collecting Social Security benefits as long as possible, I will get a larger monthly Social Security check that could help me when I m older and have used up my savings. The statement that generated the second highest level of agreement was similar in nature but addressed the risks of claiming too early. This statement, with which nearly as many non-claimers (76%) agreed, was If I start collecting Social Security benefits too early, I may end up with a monthly Social Security check that may be too small when I m older and unable to work. The third most popular statement based on the number of strongly or somewhat agree ratings was I will not be able to support myself in retirement without my Social Security benefits (72%). Although this statement was only the third most popular based on strongly and somewhat agree ratings, it elicited more strongly agree ratings (38%) than any of the other statements. (See Chart 4) 25 P age

29 Differences by Demographics, Health Status, Expected Longevity, and Financial Literacy 16 In general, these statements tended to resonate more with women, respondents with relatively less education, respondents with low-to-moderate household incomes, and those with low-to-moderate savings. The specific differences by demographics were as follows: If I wait too long before collecting Social Security benefits, Social Security may run out of funds. : This statement resonated the most with women, non-claimers with relatively less education, those with relatively low household incomes, those with relatively low savings, those in relatively poor health, and those who believe that they have a relatively small chance of living to age 85. For example, women were more likely than men (71% vs. 61%) to agree with this sentiment as were non-claimers with a bachelor s degree or less compared to non-claimers with a post-graduate degree (69% vs. 51%). Additionally, while nearly seven in ten (68%) non-claimers with household incomes of less than $100,000 agreed with this statement, only about half (52%) of those with incomes of $100,000 or more agreed. Agreement was also more common among those with savings of less than $25,000 than among those with savings of at least $100,000 (73% vs. 61%). Furthermore, non-claimers in fair or poor health (76%) were more likely than those in good or excellent health (64%) to agree with this statement as were those who believed that they had less than a 50-percent chance of living to age 78 or 85 (73% and 69%) compared to those who believed that they had at least a 75-percent chance of living until age 78 or 85 (62% for both). Additionally, agreement was also more common among respondents who believed that their spouse had less than a 50-percent chance of living to age 85 than among those who believed that their spouse had at least a 50-percent chance of living this long (70% vs. 61%). Finally, compared to non-claimers who answered all of the three financial literacy questions correctly (58%), those who answered fewer questions correctly (69%) were more likely to agree with this statement. If I wait too long before collecting Social Security benefits, I may not receive all the benefits that I have earned before I die. : Non-claimers with only a high school degree or less were more likely than those with a post-graduate degree (72% vs. 63%) to strongly or somewhat agree with this statement. Additionally, not surprisingly, non-claimers who expected to have less than a 50- percent chance of living to age 78 or 85 (76% and 74%) were more likely to agree with this statement than were those who expected to have at least a 75-percent chance of living to age 78 or 85 (67% and 65%). If I start collecting Social Security benefits too early, I may end up with a monthly Social Security check that may be too small when I m older and unable to work. : This statement resonated the most with non-claimers who had low-to-moderate household incomes, those with low-to-moderate savings, and those who believed that they had a good chance of living to age 85. Specifically, 16 This section lists only those statements for which agreement varied by demographics such as gender, education, household income, and total savings; by health status; by expected longevity, or by financial literacy. Each respondent s financial literacy was measured by how many of the three financial literacy survey questions the respondent answered correctly. Statements for which responses did not vary by any of these factors are not listed. 26 P age

30 non-claimers with household incomes of less than $100,000 were more likely than those with incomes of $100,000 or more (78% vs. 66%) to agree with this statement as were respondents with total savings of less than $250,000 compared to those with savings of $250,000 or more (80% vs. 65%). Additionally, respondents who believed that they had at least a 75-percent chance of living to age 85 were more likely to agree with this statement than were those who believed that they had less than a 50-percent chance of living this long (80% vs. 73%). If I delay collecting Social Security benefits as long as possible, I will get a larger monthly Social Security check that could help me when I m older and have used up my savings. : Not surprisingly, non-claimers with less than $250,000 in savings were more likely than those with savings of $250,000 or more (80% vs. 72%) to agree with this statement. I will not be able to support myself in retirement without my Social Security benefits. : This statement resonated the most with women, non-claimers with relatively less education, those with low-to-moderate household incomes, those with low-to-moderate savings, those in relatively poor health, and those who believed that their spouse had a small chance of living to age 85. Specifically, women were more likely than men (75% vs. 69%) to agree with this statement as were respondents with a bachelor s degree or less compared to those with a post-graduate degree (76% vs. 57%). Additionally, respondents with household incomes of less than $60,000 (83%) were more likely than those with incomes of $60,000-$100,000 (68%) and those with incomes of $100,000 or more (47%) to agree with this statement. Agreement was also more common among those with total savings of less than $25,000 (89%) compared to those with savings of $25,000 - $249,000 (77%) and least common among those with savings of $250,000 or more (48%). Furthermore, non-claimers in fair or poor health (83%) were more likely than those in good or excellent health (71%) to agree with this statement. Similarly, respondents who believed that their spouse had less than a 50-percent chance of living to age 85 were more likely to agree than were those who believed that their spouse had at least a 75-percent chance of living this long (73% vs. 63%). Finally, compared to non-claimers who answered all of the three financial literacy questions correctly (65%), those who answered fewer questions correctly (75%) were more likely to agree with this statement. Claimers Reactions Only two of the five statements elicited ratings of strongly agree or somewhat agree from the majority of claimers. These two statements related to a general sense of dependence on Social Security and a desire to ensure that all earned benefits are received prior to death. Specifically, I will not be able to support myself in retirement without my Social Security retirement benefits was the statement with which the most claimers strongly or somewhat agreed (60%). It was also the statement with which the most claimers strongly agreed (34%). The other statement with which the majority of claimers (57%) strongly or somewhat agreed was If I waited too long before collecting Social Security benefits, I may not receive all the benefits that I have earned before I die. 27 P age

