United States Government Accountability Office GAO Report to the Chairman, Special Committee on Aging, U.S. Senate April 2012 UNEMPLOYED OLDER WORKERS Many Experience Challenges Regaining Employment and Face Reduced Retirement Security GAO-12-445 April 2012 UNEMPLOYED OLDER WORKERS Many Experience Challenges Regaining Employment and Face Reduced Retirement Security Highlights of GAO-12-445, a report to the Chairman, Special Committee on Aging, U.S. Senate Why GAO Did This Study What GAO Found The number of workers age 55 and As with many other demographic groups, older workers’ unemployment overall and over experiencing long-term long-term unemployment rates have increased dramatically since the recession unemployment has grown substantially began in 2007. In December 2011, the unemployment rate for older workers was 6.0 since the recession began in 2007. percent, up from 3.1 at the start of the recession, but down from its peak of 7.6 This raises concerns about how long- percent in February 2010. In particular, long-term unemployment rose substantially, term unemployment will affect older and at a greater rate for older than younger workers. By 2011, 55 percent of workers’ reemployment prospects and unemployed older workers had been actively seeking a job for more than half a year future retirement income. (27 weeks or more). Meanwhile, the long-term trend of rising labor force participation rates among older workers has continued, with the recession possibly amplifying this In light of these developments, GAO trend. examined (1) how older workers’ employment status has changed since Long-term unemployment can put older workers at risk of deferring needed medical the recession, (2) what risks care, losing their homes, and accumulating debt. The experts and staff GAO unemployed older workers face and interviewed at some one-stop career centers, as well as the unemployed older what challenges they experience in workers who participated in GAO’s focus groups, identified employer reluctance to finding reemployment, (3) how long- hire older workers as a key challenge that older workers face in finding term unemployment could affect older reemployment. They also identified out-of-date skills, discouragement and workers’ retirement income, and (4) depression, and inexperience with online applications as reemployment barriers for what other policies might help them older workers. Some one-stop staff who serve older workers told GAO that providing return to work and what steps the the type of assistance some older workers need to address these unique challenges Department of Labor (Labor) has taken can be very time-consuming. (For audio clips from GAO’s focus groups with to help unemployed older workers. unemployed older workers, use this link: http://www.gao.gov/multimedia/video/#video_id=590295) To conduct this work, GAO analyzed Long-term unemployment can substantially diminish an older worker’s future nationally representative datasets, led retirement income in several ways. First, it can force a worker to stop working and focus groups of unemployed older stop saving for retirement earlier than the worker had planned. Second, long-term workers, modeled how job loss affects unemployment can lead individuals to draw down their retirement savings to cover retirement income, and interviewed living expenses while they are unemployed, which was a common life experience experts and federal and local officials. described by GAO’s focus group participants. GAO illustrated how a hypothetical worker who had $70,000 in retirement savings at age 55 and withdrew 50 percent of What GAO Recommends those savings during a 2 year period of unemployment, would need about another 5 To foster the employment of older ½ years of work and saving to rebuild the retirement account to the level it had been workers, we recommend that the before unemployment began. In addition, long-term unemployment can motivate Secretary of Labor consider what older workers to claim early Social Security retirement benefits, which will result in strategies are needed to address the lower monthly benefits for workers and their survivors for the rest of their lives. unique needs of older job seekers, in Experts GAO interviewed selected various policies that have been proposed to help light of recent economic and address unemployed older workers’ reemployment challenges. Experts selected technological changes. these policies from a broad list GAO compiled from previous academic studies. For Labor agreed with our example, two of the policies that experts selected would provide incentives such as recommendation. GAO received temporary wage or training subsidies for employers to hire long-term unemployed technical comments on a draft of older workers. In the current context of high unemployment and slow job creation, the this report from Labor and the Social impact of most of these policies is likely to be muted by limited job openings. After an Security Administration, and interagency Taskforce issued its report on the aging of the American workforce in incorporated them as appropriate. 2008, Labor implemented several strategies the report recommended, but since the recession started, Labor shifted focus to responding to increased demand for services. As the economy improves, Labor could refocus on older job seekers and consider what additional strategies would help address their unique reemployment View GAO-12-445. For more information, challenges, in light of recent economic and technological changes. contact Charlie Jeszeck at (202) 512-7215 email@example.com. United States Government Accountability Office Contents Letter 1 Background 3 Older Workers’ Employment Prospects Have Deteriorated 9 Unemployed Older Workers Face Many Challenges Coping with Unemployment and Regaining Employment 26 Job Loss Can Diminish Retirement Income Prospects 33 Experts Selected Policies to Consider as the Economy Recovers, and Labor Has Taken Some Steps to Help Older Workers 47 Conclusions 56 Recommendation 57 Agency Comments and Our Evaluation 57 Appendix I Objectives, Scope, and Methodology 59 Appendix II Workforce Status of Workers Displaced between 2005-2007 and 2007-2009 in January 2008 and January 2010 70 Appendix III Additional Figure Notes for Employment Figures, Including Statistical Significance Tests 71 Appendix IV Quotes from Focus Groups with Long-Term Unemployed Older Workers 74 Appendix V Percentage of the Population 50 and Older, but Less than Full Retirement Age, with Initial Dispositions and Awards of Social Security Disability Insurance, 2000-2010 76 Appendix VI Proposed Policies Presented to External Experts 77 Page i GAO-12-445 Unemployed Older Workers Appendix VII Comments from the Department of Labor 79 Appendix VIII GAO Contact and Staff Acknowledgments 81 Related GAO Products 82 Tables Table 1: Experts’ Views on Key Strengths and Limitations of Selected Policy Proposals to Help Address Perceived Employer Reluctance to Hire Older Workers 49 Table 2: Experts’ Views on the Strengths and Limitations of Selected Policy Proposals to Enhance Reemployment Assistance Specifically Targeted to Older Workers 51 Table 3: Experts’ Views on the Strengths and Limitations of Selected Policy Proposals to Encourage Older Workers to Obtain Reemployment as Quickly as Possible 53 Table 4: Comparison of Seasonally Adjusted and Unadjusted Unemployment Rates for Workers Age 55 and Older, Selected Months 61 Table 5: Selected Quotes from Focus Groups with Long-Term Unemployed Older Workers 74 Figures Figure 1: Sources of Aggregate Income for Households with Someone Aged 65 or Older, 2008 7 Figure 2: Estimated Unemployment Rates by Age, 2007-2011 11 Figure 3: Estimated Number of Unemployed and Underemployed Older Workers (55 and Over), 2007-2011 13 Figure 4: Growth in Estimated Long-Term Unemployment of Older Workers (55 and Over), 2007-2011 15 Figure 5: Difference between the Estimated Number of Unemployed of All Ages and the Estimated Number of Available Jobs, 2007-2011, and Various Explanations for High Unemployment 16 Page ii GAO-12-445 Unemployed Older Workers Figure 6: Estimated Duration of Unemployment for Older Workers (55 and Over) by Industry, 2007 and 2011 17 Figure 7: Estimated Unemployment Rates by Demographic Group for Older Workers (55 and Over), 2007 and 2011 19 Figure 8: Estimated Duration of Unemployment by Age and by Gender, 2007 and 2011 20 Figure 9: Estimated Labor Force Participation Rates by Age, 1948- 2011 22 Figure 10: Estimated Percentage of Reemployed Displaced Workers Who Earned Less on Their New Full-Time Jobs than on Their Previous Jobs, January 2010 26 Figure 11: Potential Effect That Fewer Years of Work Can Have on Retirement Benefits for Workers with Median-Level Retirement Benefits from Employer-Sponsored Plans and Median-Level Retirement Benefits from Social Security 37 Figure 12: Potential Effect That Fewer Years of Work Can Have on Social Security Retirement Benefits of Workers without Employer-Sponsored Retirement Plans 38 Figure 13: Example of the Effect of Early Claiming on Monthly Social Security Retirement Benefits 39 Figure 14: How Drawdowns from Retirement Savings during Unemployment Can Affect Amounts Saved at Time of Retirement if a Worker Became Reemployed and Resumed Saving 43 Figure 15: Estimated Levels of Retirement Savings and Types of Plan Participation for Employed Workers 55 and Over and Their Households, 2007 45 Page iii GAO-12-445 Unemployed Older Workers Abbreviations AoA Administration on Aging BLS Bureau of Labor Statistics CMS Centers for Medicare and Medicaid Services CPS Current Population Survey DB defined benefit DC defined contribution DWS Displaced Worker Supplement EBRI Employee Benefit Research Institute GATE Growing America Through Entrepreneurship HRS Health and Retirement Study IRA individual retirement account JOLTS Job Openings and Labor Turnover Survey NBER National Bureau of Economic Research OASDI Old-Age, Survivors, and Disability Insurance SBA Small Business Administration SCF Survey of Consumer Finances SCSEP Senior Community Service Employment Program SSA Social Security Administration SSI Supplemental Security Income UI Unemployment Insurance WIA Workforce Investment Act of 1998 This is a work of the U.S. government and is not subject to copyright protection in the United States. The published product may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. Page iv GAO-12-445 Unemployed Older Workers United States Government Accountability Office Washington, DC 20548 April 25, 2012 The Honorable Herb Kohl Chairman Special Committee on Aging United States Senate Dear Mr. Chairman: The recent recession in 2007-2009 was the worst since the Great Depression, and has been characterized by historically high levels of long-term unemployment. 1 While it is crucial that the nation help people of all ages return to work, long-term unemployment has particularly serious implications for older Americans (age 55 and over). Older workers’ job loss threatens not only their immediate financial security, but their ability to support themselves during retirement. While unemployed, older Americans may stop saving for retirement or tap into retirement savings to pay for critical living expenses. And while working longer may be the best solution for those approaching retirement with exhausted or otherwise inadequate savings, this strategy depends upon older workers being able to find and retain employment. To the extent that unemployed older workers face unique reemployment challenges, their ability to get the jobs they need to support themselves and protect or rebuild their retirement savings could be limited. Given your interest in the employment and retirement prospects of older workers—those aged 55 and over—since the onset of the recession, we examined (1) how the employment status of older workers has changed since the recession, (2) older workers’ financial risks from long-term unemployment and challenges in finding new jobs, (3) how periods of long-term unemployment might affect older workers’ retirement income, and (4) what other policies might help unemployed older workers regain 1 The recession of 2007-2009 started in December 2007 and ended in June 2009, according to the Business Cycle Dating Committee of the National Bureau of Economic Research (NBER). According to NBER, “a recession is a significant decline in economic activity spread across the economy, lasting more than a few months, normally visible in production, employment, real income, and other indicators. A recession begins when the economy reaches a peak of activity and ends when the economy reaches its trough.” In addition, this recession occurred in the context of a significant decline in major financial markets, which dramatically reduced the value of major assets. Page 1 GAO-12-445 Unemployed Older Workers employment and what steps the Department of Labor (Labor) has taken to help unemployed older workers. To examine changes in the employment status of older workers since the start of the recession, we analyzed nationally representative unemployment and demographic data from the Bureau of Labor Statistics (BLS), including 2007 through 2011 data from the Current Population Survey (CPS), the Job Openings and Labor Turnover Survey (JOLTS), and the 2008 and 2010 Displaced Worker Supplement (DWS). To learn about older workers’ financial risks from long-term unemployment and challenges in finding new jobs, we conducted focus groups with unemployed older workers in four metropolitan areas and also interviewed staff at one-stop career centers in each of the four areas. 2 (For audio clips from GAO’s focus groups with unemployed older workers, use this link: http://www.gao.gov/multimedia/video/#video_id=590295.) Further, we interviewed experts on older workers’ issues and reviewed studies. To assess how periods of long-term unemployment might affect older workers’ retirement income, we used microsimulation models to estimate monthly retirement income for workers who stopped work at different ages. 3 We also analyzed data on retirement savings from the nationally representative 2007 Survey of Consumer Finances and interviewed officials at the Social Security Administration (SSA). 4 To identify what policies might help unemployed older workers regain employment and what Labor has done to help older workers, we interviewed experts on policy proposals previously identified through a review of the literature and interviewed Labor officials. 2 The Workforce Investment Act of 1998 provided for the establishment of local one-stop centers to provide access to employment and training services under a number of programs, including those administered by the Departments of Labor, Education, Health and Human Services, and Housing and Urban Development. Pub. L. No. 105-220, § 121, 112 Stat. 936, 963. 3 Our simulations of retirement income include income from employer-sponsored retirement plans and Social Security retirement benefits. We used microsimulation models under license from the Policy Simulation Group, a private contractor. More information about these microsimulation models is at www.polsim.com. 4 For our analysis of retirement savings using the 2007 Survey of Consumer Finances, we counted participation in a defined benefit or defined contribution plan at a current employer, earned benefits from a defined benefit plan from a past employer, a defined contribution account from a past employer, a Keogh account, or an individual retirement account. Page 2 GAO-12-445 Unemployed Older Workers We conducted this performance audit from October 2010 through April 2012 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence we obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. For more information on our scope and methodology, see appendix I. Background Unemployment Insurance Several different programs, including the federal-state Unemployment and Employment and Insurance (UI) program and employment and training programs Training Programs Serving administered by Labor, help unemployed workers, and one program is specifically for low-income individuals 55 and over. First, the UI program Unemployed Older provides eligible unemployed individuals temporary benefits that partially Workers replace their lost wages. The UI program is a federal-state program that is generally funded through taxes on employers and, in some states, employee contributions. Eligibility requirements and benefit levels vary by state within federal guidelines, but a recently enacted federal law will require individuals to be able, available, and actively looking for work as a condition of eligibility for unemployment benefits. 5 UI benefits are generally available to eligible unemployed workers for up to 26 weeks, 6 5 This requirement, added by the Middle Class Tax Relief and Job Creation Act of 2012, Pub. L. No. 112-96, § 2101, 126 Stat. 156, 159, will take effect after the end of the first session of each state legislature that begins after the date of enactment, February 22, 2012. 6 For evidence that unemployment benefits are generally available for up to 26 weeks, see David Bradley, Benjamin Collins, Katelin Isaacs, Janemarie Mulvey, Federal Programs Available to Unemployed Workers. Congressional Research Service, RL34251 (Washington D.C., January 2012). However, the actual duration of UI benefits will vary by state and by individual. See GAO, Unemployment Insurance: Economic Circumstances of Individuals Who Exhausted Benefits, GAO-12-408 (Washington, D.C.: Feb. 17, 2012). Page 3 GAO-12-445 Unemployed Older Workers but extended benefits are sometimes made available to those who exhaust these benefits, as has occurred during the recent recession. 7 Second, the Workforce Investment Act of 1998 (WIA) established the adult and dislocated worker programs, which authorize grants to states for a broad range of employment and training activities, including job search assistance, assessment, and training for eligible individuals. 8 States that receive funds under WIA must report on the performance of these programs using performance measures that gauge program results for participants in the areas of job placement, retention, and earnings, among others. 9 States are held accountable by Labor for their performance and may receive incentive funds or financial sanctions based on whether they meet expected performance levels for each measure. 10 Last, the Senior Community Service Employment Program (SCSEP) provides subsidized, community service-based on-the-job training for low- income individuals age 55 and over who are unemployed, with an 7 Congress has extended the maximum period for receiving UI benefits through temporary federal programs during a number of economic downturns, including the recent recession. Congress also provided for increased weekly UI benefit amounts and increased federal support to states to finance extended UI benefits under the American Reinvestment and Recovery Act of 2009 (Recovery Act). Pub. L. No. 111-5, §§ 2002-05, 123 Stat. 115, 437- 44. During the recession of 2007-2009, Congress increased the maximum period for receipt of UI benefits; as a result, some long-term unemployed workers in certain states may be eligible to receive UI benefits for up to 99 weeks. The Middle Class Tax Relief and Job Creation Act of 2012, enacted on February 22, 2012, will reduce this maximum to 93 from 99. The reduction will be phased in between May and September 2012. Pub. L. No. 112-96, § 2122, 126 Stat. 156, 163-66 (2012). For more information on UI and the Recovery Act, see GAO, Worker and Family Assistance: Unemployment Insurance Measures Included in the American Recovery and Reinvestment Act of 2009, as of July 2009, GAO-09-942R (Washington, D.C.: July 27, 2009). 8 Under WIA, dislocated workers include, among others, individuals who (1) have been terminated or laid off or who have received a notice of termination or layoff, (2) are eligible for or have exhausted UI benefits or have demonstrated attachment to the workforce but are ineligible for UI benefits because of insufficient earnings or having worked for a noncovered employer, and (3) are unlikely to return to a previous industry or occupation. Dislocated workers also include individuals who were self-employed but are unemployed as a result of general economic conditions or natural disasters. For the complete definition, see 29 U.S.C. § 2801(9). 9 29 U.S.C. § 2871(b),(d). 10 29 U.S.C. § 2871(g); 20 U.S.C. § 9273. Page 4 GAO-12-445 Unemployed Older Workers emphasis on those who have poor employment prospects. 11 Labor evaluates the performance of state and local areas receiving SCSEP funds using performance measures similar to those used for the WIA adult and dislocated worker programs. 12 Structural versus Cyclical Economists often classify involuntary unemployment as either structural Unemployment or cyclical. 13 Structural unemployment arises when barriers—such as skills or geographical mismatches—prevent workers from matching their skills to available jobs. In contrast, cyclical unemployment arises when there is a decrease in the overall demand for goods and services in an economy. As a result of such decreases, employers may temporarily lay off workers or cut back their employees’ hours until the economy improves. Research suggests that much of the increase in unemployment since 2007 has been cyclical rather than structural. 14 11 To be eligible to participate in SCSEP, individuals must have a family income of no more than 125 percent of the federal poverty guidelines prepared by the Department of Health and Human Services and approved by the Office of Management and Budget. 42 U.S.C. § 3056p(a)(3), 20 C.F.R. § 641.500. In 2012, an individual living in the 48 contiguous states and D.C. must have an income of not more than $13,963 to be eligible to participate in SCSEP. Household income for an individual in a household of two must not be more than $18,913 in order for the individual to be eligible to participate in the program. 12 42 U.S.C. § 3056k. 13 Another type of unemployment—frictional unemployment—is generally due to voluntary job shifts and typically involves shorter unemployment spells. 14 Linda Levine, The Increase in Unemployment since 2007: Is it Cyclical or Structural? Congressional Research Service, R41785 (Washington, D.C.: November 2011). Page 5 GAO-12-445 Unemployed Older Workers Labor’s Taskforce Report In 2006, Labor convened an interagency 15 Taskforce on the Aging of the on the Aging of the American Workforce (the Taskforce) 16 and issued a report in 2008 American Workforce highlighting strategies for increasing older Americans’ workforce participation. 17 The Taskforce considered a broad range of issues concerning the aging of the American workforce, including legal and regulatory barriers that could prevent older workers from participating in the workforce, flexible work arrangements, and tools and technical assistance to support older workers’ employment. The Taskforce report recommended that Taskforce agencies continue to work together to implement strategies it had identified, such as to coordinate research and demonstration agendas and inventory legal and regulatory barriers that could limit older workers’ employment. The Taskforce’s recommendations mainly dealt with potential actions Taskforce agencies could undertake to help older Americans remain in or reenter the workforce or pursue self- employment. Sources of Retirement While income in retirement comes from a variety of sources, in the Income aggregate, Social Security retirement benefits are the largest source of retirement income for households with someone aged 65 or older. Other financial assets such as income from employer-sponsored retirement plans, private savings, and assets such as home equity are important sources of retirement income for many. (See fig. 1.) 