31 The statement with which claimers were least likely to agree was If I began collecting Social Security benefits too early, I may end up with a monthly Social Security check that will be too small when I m older and unable to work (33% strongly or somewhat agreed). Differences by Demographics, Health Status, Expected Longevity, and Financial Literacy 17 In the cases in which agreement with the statements did vary by demographics, it was often the case that the statements resonated the least with claimers with the highest education level, those in the highest income brackets, and those with the highest savings levels. If I began collecting Social Security benefits too early, I may end up with a monthly Social Security check that will be too small when I m older and unable to work. : This statement resonated the most with claimers who did not have a post-graduate degree, claimers with relatively little savings, those in fair or poor health, and those who believed that their spouse had a relatively small chance of living to age 85. Specifically, claimers with a bachelor s degree or less were more likely than those with a post-graduate degree to strongly or somewhat agree with this statement (35% vs. 21%). Additionally, claimers with less than $25,000 in savings were more likely than those with total savings of $250,000 or more to agree (39% vs. 24%). Furthermore, while about two in five (42%) claimers in fair or poor health agreed with this statement, only about one in four (26%) of those in excellent health agreed. This may reflect the lower savings of those in fair or poor health as more than two in five (44%) claimers in fair or poor health had less than $25,000 in total savings compared to only about one in five (19%) claimers in excellent health. Additionally, claimers who believed that their spouse had less than a 50-percent chance of living to age 85 were more likely to agree with this statement than were those who believed that their spouse had at least a 50-percent chance of living this long (55% vs. 40%). This difference may reflect the lower savings of those who feel that their spouses have less than a 50- percent chance of living to age 85 as claimers who feel this way are more likely to have less than $25,000 in savings than are claimers who believe that their spouses have at least a 50-percent chance of living this long (29% vs. 22%). If I delayed collecting Social Security benefits as long as possible, I will get a larger monthly Social Security check that could help me when I m older and have used up my savings. : Claimers with $25,000 to $249,000 in savings were more likely than those with total savings of $250,000 or more to agree with this statement (54% vs. 40%). However, claimers with less than $25,000 in savings were no more likely than those with $250,000 or more to agree, which suggests that this statement is relatively likely to resonate with respondents who have a moderate amount of savings but less likely to resonate with those who have very little savings or those who have a considerable amount of savings. 17 This section lists only those statements for which agreement varied by demographics such as gender, education, household income, and total savings; by health status; by expected longevity; or by financial literacy. Each respondent s financial literacy was measured by how many of the three financial literacy survey questions the respondent answered correctly. Statements for which responses did not vary by any of these factors are not listed. 28 P age

32 I will not be able to support myself in retirement without my Social Security benefits. : This statement resonated the most with claimers who had less than a post-graduate degree, those with low-to-moderate household incomes, those with lower financial literacy, and those with low-tomoderate savings levels. For example, while more than six in ten (62%) claimers with a bachelor s degree or less agreed with this statement, slightly less than half (47%) of those with a postgraduate degree agreed. Moreover, claimers with household incomes of less than $100,000 were more likely than those with household incomes of $100,000 or more to agree (63% vs. 34%) as were claimers with total savings of less than $250,000 compared to those with $250,000 or more in savings (66% vs. 44%). Compared to claimers who answered all financial literacy questions correctly (49%), those with lower levels of financial literacy were more likely to agree with this statement. Nearly seven in ten (69%) of those who answered no questions or only one question correctly agreed with this statement, while 63 percent of those who answered two financial literacy questions correctly agreed. Non-claimers vs. Claimers Just as in the previous section, claimers responses in this section continued to be more focused than those of non-claimers. For example, the majority of non-claimers strongly or somewhat agreed with all five of the statements about Social Security benefits whereas the majority of claimers agreed with only two of the five statements. As a result, non-claimers were more likely than claimers to strongly or somewhat agree with each of the five statements. However, ranking the statements in descending order from most popular to least popular based on the percentage of non-claimers and claimers who strongly or somewhat agreed with each statement suggests that the notion that early claiming could result in an insufficient monthly check in old age resonated considerably more with non-claimers than with claimers. Specifically, the statement that addressed this issue generated the second highest level of agreement among nonclaimers (76% agreed) but the lowest level of agreement among claimers (33%). In contrast, examining the agreement levels in this way suggests that claimers were relatively more likely than non-claimers to be concerned that waiting too long to claim benefits may cause them to die before receiving all the benefits that are due to them. Specifically, the statement that addressed the fear of dying before receiving all earned benefits produced the second highest level of agreement among claimers (57%) but the second lowest level of agreement among non-claimers (69%). Interestingly, among both non-claimers and claimers, the statement with which most respondents strongly agreed was I will not be able to support myself in retirement without my Social Security benefits. 29 P age