15 The Taskforce included senior representatives from nine federal agencies whose activities affect the lives of older Americans. These agencies were the Departments of Commerce, Education, Health and Human Services, Labor, Transportation, and the Treasury, along with the Equal Employment Opportunity Commission, the Small Business Administration, and SSA. 16 Labor officials convened this task force to respond to a request from the U.S. Senate Special Committee on Aging and GAO recommendations in two reports: GAO, Older Workers: Demographic Trends Pose Challenges for Employers and Workers, GAO-02-85 (Washington, D.C.: Nov. 16, 2001), and Older Workers: Labor Can Help Employers and Employees Plan Better for the Future, GAO-06-80 (Washington, D.C.: Dec. 5, 2005). 17 In 2008, this task force published its final report, Report of the Taskforce on the Aging of the American Workforce (Washington, D.C.: February 2008) http://www.doleta.gov/reports/FINAL_Taskforce_Report_2-11-08.pdf. Page 6 GAO-12-445 Unemployed Older Workers Figure 1: Sources of Aggregate Income for Households with Someone Aged 65 or Older, 2008 Note: “Household” here refers to what SSA identifies as aged units—either a married couple living together or an unmarried person. The age of a married couple is the age of the husband if he is 55 or older; if the husband is younger than 55, the age of the married couple is the age of the wife. Thus a married couple is considered to be 65 or older if the husband is 65 or older or if the husband is younger than 55 and his wife is 65 or older. Data reported by SSA for pension income includes regular payments from defined benefit and defined contribution plans and from individual retirement accounts (IRA) and Keogh accounts, but nonregular (nonannuitized or lump sum) withdrawals from IRA, Keogh, and defined contribution plans are not included. Social Security income includes retirement, auxiliary (such as spousal), survivors, and disability benefits. Data reported for income from assets include interest income; income from dividends, rents, or royalties; and estates or trusts. Other income includes noncash benefits, veterans’ benefits, UI benefits, workers’ compensation, and personal contributions. Income from others is excluded. The 95 percent confidence intervals for the share of aggregate income are 35.9 to 37.1 percent for Social Security, 29.1 to 30.3 for employment earnings, 17.9 to 18.9 for pension and annuity income, 12.3 to 13.1 for income from assets, 1.9 to 2.3 for other, and 0.5 to 0.7 for cash public assistance. Social Security retirement benefits are paid to eligible workers under the Old-Age, Survivors, and Disability Insurance (OASDI) program administered by SSA. The level of monthly retirement benefits an individual will receive depends on factors such as work and earnings history and the age at which the beneficiary chooses to begin receiving Page 7 GAO-12-445 Unemployed Older Workers benefits. 18 Generally, individuals may begin receiving Social Security retirement benefits at age 62; however, the payments will be lower than if they wait to receive benefits at their full retirement age, which varies from 65 to 67, depending on the individual’s birth year. 19 In contrast, the monthly benefit is increased for workers who delay receiving benefits beyond their full retirement age up to age 70. 20 Employees and employers pay payroll taxes that finance Social Security benefits. Social Security also provides benefits to eligible workers who become disabled before reaching retirement age, as well as children, spouses, and widow(er)s of eligible workers. Benefits are based upon a common formula but are calculated differently for the different beneficiary types. 21 Employer-sponsored retirement plans fall into two broad categories: defined benefit (DB) plans and defined contribution (DC) plans. DB plans promise to provide a benefit that is determined by a formula based on particular factors specified by the plan, such as salary or years of service. Typically, DB plans provide annuity payments to retirees on a monthly basis that continue as long as the recipient lives. 22 Under DC plans, workers and employers may make contributions into individual accounts. 23 At retirement, participants’ distribution options vary depending on the plan, but often include leaving their money in the plan, taking a full or partial distribution, or purchasing an annuity. In order to preserve the tax benefits from their DC plan savings, many participants chose to roll plan savings into an individual retirement account (IRA). IRAs are personal retirement savings arrangements that allow individuals to make 18 42 U.S.C. §§ 402, 415. 19 42 U.S.C. § 402(q)(1); 20 C.F.R. §§ 404.409 to 404.410. 20 42 U.S.C. § 402(w); 20 C.F.R. § 404.313. 21 In this report we use the term “Social Security retirement benefits” to refer to an individual’s retirement (old-age) benefits, and not other Social Security benefits such as spousal benefits or disability benefits, unless otherwise noted. 22 A DB plan may also provide benefits to a surviving spouse, if the plan participant is married and took these benefits. 23 The most common type of DC plan is the 401(k) plan, which typically allows workers to choose to contribute a portion of their pretax compensation to the plan. Some 401(k) plans may also provide for employer contributions, and Roth 401(k) plans may accept after-tax employee contributions. Page 8 GAO-12-445 Unemployed Older Workers contributions to an individual account and receive favorable tax treatment. 24 Like many other demographic groups, older workers have faced dramatic Older Workers’ increases in unemployment and long-term unemployment since the Employment recession began in 2007. For example, only about a third of older workers displaced from 2007 to 2009 had found full-time work by 2010, and those Prospects Have who did sustained greater earnings losses than did reemployed younger Deteriorated workers. Nonetheless, older workers’ labor force participation has continued to increase despite the worst labor market in decades. For example, the proportion of displaced workers age 65 or over who retired was substantially lower among those displaced during the recession than among those displaced prior to the recession. 25 Multiple Measures Show Unemployment rates for workers of all ages, including those 55 and over, Unemployment among have risen dramatically since the start of the recent recession in Older Workers Rose December 2007. As shown in figure 2, the seasonally unadjusted unemployment rate for older workers increased from 3.1 percent in Dramatically after 2007 December 2007 to a high of 7.6 percent in February 2010, before it 24 The tax treatment differs depending on the type of IRA. For example, with traditional IRAs, individuals who meet certain conditions can take an income tax deduction on some or all of the contributions they make to their IRAs, but they must pay taxes on amounts they withdraw from the IRA. Individuals below certain income limits may also contribute to Roth IRAs, which do not provide an income tax deduction on contributions, but permit tax- free withdrawals. 25 In this report, displaced workers are defined as persons 20 years or older who lost or left jobs because their plant or company closed or moved, there was insufficient work for them to do, or their position or shift was abolished. We analyzed displaced workers using the CPS Displaced Worker Supplement (DWS). Displaced workers have lost a job in the past 3 years; however, they may be unemployed, employed, or not in the labor market at the time of the survey. See appendix I for more information about displaced workers and the DWS. In our analysis of displaced workers, “retired” is defined to include only people who are no longer in the labor force. Therefore, this definition does not include retired persons who have part-time work. See appendix II for more data on the labor force status of displaced workers in 2008 and 2010. Page 9 GAO-12-445 Unemployed Older Workers decreased to 6.0 percent in December 2011. 26 Further, as shown in figure 3, the seasonally unadjusted number of unemployed older workers peaked at 2.3 million in February 2010 and decreased to approximately 1.9 million by December 2011, compared with around 839,000 when the recession began in December 2007. As in prior recessions, smaller percentages of workers age 55 and over became unemployed. Some researchers attribute older workers’ lower unemployment rates to the fact that older workers tend to have longer job tenure and are consequently less likely to be laid off than younger workers. 27 26 This figure, along with all others describing characteristics of workers, is based on sample data and subject to sampling error. For example, we are 95 percent confident that the unemployment rate for workers age 55 and older was between 5.5 and 6.4 percent in December 2011. See appendix I for a description of sampling error and the surveys relied upon. In this report, confidence intervals for estimates and statistical tests for differences between estimates are presented in graphs, tables, or footnotes in the more detailed sections presenting the estimates, or in appendix III when indicated. Estimated labor force statistics in this report are based on analysis of microdata, which beginning in January 2011 may diverge slightly from BLS published estimates; see appendix I for more information. Because we analyze a variety of labor force outcomes for several subgroups of the population in this report, we did not attempt to seasonally adjust any of the estimates. See appendix I for more information about seasonally unadjusted data. 27 A recent study, however, suggests that older workers with less than 4.6 years of tenure are actually more likely to be laid off than their otherwise similar younger counterparts. See Richard Johnson and Corinna Mommaerts, Age Differences in Job Loss, Job Search, and Reemployment, the Urban Institute (Washington D.C.: January 2011). Page 10 GAO-12-445 Unemployed Older Workers Figure 2: Estimated Unemployment Rates by Age, 2007-2011 Note: Estimates have 95 percent confidence intervals within plus or minus 0.5 percentage points of the estimate itself. Recession dates obtained from NBER. Estimates are not seasonally adjusted. Alternative measures show that the number of discouraged and underemployed older workers has also increased dramatically since the start of the 2007-2009 recession. According to the broadest measure, approximately 4.2 million older workers were unemployed or underemployed in January 2011, and approximately 3.65 million remained unemployed or underemployed in December 2011 (see fig. 3). In addition to the officially unemployed, this measure includes individuals who wanted to work and were available for work but did not actively seek employment in the last month for various reasons, 28 such as believing no 28 BLS considers workers who want work and are available for work but who did not actively seek work in the past month as “marginally attached to the labor force.” BLS records the reason given for not actively seeking work in the past month: for example, some marginally attached workers indicate family obligations, such as caring for an aging parent or a sick child as the reason they did not look for work. Page 11 GAO-12-445 Unemployed Older Workers jobs were available. 29 It also includes part-time workers who would prefer full-time work. 30 29 BLS defines workers who want work and are available for work, but did not actively seek work in the past month because they thought no jobs were available, as “discouraged workers.” Discouraged workers are a subset of those marginally attached to the labor force. 30 BLS identifies such workers as part-time for economic reasons, defined as those employed less than 35 hours per week who want and are available for, but are unable to find, full-time work, as well as people who prefer full-time hours of work but had their hours reduced by their employer. In this report GAO includes these workers in the term “underemployed,” along with marginally attached workers. Page 12 GAO-12-445 Unemployed Older Workers Figure 3: Estimated Number of Unemployed and Underemployed Older Workers (55 and Over), 2007-2011 Note: (1) Estimates have 95 percent confidence intervals within plus or minus 12 percent of the estimate itself. (2) See appendix II for BLS’s definitions of unemployed workers, discouraged workers, workers marginally attached to the labor force, and part-time for economic reasons. Our “broader measure of people affected by unemployment and underemployment” is the same as measure U-6 in BLS’s alternative measures of labor underutilization. Estimates are not seasonally adjusted. Page 13 GAO-12-445 Unemployed Older Workers Long-Term Unemployment Although older workers are less likely than younger workers to lose their for Older Workers Has jobs, it generally takes older job seekers longer to find new work. 31 Since Increased Substantially 2007, many job seekers of all ages have experienced long-term unemployment, 32 but individuals age 55 and over have consistently since the Start of the experienced longer durations of unemployment than younger workers. Recession Moreover, the median length of unemployment has more than tripled for older workers since the recession started, increasing at a greater rate than that of younger workers. Prior to the recession, the median duration of unemployment for job seekers age 55 and over was 10 weeks compared with 9 weeks for job seekers aged 25-54. By 2011, the median duration of unemployment for older job seekers had increased to 35 weeks compared with 26 weeks for younger job seekers. Since 2007, the number and percentage of long-term unemployed older workers—those out of work for more than half a year (27 weeks or more)—has increased substantially. In 2007, less than a quarter of unemployed older workers were unemployed for longer than 27 weeks, as shown in figure 4. By 2011, this number had increased to 55 percent. Moreover, by 2011 over one-third of all unemployed older workers had been unemployed for over a year. Data from the Displaced Worker Supplement of the CPS corroborate that many older workers who lose their jobs struggle to regain employment. Specifically, the data show that only 31 percent of those older workers age 55-64 who were displaced between 2007 and 2009 had regained full-time employment by January 2010 (see app. II). 31 In this report, we use the phrase “job seekers” to refer to the unemployed. 32 BLS defines long-term unemployment as being unemployed for more than half a year (27 weeks or more). Page 14 GAO-12-445 Unemployed Older Workers Figure 4: Growth in Estimated Long-Term Unemployment of Older Workers (55 and Over), 2007-2011 Note: All estimates in this figure have 95 percent confidence intervals within plus or minus 3 percentage points of the estimate itself. See appendix III for statistical comparisons of estimates across years. Some bars do not sum to 100 percent because of rounding. The dramatic increase in long-term unemployment bodes poorly for the reemployment prospects of older workers; several studies suggest that unemployment erodes workers’ skills and reemployment prospects, 33 and several experts we interviewed said long-term unemployment diminishes the likelihood older workers will ever be reemployed. 34 Thus, older workers who lose their jobs may face both immediate and long-term financial challenges. Long-term unemployed older workers who exhaust unemployment benefits before turning 62 are particularly at risk of 33 Rob Valetta and Katherine Kuang. Why is Unemployment Duration So Long? Federal Reserve Bank of San Francisco Economic Letter 2012-03 (San Francisco, CA, January 2012), 1, and Linda Levine, The Increase in Unemployment since 2007: Is it Cyclical or Structural? Linda Levine and Gerald Meyer, Long-term Unemployment and Recessions (Washington, D.C.: May 2010). 34 One expert believed that many older workers “face the real prospect of never working again, certainly not at their former wage levels.” See What to do about the New Unemployment, Urban Institute (Washington, D.C.: June 2011). Page 15 GAO-12-445 Unemployed Older Workers compromising their future retirement security. (For some potential causes of long-term unemployment, see fig. 5.) Figure 5: Difference between the Estimated Number of Unemployed of All Ages and the Estimated Number of Available Jobs, 2007-2011, and Various Explanations for High Unemployment Note: Estimates for the number of unemployed workers have 95 percent confidence intervals within plus or minus 5 percent of the estimate itself. Estimates for the number of job openings have confidence intervals within plus or minus 10 percent of the estimate itself. Some Subgroups of Older The length of time older workers remained unemployed varied across Workers Continue to Face industries (see fig. 6). Nearly two-thirds of unemployed older workers in Longer Spells of the manufacturing and financial service industries were out of work for 27 or more weeks in 2011. In fact, in 2011, 20 percent of unemployed older Unemployment than workers within manufacturing had been out of work for more than 2 years. Others In contrast, less than half of unemployed older workers in the education and health services or leisure and hospitality industries had been out of work for 27 weeks or more. Even though the education and health Page 16 GAO-12-445 Unemployed Older Workers services and leisure and hospitality industries had lower levels of long- term unemployment, the percentage of older workers in these industries who experienced long-term unemployment also grew significantly from 2007 to 2011. Figure 6: Estimated Duration of Unemployment for Older Workers (55 and Over) by Industry, 2007 and 2011 Note: The “all other industries” category includes public administration; “other services”; agriculture, forestry, fishing and mining; and information and a small number of civilians employed in the armed forces. All estimates in this figure have 95 percent confidence intervals within plus or minus 11 percentage points of the estimate itself. For information about the reliability of estimate and statistical comparisons of the estimates across different groups and years, see appendix III. Page 17 GAO-12-445 Unemployed Older Workers Rates of unemployment differed across demographic groups. For example, as shown in figure 7, unemployment rates for older men and women were comparable in 2007 but increased more for men than women after 2007, and were significantly higher for men than women by 2011. 35 In addition, older workers who were black or Hispanic had significantly higher unemployment rates than white older workers in both 2007 and 2011. Regarding education level, older workers without a high school diploma were more likely to be unemployed before and after the recession than those with a high school diploma. 36 However, the unemployment rate for workers with at least a bachelor’s degree approximately doubled by 2011 from its 2007 level, just as it did for those older workers with less education. 35 One possible explanation for men’s greater increase in unemployment since 2007 is the particularly steep increase in unemployment in the manufacturing and construction industries, which tend to employ higher percentages of men than women. 36 One possible explanation of the unemployment increase among less educated older workers is that unemployment rates in manufacturing and construction increased dramatically in the recent recession, and these industries tend to employ a higher percentage of less educated workers than do many other industries. Also, a recent study of the long-term unemployed aged 18-64 also found that the long-term unemployed are less likely to hold a college degree. Kaiser Family Foundation/NPR Long-Term Unemployed Survey. Page 18 GAO-12-445 Unemployed Older Workers Figure 7: Estimated Unemployment Rates by Demographic Group for Older Workers (55 and Over), 2007 and 2011 Note: All estimates in this figure have 95 percent confidence intervals within plus or minus 1 percentage point of the estimate itself. For statistical comparisons of the estimates across different groups and years, see appendix III. Across several different demographic groups, once unemployed, older workers were similarly likely to remain unemployed for more than half a year (27 weeks or more) in 2011. For example, in 2011, older unemployed workers with at least a bachelor’s degree were similarly likely Page 19 GAO-12-445 Unemployed Older Workers to face long-term unemployment as those older workers with less education. In addition, older workers in each racial or ethnic group who became unemployed were equally likely to face long-term unemployment in 2011. Even older women—who in 2007 had lower rates of long-term unemployment than men—were similarly likely to face long-term unemployment after the recession, as shown in figure 8. Finally, while long-term unemployment increased for both younger and older workers, a higher percentage of older workers were long-term unemployed— approximately 55 percent of unemployed older workers were out of work for over half a year, compared with approximately 47 percent of workers age 25-54 in 2011. Figure 8: Estimated Duration of Unemployment by Age and by Gender, 2007 and 2011 Note: Estimates in this figure have 95 percent confidence intervals within plus or minus 3 percentage points of the estimate itself. For statistical comparisons of the estimates across different groups and years, see appendix III. Page 20 GAO-12-445 Unemployed Older Workers Older Workers’ Labor Despite high levels of unemployment and longer spells of unemployment, Force Participation older workers’ labor force participation rate—the proportion of the Nonetheless Continued to population that is employed or actively seeking employment—increased throughout the 2007-2009 recession, continuing historic trends. In Rise contrast, as shown in figure 9, the labor force participation rate for younger workers aged 16-24 has decreased since the recession began in 2007, while the participation of workers aged 25-54 generally decreased, but to a lesser degree. 37 Older workers’ increased labor force participation during the recession continued a long-term trend that began in the 1990s, and thus cannot be attributed solely to the 2007-2009 recession or declines in financial markets. Researchers have identified a number of factors contributing to this historic increase. For example, research indicates that improved health and longer life expectancies could increase older workers’ labor force participation. Researchers also note that rising labor force participation among older women is an important factor in the increase in labor force participation among older workers in recent years. In addition, researchers have noted that some older workers may remain in the labor force to retain health care benefits until they become eligible for Medicare at age 65—particularly since fewer employers now provide retiree health care coverage. 