33 Chart 4: Agreement with statements about Social Security benefits and claiming age.* (Non-Claimers n=1,988) (Claimers n=1,057) I will not be able to support myself in retirement without my Social Security benefits. 34% 38% 26% 34% If I delay collecting Social Security benefits as long as possible, I will get a larger monthly check that could help me when I'm older and have used up my savings. If I start collecting Social Security benefits too early, I may end up with a monthly check that may be too small when I'm older and unable to work. 20% 26% 33% 10% 22% 36% 44% 40% Strongly agree (Non Claimers) Somewhat agree (Non Claimers) If I wait too long before collecting Social Security benefits, I may not receive all of the benefits that I have earned before I die. 20% 23% 37% 47% Strongly agree (Claimers) Somewhat agree (Claimers) If I wait too long before collecting Social Security benefits, Social Security may run out of funds. 13% 26% 29% 40% 0% 20% 40% 60% 80% 100% *Question SS11 (Non-Claimers). We would like to know more about which factors may influence your decision regarding when to start collecting Social Security retirement benefits. For each of the following, please indicate whether you strongly agree, somewhat agree, somewhat disagree, or strongly disagree. *Question SS13 (Claimers). We would like to know more about which factors influenced your decision regarding when to start collecting Social Security retirement benefits. For each of the following statements, please indicate whether you strongly agree, somewhat agree, somewhat disagree, or strongly disagree that the statement represents a thought that you had at the time you decided to start collecting Social Security retirement benefits. SURVIVAL PROBABILITY Because the age at which someone claims Social Security benefits may be based in part on how long they expect to live, respondents were asked to indicate what they think the chances are that they will live to be age 78 or more as well as the chances that they will live to be age 85 or more. 30 P age

34 They were instructed to respond on a scale from 0 to 100 where 0 means that they believe that they have absolutely no chance of living to that age and 100 means that they are absolutely certain that they will live to that age. In addition, respondents who were married or living with a partner were asked to indicate what they think the chances are that their spouse/partner will live to these ages. Chances of Living to 78 or More Approximately eight in ten (81%) respondents indicated that they believe that they have at least a 50-percent chance of living to age 78 or more, and slightly more than half (56%) said that they think that they have at least a 75-percent chance of living this long. On average, respondents expressed the belief that they have a 70-percent chance of living to age 78. (See Chart 5) When estimating their spouse s or partner s chances of living to age 78 or more, respondents provided very similar estimates. Specifically, about eight in ten (82%) respondents who were married or living with a partner believed that their spouse/partner has at least a 50-percent chance of living to age 78 or more, including slightly more than half (56%) of respondents who believed that their spouse/partner has at least a 75-percent chance of living this long. On average, respondents who were married or living with a partner felt that their spouse/partner had a 69- percent chance of living to age 78. Across all respondents, nearly nine in ten (87%) felt that either they or their spouse/partner had at least a 50-percent chance of living to age 78, and nearly two in three (65%) felt that either they or their spouse/partner had at least a 75-percent chance of living this long. Chances of Living to 85 or More Not surprisingly, respondents were less likely to believe that they will live to age 85 than to believe that they will live to age 78. Specifically, just over six in ten (62%) respondents felt that they have at least a 50-percent chance of living to age 85 or more, and only about three in ten (31%) respondents felt that they have at least a 75-percent chance of living this long. On average, respondents believed they have a 54-percent chance of living to age 85. When estimating their spouse s or partner s chances of living to age 85 or more, respondents continued to provide estimates that were similar to the chances that they gave to themselves for living that long. For example, about two in three (67%) respondents who were married or living with a partner believed that their spouse/partner has at least a 50-percent chance of living to age 85, with 35 percent of respondents believing that their spouse/partner has at least a 75-percent chance of living this long. On average, respondents who were married or living with a partner felt that their spouse/partner had a 58-percent chance of living to age 85. Across all respondents, just over seven in ten (72%) felt that either they or their spouse/partner had at least a 50-percent chance of living to age 85, and about four in ten (41%) felt that either they or their spouse/partner had at least a 75-percent chance of living this long. 31 P age

35 Chart 5: What do you think are the chances you will live to be * (N=3,045)...age 78 or more...age 85 or more % 56% % 31% 50 70% 25% 50 70% 31% 25 45% 8% 25 45% 20% 0 20% 6% 0 20% 15% 0% 20% 40% 60% 80% 100% 0% 20% 40% 60% 80% 100% *Question SP1. What do you think are the chances you will live to be age 78? *Question SP2. What do you think are the chances you will live to be age 85? ATTITUDES ABOUT PLANNING, RETIREMENT, AND SOCIAL SECURITY Because an individual s attitudes about financial planning, retirement confidence, and confidence in the Social Security system may affect the age at which they claim Social Security benefits, respondents were asked to indicate the extent to which they agreed or disagreed with various statements about these issues. Attitudes About Planning First, they were asked to indicate their agreement with seven different statements about finances and planning. Overall, the statements that elicited agreement from the largest number of respondents were those that addressed the benefits of gathering information and planning ahead. For example, the statement with which the most respondents agreed was When making decisions about my personal finances, I like to get as many facts as I can before making the final decision even if it takes a while (92% strongly or somewhat agreed). The next most popular statement was If I plan ahead, things will often work out (86%). (See Table 8) Based on their agreement with these statements, it appears that few respondents make decisions about their personal finances quickly and few believe that planning is a waste of time. Specifically, the statements that elicited the least agreement were Planning is usually a waste of time because many things are a matter of luck anyhow (18% strongly or somewhat agreed) and When making decisions about my personal finances, I tend to make decisions quickly (24%). 32 P age

36 Table 8: Attitudes About Planning (Base: All Respondents) Strongly Agree Somewhat Agree Strongly / Somewhat Agree When making decisions about my personal 39% 53% 92% finances, I like to get as many facts as I can before making the final decision even if it takes a while. If I plan ahead, things will often work out. 21% 65% 86% I consider myself to be a planner and try to plan 21% 55% 76% carefully for the future. I consider myself to be knowledgeable about 11% 63% 74% financial issues. I prefer to focus on the present rather than 5% 34% 39% spending too much time planning for the future. When making decisions about my personal 2% 22% 24% finances, I tend to make decisions quickly. Planning is usually a waste of time because many things are a matter of luck anyhow. 2% 15% 18% Source: Question A1. Please indicate your level of agreement with each of the statements below. Response options were strongly agree, somewhat agree, somewhat disagree, and strongly disagree. Retirement Confidence When asked how confident they were that they and their spouse or partner will have enough money to live comfortably throughout their retirement years, about seven in ten (71%) respondents indicated that they were very confident (16%) or somewhat confident (54%). About three in ten (29%) were not confident, including about two in ten (19%) who were not too confident and one in ten (10%) who were not at all confident. Older respondents, those with higher educational attainment, and respondents with higher household incomes were most confident. Specifically, for respondents ages 62-66, 78 percent were either very or somewhat confident compared to 67 percent of year olds. Individuals with a post-graduate degree (79%) were more confident than those with less education. Specifically, just 67 percent of those with a high school degree or less, 69 percent of those with some college or an associate s degree, and 71 percent of those with a bachelor s degree were very or somewhat confident that they would have enough money to live comfortably throughout retirement. Further, confidence levels increased in a stepwise fashion with increases in household income. Just over half (52%) of those with an annual household income under $30,000 were confident while those with household incomes of $30,000-60,000 (67%), $60, ,000 (79%), and over $100,000 (85%) were more likely to be confident. The same pattern was seen for amount in savings. The most confidence was exhibited by those with $250,000 or more in savings (93%), followed by those with $100,000-$250,000 (77%), $25,000-$100,000 (65%), and finally less than $25,000 (49%). 33 P age