38 37 Research suggests the long-term decline in young adults’ (aged 20-24) labor force participation is associated with increased school enrollment. Regarding individuals aged 25-54, the decline in labor force participation is driven by men, particularly those with less education. The demand for less educated workers has fallen significantly over the past three decades. Similarly, inflation-adjusted wages for men with less than a high school diploma have also fallen. Some studies have suggested that increased access to Social Security disability benefits might also explain some of the historic decline in the labor force participation of men aged 25-54. See Abraham Mosisa and Steven Hipple, “Trends in Labor Force Participation in the United States,” Monthly Labor Review, October 2006, 35- 57. Also, Chinhui Juhn and Simon Potter, “Changes in Labor Force Participation in the United States,” Journal of Economic Perspectives, Vol. 20, No. 3, Summer 2006, 27-46. 38 Mosisa and Hipple, “Trends in Labor Force Participation in the United States,” 35-57; Craig Copeland, Labor-Force Participation Rates of the Population Age 55 and Older: What did the Recession Do to the Trends? EBRI Notes Vol. 32, No. 2, Employee Benefit Research Institute (Washington, D.C.: February 2011), and Murray Gendell, “Older Workers: Increasing Their Labor Force Participation and Hours of Work,” Bureau of Labor Statistics, Monthly Labor Review, January 2008. Page 21 GAO-12-445 Unemployed Older Workers Figure 9: Estimated Labor Force Participation Rates by Age, 1948-2011 Note: All estimates in this figure have 95 percent confidence intervals within plus or minus 0.7 percentage points of the estimate itself. One recent report found that as a result of the recession, some older workers decided to remain in the labor force longer than previously planned, while others reentered the labor force, likely to bolster their income after the financial crisis. 39 In fact, several one-stop career center staff we interviewed told us that they were serving increasing numbers of older individuals who had reentered the workforce from retirement. (For 39 Kristie M. Engemann and Howard J. Wall, “The Effects of Recessions Across Demographic Groups,” Federal Reserve Bank of St. Louis Review, St. Louis, MO, January/February 2010, 92(1), 1-26. Page 22 GAO-12-445 Unemployed Older Workers more information on the potential implications of older workers’ increased labor force participation on younger workers, see the following text box.) Does Labor Force Participation by Older Workers Diminish Employment Opportunities for Younger Workers? Results of recent academic studies contradict a popular notion that labor force participation by older workers diminishes employment opportunities for younger workers. This notion assumes a zero sum game with regard to employment; that is, that there are a fixed number of jobs available at any given time and employment of one group results in unemployment for another group. However, according to a recent study of employment a data from numerous countries over several decades, increased employment of older workers has not been associated with decreased employment of younger workers. The study found that when employment of older workers increased, employment of younger workers also increased, and this relationship remained even when the researchers took account of overall economic growth. The study also analyzed what happened to younger workers’ employment rates in some European countries during specific time periods after increased numbers of older workers retired early, not because of changes in the economy, but because of changes in national retirement policies. The researchers found the opposite of what the popular notion of a zero sum game would assume; they found that when more older workers left their jobs to retire early, more younger workers became unemployed. The researchers concluded that “the evidence suggests that greater labor force participation of older persons is associated with greater youth employment and with b reduced youth unemployment.” An additional study of multiple countries’ economic data also found that employment of older workers does not adversely affect the employment of c younger workers. Two possible explanations for the results of these studies are that (1) over time, entire economies grow as more workers enter the workforce, increasing the demand for all goods and services and also for workers, and (2) some jobs held by younger workers complement jobs held by older workers so that having one position filled leads to hiring for another position. For example, having a senior researcher may create the need for research assistants. a Jonathan Gruber, Kevin Milligan, and David A. Wise, Social Security Programs and Retirement Around the World: The Relationship to Youth Employment, Introduction and Summary, Working Paper 14647, http://www.nber.org/papers/w14647, NBER (Cambridge, MA: 2009). The 12 countries covered in this study were Belgium, Canada, Denmark, France, Germany, Italy, Japan, the Netherlands, Spain, Sweden, the United Kingdom, and the United States. b For more information on this research, see Jonathan Gruber and David A. Wise, eds., Social Security Programs and Retirement Around the World: The Relationship to Youth Employment, University of Chicago Press, February 2010. c Adriaan Kalwij, Arie Kapteyn, and Klaas De Vos, “Retirement of Older Workers and Employment of the Young” De Economist (2010) 158:341-359. The 22 countries covered by this study were Australia, Austria, Belgium, Canada, Denmark, Finland, France, Germany, Greece, Iceland, Ireland, Italy, Japan, the Netherlands, New Zealand, Norway, Portugal, Spain, Sweden, Switzerland, the United Kingdom, and the United States. Page 23 GAO-12-445 Unemployed Older Workers Among displaced workers 40 aged 65 and over, a significantly larger fraction of those who lost their job during the recession years chose to remain in the workforce, compared with those who lost their jobs during the prerecession period. Specifically, the percentage of displaced workers aged 65 and over who retired and left the labor force declined from 61 percent in the prerecession period to 35 percent in the recession years. At the same time, the percentage of displaced workers age 65 and over who chose to remain in the workforce but were unemployed increased from 15 percent prior to the recession to 33 percent during the recession (see app. II). 41 This could indicate that increasing numbers of older workers recognize that they are not well positioned for retirement. In addition, the increased availability of extended UI benefits could have caused some workers to remain in the labor force, 42 although several researchers estimate this has had only a modest effect. 43 Those older workers displaced from 2007 to 2009 who successfully regained employment by January 2010 generally sustained greater 40 Displaced workers are those who indicated that they lost a job for economic reasons (such as plant closures or their position being eliminated) during the previous 3 calendar years. Displaced workers are surveyed by the Census Bureau every 2 years, with the most recent survey interviewing people who lost their job during the recession period (January 2007-December 2009), and the previous survey interviewing people who predominantly lost their jobs prior to the recent recession (January 2005-December 2007). Older displaced workers may choose to stay in the workforce—becoming either employed or unemployed—or to exit the workforce, possibly into retirement. 41 One study indicates this increase suggests that growing concerns about the adequacy of retirement savings and the 2008 stock market collapse may have discouraged early retirement and prompted more older workers to remain in the labor force after losing their jobs. Richard Johnson, Rising Senior Unemployment and the Need to Work at Older Ages, Urban Institute (Washington, D.C.: September 2009). 42 Research indicates extended UI benefits could increase the unemployment rate in two ways: (1) by keeping older workers who would otherwise have dropped out of the labor force attached to the labor market or (2) by allowing the unemployed to turn down job offers they would have accepted had UI benefits not been available. See Jesse Rothstein, Unemployment and Job Search in the Great Recession, NBER (Cambridge, MA: October 2011). 43 See Gary Burtless and Adam Looney, Growth through Innovation: The Immediate Jobs Crisis and Our Long-Run Labor Market Problem. Brookings Institution (Washington, D.C.: January 2012); Jesse Rothstein, Unemployment and Job Search in the Great Recession NBER (Cambridge, MA: October 2111); and Rob Valetta and Katherine Kuang, Why is Unemployment Duration So Long? Federal Reserve Bank of San Francisco, (San Francisco, CA: January 2012); The Increase in Unemployment since 2007: Is it Cyclical or Structural? Levine and Meyer, Long-term Unemployment and Recessions. Page 24 GAO-12-445 Unemployed Older Workers earnings losses than did younger workers. When comparing earnings before and after displacement, the median earnings replacement rate for workers aged 55-64 who were displaced from 2007 to 2009 was only 85 percent, compared with approximately 95 percent for workers aged 25-54 and over 100 percent for workers aged 20-24. 44 Indeed, as shown in figure 10, an estimated 70 percent of reemployed displaced older workers sustained earnings losses (an earnings replacement rate of less than 100 percent) compared with 53 percent of reemployed individuals aged 25-54. Other researchers have also found that displaced older workers suffer greater wage losses than younger workers, 45 and that the effects of job loss are likely to be long-lasting—including being more likely to lose subsequent jobs and experience additional unemployment spells. 46 44 This analysis is restricted to long-tenured displaced workers (workers with 3 or more years of tenure on the job they lost or left) who lost full-time, salaried jobs and were reemployed in full-time, salaried jobs at the time of the survey. 45 A recent study using data from the Survey of Income and Program Participation found that between 1996 and 2007, the median hourly wage earned by displaced men aged 50 to 61 who become reemployed at ages 50 to 61 was 20 percent lower than the median wage on the prior job. For those reemployed at age 62 or older, the median wage for the new job was 36 percent below the median wage for the prior job. In contrast, the study found that younger men’s median wages fell by only 4 percent for men aged 35 to 49 and 2 percent for those aged 25 to 34. The study found that reemployed older displaced women also suffered wage losses, but these differences were not as significant as those for men. See Johnson and Mommaerts, “Age Differences in Job Loss, Job Search, and Reemployment.” 46 A recent study using data from the Health and Retirement Study (HRS) found that individuals who lost a job between ages 50 and 56 had “messy post-displacement employment histories.” Specifically, such workers were “more likely to job-hop, to suffer further involuntary job losses, and to experience subsequent unemployment than those who were still working for their age-50 employer at age 56.” See Steven A. Sass and Anthony Webb, Is the Reduction in Older Workers’ Job Tenure a Cause for Concern? Center for Retirement Research at Boston College (December 2010). Page 25 GAO-12-445 Unemployed Older Workers Figure 10: Estimated Percentage of Reemployed Displaced Workers Who Earned Less on Their New Full-Time Jobs than on Their Previous Jobs, January 2010 Note: Estimates in this figure have 95 percent confidence intervals within plus or minus 7 percentage points of the estimate itself. Analysis is restricted to long-tenured displaced workers (workers with 3 or more years of tenure on the job they lost or left) who lost full-time, salaried jobs and were reemployed in full-time, salaried jobs at the time of the survey. For statistical comparisons of the estimates across different groups and years, see appendix III. Since the start of the recession, workers of all ages have struggled to Unemployed Older cope with unemployment. However, older workers generally face longer Workers Face Many periods of unemployment than do younger workers and may face unique reemployment challenges. According to experts and one-stop career Challenges Coping center staff we interviewed, such challenges include employer reluctance with Unemployment to hire older workers, out-of-date skills, and unfamiliarity with online and Regaining applications, which can be particularly common for older workers with limited technological skills. Employment Page 26 GAO-12-445 Unemployed Older Workers Unemployed Older Unemployed older workers in our focus groups wanted to return to work Workers Have Difficulty so they could pay for critical living expenses and contribute to their Meeting Financial families and communities. 47 They told us that losing their jobs had taken a toll on their sense of self-worth, reduced their standard of living, and put Obligations them at risk of long-term financial hardship. Older workers cited difficulty paying for health care and health insurance as a key financial challenge since they lost their jobs. 48 Specifically, many focus group participants described struggling to pay health insurance premiums and some said they had found it difficult to secure private insurance because of high costs or preexisting conditions. Many focus group participants said they had forgone seeking medical care or taking prescribed medications because they could not afford them. See appendix IV for examples of specific quotes from focus group participants regarding risks associated with long-term unemployment. Many focus group participants said that being unemployed had made it difficult to afford mortgage or rent payments. 49 In addition, some focus group participants told us they had lost or were at risk of losing their homes. Others had taken in roommates, moved in with friends or family, 47 GAO conducted focus groups in Baltimore, Maryland; Falls Church, Virginia; San Jose, California; and St. Louis, Missouri. GAO selected these locations based on metropolitan areas’ unemployment rates, geographic diversity, and the estimated costs for travel and securing focus group facilities. In total GAO conducted 10 focus group sessions. Methodologically, focus groups are not designed to (1) demonstrate the extent of a problem or to generalize results to a larger population, (2) develop a consensus to arrive at an agreed-upon plan or make decisions about what actions to take, or (3) provide statistically representative samples or reliable quantitative estimates. Instead, they are intended to generate in-depth information about the reasons for the focus group participants’ attitudes on specific topics and to offer insights into their concerns about and support for an issue. Please see appendix I for further details on our focus group methodology. 48 A survey of workers unemployed during the recession also found that some unemployed individuals age 55 and over had no health insurance or had gone without medical care. See M. Heidkamp, N. Corre, and C. Van Horn, The “New Unemployables” Older Jobseekers Struggle to Find Work During the Great Recession, Sloan Center on Aging and Work, Boston College (Chestnut Hill, MA: 2010). A recent survey of long-term unemployed and underemployed individuals age 18-64 also found that many of the long- term unemployed individuals reported difficulty paying for health insurance or health care. See Kaiser Family Foundation/ NPR Long-Term Unemployed Survey. 49 A recent survey of long-term unemployed and underemployed individuals aged 18-64 also found that many long-term unemployed individuals reported difficulty paying for housing. See Kaiser Family Foundation/ NPR Long-Term Unemployed Survey. Page 27 GAO-12-445 Unemployed Older Workers or moved to a more affordable apartment to reduce their housing expenses. For example, one focus group participant told us his son had moved back home so that he could give his parents money for rent. Most of our focus group participants had relied on UI benefits to help pay for critical living expenses while they were unemployed, and many said that without UI benefits they would have been in much greater financial jeopardy. Specifically, many focus group participants said that without UI benefits they would have lost their homes or even become homeless. However, some participants said they had already exhausted their UI benefits or would soon exhaust their benefits. Some focus group participants indicated that, since losing their jobs, they had received public assistance, such as through the Supplemental Nutrition Assistance Program (formerly known as food stamps). 50 Even with UI, many focus group participants said they had drawn down retirement savings, increased their credit card debt, or tapped into their home equity to cover living expenses while unemployed. Others reported borrowing money from family or friends or selling possessions to meet their financial needs. 51 A majority of those who were old enough to claim early Social Security retirement benefits had already done so. In addition, focus group participants said they had struggled to fulfill family financial obligations they had been able to meet in the past. Several mentioned that they were attempting to pay their children’s college tuition 50 A recent GAO report examines how many workers who lost their jobs during the recent recession received and exhausted UI benefits, the economic circumstances of those UI recipients who exhausted their benefits and whether they received support from other government programs, and the extent to which UI agencies refer those exhausting UI to other support programs. According to the report, 15 percent of these UI exhaustees had received benefits through the Supplemental Nutrition Assistance Program. See GAO, Unemployment Insurance: Economic Circumstances of Individuals Who Exhausted Benefits, GAO-12-408 (Washington, D.C.: Feb. 17, 2012). 51 A survey of unemployed older workers age 55 and over also found that unemployed older workers have increased credit card debt; sold possessions; borrowed money from family, friends, or adult children; or moved to a different house or apartment. See Heidkamp, M., Corre, and Van Horn, The “New Unemployables:” Older Jobseekers Struggle to Find Work During the Great Recession. Another survey of long-term unemployed and underemployed individuals ages 18-64 also found that many had taken money out of savings or retirement funds to pay bills, sold personal belongings, or borrowed money from relatives and friends. See Kaiser Family Foundation/ NPR Long- Term Unemployed Survey. Page 28 GAO-12-445 Unemployed Older Workers or were helping to pay unemployed adult children’s living expenses. They also frequently expressed general concern about high unemployment among younger generations. Some said they were supporting their children’s employment by taking care of their grandchildren during the workday. Others mentioned that they were caring for their aging parents. Unemployed Older Workers Face Many Challenges in Becoming Reemployed Perceived Employer Many experts, one-stop career center staff, and other workforce Reluctance to Hire Older professionals we interviewed said that some employers are reluctant to Workers hire older workers. 52 Because of legal prohibitions against age discrimination, employers are unlikely to explicitly express a lack of interest in hiring older workers; 53 however, one workforce professional told us that local employers had asked her to screen out all applicants over the age of 40. 54 Focus group participants perceived employer reluctance to hire older workers as their primary reemployment challenge, and several cited job interview experiences that had convinced them that age discrimination was limiting their ability to find a new job. According to experts we interviewed, a key reason employers are reluctant to hire older workers is that they expect providing health benefits to older workers would be costly. Several employer surveys corroborate 52 In some cases, our interviews with one-stop career center staff also included individuals who worked for other organizations, such as nonprofits, to help older workers overcome their employment challenges. In this report, we refer to these individuals and the one-stop career center staff as “workforce professionals.” 53 The Age Discrimination in Employment Act of 1967, as amended, prohibits employment practices that discriminate against people who are age 40 or older. Pub. L. No. 90-202, 81 Stat. 602, codified at 29 U.S.C. §§ 621-634. 54 For information about evidence that employers discriminate against older job applicants, see Joanna N. Lahey, “Do Older Workers Face Discrimination?” Center for Retirement Research at Boston College, Issue Brief Number 33, July 2005. Page 29 GAO-12-445 Unemployed Older Workers this concern. 55 In addition, a few focus group participants who had handled their previous employer’s health insurance or had been involved in hiring decisions said they had seen that older workers substantially increased insurance costs, which provided a disincentive to hire older workers. For example, one focus group participant told us that his prior employer had told him not to hire anyone older than him. The higher wages that older workers previously earned also make some employers hesitant to hire them, according to experts, workforce professionals, and our focus group participants. 56 For example, employers may expect that an older worker who accepts a job paying significantly less than the worker had previously earned might continue to search for a higher-paying job and might leave the job if a better offer became available. Also, according to experts we interviewed, employers may believe that an older worker who previously held a high-level position will be overqualified and therefore unhappy in a lower-level position. Further, some experts we interviewed said employers may believe that an older worker may not be happy working for a younger, less experienced supervisor, which could cause interpersonal conflicts. Some focus group participants said potential employers had specifically mentioned these concerns during job interviews. Many of our focus group participants said they had great financial need for a new job and expected to take a significant pay cut to get one. See appendix IV for examples of specific quotes from focus group participants regarding challenges they experienced in becoming reemployed. Lack of Up-to-Date Skills and Workforce professionals we interviewed said that many older workers Diminished Skills lack up-to-date skills with computers and other technology, and this puts them at a disadvantage in becoming reemployed. Some noted that after a long spell of unemployment, even those older workers who had previously been proficient with computer technology might find their 55 See Marcie Pitt-Catsouphes, Michael A. Smyer, Christina Matz-Costa, and Katherine Kane, “The National Study Report: Phase II of the National Study of Business Strategy and Workforce Development,” Center on Aging and Work/Workplace Flexibility at Boston College Research Highlight 04, March, 2007, 21. Also, see The Real Talent Debate: Will Aging Boomers Deplete the Workforce? A WorldatWork Research Report, April 2007, 4. 