37 Among respondents age 62+, those who have already claimed their Social Security benefits are more confident in their retirement financial security than those age 62+ who have not claimed (81% versus 73%). Confidence in Social Security Because confidence in the Social Security system may also affect claiming age, respondents were also asked how confident they were that the Social Security system would be able to provide the benefits that are owed to them. Just under six in ten (59%) reported that they were either very or somewhat confident in the Social Security system, with only one in ten (10%) describing themselves as very confident and the remaining 48 percent describing themselves as somewhat confident. Older, most educated, and wealthier participants were more confident in the Social Security system s viability. Respondents ages were more likely than those ages (68% versus 54%) to indicate that they are very or somewhat confident that the Social Security system will be able to provide the benefits that are owed to them. Those with a post-graduate degree (64%) were more confident than those with only some college or an associate s degree (56%). Respondents with annual household incomes of $100,000 or more (64%) and from $60,000 to $100,000 (62%) were more likely to say they were somewhat or very confident than respondents with annual household incomes under $30,000 (54%). Similarly, respondents with higher levels of savings ($100,000-$250,000: 61%; $250,000+: 65%) were more confident than those with less than $25,000 in savings (53%). Concerns Related to Social Security Claiming Respondents were also asked to indicate which of two statements represented a more serious concern for them personally as they thought about their Social Security retirement benefits. One statement addressed a concern that waiting too long to claim benefits may cause one to miss out on receiving all of one s benefits, while the other statement addressed a concern that claiming benefits too early may lead to inadequate financial resources in old age. The majority of respondents (59%) indicated that the risk of having inadequate financial resources in old age represented a more serious concern to them, while four in ten (40%) were more concerned about the risk of not receiving all of the benefits that are owed to them. (See Table 9) Those who identified the risk of having inadequate financial resources in my old age as most pressing were more likely to have a bachelor s degree (64%), lower annual household incomes (under $30,000: 69%; $30,000-$60,000: 64%), or lower savings (less than $25,000: 71%; $25,000- $100,000: 67%). Additionally, as might be expected, this risk resonated more with respondents age 62 or older who had not yet claimed benefits than with respondents age 62 or older who had already claimed benefits. Conversely, compared to respondents with less education, lower incomes, and less savings, respondents with a post-graduate degree, those who make $60,000 or more in household income per year, and those who have $100,000 or more in savings were more likely to select I may miss out on getting all of the Social Security benefits that are owed to me if I wait too long. 34 P age

38 Table 9: Concerns Related to Social Security Claiming (Base: All Respondents) N=3,045 I may end up with inadequate financial resources in my old age if my 59% monthly Social Security check is too small to support me and I am unable to work I may miss out on getting all of the Social Security benefits that are 40% owed to me if I wait too long to start receiving benefits Question A4. Of the two issues below, which of the following is a more serious concern for you personally as you think about your Social Security retirement benefits? FINANCIAL PLANNING AND FINANCIAL LITERACY Planning Behaviors Determining Savings Needed for Retirement Just under half (46%) of respondents who had not yet retired indicated that they or their spouse/partner had tried to figure out how much money they would need to have saved in order to live comfortably in retirement, while just over half (53%) reported that they had not tried to figure this out. When asked to describe how they had tried to figure out the amount of savings that they would need, respondents who had tried to figure this out were most likely to indicate that they had arrived at an estimate on their own (58%) or received advice from a professional financial advisor (44%). Less common methods included receiving advice from family or friends (12%), guessing (12%), using an online calculator (17%), and reading or hearing that a certain amount is needed (17%). Determining Annual Income Needed in Retirement Respondents who had not yet retired were just as likely to have tried to figure out how much annual income they will need in retirement as they were to have tried to figure out how much savings they will need. Specifically, just under half (48%) of respondents who had not yet retired indicated that they had tried to determine their annual income needs in retirement, while just over half (52%) had not. When respondents were asked to describe how they had tried to figure out their annual income needs in retirement, the most common methods identified were similar to the methods used most frequently for determining savings needs. For example, the method identified most frequently was estimating on my own (65%) followed by receiving advice from a professional financial advisor (38%). Less common methods included guessing (10%), receiving advice from a family member (11%), reading or hearing that a certain amount is needed (15%), and using an online calculator (16%). 35 P age

39 Furthermore, most respondents who had taken one of these two steps determining savings needed for retirement or determining annual income needed in retirement had actually taken both steps. Specifically, of respondents who had tried to figure out their savings needs for retirement, over eighty percent (87%) had also tried to figure out their annual income needs in retirement. The same was true for respondents who had tried to figure out their annual income needs. Specifically, of respondents who had tried to figure out their annual income needs in retirement, over eighty percent (83%) had also tried to figure out how much savings they would need for retirement. Planning Household Finances Because the time horizon that people consider when making financial decisions may affect their decision about when to claim Social Security benefits, respondents were asked to indicate the time period that is most important to them when planning their household s saving and spending. The majority of respondents indicated that a time horizon beyond the next year was most important to them when planning their household s finances. Of the time horizons from which respondents were able to choose, the next 5-10 years was the time horizon selected by the largest share of respondents (35%) as most important to their planning. The next few years was the time horizon selected next most often (26%) as most important. As might be expected, attitudes concerning the most important time horizon varied by household income, savings, and age. Specifically, as income, savings, and age increased, the time horizon deemed most important became longer. For example, respondents with household incomes under $30,000 (25%) were more likely than those with higher incomes (9%) to identify the next few months as most important to their planning. Furthermore, those ages 62 and older (19%) were more likely than those ages (14%) to consider a time period longer than 10 years to be most important. The time horizon deemed most important also became longer as educational attainment increased. Financial Literacy Respondents were asked a few questions in an effort to gauge their financial literacy as financial literacy may also affect financial decisions. These questions dealt with compound interest, inflation, and asset diversification. Of these three concepts, the benefits of asset diversification appeared to be the most widely understood followed by the effect of compound interest and the effect of inflation. Asset Diversification The vast majority (86%) of respondents correctly answered that it is false to claim that buying a company stock usually provides a safer return than a stock mutual fund. 36 P age