56 A recent study using data from the Survey of Income and Program Participation found that between 1996 and 2007, the median hourly wage for reemployed displaced workers was lower at ages 50 to 61 than at ages 35 to 49. The authors of the study suggest that “concern over the expense of hiring older workers may be overblown.” See Johnson and Mommaerts, Age Differences in Job Loss, Job Search, and Reemployment. Page 30 GAO-12-445 Unemployed Older Workers technology skills outdated. Some experts we interviewed said that employers might hesitate to hire and retrain older workers because they assume that older workers will not want to work much longer, so the employer would not get a good return on the training investment. Some workforce professionals also said that, according to their observations, employers have increased the number of skills they require applicants to possess, even for low-level positions. For example, a workforce professional told us one job posting for a receptionist stated that the applicant needed to be able to manage the employer’s website in addition to completing more typical administrative duties. Some focus group participants said that they were at a disadvantage in finding new jobs because injuries or other health problems prevented them from performing the type of work they had done in the past or meeting requirements of available jobs in other fields. Challenges in the Online Job According to workforce professionals, an ongoing trend among Application Process employers—to require job seekers to submit all applications and résumés online—creates difficulties for many older workers, particularly those with few or no computer skills. Further, workforce professionals told us that many online job applications require applicants to disclose information that readily reveals the applicant’s age, such as the year the job seeker graduated from high school, and that applications cannot be submitted until such fields are completed. Such information would make it possible for employers to screen out older workers, if the employer wanted to do so. Workforce professionals also said that even workers seeking jobs that require little or no computer use could get those jobs only by completing a long online application. For example, workforce professionals in two locations told us that individuals seeking positions as maids and janitors in national chain hotels could apply for these positions only online and that the older workers seeking these positions often were unfamiliar with such applications. Moreover, workforce professionals and focus group participants noted that online applications can vary widely among employers, cutting and pasting is sometimes not an option, and each application can take hours to complete. They also told us that many online applications cut off if an applicant has not completed or saved the work within a certain amount of time. This makes applying for jobs more difficult for workers with limited computer skills. Emotional Challenges That Workforce professionals at the four one-stop career centers we visited Result from Long-Term and some other experts said that some older workers experience Unemployment depression and discouragement because of their long-term Page 31 GAO-12-445 Unemployed Older Workers unemployment. While this challenge is not limited to older workers, depressed and discouraged older workers may not search for jobs as intensely as they might have otherwise and their performance in interviews may also be compromised. 57 For example, workforce professionals said that depressed or discouraged job seekers may show up at interviews looking disheveled or may become short-tempered during interviews. A workforce professional in Falls Church, Virginia, said that employers have told him they had decided against hiring an older job applicant because the applicant had appeared too desperate for a job. Many of our focus group participants said they had been discouraged or depressed because of their continued unemployment. Older Workers Need Workforce professionals identified different types of services that older Reemployment Services workers need from the workforce system to help address their That Address Their Unique reemployment challenges, such as employers’ reluctance to hire them. For example, workforce professionals said older workers need services Challenges that help them • learn how to present their skills and experiences to potential employers in a way that does not draw attention to their age, extensive years of experience, and past high-level positions; • develop interview responses that can diffuse employer concerns about hiring older workers, such as whether the job seeker would be a good fit, be willing to work for less pay, or be okay with reporting to a younger manager; • understand how to adjust their physical appearance to make a better impression on prospective employers; • develop skills, including technological skills, that employers currently expect their employees to possess; and • complete and submit online job applications. 57 A recent survey of long-term unemployed and underemployed individuals ages 18-64 also found that many of these long-term unemployed individuals reported negative impacts on their mental and physical health. See Kaiser Family Foundation/ NPR Long- Term Unemployed Survey. Page 32 GAO-12-445 Unemployed Older Workers One-stop career center staff told us they offer classes and other support for workers to help with résumé writing, job interviewing, and computer skills. Some one-stop career center staff members told us that they perform mock interviews to help older job seekers learn how to gracefully respond to blunt questions that employers may ask older job applicants. Staff at two of the one-stop career centers we visited told us they had studied the details of some large employers’ online applications to help job seekers avoid having their applications automatically rejected because of blank fields or inappropriate responses. One-stop career center staff also told us that providing the type of assistance that some older workers need can be very time-consuming. For example, one-stop career center staff said that helping some older workers understand that their physical appearance or discouraged demeanor hurts their reemployment prospects usually requires sensitive one-on-one discussions. Also, one-stop career center staff told us they sometimes have to work individually with older workers for long periods of time to help them complete online applications. Older workers who saved for retirement but lost their jobs following the Job Loss Can recession could face reduced retirement security because of long-term Diminish Retirement unemployment, in part because they have fewer years to accrue additional benefits or make additional contributions, and they might rely Income Prospects upon their retirement savings to cover expenses incurred while they are unemployed. Furthermore, a long period of unemployment could lead older workers to claim early Social Security retirement benefits, which would reduce their monthly benefits for the rest of their lives. Job Loss Can Lead to Long-term unemployment can reduce an older worker’s future monthly Lower Private Retirement retirement income in numerous ways, such as by reducing the number of Income, Early Social years the worker can accumulate DB plan retirement benefits or DC plan savings, by motivating Social Security claims at an earlier age than the Security Claims, and worker otherwise would have chosen, and by leading workers to draw Exhaustion of Retirement down retirement savings to pay for expenses during unemployment. Savings Fewer Years of Work Can Job loss can result in fewer years of work over a worker’s lifetime, which Lower Retirement Income can lower the worker’s retirement income in several ways. For example, fewer years of work can prevent a worker covered by a traditional DB plan from having enough years of work with an employer to vest in (that is, earn a nonforfeitable right to receive) employer-funded retirement benefits, thus preventing the worker from having any retirement benefits Page 33 GAO-12-445 Unemployed Older Workers from the employer. 58 Fewer years of work can also reduce a worker’s final retirement benefit from a traditional DB plan if the number of years worked is used in the formula for calculating retirement benefits. For workers with DC plans, having fewer years of work can limit the amount of yearly employee and employer contributions that accumulate in a worker’s account and reduce the earnings from those contributions. Further, having fewer years of work gives a worker less time to move up the salary ladder and achieve higher levels of pay. For a worker with a traditional DB plan, this will result in lower benefits if salary levels are included in the formula for calculating benefits, and, for a worker with a DC plan, this can reduce the worker’s ability to increase contributions to the plan over time. Social Security retirement benefits may be reduced as a result of fewer years of work because the benefits are based, in part, on a calculation of the worker’s average monthly earnings over 35 years. The 35 years used for the calculation are the worker’s highest earnings years, adjusted for changes in wage levels. If a worker has less than 35 years of earnings, then zeros would be used for earnings in the missing years, and this will result in a lower calculated benefit. Even if the worker had already worked for 35 years, losing work could reduce the worker’s Social Security retirement benefits because the worker did not have the opportunity to achieve higher earnings to replace low-earnings years in the benefit calculation. 59 Our simulations of how job loss and a forced early retirement would affect an older worker’s retirement income show that workers who had been participating in an employer-sponsored retirement plan would lose more retirement income because of job loss than workers who relied 58 The terms of an employer-sponsored retirement plan may specify when the employee has earned a nonforfeitable right to employer-funded benefits (called vesting), typically after the employee reaches a certain age or has completed a certain period of service. Federal vesting requirements may apply to some plans. For example, to qualify for favorable tax treatment, private sector DB plans are generally required to vest within a maximum of 7 years if they use graded vesting, in which the employee is vested in an increasing percentage of the benefits over time. If the plan does not use graded vesting, employees must be 100 percent vested within 5 years. In addition, employees must be vested upon reaching retirement age (typically age 65 or earlier, if defined by the plan), and federal law limits the ability of plans to disregard an employee’s prior years of service after breaks in service of less than 5 years. 29 U.S.C. § 1053(a)-(b). However, plans sponsored by public sector employers are not generally subject to these requirements, although state laws may apply. 59 For more information on how Social Security retirement benefits are calculated, see online illustration at http://www.ssa.gov/oact/ProgData/retirebenefit1.html. Page 34 GAO-12-445 Unemployed Older Workers exclusively on Social Security retirement benefits. 60 Workers with employer-sponsored retirement plans are in the best position to save for retirement because their retirement saving is facilitated and may be supplemented by their employers. Also, workers with access to employer- sponsored plans typically have higher average earnings than workers without access to such plans, and higher earnings will generally result in higher Social Security retirement benefits and generally in a greater ability to save for retirement. These workers also have the most retirement income to lose by becoming unemployed. We simulated how losing years of work as well as losing coverage by an employer-sponsored retirement plan between ages 55 and 62 could affect retirement income. Specifically our simulation estimated pretax monthly retirement income beginning at age 62 for groups forced to leave work at ages 55 and 58, and we compared future retirement income of these groups with our simulation results for a group who stopped work at age 62. The simulations show that retirement benefits from employer- sponsored plans are reduced much more than Social Security retirement benefits as a result of having fewer years of work. For example, median- level retirement benefits from employer-sponsored plans are 39 percent lower—$500 compared with $817—for workers with a DC plan (and no DB plan) who leave work at age 55 compared with similar workers who work until age 62. For these same groups of workers, median Social Security retirement benefits are only 13 percent lower—$1,273 compared with $1,467—for those who stopped working at age 55. Also, according to our simulations, the median Social Security retirement benefit for workers who leave work at age 55 or 58 with only a DC or a DB plan (but not both) is higher than the median benefits for such workers from an employer- sponsored retirement plan. 61 60 Our simulations of how job loss and a forced early retirement would affect an older worker’s retirement income cover retirement income from employer-sponsored retirement plans and Social Security retirement benefits but not retirement income from other sources. 61 Over time, the percentage of a worker’s retirement income coming from Social Security compared with private sources may increase because Social Security retirement benefits are subject to cost-of-living adjustments based on the Consumer Price Index for Urban Wage Earners and Clerical Workers published by the Bureau of Labor Statistics; however, retirement benefits from private sources may not be inflation-adjusted. Page 35 GAO-12-445 Unemployed Older Workers The simulations also show that some workers with employer-sponsored retirement plans who lose their jobs in their 50s are likely to end up either not vesting in the plans or cashing out their savings when they lose their jobs. On the other hand, the simulations demonstrate that workers who have benefits from both a DB and a DC plan are in the best financial position for retirement. At the median level of benefits, even those who were forced to retire at age 55 would have higher monthly retirement income than workers with only a DB or DC plan who worked until age 62. Figure 11 shows the simulations’ results at the median level (middle value in distribution) of benefits for workers who had been participating in employer-sponsored plans. The older workers from our simulations in figure 11 are better positioned for retirement than many other older workers because (1) they were fortunate enough to be working for an employer that offered a retirement plan before they lost their job, and (2) their retirement benefits are at the median level, which means their benefits are higher than those of almost half of the other workers in their group with similar retirement plans. Workers with benefits in the lower range (25th percentile) have substantially lower benefits. For example, a worker with benefits from a DC plan at the 25th percentile level will receive only about $136 per month from the DC plan at age 62 if the worker stops working at age 58, and such a worker will receive only $239 if he or she stops working at age 62. The results of our simulation show a best-case scenario for workers with DC plans because the simulation assumes the workers do not draw down any DC plan savings before age 62. Finally, the simulations demonstrate that workers with both a DB and DC plan are in the best financial position for retirement. At the median level of benefits, even those who were forced to retire at age 55 would have higher monthly retirement income than workers with only a DB or DC plan who worked until age 62. Page 36 GAO-12-445 Unemployed Older Workers Figure 11: Potential Effect That Fewer Years of Work Can Have on Retirement Benefits for Workers with Median-Level Retirement Benefits from Employer-Sponsored Plans and Median-Level Retirement Benefits from Social Security Note: The cohort used from the simulation models comprises individuals who were age 55 in 2010 and were participating in a retirement plan at their current employer. The models simulated retirement benefits at age 62 after all individuals in the cohort stopped working at ages 55, 58, and 62. The graphic shows the median level of benefits from employer-sponsored plans (including benefits from prior employers) based on the type of plan the individuals were participating in at the time of job loss. The medians for private retirement benefits and for Social Security retirement benefits were calculated separately and cannot be combined to get a median for total retirement income. The models assumed that all participants who vested in benefits and did not cash out their benefits when they left the job before age 62 used DC plan savings to purchase an annuity at age 62. Our simulations showed that workers with only Social Security retirement benefits have a comparatively small reduction in retirement income because of job loss, but, given their low monthly retirement income levels, the approximately $30 to $60 reduction could become problematic as retirees age and if health care costs and premiums continue to increase. Also, these workers may have great difficulty paying for living expenses while they are unemployed before they are old enough to claim Social Security retirement benefits because, if they do not have retirement savings, they may not have other savings to help them through a period of long-term unemployment. Figure 12 shows the simulations’ results at the median level of benefits for workers who stopped working at different ages and had only Social Security retirement benefits. Page 37 GAO-12-445 Unemployed Older Workers Figure 12: Potential Effect That Fewer Years of Work Can Have on Social Security Retirement Benefits of Workers without Employer-Sponsored Retirement Plans Note: Benefit amounts shown are for claims filed at age 62. The cohort used from the simulation models comprises individuals who were age 55 in 2010 and did not have retirement benefits from an employer-sponsored plan or IRA at age 62. Unemployment Can Motivate The responses of our focus group participants, academic research, and a Older Workers to Claim Social recent spike in Social Security claims indicate that long-term Security Retirement Benefits unemployment can motivate older workers to file for early Social Security Early retirement benefits, which will result in those workers and their survivors receiving lower monthly retirement benefits for the rest of their lives than if the workers had waited to claim benefits at full retirement age. 62 Many unemployed older workers in our focus groups said that they were planning to claim Social Security retirement benefits as soon as they were eligible or had already done so. They said that they could not find a job and needed a source of income to help pay for living expenses, and therefore decided to claim Social Security retirement benefits early. Some focus group participants also expressed the notion that they needed to claim Social Security retirement benefits as soon as possible because they were afraid of changes Congress might make in Social Security benefits and wanted to ensure they were grandfathered into the program before the changes were made. A few older workers in our focus groups who were too young to claim Social Security retirement benefits but had health problems that affected their ability to work said they applied for Social Security disability benefits to help pay for living expenses after they became unemployed. 62 The full retirement age is the age at which an individual is eligible to receive unreduced retirement benefits. This age ranges from 65 to 67 depending on the year the recipient was born. 42 U.S.C. § 416(l); 20 C.F.R. § 404.409(a). Page 38 GAO-12-445 Unemployed Older Workers Claiming Social Security retirement benefits at age 62 will cause an older worker to receive lower monthly benefits than if the worker had waited until full retirement age to claim benefits. Because Social Security retirement benefits are adjusted to provide approximately the same estimated value of lifetime benefits regardless of when the benefits are claimed, claiming benefits early results in a reduction to the monthly benefit amount because the benefits are paid out over a longer period of time. Therefore, older workers who claim Social Security retirement benefits early will have less income from Social Security each month to pay for living expenses for the rest of their lives. Figure 13 shows how an older worker’s monthly Social Security retirement benefit would differ depending on the age at which the worker claimed the benefits. Figure 13: Example of the Effect of Early Claiming on Monthly Social Security Retirement Benefits Note: For this analysis, GAO applied Social Security retirement benefit formulas for delayed and early claiming to the age 62 benefit amount derived from the analysis used in the preceding figure (fig. 12). The illustration is based on an individual who worked until age 62. The individual used in the illustration (1) is from a subset of the cohort used in the simulation models comprising individuals who were age 55 in 2010 and did not have retirement benefits from an employer-sponsored plan at age 62, and (2) had Social Security retirement benefits at the median level for the subset. Academic studies have found that unemployment can lead workers to retire. A study published in 2009 covering 30 years of data on employment and retirement decisions found that during times when obtaining a new job is difficult, older workers were likely to retire in Page 39 GAO-12-445 Unemployed Older Workers response to becoming unemployed. 63 Regarding the then-ongoing recession of 2007-2009, the study estimated that more older workers would retire early because of unemployment than would delay retirement in an attempt to rebuild savings after the downturn in the financial market. Also, a 2012 study found that high unemployment increases Social Security retirement claims among men with limited education. 64 The spike in claims for Social Security retirement benefits that occurred in 2009 after large increases in unemployment rates offers support for the study’s findings. According to estimates from SSA’s Office of the Chief Actuary, in fiscal year 2009 about 139,500 (about 6 percent) more older workers applied for Social Security retirement benefits than would have been expected in the absence of a recession. 65 The recession also led to an increase in applications for disability benefits from the Social Security Disability Insurance program. In 2009, SSA received approximately 205,000 (12 percent) more applications for disability benefits because of the recession, and applications also rose in 2010 because of the recession, according to estimates by the Office of the Chief Actuary. 66 While we do not know the percentage of these 63 Courtney Coile and Phillip B. Levine, The Market Crash and Mass Layoffs: How the Current Economic Crisis May Affect Retirement, Working Paper 15395, http://www.nber.org/papers/w15395, NBER (Cambridge, MA: 2009) . 64 The researchers estimate that the recession of 2007-2009 increased Social Security retirement claiming for men with limited education by about 40 percent. See Owen Haaga and Richard W. Johnson, Social Security Claiming: Trends and Business Cycle Effects, Center for Retirement Research at Boston College (Chestnut Hill, MA: February 2012). 65 When the Office of the Chief Actuary made estimates in December 2008 for the number of retirement benefit claims SSA would receive in fiscal year 2009, it did not factor recessionary effects into the estimates because, at that time, it did not know if the recession would increase or reduce the number of applications SSA would receive for retirement benefits. Therefore, according to the Office of the Chief Actuary, comparing the estimates for retirement benefits applications for fiscal year 2009 that were made in December 2008 with the actual number of applications received in fiscal year 2009 provides a reasonable estimate of the effect of the recession on Social Security applications in fiscal year 2009. 66 When the Office of the Chief Actuary made estimates in April 2008 for the number of OASDI disability applications SSA would receive in fiscal year 2009, it was not recognized within the United States that the economy was in recession, and a recession was not expected. Therefore, the Office of the Chief Actuary did not factor recessionary effects into the April 2008 estimates. According to the Office of the Chief Actuary, comparing the estimates for OASDI disability benefits applications for fiscal year 2009 that were made in April 2008 with the actual number of applications received provides a reasonable estimate of the increase in applications due to the recession. Page 40 GAO-12-445 Unemployed Older Workers additional applications that were filed by older workers, the majority of disability awards (approved applications) are for individuals age 50 and over. Also, for individuals age 50 and over, awards for disability benefits, as well as the percentage of individuals in the population who have been awarded disability benefits have increased since the recession started. 67 (See app. V for information on the increase in disability benefit awards.) Older workers who lost their jobs in the recession had significant injuries or health problems, and were not old enough to claim Social Security retirement benefits have strong incentive to apply for Social Security disability benefits. If they are awarded benefits, they will receive monthly payments and, after a 24-month waiting period, they will be eligible for health insurance from the Medicare program. 68 Also, receiving Social Security disability benefits gives unemployed older workers an alternative to claiming Social Security retirement benefits early. This is because individuals who are awarded Social Security disability benefits and remain eligible for those benefits can continue receiving them up until full retirement age, when they can begin receiving full retirement benefits. An increase in the number of individuals receiving disability benefits is costly for the OASDI trust funds. 69 According to the Office of the Chief Actuary, some workers who applied for disability benefits as a result of the recession probably would have applied eventually, but job loss or other effects of the recession motivated them to apply earlier. However, other workers may never have applied for disability benefits if they had not lost their jobs, and the trust fund would not have made disability payments to those workers. 67 According to the Office of the Chief Actuary, applications did not increase as a result of the recession for Aged benefits under the Social Security Supplemental Security Income (SSI) program. To be eligible for SSI Aged benefits, individuals must be 65 or over and have very low income and few assets. Such individuals may have already been unemployed before the recession, which could help explain why the recession did not increase applications for SSI Aged benefits. 68 Receipt of disability benefits is generally subject to a 5-month waiting period beginning with the month the applicant was both insured for disability and disabled, as defined by statute. 42 U.S.C. § 423, 20 C.F.R. § 404.315. 69 Unlike Social Security disability benefits, early Social Security retirement benefits do not increase expenditures from the OASDI trust funds. This is because retirement benefits are actuarially adjusted based on the age the benefits are claimed. Page 41 GAO-12-445 Unemployed Older Workers Unemployment Can Lead Older Unemployment can also lead an older worker who has a retirement Workers to Use Their Private account to use some or all of those savings to cover living expenses while Retirement Savings Early unemployed. Slightly over half of the older workers in our focus groups who reported having retirement savings in an IRA or DC plan also reported that they had used some or all of these savings to pay for expenses while they were unemployed. For example, focus group participants described using retirement savings to cover expenses such as mortgage and car payments, medical bills, a child’s college tuition, and moving to more affordable housing. A survey of unemployed workers conducted in March 2010 also found that a high percentage reported using savings set aside for retirement or other purposes to help make ends meet. 70 In addition, an October 2010 survey of workers age 50 and over found that nearly a quarter reported that they had used all their savings during the past 3 years. 71 The earlier a worker stops working and cashes out DC plan savings, the lower the savings will be and the shorter the period that the savings are likely to last. Depending on the level of savings, the length of time the worker spends unemployed, and the worker’s other financial resources, a worker may be at risk of using a large percentage of DC savings during unemployment. If the worker is fortunate enough to find another job that includes an employer-sponsored retirement plan or pays enough to enable the worker to save some earnings in an IRA, the worker will be able to resume saving for retirement. Figure 14 illustrates how a worker’s retirement savings of $70,000 in a 401(k) plan could change after 2 years of unemployment, depending on how much the worker withdrew from the account while unemployed. 72 The figure also shows how the account value could increase if the worker became reemployed and resumed 70 Heidkamp, Corre, and Van Horn, The “New Unemployables”, Older Job Seekers Struggle to Find Work During the Great Recession. 71 Sara E. Rix, AARP Public Policy Institute, “Recovering from the Great Recession: Long Struggle Ahead for Older Americans” (Washington, D.C.: May 2011). This study surveyed adults aged 50 and over who had been in the labor force at some point during the previous 3 years. 72 We used $70,000 as the starting point for this illustration because it is about the median level of DC plan savings for employed workers age 55 and over who have a DC plan account from a current or past employer. For purposes of this illustration, we decided to round this median to the nearest $10,000. Based on 2007 Survey of Consumer Finances data, the estimated median is $70,800 and its 95 percent confidence interval is within plus or minus $13,204, or between $57,596 and $84,004. Page 42 GAO-12-445 Unemployed Older Workers saving for retirement. As shown in figure 14, if the worker did not make any withdrawals during the period of unemployment, savings could have reached nearly $110,000 by age 62, after becoming reemployed. On the other hand, if the worker withdrew 50 percent of the retirement account balance while unemployed but then found a job and saved for another 5 years, the worker would still have less savings than before unemployment began. Figure 14: How Drawdowns from Retirement Savings during Unemployment Can Affect Amounts Saved at Time of Retirement if a Worker Became Reemployed and Resumed Saving Note: This illustration is based on an individual who was born at the beginning of 1953, turns 55 in 2008, and retires at age 62 in 2015. To calculate changes in the account balance over time, we used the interest and rate-of-return assumptions as reported in past and projected under the intermediate cost assumptions in the 2011 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Federal Disability Insurance Trust Funds (also known as the OASDI Trustees’ Report). We used scaled earnings for medium annual earners as reported in past and projected in the 2011 OASDI Trustees’ Report. We assumed the employee contributions to the retirement account are 6 percent of the individual’s wages and received a 3 percent employer matching contribution. Page 43 GAO-12-445 Unemployed Older Workers If the individual shown in figure 14 worked beyond the age of 62, the worker could continue to increase the savings and postpone the time when the savings would need to be used. If the worker worked until age 65, savings could reach about $139,500, if no withdrawals were made while unemployed. But, if substantial withdrawals were made while unemployed, the worker would need to work past age 62 just to get back to the level of savings in the account before the unemployment began. On the basis of our estimates, if the worker withdrew 50 percent of the retirement account balance while unemployed and became reemployed at age 57, it would take about 5 ½ more years of saving (age 62 ½) until the account balance got back to the level it was when the worker was 55. Before the Recession, In the period shortly before the recession started, we estimate that 40 Many Older Workers Had percent of employed individuals age 55 and over had no DB plan and no Little or No Retirement retirement savings or savings below $50,000. 73 Specifically, we estimate that 22 percent of older workers had no private retirement savings and did Savings not participate in a retirement plan, and an additional 18 percent only had retirement savings of less than $50,000 in a DC plan or IRA. Another 23 percent had total retirement savings of $50,000 or more in a DC plan or IRA and did not participate in a DB plan. We estimate that 37 percent of older workers were participating in a DB plan at their current employer or had earned the right to receive benefits from a past employer’s DB plan, and about half of these (19 percent of all older workers) had both a DB and a DC plan. 74 See figure 15 for the results of our analysis, based on data from the 2007 Survey of Consumer Finances. When older workers’ retirement plans and retirement savings are combined with those of their spouses, the resulting estimates indicate that more households than individuals have private retirement savings or plan participation. For households with at least one spouse over age 55, an estimated 16 percent had no retirement savings or plan participation, and an additional 13 percent had savings of less than $50,000. Figure 15 shows the estimated percentage of part- or full-time employed individuals 73 Percentage estimates based on the 2007 Survey of Consumer Finances have 95 percent confidence intervals of within plus or minus 3.4 percentage points of the estimate itself. See appendix I for additional information about this survey and estimates. 74 Our analysis does not cover the dollar value of benefits that individuals expect to receive from DB plans because these data are not collected in the Survey of Consumer Finances. Page 44 GAO-12-445 Unemployed Older Workers age 55 and older and their households according to the type of retirement plans or savings they had at the time of the 2007 survey. The percentage of those shown as participating in a DC or DB plan includes those who were participating at their current employers or who had accumulated benefits from a prior employer. Figure 15: Estimated Levels of Retirement Savings and Types of Plan Participation for Employed Workers 55 and Over and Their Households, 2007 Page 45 GAO-12-445 Unemployed Older Workers Note: For the purpose of this figure, retirement savings are considered to be participation in a DB or DC plan at a current employer, earned DB benefits from a past employer, a DC account from a past employer, a Keogh account, or an IRA. In this figure, rollover IRAs—IRAs that were started with funds rolled over from a DC plan account—are counted as DC savings, not IRAs. Also, Keogh accounts are counted as IRAs. Our analysis of households only includes the retirement savings and plan participation of the head of household and the spouse and does not include the retirement savings or retirement plan participation of additional family members. Percentage estimates in this figure have 95 percent confidence intervals that are within plus or minus 3.2 percent of the estimate itself. Because of rounding, percentages for “total value of retirement savings” for households does not sum to the corresponding combined value for categories “IRA only” and “DC.” Changes that have occurred since 2007, including job losses and financial market declines, have likely reduced retirement savings for some older workers. 75 Also, after the start of the recession in 2007, to cut costs, some employers suspended contributions to their employees’ DC accounts, which will ultimately reduce employees’ retirement income. According to the Employee Benefit Research Institute’s (EBRI) annual Retirement Confidence Survey, workers’ confidence in being able to afford a comfortable retirement has eroded since 2007. 76 For example, in 2007, according to EBRI estimates, 11 percent of workers age 55 and over said they were not at all confident about having enough money to live comfortably in retirement. By 2011, that percentage had increased to 22 percent. In addition to the impact of the 2007-2009 recession and ensuing financial downturn, these changes may indicate that more older workers are becoming aware that they do not have enough savings to retire and need to keep working to increase their income or save more for retirement. This may help explain why the proportion of older workers participating in the workforce has continued to increase. Also, since the onset of the recession, more workers have reported that they expect to retire at older ages than in the past. EBRI conducted an analysis of expected retirement ages of workers age 50 and over using 2006, 2008, and 2010 data from the University of Michigan’s Health and Retirement Study. EBRI found increases from 2006 to 2010 in the percentage of 75 According to a recent study using data from the Health and Retirement Study, “The retirement wealth held by those ages 53 to 58 before the onset of the recession in 2006 declined by a relatively modest 2.8 percentage points by 2010. In more normal times, their wealth would have increased over these 4 years.” See Alan L. Gustman, Thomas L. Steinmeier, and Nahid Tabatabai, “How Did the Recession of 2007-2009 Affect the Wealth and Retirement of the Near Retirement Age Population in the Health and Retirement Study?” NBER Working Paper 17547, http://www.nber.org/papers/w17547 (Cambridge, MA: October 2011). 76 Ruth Helman, Craig Copeland, and Jack VanDerhei, Issue Brief No. 355: The 2011 Retirement Confidence Survey: Confidence Drops to Record Lows, Reflecting “the New Normal,” Employee Benefit Research Institute (Washington, D.C.: 2011). Page 46 GAO-12-445 Unemployed Older Workers workers age 50 and over who expected to retire after age 65. 77 However, workers’ reports about plans to delay retirement should be considered with some skepticism because EBRI has found that a large percentage of workers retire earlier than they expected for various reasons, including health problems, downsizing or closure of their company, and having to care for a spouse or other family member. Experts we interviewed most frequently selected a range of policy Experts Selected proposals that could potentially help unemployed older workers regain Policies to Consider employment as the economy improves. 78 In the current context of high as the Economy unemployment and slow job creation, the impact of such policies is likely to be muted by limited job openings. Nonetheless, experts we spoke to Recovers, and Labor said these policies could potentially help some older workers obtain Has Taken Some reemployment. Labor has taken some steps to help older workers by implementing several strategies proposed in its 2008 Taskforce report. Steps to Help Older Workers 77 Sudipto Banerjee, “Retirement Age Expectations of Older Americans Between 2006 and 2010,” Employee Benefit Research Institute, Notes Vol. 32, No. 12, Dec. 2011, 2-12. 78 The 12 experts we interviewed primarily work in academia and think tanks, and were selected based on their expertise on older workers, workforce development, or retirement policy. We selected a group of experts to help ensure a range of viewpoints. We asked these experts if they would be willing to assess the strengths and weaknesses of policy proposals to help long-term unemployed older workers regain employment. Prior to interviewing the experts, we sent them a list of the 21 policy proposals we had compiled from previously published studies and reports. We asked the experts to select those 5 policies they believed merited serious consideration. The 8 policies addressed in this section are those that received votes from 4 or more of the 12 experts we interviewed. The expert discussion cited in this report should be interpreted in the context of two key limitations and qualifications. First, although we were able to secure the participation of a balanced, highly qualified group of experts, other experts in this field could not be included because we needed to limit the size of the panel. Second, although many points of view were represented, the panel was not representative of all potential views. Page 47 GAO-12-445 Unemployed Older Workers Proposed Policies to Help Experts we interviewed most frequently selected a variety of policy Older Workers as the options that could help address unemployed older workers’ reemployment Economy Recovers challenges. Experts selected these policies from a broader list of proposals we compiled from previous studies, 79 many of which were conducted before the recession. (For a complete list of the proposals we presented to experts, see app. VI.) While there was no consensus among experts and each proposal had advantages and disadvantages, several experts we spoke to said implementing several policies in combination would likely improve older workers’ employment levels more than implementing any of the policies in isolation. Experts differed on whether they believed the policies should be narrowly targeted at older workers or more broadly applied to all long-term unemployed workers. Some experts suggested all long-term unemployed workers—not just those who are older—should qualify to receive assistance under some proposed policies. Others said that policies specifically targeted to older workers could help “level the playing field” and contain federal costs. Finally, experts said that implementing many of the policies would increase federal spending and involve legislative changes, and would therefore need to be considered carefully in the context of our nation’s current long- term fiscal challenges. The eight policy proposals experts most frequently selected fall into three categories, organized by the underlying issue they are meant to address: (1) employer reluctance to hire older workers, (2) the need to enhance reemployment assistance targeted specifically to older workers, and (3) the need to encourage older workers to obtain reemployment as quickly as possible. However, some of the policy proposals experts selected will be of limited effectiveness as long as the number of job seekers greatly outnumbers the number of available jobs. Further, the effectiveness of some proposed policies could be limited if older workers’ unemployment is caused by structural rather than cyclical changes. However, experts we spoke to said these policies could potentially help older workers obtain reemployment. 79 Before interviewing experts, we reviewed the literature, including past GAO reports, academic studies, and federal agency reports, to identify policies that have been proposed to help unemployed older workers regain employment. The eight policies that received votes from at least 4 of the 12 experts are listed in tables 1 to 3 along with experts’ comments on the policies’ strengths and limitations. Page 48 GAO-12-445 Unemployed Older Workers Of the eight proposed policies most frequently selected, three aim to address employer reluctance to hire older workers (see table 1). As discussed previously, almost all of the experts and workforce professionals we interviewed said that some employers are reluctant to hire older workers. When explaining why this might be the case, many experts cited older workers’ higher health care costs and salary expectations. Many experts we interviewed said that eliminating the requirement that Medicare generally be the secondary payer for benefits for workers age 65 and over covered by an employer group health plan could improve employers’ willingness to hire older workers. 80 However, the experts also acknowledged that implementing such a policy would shift health insurance costs from employers to the public system and exacerbate Medicare’s current financial challenges. Two of the other proposed policies experts most frequently selected would provide incentives to employers to encourage them to hire long-term unemployed older workers, such as by offering temporary wage or training subsidies. Table 1: Experts’ Views on Key Strengths and Limitations of Selected Policy Proposals to Help Address Perceived Employer Reluctance to Hire Older Workers Experts’ favored policy proposals to help address perceived employer reluctance to hire older workers Eliminate the requirement that Medicare generally be Key strengths as identified by experts the secondary payer for workers covered by an Could minimize younger workers’ cost advantage to employers because of a employer group health plan. their lower health care costs, and by doing so would make hiring older workers more attractive to employers. Could be justified on the grounds of fairness, since working individuals 65 and over have paid their Medicare taxes and are entitled to the benefit. Key weaknesses as identified by experts Might not help that many older workers, or the neediest among them, because it would apply only to those 65 and over, who can typically collect Social Security retirement benefits, and not those 55-64. This would represent a cost to the government and exacerbate Medicare’s financial challenges, although it is unclear by how much. The cost might be limited because workers 65 and over tend to be in better health—and therefore less expensive to cover—than nonworkers 65 and over. 80 For workers aged 65 or over and covered by their own or their spouse’s employer’s group health plan, federal law generally requires that the employer’s group health plan be the primary payer and Medicare be the secondary payer for benefits. 