40 Compound Interest Respondents were told to suppose that they had $100 in a savings account that was earning interest at a rate of 5 percent each year. They were then asked how much they should expect to have in this account after five years assuming that they never withdrew any money or interest from this account during the five-year period. Two thirds (66%) of respondents selected the correct answer of more than $125, which was significantly more than the share who selected the incorrect response options of exactly $125 (25%) or less than $125 (9%). Inflation Only about half (51%) of respondents understood that, if their after-tax income doubles and prices for everything also double at the same time, their buying power will remain the same. Interestingly, almost as many respondents (44%) incorrectly believed that these circumstances would lead to a reduction in their buying power. (See Table 10) Table 10: Suppose that, in the year 2015, your after-tax income has doubled and prices for everything have doubled too. In 2015, how much will you be able to buy with your income? (Base: All Respondents) More than today 5% The same (correct) 51% Less than today 44% Source: Question F8. SOURCES OF INFORMATION ABOUT SOCIAL SECURITY BENEFITS Respondents were asked whether they had used certain sources of information to learn about either Social Security benefits generally or when to start collecting benefits. The sources of information that they were asked about included the Social Security administration, an employer, AARP, a labor union, friends or family, a financial advisor, a financial services firm, financial publications, financial television shows, college classes, and the public library. For each information source that they had used, respondents were then asked to rate how useful it had been. Information Sources Used Of the eleven information sources about which respondents were asked, the Social Security Administration was by far the source that had been consulted by the largest share of respondents (62%). Friends or family, the second most commonly used information source, had been consulted by nearly half (47%) of respondents. Other than the Social Security Administration and friends or family, the next most popular information sources were a professional financial advisor (30%), financial magazines or books (27%), AARP (25%), and your current or former employer (23%). (See Chart 6) 37 P age

41 Chart 6: Use of information sources to learn about Social Security retirement benefits* (N=3,045) The Social Security Administration 62% Friends or Family Members 47% A Professional Financial Advisor 30% Financial Magazines or Books 27% AARP 25% Your Current or Former Employer 23% Financial Shows on TV 21% A Financial Services Firm 18% Public Library 8% Your Current or Former Labor Union 6% A Class or Seminar At A Local College 5% 0% 20% 40% 60% 80% 100% *Question SI1. Have you used any of the following information sources to learn about Social Security retirement benefits or when you should start collecting your benefits? 38 P age

42 The degree to which respondents had used the following sources of information varied by demographics: Social Security Administration Respondents ages as well as those with more education, higher household incomes, and more savings were most likely to have consulted the Social Security Administration for information about Social Security benefits. For example, over three in four (77%) respondents ages had consulted the Social Security Administration compared with just over half (55%) of respondents ages Similarly, about two in three (67%) respondents with at least some college education had used the Social Security Administration compared with just about half (52%) of respondents with a high school degree or less. Current or Former Employer, AARP, Financial Advisor, Financial Services Firm, Financial Shows on Television Respondents with more education, higher household incomes, and more savings were more likely than their counterparts with less education, lower incomes, and less savings to have consulted each of the following information sources to learn about Social Security benefits: an employer, AARP, a financial advisor, a financial services firm, or financial shows on television. For example, while only about one in six (16%) respondents with a high school degree or less had consulted an employer for this information, more than one in four (27%) respondents with at least some college education had consulted an employer. Similarly, over two in five (42%) respondents with a post-graduate degree had consulted a professional financial advisor compared with just over one in five (22%) respondents with a high school degree or less. Friends or Family Women (50%) were more likely than men (43%) to have consulted friends or family as were respondents ages (56%) compared with those ages (42%). Additionally, respondents with household incomes between $30,000 and $60,000 (52%) were more likely than those with incomes less than $30,000 (40%) as well as those with incomes above $60,000 (45%) to have consulted friends or family. Financial Magazines or Books Respondents with more education, higher household incomes, and more savings were more likely than their counterparts with less education, lower incomes, and less savings to have used financial magazines or books. Men were also more likely than women to have used these resources. For example, while about one in three (32%) men had used financial magazines or books, only one in five (21%) women had done so. Additionally, nearly half (47%) of respondents with a post-graduate degree had used financial publications compared with only about one in ten (11%) respondents with a high school degree or less, about one in four (26%) of those with some college education, and just over one third (37%) of those with a bachelor s degree. 39 P age

43 Public Library Use of public libraries to gather information about Social Security benefits was relatively rare (8% of respondents). Libraries were most likely to have been used by respondents with more education and by respondents with lower incomes. For example, respondents with post-graduate degrees (14%) were more likely than those without a bachelor s degree (5%) to have used a public library for this information, as were respondents with household incomes under $30,000 (11%) compared to respondents with incomes of $60,000 or more (6%). Current or Former Labor Union Very few respondents (6%) had consulted a labor union for information about Social Security, which may suggest that relatively few respondents had ever belonged to a labor union. Men (8%) were more likely than women (5%) to have consulted a labor union as were respondents with household incomes above $30,000 (7%) compared to those with incomes less than $30,000 (2%). Class or Seminar at a Local College Very few respondents (5%) had taken a class or seminar at a local college to learn about Social Security benefits. Likelihood of having done this increased with education and savings level. For example, as should be expected, those with at least some college education (7%) were more likely than those with a high school degree or less (1%) to have taken a college class or seminar for this purpose. Usefulness of Information Sources Respondents tended to rate the information sources that they had used very highly when asked to indicate how useful each source had been to them in their efforts to learn about either Social Security benefits generally or when they should start collecting benefits. In fact, all of the information sources were rated as either very useful or somewhat useful by nearly nine in ten of the respondents who had used them. Of all 11 information sources examined, the Social Security Administration was considered to be useful ( very or somewhat ) by the largest share of respondents who had used it (96%). However, each of the other information sources was viewed as useful by nearly as large a share of users. For example, in addition to the Social Security Administration, the following information sources were each considered to be useful by at least nine in ten users: a professional financial advisor (94%), a class or seminar at a local college (94%), AARP (93%), your current or former labor union (93%), financial magazines or books (92%), the public library (91%), and a financial services firm (90%). Furthermore, the remaining three information sources --your current or former employer, friends or family, and financial shows on television were each viewed as useful by at least 88 percent of users. Among users of each of these information sources, perceptions of usefulness varied little by demographics except in the following cases: 40 P age

44 Women who had consulted friends or family were more likely than their male counterparts to consider friends or family to be a useful source of information (92% of female users vs. 87% of male users). Among individuals who had used a professional financial advisor, those with savings levels of at least $25,000 were more likely than those with less savings to view the advisor as useful (95% of users with savings of $25,000+ vs. 81% of users with savings less than $25,000). Men who had viewed financial shows on television were more likely than their female counterparts to consider the shows to be useful (93% of male viewers vs. 85% of female viewers). Additionally, among financial television show viewers, individuals with household incomes of less than $100,000 were more likely than those with household incomes of at least $100,000 to consider the shows to be useful (91% of viewers with incomes under $100,000 vs. 79% of viewers with incomes of $100,000+). Methods Used to Learn About Social Security In addition to asking respondents which sources of information they had used to learn about Social Security benefits, the survey also asked respondents which methods they had used. Specifically, respondents were asked whether they had looked for information on the Internet, talked with people face to face, and/or placed telephone calls in an effort to gather information. Talking with people face to face appears to be the most common method used to learn about Social Security, with just over half (53%) of respondents reporting that they have used this technique. Looking for information on the Internet was the next most popular method, utilized by just over two in five (42%) respondents. The least-utilized method was the telephone, with fewer than one in five (18%) respondents indicating that they have used this approach. Methods Likely to Use to Learn about Social Security Respondents who had not yet used any of these three methods to gather information about Social Security benefits were asked how likely they might be to use these methods in the future if they were to try to learn more about Social Security. The Internet was the method that respondents most often assumed that they would use (73% of those who had not yet used any of these methods), followed by face-to-face conversations (67%) and telephone (43%). Trustworthiness of Information Sources As might be expected, the Social Security Administration was viewed as the most trustworthy source of information about Social Security benefits. Specifically, of the 11 information sources addressed by the survey, it was the only source deemed very or somewhat trustworthy by more than nine in ten respondents (93%). The next most trusted sources of information on Social Security included the public library (88%), a professional financial advisor (84%), and a class or seminar at a local college (84%). (See Chart 7) The sources of information receiving the lowest marks on trustworthiness included financial shows on television (54%) and your current or former labor union (54%) It is worth noting that all respondents, regardless of any current or former affiliation with a labor union, were asked to rate how trustworthy they viewed labor unions as a source of information on Social Security. The survey did not restrict this question to respondents who had ever belonged to a labor union and such a question about current or former membership in labor unions was not included in the survey. Therefore, it is 41 P age

45 Chart 7: Trustworthiness rating* The Social Security Administration 38% 55% AARP 27% 50% Public Library 26% 62% A Professional Financial Advisor 26% 58% A Class or Seminar At A Local College 20% 64% Your Current or Former Employer 18% 54% Somewhat Trustworthy A Financial Services Firm Friends or Family Members 17% 16% 59% 59% Very Trustworthy Your Current or Former Labor Union 9% 45% Financial Magazines or Books 8% 65% Financial Shows on TV 4% 50% 0% 20% 40% 60% 80% 100% *Question SI5. Please rate how trustworthy you feel each of the following organizations would be as a source of information about Social Security benefits. possible that the lower trust ratings of labor unions are attributable in part to the fact that many respondents may have never belonged to a labor union rather than a reflection of direct experiences with labor unions. 42 P age