42 U.S.C. § 1395y(b)(2)(A); 42 C.F.R. §§ 411.170, .172, .175. Page 49 GAO-12-445 Unemployed Older Workers May not increase the employment rate of those 65 and over very much, since many work part-time and are not covered by their employers’ health insurance plans. Congress could offer temporary wage subsidies to Key strengths as identified by experts employers that hire older workers who have Could influence employers’ behavior to make them more likely to hire older experienced long-term unemployment (27 or more workers. weeks). Limiting eligibility to older workers makes the policy more efficient by reducing the likelihood that the government would be subsidizing individuals that employers would have hired anyway. Could be cost-effective when the costs of long-term unemployment and entitlement programs are taken into account. Key weaknesses as identified by experts May lead some employers to postpone hiring to get the benefit, because it would make older workers unemployed for 27 or more weeks cheaper than those unemployed for a shorter period. May influence the type of workers employers hire, rather than increasing overall employment levels. May want to expand eligibility to all unemployed older workers, not just those unemployed for 27 or more weeks, because evidence indicates that unemployed older workers tend to remain unemployed for long periods of time. Potential for employer windfalls if employers would have hired the older worker even without the subsidies. Congress could offer training subsidies to employers Key strengths as identified by experts that hire older workers who experienced long-term May increase older workers’ employment levels because the training unemployment (27 weeks or more). To be eligible for subsidies would be led by employer demand and tied to an actual job. the training subsidy, employers would have to commit to retaining these workers for a certain amount of time Could help prevent older workers’ skills from atrophying because of (e.g., 6 months to a year). prolonged unemployment and help return them to their pre-unemployment productivity levels. Limiting eligibility to older workers makes the policy more efficient by reducing the likelihood that the government would be subsidizing individuals that employers would have hired anyway. Key weaknesses as identified by experts Would need to ensure that current employees of organizations receiving subsidies are given comparable levels of training so they are not at risk of subsequent displacement. Could pose equity issues, since workers of all ages have been affected by increased long-term unemployment. Potential for employer windfalls if employers would have provided this training even without the subsidies. Source: GAO summary of experts’ views on policy proposals selected by four or more experts. a For workers aged 65 or over and covered by their own or their spouse’s employer’s group health plan, federal law generally requires that the employer’s group health plan be the primary payer and Medicare be the secondary payer for benefits. 42 U.S.C. § 1395y(b)(2)(A); 42 C.F.R. §§ 411.170, .172, .175. Page 50 GAO-12-445 Unemployed Older Workers Experts most frequently selected three proposed policies to enhance the reemployment assistance the federal workforce development system currently provides to older workers (see table 2). One of these policies proposes that Labor develop a job search assistance program specifically targeted to older workers to provide training in basic computer skills, résumé writing, and online application filing. The second proposed policy would involve changing the WIA and SCSEP performance measures to remove a potential disincentive for serving older workers. 81 The third policy proposes increasing funding for the SCSEP program to reflect increases in the older worker population. Table 2: Experts’ Views on the Strengths and Limitations of Selected Policy Proposals to Enhance Reemployment Assistance Specifically Targeted to Older Workers Experts’ favored policy proposals to help enhance reemployment assistance specifically targeted to older workers Labor could develop a job search assistance program Key strengths as identified by experts specifically targeted to older workers that provides Could help older workers navigate the new, more advanced job search training in basic computer skills, résumé writing, online technologies that may be unfamiliar to them because they have not searched application filing, and other areas where older workers for a job in many years. may require specialized assistance. Job search assistance has proven beneficial to older workers and is generally more valuable to them than training, given their time left in the labor force. Could be efficient because of a potentially large take-up rate, ease of program development and implementation, and low costs. 81 Each state receiving funds under WIA is required to report annually the state’s and local areas’ progress on several performance measures based on indicators such as participants’ earnings, and Labor uses these measures to determine fiscal incentives and sanctions. 29 U.S.C. § 2871, 20 U.S.C. § 9273. The SCSEP program is subject to similar performance measures, 42 U.S.C. § 3056k, but does take the unique characteristics of the SCSEP population into account when setting performance goals for SCSEP grantees, according to Labor officials. For example, Labor officials said the average earnings performance goals set for SCSEP grantees reflects average earnings that result from more likely part-time work. In a previous GAO report, we found that Labor’s calculation of the earnings measure, which compares pre- and post enrollment earnings for participants in WIA adult and dislocated worker programs, could be a barrier to enrolling older workers because older workers’ high prior wages and their tendency to work part-time could negatively affect a local area’s performance on the earnings measure. Although Labor has changed the earnings measure so it no longer compares pre- and post enrollment earnings, the measure could still provide a disincentive for serving older workers because older workers may be more likely to work part-time and could still have lower earnings after exiting the program than other adults, since older dislocated workers generally suffer greater earnings losses than do younger workers. See GAO, Older Workers: Employment Assistance Focuses on Subsidized Jobs and Job Search, but Revised Performance Measures Could Improve Access to Other Services, GAO-03-350 (Washington, D.C.: Jan. 24, 2003). Page 51 GAO-12-445 Unemployed Older Workers Key weaknesses as identified by experts May overlap somewhat with services already offered to the general unemployed population. Does not address what is perceived as employers’ reluctance to hire older workers, which likely affects their reemployment prospects more than whether their job search skills are up to date. Congress could change WIA and SCSEP performance Key strengths as identified by experts measures to eliminate any disincentives to placing Could level the playing field for older workers, who may be penalized by the older workers in part-time employment. current measures, and make it easier to serve older workers through the one- stop career center system, especially those seeking part-time work. Could be implemented at a low cost because it primarily involves an administrative or technical change. Key weaknesses as identified by experts Unlikely to reduce unemployment by very much, since WIA and SCSEP serve only a fraction of eligible individuals in need of job search assistance. Could mean fewer younger workers will get services at one-stop career centers if funding levels do not increase; however, this is not a serious limitation if current measures unfairly penalize older workers. Congress could expand funding for SCSEP to take into Key strengths as identified by experts account increases in the older worker population. Increased funding could help meet the employment needs of a very disadvantaged and underserved population that many employers are unlikely to employ in the absence of severe labor shortages. Only a small percentage of eligible individuals are currently served by SCSEP; expanding funding to keep up with the aging population could help keep this percentage from declining. Research has shown that SCSEP has done a reasonably good job at accomplishing its goals. Key weaknesses as identified by experts Expanding the SCSEP program would cost money and could be difficult given the current budget environment. SCSEP is designed to serve the neediest older workers, and as a result, other subgroups of older workers that have been negatively affected by the recession would not benefit from this policy option. Source: GAO summary of experts’ views on policy proposals selected by four or more experts. Finally, as previously discussed, the longer older workers remain unemployed, the greater their risk of losing relevant skills or of dropping out of the labor force. Consequently, several proposed policies aim to encourage older workers to obtain reemployment as quickly as possible or build their job skills to better position them for reemployment. Experts we interviewed most frequently selected two such policy proposals (see table 3). Specifically, a number of experts we interviewed believed that implementing a wage insurance program could help unemployed older workers accept new full-time jobs that pay less than they had previously earned. In addition, a number of experts favored a policy that would Page 52 GAO-12-445 Unemployed Older Workers require the long-term unemployed to enroll in training to remain eligible for UI benefits. Table 3: Experts’ Views on the Strengths and Limitations of Selected Policy Proposals to Encourage Older Workers to Obtain Reemployment as Quickly as Possible Experts’ favored policy proposals to encourage older workers to regain employment as quickly as possible Congress could enact a wage insurance program that Key strengths as identified by experts temporarily compensates older workers—up to a specified Could help older workers transition to the lower wages they will maximum benefit—for accepting new full-time jobs that pay less likely receive upon becoming reemployed and reduce the likelihood than their previous jobs within a given time frame. of premature withdrawal from the labor force. Could help protect older workers’ retirement security and lessen their reliance on government assistance by reducing the likelihood that they will use retirement savings to replace lost income because of lower wages. Could result in less spending on UI for older workers, and those savings could help fund the wage insurance program. Key weaknesses as identified by experts Could be expensive for the government, since unemployed older workers almost always take a pay cut when they become reemployed. Might not affect older workers’ reemployment significantly, since the problem may be that they are unable to find work at any wage. May encourage older workers to settle for low-wage, less promising employment that increases the probability of a subsequent period of unemployment. May stigmatize unemployed older workers by putting them in a separate class of workers than everyone else. Congress could require long-term unemployed individuals (27 or Key strengths as identified by experts more weeks) to enroll in publicly funded retraining programs or Could help prevent older workers’ skills from eroding because of publicly subsidized on-the-job training programs as a condition long-term unemployment and alleviate concerns that UI benefits are of receiving UI benefits. Some of the training funds could be simply a government handout. obtained by redirecting a portion of individuals’ UI benefits for these purposes. Could potentially be less expensive for the government in the long run than providing UI benefits without conditions for retraining. Could help older workers adjust to the fact that today many job losses are permanent rather than temporary and that they may need to obtain new skills to become reemployed. Page 53 GAO-12-445 Unemployed Older Workers Key weaknesses as identified by experts The United States may lack the necessary training infrastructure and funding to launch a major training program that sufficiently prepares workers for jobs that pay decent wages and benefits. Would require an enhanced understanding of reemployment barriers and employers’ training needs before the policy could be implemented. Would need to ensure that the training provided is linked to actual employment opportunities to justify the financial investment in training. Requiring all long-term unemployed older workers to enroll in training may not make sense financially, particularly for those with a short remaining work life. Source: GAO summary of experts’ views on policy proposals selected by four or more experts. Labor Implemented Some Labor officials told us that Labor has taken several steps to implement Proposed Strategies to selected strategies recommended in 2008 by the Taskforce on the Aging Help Older Workers, but of the American Workforce. These steps included awarding approximately $10 million in grants to 10 organizations in 2009 through the Aging Has Shifted Priorities since Worker Initiative demonstration project to test new ways of providing the Recession training and other services to connect older Americans with employment opportunities in high-growth, high-demand industries. Labor is currently in the process of evaluating these grants. 82 Also, in 2008, Labor expanded a demonstration project designed to assist individuals in creating or expanding their own businesses, which addresses the Taskforce’s recommendation to facilitate self-employment for older workers. Specifically, Labor awarded a second round of grants to four demonstration sites through this project—Project GATE (Growing America Through Entrepreneurship). In addition, Labor officials told us that they had taken steps to connect one-stop career centers to Aging and Disability Resource Centers (which serve adults with disabilities, 82 According to Labor officials, the evaluation examines the implementation of the grants, documents the various types of interventions, assesses attributes of the treatments, estimates how successful they were in assisting aging workers in becoming employed or reemployed, and determines the potential for implementation of various methods in the broader workforce system. The final report is expected in December 2012, according to Labor officials. Page 54 GAO-12-445 Unemployed Older Workers including the elderly) to enhance communication. 83 In addition, Labor officials told us that since the recession began, the department has taken various steps to help the long-term unemployed, many of whom are older job seekers. Labor also continues to sponsor National Employ Older Workers Week, an event held annually in September in localities throughout the country. 84 As Labor officials we interviewed noted, the Taskforce conducted its work under different economic circumstances, when the workforce community was primarily focused on avoiding potential labor shortages. Since then, the number of unemployed individuals per job opening has greatly increased and technology has continued to change the job search and application process. Consequently, some issues may have assumed greater importance since the Taskforce issued its report, especially employer reluctance to hire older workers and the prevalence of online applications. According to Labor officials, the onset of the 2007-2009 recession shifted Labor’s focus away from implementing strategies recommended in the Taskforce report to responding to greatly increased demand for services. As more Americans lost jobs and struggled to find reemployment, increasing numbers sought reemployment services through the one-stop career center system, according to Labor officials. Administrative data Labor provided show that from 2007 to 2010 the overall number of all WIA adult and dislocated worker program participants the one-stop career centers served nearly tripled—from over a million participants in 2007 to 83 Aging and Disability Resource Centers are designed to serve as “one-stop shops” for individuals and their families who need information about or access to long-term support services. The centers are part of a collaborative effort led by the Administration on Aging (AoA) and Centers for Medicare and Medicaid Services (CMS). Other Aging and Disability Resource Center partners include the Department of Health and Human Services Office on Disability, the Administration for Developmental Disabilities, the Department of Education, and the Veterans Administration. 84 According to Labor officials, National Employ Older Workers Week “provides Senior Community Service Employment Program grantees with an opportunity to reach out to employers and the whole community as they recognize the vital role of older workers and their employers in the workforce.” Page 55 GAO-12-445 Unemployed Older Workers over 3 million in 2010. 85 The number of older workers served by these programs from 2007 to 2010 increased at an even greater rate—over 3.5 times—from approximately 124,000 in 2007 to around 441,000 in 2010. Although long-term unemployment hurts job seekers of all ages, it poses Conclusions some greater challenges for older workers. Specifically, once unemployed, older workers tend to stay unemployed longer, and those who regain employment generally sustain greater wage losses than do younger workers. The challenges older workers face once they lose their jobs also highlight the increased fragility of retirement security in this country. Long-term unemployment can reduce retirement income, and older Americans have fewer years to recover from such losses. A long spell of unemployment may even force some older Americans to leave the labor market and retire earlier than they had hoped. The high costs of long-term unemployment—overlaid upon the retirement insecurity facing so many workers—explain the different paths older workers are taking. Thus, some are using their retirement savings and taking Social Security retirement benefits early in response to this extended joblessness. Yet it is striking that other older workers are choosing to remain in the labor force longer as older workers’ labor force participation continues to rise despite the worst labor market in generations. At least part of this trend may be due to inadequate retirement savings or accounts that suffered losses from the financial crisis. According to experts we interviewed, some proposed policy options could help older workers regain employment as the economy continues to improve. In addition, a renewed focus by Labor on older workers’ needs could help workforce professionals better address the unique needs of older job seekers. While Labor took steps to implement some of the 2008 Taskforce recommendations, Labor officials understandably shifted their focus away from the report’s findings when the recent recession caused a dramatic increase in demand for workforce services. Now, a renewed 85 The administrative data from Labor for WIA adult and dislocated worker program participation is for program year 2007, starting April 1, 2007, through program year 2010, ending on March 31, 2011. In addition to increased demand for services, additional funding provided under the Recovery Act contributed to this increase in the number of individuals the programs served. Specifically, the Recovery Act provided an additional $500 million in funding for grants to the states for adult employment and training activities, and $1.25 billion for grants to the states for dislocated worker employment and training activities. Pub. L.No. 111-5, tit. VIII, 123 Stat. 115, 172-73 (2009). Page 56 GAO-12-445 Unemployed Older Workers focus on the needs of unemployed older workers is needed to identify strategies to help address older workers’ significant reemployment challenges because older workers remain a critical and growing segment of the workforce. This effort could include examining what has been learned since 2008 about addressing older workers’ employment needs in light of a changed economy, the shift to online employment applications, and employers’ altered expectations. Without a renewed focus on the unique needs of older job seekers, many unemployed older Americans may face difficulty regaining the employment they need to support themselves and their families in the short term, while also facing long- term financial hardship in retirement. To foster the employment of older workers, we recommend that the Recommendation Secretary of Labor consider what strategies are needed to address the unique needs of older job seekers, in light of recent economic and technological changes. We provided a draft of this report to the Department of Labor and the Agency Comments Social Security Administration. Labor provided a written response (see and Our Evaluation app. VII). Both agencies provided technical comments, which we incorporated as appropriate. Labor agreed with our recommendation and noted a couple of its initiatives focused on the employment of older workers. Specifically, Labor cited its current evaluation of the Aging Worker Initiative demonstration project, which will assess the success of new interventions used by 10 local grantees to help connect aging workers with employment opportunities. In addition, Labor cited its sponsorship of the annual National Employ Older Workers Week that provides outreach opportunities for SCSEP grantees. Page 57 GAO-12-445 Unemployed Older Workers As agreed with your office, unless you publicly announce its contents earlier, we plan no further distribution until 30 days after the date of this letter. At that time, we will send copies of this report to the Secretary of Labor and the Commissioner of Social Security, and other interested parties. In addition, this report will be available at no charge on GAO’s website at http://www.gao.gov. If you or your staff have any questions concerning this report, please contact me at (202) 512-7215 or firstname.lastname@example.org. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff who made key contributions to this report are listed in appendix VIII. Sincerely yours, Charles Jeszeck Director, Education, Workforce, and Income Security Page 58 GAO-12-445 Unemployed Older Workers Appendix I: Objectives, Scope, and Appendix I: Objectives, Scope, and Methodology Methodology The objectives of this study were to examine (1) how the employment status of older workers has changed since the recession, (2) older workers’ financial risks from long-term unemployment and challenges in finding new jobs, (3) how periods of long-term unemployment might affect older workers’ retirement income, and (4) what other policies might help unemployed older workers regain employment and what steps the Department of Labor (Labor) has taken to help unemployed older workers. We conducted this performance audit from October 2010 through April 2012 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence we obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. We determined that the data that we analyzed were sufficiently reliable for the purposes of this report. To obtain background information on older workers’ employment and retirement prospects, we reviewed numerous studies, reports, and surveys of older workers. We interviewed officials from Labor and the Social Security Administration (SSA) and reviewed relevant data from those agencies. Analysis of Employment To examine changes in the employment prospects of older workers since Data on Older Workers the start of the recession, we analyzed monthly data for 2007 through 2011 from the Current Population Survey (CPS) produced by the Bureau of Labor Statistics (BLS). We also analyzed data from the 2008 and 2010 Displaced Worker Supplement to the CPS. We selected the CPS mainly because it is nationally representative and contains large sample sizes, demographic and industry information, and data directly relevant to unemployment and underemployment. Similarly, we analyzed the 2008 and 2010 Displaced Worker Supplement to the CPS because this supplement contains data on the employment and earnings status of displaced workers before and after their job loss. Displaced workers are defined as persons 20 years or older who lost or left jobs within the past 3 years for the following reasons: because their plant or company closed or moved, there was insufficient work for them to do, their position or shift was abolished, or similar economic reasons. The Displaced Worker Survey (DWS) is administered every 2 years as a supplement to the CPS. In the 2008 DWS, people are identified as Page 59 GAO-12-445 Unemployed Older Workers Appendix I: Objectives, Scope, and Methodology displaced if they lost or left their jobs for one of the specified reasons between January 2005 and December 2007. In the 2010 DWS, people are identified as displaced if they lost or left their jobs for one of the specified reasons between January 2007 and December 2009. Displaced workers have lost a job in the past 3 years; however, they may be unemployed, employed, or not in the labor market at the time of the survey. We use the following labor force definitions in this report: • Unemployed workers are all jobless persons who are available to take a job and have actively sought work in the past 4 weeks. • Marginally attached workers are persons who are not in the labor force, who want and are available for work, and have looked for work in the past 12 months. They are not counted as unemployed because they had not searched for work in the prior 4 weeks. • Discouraged workers are a subset of the marginally attached who indicate that they have not searched for work in the prior 4 weeks for the specific reason that they believed no jobs were available for them. • Workers employed part-time for economic reasons are those employed less than 35 hours per week who want and are available for, but are unable to find, full-time work, and those who prefer full- time work but had their hours reduced by their employer because of business conditions. We assessed the reliability of the CPS generally and of data elements that were critical to our analyses and determined that they were sufficiently reliable for our analyses. Specifically, we • reviewed documentation on the general design and methods of the CPS and on the specific elements of the CPS data that were used in our analyses, • interviewed BLS officials knowledgeable about the CPS data and consulted these officials periodically throughout the course of our study, and • completed our own electronic data testing to assess the accuracy and completeness of the data used in our analyses. To the extent Page 60 GAO-12-445 Unemployed Older Workers Appendix I: Objectives, Scope, and Methodology possible, we compared our estimates against published reports using the CPS data, such as BLS reports. As a result of our assessment, we identified a limitation with the CPS basic monthly data. Specifically, beginning in January 2011, labor force estimates based on microdata diverge slightly from BLS published labor force estimates. According to Census Bureau officials, these discrepancies result from new steps that CPS implemented in 2011 to help prevent the inadvertent disclosure of individuals in the public use files. While some primary topside labor force estimates will agree, all others will be slightly off. This is solely the result of these disclosure protection procedures. Census Bureau officials told us that the masking procedures implemented in 2011 are random and are applied to all records on the file identically; they do not create any systematic biases that would affect our analysis of older workers. Throughout this report, when monthly data are presented, the estimates are not seasonally adjusted. We were advised by Census Bureau officials not to attempt to seasonally adjust our analyses of CPS microdata, because the sample sizes in the age groups we present are relatively small. Most of the data in this report are annual averages, for which seasonal adjustments are irrelevant. Where monthly data are presented, unadjusted estimates and seasonally adjusted estimates do not necessarily diverge by a large amount. For example, this table shows seasonally adjusted data for figure 2, compared with the unadjusted data shown in table 4. Table 4: Comparison of Seasonally Adjusted and Unadjusted Unemployment Rates for Workers Age 55 and Older, Selected Months Unemployment rate, age 55 and Unemployment rate, age 55 and older, unadjusted older, seasonally adjusted December 2007 3.1% 3.2% June 2009 6.8% 7.0% February 2010 7.6% 7.2% December 2011 6.0% 6.2% Source: GAO analysis of CPS public use microdata and published data series downloaded from BLS website. Note: Estimates of unadjusted unemployment rates calculated by GAO using CPS public use microdata. These estimates are identical to BLS’s published unadjusted unemployment rates for workers age 55 and older. Estimates of seasonally adjusted unemployment rates for workers age 55 and older are downloaded from BLS’s website. Page 61 GAO-12-445 Unemployed Older Workers Appendix I: Objectives, Scope, and Methodology We also use nationally representative data from the Job Openings and Labor Turnover Survey (JOLTS), a monthly survey developed by BLS to address the need for data on job openings, hires, and separations. We use the JOLTS data to present a comparison of the number of unemployed persons with the number of job openings over the period January 2007 to December 2011. This is the only national data source available to measure job openings in the United States. Focus Groups with Older To learn about older workers’ financial risks from long-term Workers and Interviews unemployment and challenges in finding new jobs, we conducted 10 with One-Stop Staff and focus group sessions with a total of 77 long-term unemployed older workers aged 55 or over at four locations. These sessions involved Other Experts structured small-group discussions designed to gain more in-depth information about specific issues that cannot easily be obtained from another method, such as a survey or individual interviews. Consistent with typical focus group methodologies, our design included multiple groups with varying characteristics but some similarity on one or two homogeneous characteristics. All but one of the groups involved 8 to 10 participants; the remaining group had only 5 participants because of poor attendance. Our overall objective in using a focus group approach was to obtain views, insights, and feelings of older workers who had been looking for a job for more than half a year. Specifically, we wanted to learn what challenges they had faced since becoming unemployed, what barriers they perceived as hindering their ability to become reemployed, and their views on how their spell of unemployment had affected or would likely affect their retirement income and decisions. By including long-term unemployed older workers with and without employer-sponsored retirement plans, we intended to gather a range of perspectives regarding how unemployment might affect retirement prospects. We conducted 10 separate focus group sessions with long-term unemployed older workers. Specifically, focus group participation was limited to individuals 55 or older who had been unemployed for 27 weeks or more and who had worked for their previous employer for at least 3 years before losing their job. We held three sessions with individuals aged 55-61 who did not have an employer-sponsored retirement plan at the job they lost, three sessions with individuals aged 55-61 who did have an employer-sponsored retirement plan at the job they lost, and three sessions with individuals aged 62-67 regardless of their retirement plan coverage. We also conducted one focus group session in Falls Church, Page 62 GAO-12-445 Unemployed Older Workers Appendix I: Objectives, Scope, and Methodology Virginia. We conducted this session as a pretest, but because we did not need to significantly change our focus group guide after the session, we decided to include the results of the pretest in our focus group analysis. We selected three cities—in addition to our pretest location in Falls Church, Virginia—in which to conduct focus groups. We selected these locations based on metropolitan areas’ unemployment rates, geographic diversity, and the estimated costs for travel and securing focus group facilities. We conducted three sessions in each of the following three cities—Baltimore, Maryland; San Jose, California; and St. Louis, Missouri. Additionally, we used criteria in selecting participants that ensured a mix of gender and that accounted for the race and ethnicity of the area in which the focus groups were located. Discussions were structured, guided by a moderator who used a standardized list of questions to encourage participants to share their thoughts and experiences. During the sessions, we informed participants that their names would not be used in the published report. We conducted one pretest focus group session prior to beginning our travel for the sessions. Each of the 10 focus groups was recorded and transcriptions were created, which served as the record for each group. Those transcripts were then evaluated using content analysis to develop our findings. The analysis was conducted in two steps. In the first step, three analysts jointly developed a set of codes to track the incidence of various responses and themes during focus group sessions. In the second step, each transcript was coded by an analyst and then those codes were verified by a second analyst. Any coding discrepancies were resolved by both analysts agreeing on what the codes should be. In addition to focus group sessions, we conducted one-on-one interviews with selected long- term unemployed individuals in three of our four focus group locations. Methodologically, focus groups are not designed to (1) demonstrate the extent of a problem or to generalize results to a larger population, (2) develop a consensus to arrive at an agreed-upon plan or make decisions about what actions to take, or (3) provide statistically representative samples or reliable quantitative estimates. Instead, they are intended to generate in-depth information about the reasons for the focus group participants’ attitudes on specific topics and to offer insights into their concerns about and support for an issue. The projectability of the information produced by our focus groups is limited for several reasons. First, the information includes only the responses from long-term Page 63 GAO-12-445 Unemployed Older Workers Appendix I: Objectives, Scope, and Methodology unemployed older workers from the 10 selected groups. Second, while the composition of the groups was designed to ensure a range of age, retirement plan coverage, and racial background, the groups were not randomly sampled. Third, participants were asked questions about their experiences or expectations, and other long-term unemployed older workers not in the focus groups may have had other experiences or expectations. Because of these limitations, we did not rely entirely on focus groups, but rather used several different methods to corroborate and support our conclusions. We also interviewed staff at one-stop career centers in each of the locations where we conducted focus groups to learn more about challenges unemployed older workers face in finding employment. Further, we interviewed experts about older workers’ reemployment challenges. We selected these experts based on their knowledge of older workers’ issues, labor economics, and the workforce development system. Specifically, we used several criteria to select experts to interview, such as (1) having conducted research and published studies on relevant topics (including older workers, the workforce development system, labor economics, or retirement issues) or (2) representing associations with highly established awareness or knowledge of issues relevant to the employment and retirement prospects of older workers. How Unemployment Might To assess how periods of long-term unemployment might affect older Affect Retirement Income workers’ retirement income, we used the Policy Simulation Group’s (PSG) microsimulation models to simulate Social Security benefits and retirement plan income. For our simulations, we used PSG’s Social Security and Accounts Simulator (SSASIM), Genuine Microsimulation of Social Security Acounts (GEMINI), and Pension Simulator (PENSIM) simulation models. GEMINI simulates Social Security benefits and taxes for large, representative samples of people born in the same year. GEMINI simulates all types of Social Security benefits, including retired workers’, spouses’, survivors’, and disability benefits. It can be used to model a variety of changes to Social Security. GEMINI uses inputs from SSASIM, which has been used in numerous GAO reports, and PENSIM, which was developed for the Department of Labor. GEMINI relies on SSASIM for economic and demographic projections and relies on PENSIM for simulated life histories of large, representative samples of Page 64 GAO-12-445 Unemployed Older Workers Appendix I: Objectives, Scope, and Methodology people born in the same year and their spouses. 1 Life histories include educational attainment, labor force participation, earnings, job mobility, marriage, disability, childbirth, retirement, and death. Life histories are validated against data from the Survey of Income and Program Participation, the Current Population Survey, Modeling Income in the Near Term (MINT3), 2 and the Panel Study of Income Dynamics. Additionally, any projected statistics (such as life expectancy, employment patterns, and marital status at age 60) are, where possible, consistent with intermediate cost projections from the Social Security Administration’s Office of the Chief Actuary. At their best, such models can provide only very rough estimates of future incomes. However, these estimates may be useful for comparing future incomes across alternative policy scenarios and over time. In order to compare pretax retirement benefits accumulated at different ages, we simulated and analyzed retirement benefits for three cohorts of individuals born in 1955. For the first cohort, we conducted the simulation to have all individuals stop work at age 55 and estimated the retirement benefits earned by that age for each individual in the cohort. Similarly, for the second and third cohorts, we simulated that all individuals in the cohort stopped working at ages 58 and 62, respectively, and estimated pretax retirement benefits earned by those ages. In the simulations, for each cohort, we had all individuals claim their benefits at age 62, regardless of when they stopped working. We determined the median level of benefits from employer-sponsored retirement plans and Social Security using all individuals in each cohort, including those who would have had no retirement benefits from an employer-sponsored plan because they did not vest in benefits or they cashed out of their retirement plan when they stopped working. To compare how job loss at different ages may affect retirement benefits, we compared the retirement benefits for the individual at the median level of retirement benefits in each cohort. For our analysis of how job loss might affect retirement benefits for those participating in employer-sponsored plans, we only 1 While these models use sample data, our report, like others using these models, does not address the issue of sampling errors. The results of the analysis reflect outcomes for individuals in the simulated populations and do not attempt to estimate outcomes for an actual population. 2 MINT3 is a detailed microsimulation model developed jointly by SSA, the Brookings Institution, RAND, and the Urban Institute to project the distribution of income in retirement for the 1931 to 1960 birth cohorts. Page 65 GAO-12-445 Unemployed Older Workers Appendix I: Objectives, Scope, and Methodology included individuals who were participating in an employer-sponsored retirement plan at the job they had when they stopped working. Our analysis took into consideration retirement benefits earned by individuals in the cohort and retirement benefits for the surviving spouse of someone who vested in a DB plan. Our analysis includes retirement income from employer-sponsored retirement plans and Social Security retirement benefits and does not include retirement income from other sources. PENSIM uses several different asset allocations for defined contribution (DC) accounts and assigns each type of allocation to a portion of individuals in the cohort, based on existing research on actual use of different asset allocations. For our analysis, the PENSIM model used the assumption that all individuals who vested in employer-sponsored retirement benefits and did not cash out their benefits when they left the job before age 62 used DC plan savings to purchase an annuity at age 62. In the model the annuity prices are based on projected mortality and interest rates using annuity price loading factors that ensure that the cost of providing these annuities equals the revenue generated by selling them at those prices. We used nationally representative 2007 Survey of Consumer Finances (SCF) data from the Board of Governors of the Federal Reserve to estimate the percentage of employed individuals age 55 and older (working full- or part-time) who were participating in employer-sponsored retirement plans or had private retirement savings before the recession caused substantial job losses. 3 In addition to providing household-level data, the SCF also provides detailed individual-level economic information about an economically dominant single individual or couple in the household. To estimate the percentage of older workers with defined benefit (DB) plans, we included in our analysis any employed older worker who (1) was participating in an employer-sponsored DB plan at a current job, (2) had participated in an employer-sponsored DB plan at a past job and was expecting to receive retirement benefits from the plan, or (3) was already receiving payments from an employer-sponsored DB plan. Similarly, to estimate the percentage of older workers with a DC plan, we included any older workers with an account from a DC plan from 3 The Survey of Consumer Finances data that we used were collected from May 2007 to March 2008. The collection period includes a few months of the recession that started in December 2007; however, the significant increases in unemployment that followed the onset of the recession had not yet occurred. Page 66 GAO-12-445 Unemployed Older Workers Appendix I: Objectives, Scope, and Methodology a current or past employer, and we counted rollover individual retirement accounts (IRA) as DC accounts. We produced analogous estimates for households with an employed head of household or spouse age 55 or over. Our estimates for households only include retirement benefits and savings of the head of household and a spouse or partner and do not include retirement benefits or savings held by additional family members. We assessed the reliability of the SCF generally and of data elements that were critical to our analyses and determined that they were sufficiently reliable for our analyses. Specifically, we reviewed documentation on the general design and methods of the SCF and on the specific elements of the SCF data that were used in our analyses and completed our own electronic data testing to assess the accuracy and completeness of the data used in our analyses. To illustrate how drawdowns from retirement savings before retirement can affect amounts saved by the time of retirement, we developed a model to tabulate retirement account balances for a hypothetical individual. For this, we used an individual who was born at the beginning of 1953, turns 55 in 2008, and retires at age 62 in 2015. We used $70,000 as the starting point for this illustration because it is about the median level of DC plan savings for employed workers age 55 and over who have a DC plan account from a current or past employer, based on 2007 SCF data. To calculate changes in the account balance over time, we used the intermediate interest and rate-of-return assumptions as reported in past and projected under the intermediate cost assumptions in Social Security’s 2011 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Federal Disability Insurance Trust Funds (also known as the OASDI [Old-Age, Survivors, and Disability] Trustees’ Report). We used scaled earnings for medium annual earners as reported in past and projected in the 2011 OASDI Trustees’ Report. 4 We assumed the employee’s contributions to the retirement account to be 6 percent of the individual’s wages and the employee received a 3 percent employer matching contribution. 