46 EXPECTED SOURCES OF INCOME IN RETIREMENT When respondents were asked to indicate the degree to which they expected to rely on certain sources of income in retirement, Social Security was the source of income most commonly expected to be a major source of retirement income. In fact, Social Security was the only source from which a majority of respondents (57%) expected to draw a major portion of their retirement income. After Social Security, an employer-provided pension or cash balance plan was the next most frequently expected major source of retirement income (38%) followed by an employersponsored retirement savings plan (30%). (See Chart 8) Chart 8: Major source of income for retirement* (n=3,045) Social Security 57% An employer provided traditional pension or cash balance plan, with the benefit based on salary and years of service. An employer sponsored retirement savings plan, such as a 401(k), tax deferred annuity or 403(b), thrift savings, money purchase, or profit sharing plan 38% 30% Employment 25% Other personal savings or investments outside of a retirement account, such as mutual funds, stocks, certificates of deposit, or annuities 24% An Individual Retirement Account or IRA 18% 0% 20% 40% 60% 80% 100% *Question D. 16. For each of the following possible sources of income in retirement, please indicate whether you expect it to be a major source of income, a minor source of income or not a source of income in your (your and your spouse s) retirement. 43 P age

47 Of those who expect that Social Security will be a source of their income (major or minor source) in retirement, about one in five (21%) expect that it will represent less than a quarter of their retirement income, two in five (41%) expect that it will represent at least a quarter but less than half of their retirement income, just over three in ten (32%) expect that it will contribute more than half of their retirement income, and a few (7%) don t know how much of their retirement income it will represent. Those who expect Social Security to contribute more than 75 percent of their retirement income are more likely to be female, have a high school degree or less, have relatively low household incomes, or have relatively low savings. These same individuals are also more likely than others to say that they don t know what percent of their retirement income will come from Social Security. For example, among individuals who expect Social Security to be a source of their retirement income, more than one in four (27%) of those with household incomes under $30,000 expect that Social Security will make up more than 75 percent of their retirement income compared with just 8 percent of those with incomes between $30,000 and $60,000, 4 percent of those with incomes between $60,000 and $100,000, and just 2 percent of those with household incomes of $100,000 or more. Additionally, another 14 percent of these individuals with household incomes under $30,000 say that they don t know how much of their retirement income will come from Social Security, compared with just 5 percent or less of those with higher incomes. RECENT LIFE EXPERIENCES Because recent life changes or events may affect decisions about when to claim Social Security benefits, respondents were asked whether they had experienced certain life changes or events within the past three years. Of the 17 different experiences about which respondents were asked, the loss of a close relative or friend was the most common experience, with just over half (51%) of respondents indicating that they had endured such a loss within the past three years. Serious financial problems represented the next most common experience (21%), followed by losing your job (15%), losing your health benefits or having your health benefits significantly reduced (14%), becoming responsible for the care of a parent or an adult relative other than a spouse or partner (13%), having an adult child move back home (13%), and making a major career change (11%). (See Chart 9) ASSISTANCE PROVIDED TO RELATIVES OR FRIENDS Previous research has suggested that caregiving responsibilities can cause people to leave the workforce earlier than they otherwise would. For this reason, individuals with caregiving responsibilities may be more inclined to take Social Security benefits at an early age. Similarly, people with financial responsibility for others may also be inclined to take benefits early if those benefits help them meet those financial obligations. In order to examine these hypotheses, the survey asked respondents whether they provided financial assistance to any relatives or close friends on a regular basis as well as whether they provided unpaid help or care to any relatives or close friends on a regular basis. 19 The MetLife Caregiving Cost Study: Productivity Losses to U.S. Businesses. MetLife Mature Market Institute & National Alliance for Caregiving, July Caregiving in the US: National Alliance for Caregiving & AARP, P age

48 Chart 9: Significant life changing events in the past three years* (n=3,045) Had a close relative or friend die 51% Experienced serious financial problems Lost your job Lost your health benefits or had your health benefits significantly reduced Became responsible for the care of a parent or an adult relative other than a spouse or parent Had an adult child move back home Made a major career change Retired from job earlier than you expected Experienced a significant decline in health or developed a major illness Lost your pension or had your pension benefits significantly reduced Started providing care to a grandchild or other child Experienced a significant improvement in health or recovered from a major illness Experienced a significant increase in income or wealth Began to take care of spouse or partner on a regular basis Had a spouse or partner die Gotten a divorce Gotten married 21% 15% 14% 13% 13% 11% 9% 9% 8% 8% 8% 6% 5% 3% 3% 2% 0% 20% 40% 60% 80% 100% *Question D.4. In the past three years, have you experienced any of the following major life changes or events? 45 P age