4 Scaled earnings are earnings levels that have been scaled up or down to reflect the average patterns of work and earnings of actual insured workers over their careers. Page 67 GAO-12-445 Unemployed Older Workers Appendix I: Objectives, Scope, and Methodology Experts’ Views on Policies The 12 experts we interviewed primarily work in academia and think to Help Unemployed Older tanks, and were selected based on their expertise on older workers, Workers and Steps Taken workforce development, or retirement policy. We selected a group of experts to help ensure a range of viewpoints. To identify policies that by Labor to Help could help unemployed older workers become reemployed, we compiled Unemployed Older a list of 21 proposed policy options from the relevant literature. (See app. Workers VI for the list of 21 policy options.) In many cases, the policy options on the list we compiled were proposed before the recession started in 2007. Prior to interviewing the experts, we sent them a list of the 21 policy proposals we had compiled and asked them to select the five policies that merited the most serious consideration because they were either relevant in the current economy or could be relevant as the economy recovers. We also asked these experts to assess the strengths and weaknesses of policy proposals to help long-term unemployed older workers regain employment. After interviewing all 12 experts, we tallied the number of votes each policy proposal had received. The 8 proposed policies addressed in the body of the report are those that received votes from 4 or more of the 12 experts we interviewed. The expert discussion cited in this report should be interpreted in the context of two key limitations and qualifications. First, although we were able to secure the participation of a balanced, highly qualified group of experts, other experts in this field could not be included because we needed to limit the size of the group. Although many points of view were represented, the group of experts we interviewed was not representative of all potential views. While we conducted preliminary research and heard from national experts in their fields by conducting these expert interviews, these discussions cannot represent the full variety of opinions on the policy proposals. More thought, discussion, and research must be done to develop greater agreement on what we really know, what needs to be done, and how to do it. These two key limitations and qualifications provide contextual boundaries. Nevertheless, the experts we interviewed provided insightful comments in responding to the questions they were asked. To identify what steps Labor has taken to help unemployed older workers, we interviewed Labor officials. Sampling Variability As noted above, we relied on estimates from several surveys, including the CPS, the Displaced Worker Supplement to the CPS, SCF, and JOLTS. Because these surveys are probability samples based on random selections, the specific sample selected is only one of a large number of Page 68 GAO-12-445 Unemployed Older Workers Appendix I: Objectives, Scope, and Methodology samples that might have been drawn. Since each sample could have provided different estimates, we express our confidence in the precision of our particular sample’s results as a 95 percent confidence interval (for example, plus or minus 7 percentage points). This is the interval that would contain the actual population value for 95 percent of the samples that could have been selected. In this report, 95 percent confidence intervals are provided along with sample-based estimates where used. We calculated standard errors for estimates from the CPS using the Census generalized variance functions, as published in the BLS technical notes to the household survey data published in the Employment and Earnings monthly publication. Page 69 GAO-12-445 Unemployed Older Workers Appendix II: Workforce Status of Workers Appendix II: Workforce Status of Workers Displaced between 2005-2007 and 2007-2009 in January 2008 and January 2010 Displaced between 2005-2007 and 2007-2009 in January 2008 and January 2010 Note: (1) Estimates for 25- to 54-year-olds and 55- to 64-year-olds have 95 percent confidence intervals within 5 percentage points of the estimate itself; estimates for people 65 and older have confidence intervals within 10 percentage points of the estimate itself. The following estimates have margins of error greater than or equal to 30 percent of the estimates: for people 55-64, the “percentage retired” in 2008, for people 65 and older, the percentage employed full-time, percentage employed part-time, and percentage not in the labor force (other) in both 2008 and 2010. For statistical comparisons of the estimates across different groups and years, see appendix III. Page 70 GAO-12-445 Unemployed Older Workers Appendix III: Additional Figure Notes for Appendix III: Additional Figure Notes for Employment Figures, Including Statistical Significance Tests Employment Figures, Including Statistical Significance Tests Figure 3: Estimated Number of Unemployed and Underemployed Older Workers (55 and Over), 2007-2011 • Appendix I includes the definitions used in the report of unemployed workers, marginally attached workers, discouraged workers, and workers employed part-time for economic reasons. Figure 4: Growth in Estimated Long-Term Unemployment of Older Workers (55 and Over), 2007-2011 • There was a statistically significant change in the proportion of unemployed older workers in each of the categories shown in the figure between 2007 and 2011. Specifically: • The proportion of unemployed older workers who were unemployed for under 5 weeks, for 5-14 weeks, and for 15-26 weeks each declined significantly from 2007 to 2011. • The proportion of unemployed older workers who were unemployed for 27 weeks to a year, and for more than 1 year, each increased significantly from 2007 to 2011. Figure 6: Estimated Duration of Unemployment for Older Workers (55 and Over) by Industry, 2007 and 2011 • Because of the smaller number of persons with long durations of unemployment in 2007, some of the estimates for duration of unemployment by industry are unreliable in 2007 (the margins of error for the estimates are high relative to the value of the estimates.) For 2007, the following estimates have margins of error that exceed 30 percent of the estimate itself: the percentage unemployed for 27 weeks or more for all industries except wholesale and retail trade, manufacturing, and “all other industries”; and for the percentage employed for 27-52 weeks, 53-104 weeks, and 105 or more weeks for all industries. In 2011 the following estimates in the figure have margins of error that exceed 30 percent of the estimate itself: the percentage unemployed for 53-104 weeks in leisure and hospitality, and the percentage unemployed for 105 or more weeks in: transportation and utilities, financial activities, and leisure and hospitality. Page 71 GAO-12-445 Unemployed Older Workers Appendix III: Additional Figure Notes for Employment Figures, Including Statistical Significance Tests • For each industry in the figure, the percentage of unemployed workers who have been unemployed for 27 weeks or more is significantly higher in 2011 than in 2007. Figure 7: Estimated Unemployment Rates by Demographic Group for Older Workers (55 and Over), 2007 and 2011 • Differences in unemployment rates between 2007 and 2011 are statistically significant for all groups in this figure. • Differences in unemployment rates between men and women are statistically significant in 2011, but not in 2007. • Differences in unemployment rates between whites and blacks, and between whites and Hispanics, are statistically significant in both 2007 and 2011; however, differences between whites and the “all other races” group are significant only in 2011. • Differences in unemployment rates between those with no high school diploma and each of the other education groups are statistically significant in both 2007 and 2011. Figure 8: Estimated Duration of Unemployment by Age and by Gender, 2007 and 2011 • Differences between younger and older workers in the percentage unemployed for 27 weeks or longer were significant in both 2007 and 2011. • Differences between men and women in the percentage unemployed for 27 weeks or longer were statistically significant in 2007, but not in 2011. Figure 10: Estimated Percentage of Reemployed Displaced Workers Who Earned Less on Their New Full-Time Jobs than on Their Previous Jobs, January 2010 • Differences between younger workers and older workers are statistically significant in the 2010 survey, but not in the 2008 survey. Page 72 GAO-12-445 Unemployed Older Workers Appendix III: Additional Figure Notes for Employment Figures, Including Statistical Significance Tests • For both younger workers and older workers, there was a statistically significant increase in the percentage of workers with earnings replacement rates less than 100 percent between the 2008 survey and the 2010 survey. Appendix II: Workforce Status of Workers Displaced between 2005-2007 and 2007-2009 in January 2008 and January 2010 • For 25- to 54-year-olds and 55- to 64-year-olds, the following estimates are significantly different between 2008 and 2010: • the percentage unemployed, • the percentage working full-time, and • the overall percentage working/not working. • For people 65 and older, the following estimates are significantly different between 2008 and 2010: • the percentage unemployed, • the percentage retired, and • the percentage not in the labor force (other). Page 73 GAO-12-445 Unemployed Older Workers Appendix IV: Quotes from Focus Groups with Appendix IV: Quotes from Focus Groups with Long-Term Unemployed Older Workers Long-Term Unemployed Older Workers The following table provides examples of specific focus group quotes organized by topic. These selected quotes further exemplify our findings on the risks long-term unemployed older workers face, the challenges they experience becoming reemployed, and how long-term unemployment has affected their plans for retirement. (For audio clips from GAO’s focus groups with unemployed older workers, use this http://www.gao.gov/multimedia/video/#video_id=590295: ) Table 5: Selected Quotes from Focus Groups with Long-Term Unemployed Older Workers Focus group participant Topic/quotes characteristics Some unemployed older workers want to work and have difficulty meeting financial obligations That’s our culture. We want to work. We need to work. 56-year-old man You’ve got to support a family, so you’re still out there on the pavement or the Internet looking for a 55-year-old man job. I was used to a 70-hour week, and then to go from that to nothing, your sense of self-worth isn’t there. 58-year-old woman Medical can cost you $1,000 a month. When you think about Social Security at my age group—my 64-year-old woman Social Security is $1,300 a month. Medical is $1,000, where is my mortgage and food? I don’t have medical insurance right now. Any medical insurance. Because it is too expensive. 63-year-old woman I’ll never retire. I still have a kid in college. My son lost his job about 6 months after I did, so my wife 55-year-old man and I are paying his mortgage, his car insurance, and everything he has, electric, all that stuff. Perceived employer reluctance to hire older workers The interview was going really, really well and … she went, “Oh no!” and I said, “What is it?” [And she 59-year-old man said], “Oh nothing, nothing, I just noticed what year you graduated high school.” They asked me one question that always knocked me out of the game—they’re not allowed to ask 62-year-old man how old you are, but they wanted to know when I graduated from college. I would sacrifice pay to start somewhere else. I say, “I’ll start at rock bottom . . . whatever it takes.” 59-year-old man I need a job. I have to pay insurance for me and my wife. I have seven kids. I’m trying to get them 64-year-old man through college. I’ll take that $10 job. I need a job that bad. Emotional challenges that result from long-term unemployment When you’re not working, you don’t feel very good, you are depressed. You feel discouraged. Your 62-year-old woman self-esteem is about an inch high. One of the most devastating things with becoming unemployed was with losing my identity. I was one 58-year-old woman thing and then I was nothing. Claiming Social Security retirement benefits as soon as possible I’m turning 62 in a few days, and I’m like “What am I going to do?” You can’t find a job out there so I 61-year-old man might as well retire. I just turned 62 in January, and I just filed for my Social Security so that I would have some kind of 62-year-old woman income to fall back on, because unemployment was exhausted. If I don’t find a job, I’ll claim at 62 because of, you know, unemployment will be long gone by then. And 59-year-old woman if I do sell that house, anything I have left after the sale will be gone by then, too. So, if I find a job, I won’t claim unless or until I have to for health reasons or whatever. Page 74 GAO-12-445 Unemployed Older Workers Appendix IV: Quotes from Focus Groups with Long-Term Unemployed Older Workers Focus group participant Topic/quotes characteristics I want to secure my place because if the eligibility requirement on that changed, then you’re out of 58-year-old woman luck, too. It’s scary. They’re talking about cutting it off, so I think anybody that’s eligible for it, they need to try and go and 66-year-old woman apply for it. Using retirement savings to cover expenses during long-term unemployment Coping financially has been really rough, because even though I had a retirement plan, it’s not enough 59-year-old woman to pay my mortgage and my car payment. . . . So then you tap into what you have saved . . . you see that dwindle down to nothing and then you still got to make it to 62, and you don’t have enough to make it to 62. When I got laid off . . . my daughter was in college at the time . . . I wanted her to finish school so, of 55-year-old woman course, I depleted all of my savings, my 401(k). Source: Transcripts of focus groups GAO conducted with long-term unemployed older workers. Page 75 GAO-12-445 Unemployed Older Workers Appendix V: Percentage of the Population 50 and Older, Appendix V: Percentage of the Population 50 and Older, but Less than Full Retirement Age, with Initial Dispositions and Awards of Social but Less than Full Retirement Age, with Initial Security Disability Insurance, 2000-2010 Dispositions and Awards of Social Security Disability Insurance, 2000-2010 Page 76 GAO-12-445 Unemployed Older Workers Appendix VI: Proposed Policies Presented to Appendix VI: Proposed Policies Presented to External Experts External Experts Selected by at least Policy option four experts Policies to help address perceived employer reluctance to hire older workers 1. Labor could develop a comprehensive and highly visible campaign to educate employers on the strengths NO of older workers as potential employees and their important role in the 21st century economy. 2. Congress could offer temporary wage subsidies to employers that hire older workers who have YES experienced long-term unemployment (27 or more weeks). 3. Congress could offer training subsidies to employers that hire older workers who experienced long-term YES unemployment (27 weeks or more). To be eligible for the training subsidy, employers would have to commit to retaining these workers for a certain amount of time, for example, 6 months to a year. 4. Congress could offer a tax credit—that applies to the first year of employment—to employers that hire NO older workers who experienced long-term unemployment (27 weeks or more). 5. To reduce the potentially high cost to employers of providing health insurance benefits to older workers, YES Congress could eliminate the requirement that Medicare generally be the secondary payer for workers 65 and over who are covered by an employer group health plan. 6. Congress could allow older workers and their employers to opt out of paying the Social Security payroll NO tax once the worker has accumulated 35 years of covered earnings. 7. Congress could provide tax credits to new small businesses that are owned by older workers and employ NO a high percentage of older workers (such as 50 percent or more). 8. Congress could pass legislation (such as the proposed Fair Employment Opportunity Act of 2011) that NO prohibits employers and employment agencies from screening workers out of the candidate pool solely because they are unemployed. 9. To alleviate employer concerns about declining productivity and increased health care costs at older ages NO (at or near Social Security’s full retirement age), Congress could allow employers to set a mandatory retirement age (of 66 or greater) for new hires who were within 10 years of reaching this age (e.g., 56 or older). 10. In partnership with other federal agencies, Labor could identify the legal and regulatory barriers to the NO employment of older workers and determine (1) whether legal and regulatory changes are needed and (2) the impact of potential changes. Policies to help enhance reemployment assistance specifically targeted to older workers 11. Labor could collect and disseminate information on the most effective strategies for serving older workers NO through the workforce investment system. 12. Labor could develop job search assistance programs that address skill deficiencies common among YES seniors, such as deficiencies in basic computer skills, résumé writing, and online application filing. 13. Labor could encourage partnerships between one-stop career centers and the Small Business NO Administration (SBA) to provide entrepreneurial development services to older job seekers. As part of this effort to link older workers with SBA’s programs and services, SBA could provide targeted information to older workers regarding entrepreneurial resources, counseling, and training on its website. 14. To directly assist older workers who wish to start their own businesses, Labor could replicate Project NO GATE (Growing America Through Entrepreneurship) at one-stop career centers, targeted specifically to older workers. 15. Labor could direct the Bureau of Labor Statistics to work with the Census Bureau to (1) add specific NO questions to the CPS about older workers and their labor force participation, and (2) request a special tabulation of census data on older workers designed to match workforce investment areas. These data would allow one-stop staff to better target services to the different types of older workers within their jurisdiction. Page 77 GAO-12-445 Unemployed Older Workers Appendix VI: Proposed Policies Presented to External Experts Selected by at least Policy option four experts 16. Congress could change Workforce Investment Act of 1998 (WIA) and Senior Community Service YES Employment Program (SCSEP) performance measures to eliminate any disincentives to placing older workers in part-time employment. 17. Congress could expand funding for SCSEP to take into account increases in the older worker population. YES Policies to help encourage older workers to obtain reemployment as quickly as possible 18. Congress could enact a reemployment bonus program that provides a low-value bonus (for example, NO approximately $1,000) to Unemployment Insurance (UI) claimants who accept new jobs within a given time period (for example, 3-6 months). The program could target UI claimants with an above average likelihood of exhausting their UI benefits. 19. Congress could enact a wage insurance program that temporarily compensates older workers for YES accepting new full-time jobs that pay less than their previous jobs within a specified time frame. For example, the program could pay older workers 50 percent of the difference between their old and new wages over a period of 2 years (up to a specified maximum benefit). The subsidy could be limited to individuals making less than $50,000 at their new jobs and who accepted their new jobs within 27 weeks of filing for UI. 20. Labor could encourage one-stop career centers to prioritize job-matching services (i.e., matching older NO workers’ existing skills with available jobs) rather than training or retraining services. 21. Congress could require long-term unemployed individuals (those unemployed for 27 or more weeks) to YES enroll in publicly funded retraining programs or publicly subsidized on-the-job-training programs as a condition of receiving Unemployment Insurance. Some of the training funds could be obtained by redirecting a portion of individuals’ unemployment benefits for these purposes. Source: GAO. Note: To compile this list of policy options, we reviewed the literature, including academic studies, past GAO reports, and federal agency reports, to identify policies that have been proposed to help unemployed older workers regain employment. We presented these options to 12 experts in areas such as older workers’ issues, labor economics, and the workforce development system, and asked the experts to select the five policies that they believed merited serious consideration. Policies that were selected by 4 or more experts are described in more detail in the body of the report. Page 78 GAO-12-445 Unemployed Older Workers Appendix VII: Comments from the Appendix VII: Comments from the Department of Labor Department of Labor Page 79 GAO-12-445 Unemployed Older Workers Appendix VII: Comments from the Department of Labor Page 80 GAO-12-445 Unemployed Older Workers Appendix VIII: GAO Contact and Staff Appendix VIII: GAO Contact and Staff Acknowledgments Acknowledgments Charles Jeszeck, (202) 512-7215 or email@example.com GAO contact In addition to the contact named above, Laura J. Heald, Assistant Staff Director; Lucas Alvarez; Laurel E. Beedon; James E. Bennett; Amy Buck; Acknowledgments David M. Chrisinger; William Colvin; Sarah C. Cornetto; Cynthia L. Grant; Gene G. Kuehneman Jr.; Kathy D. Leslie; Douglas A. Manor; Jaclyn Nidoh; Rhiannon Patterson; Kathy Peyman; Mark F. Ramage; David M. Reed; Nyree M. Ryder Tee; Aron E. Szapiro, Frank Todisco; and Walter Vance made key contributions to this report. Page 81 GAO-12-445 Unemployed Older Workers Related GAO Products Related GAO Products Income Security: Older Adults and the 2007-2009 Recession. GAO-12-76. Washington, D.C.: October 17, 2011. Retirement Income: Ensuring Income throughout Retirement Requires Difficult Choices. GAO-11-400. Washington, D.C.: June 7, 2011. Private Pensions: Some Key Features Lead to an Uneven Distribution of Benefits. GAO-11-333. Washington, D.C.: March 30, 2011. Social Security Reform: Raising the Retirement Ages Would Have Implications for Older Workers and SSA Disability Rolls. GAO-11-125. Washington, D.C.: November 18, 2010. Private Pensions: Alternative Approaches Could Address Retirement Risks Faced by Workers but Pose Trade-Offs. GAO-09-642. Washington, D.C.: July 24, 2009. Employment and Training: Most One-Stop Career Centers Are Taking Multiple Actions to Link Employers and Older Workers. GAO-08-548. Washington, D.C.: April 21, 2008. 401(k) Plans: Policy Changes Could Reduce the Long-term Effects of Leakage on Workers’ Retirement Savings. GAO-09-715. Washington, D.C.: August 28, 2009. Private Pensions: Low Defined Contribution Plan Savings May Pose Challenges to Retirement Security, Especially for Many Low-Income Workers. GAO-08-8. Washington, D.C.: November 29, 2007. Retirement Security: Women Face Challenges in Ensuring Financial Security in Retirement. GAO-08-105. Washington, D.C.: October 11, 2007. Highlights of a GAO Forum: Engaging and Retaining Older Workers. GAO-07-438SP. Washington, D.C.: February 28, 2007. Senior Community Service Employment Program: Labor Has Made Progress Implementing Older Americans Act Amendments of 2000, but Challenges Remain. GAO-06-549T. Washington, D.C.: April 6, 2006. Page 82 GAO-12-445 Unemployed Older Workers Related GAO Products Older Workers: Employment Assistance Focuses on Subsidized Jobs and Job Search, but Revised Performance Measures Could Improve Access to Other Services. GAO-03-350. Washington, D.C.: January 24, 2003. (131032) Page 83 GAO-12-445 Unemployed Older Workers GAO’s Mission The Government Accountability Office, the audit, evaluation, and investigative arm of Congress, exists to support Congress in meeting its constitutional responsibilities and to help improve the performance and accountability of the federal government for the American people. GAO examines the use of public funds; evaluates federal programs and policies; and provides analyses, recommendations, and other assistance to help Congress make informed oversight, policy, and funding decisions. GAO’s commitment to good government is reflected in its core values of accountability, integrity, and reliability. The fastest and easiest way to obtain copies of GAO documents at no Obtaining Copies of cost is through GAO’s website (www.gao.gov). Each weekday afternoon, GAO Reports and GAO posts on its website newly released reports, testimony, and correspondence. To have GAO e-mail you a list of newly posted products, Testimony go to www.gao.gov and select “E-mail Updates.” Order by Phone The price of each GAO publication reflects GAO’s actual cost of production and distribution and depends on the number of pages in the publication and whether the publication is printed in color or black and white. Pricing and ordering information is posted on GAO’s website, http://www.gao.gov/ordering.htm. Place orders by calling (202) 512-6000, toll free (866) 801-7077, or TDD (202) 512-2537. Orders may be paid for using American Express, Discover Card, MasterCard, Visa, check, or money order. Call for additional information. Connect with GAO on Facebook, Flickr, Twitter, and YouTube. Connect with GAO Subscribe to our RSS Feeds or E-mail Updates. Listen to our Podcasts. Visit GAO on the web at www.gao.gov. Contact: To Report Fraud, Waste, and Abuse in Website: www.gao.gov/fraudnet/fraudnet.htm E-mail: firstname.lastname@example.org Federal Programs Automated answering system: (800) 424-5454 or (202) 512-7470 Katherine Siggerud, Managing Director, email@example.com, (202) 512- Congressional 4400, U.S. Government Accountability Office, 441 G Street NW, Room Relations 7125, Washington, DC 20548 Chuck Young, Managing Director, firstname.lastname@example.org, (202) 512-4800 Public Affairs U.S. Government Accountability Office, 441 G Street NW, Room 7149 Washington, DC 20548 Please Print on Recycled Paper.
Unemployed Older Workers: Many Experience Challenges Regaining Employment and Face Reduced Retirement Security
Published by the Government Accountability Office on 2012-04-25.
Below is a raw (and likely hideous) rendition of the original report. (PDF)