A REVISED MINIMUM BENEFIT TO BETTER MEET THE ADEQUACY AND EQUITY STANDARDS IN SOCIAL SECURITY. January 2018

Transcription

1 January 2018 A REVISED MINIMUM BENEFIT TO BETTER MEET THE ADEQUACY AND EQUITY STANDARDS IN SOCIAL SECURITY Kimberly J. Johnson, School of Social Work, Indiana University Elizabeth Johns, Gerontology Institute, University of Massachusetts Boston and University of Maine Center on Aging

2 The authors thank Karen E. Smith, senior fellow at the Urban Institute, for running the Dynamic Simulation of Income Model (DYNASIM) and for providing insightful comments on our proposal. This paper represents the views of the authors and does not necessarily reflect the views or policy of AARP or the opinions or policy of any agency of the federal government nor of any of the educational and research institutions that sponsor their work.

3 CONTENTS Introduction 1 Economic Insecurity among Older Americans 2 Multiple Measures of Poverty 4 New Social Risks and Economic Insecurity in Old Age 5 How Does a Minimum Benefit Work? 7 Minimum Pension Benefits in International Contexts 9 Design for a New Minimum Benefit in Social Security 11 How Could a New Minimum Benefit Be Financed? 12 The DYNASIM Analysis 13 Key Findings from the DYNASIM Simulation 15 Impact on Poverty 15 Income Change by Quintile 16 Income Change by Age 16 Income Change by Sex 17 Income Change by Marital Status 17 Income Change by Race and Ethnicity 18 Social Security Benefits in Proportion to OASI Taxes 18 Other Retirement Financial Assets 19 Net Cash Income 20 Conclusion 21 References 22 A REVISED MINIMUM BENEFIT TO BETTER MEET THE ADEQUACY AND EQUITY STANDARDS IN SOCIAL SECURITY i

4 FIGURE FIGURE 1. Projected Official Poverty Rates among Individuals Ages 62 and Older under Current Law and the Revised Minimum Benefit Option, TABLES TABLE 1. Change in Mean Net per Capita Cash Income of Persons Ages 62 and over by Shared Income Quintiles, (2015 Dollars) 16 TABLE 2. Change in Mean Net per Capita Cash Income of Persons Ages 62 and over by Age, (2015 Dollars) 17 TABLE 3. Change in Mean per Capita Net Cash Income of Persons Ages 62 and over by Sex, (2015 Dollars) 17 TABLE 4. Change in Mean per Capita Net Cash Income of Persons Age 62 and over by Marital Status, (2015 Dollars) 18 TABLE 5. TABLE 6. TABLE 7. Change in Mean Net per Capita Income of Persons Ages 62 and over by Race and Ethnicity, (2015 Dollars) 18 Median Ratio of per Capita Lifetime Social Security Benefits to per Capita Lifetime OASDI Payroll Tax at Age 65 by Birth Year and Shared Lifetime Earnings Quintile 19 Mean per Capita Retirement Account Assets by Lifetime Shared Earnings Quintile, (2015 Dollars) 20 TABLE 8. Average per Capita Net Cash Income among Persons Ages 62 and over by Shared Income Quintile, (2015 Dollars) 20 ii A REVISED MINIMUM BENEFIT TO BETTER MEET THE ADEQUACY AND EQUITY STANDARDS IN SOCIAL SECURITY

5 INTRODUCTION This research analyzes a revised minimum benefit in the Old-Age and Survivors Insurance (OASI) program that would bring the lowest-income beneficiaries above the federal poverty threshold. While not a new idea, the revised minimum benefit could make the greatest difference in reducing economic insecurity among older Americans. The following sections discuss the background and rationale for revising Social Security s existing minimum benefit. Particular attention is paid to the challenges faced by an increasingly diverse American workforce and to retirees risk of living in poverty. The paper details the proposed new benefit structure and how a revised minimum benefit could be financed. While not a new idea, The Urban Institute s the revised minimum Dynamic Simulation of benefit could make Income Model (DYNASIM) the greatest difference is used to simulate the varied effects of a revised in reducing economic minimum benefit as insecurity among older compared with no change Americans. to current Social Security law. We evaluate the impact of the revised minimum benefit and briefly discuss implications of this policy change. A REVISED MINIMUM BENEFIT TO BETTER MEET THE ADEQUACY AND EQUITY STANDARDS IN SOCIAL SECURITY 1

6 ECONOMIC INSECURITY AMONG OLDER AMERICANS Core objectives of the Social Security program are to protect older Americans from economic insecurity (adequacy) while linking benefits to workers contributions to the system (equity). These principles are implicit in the founding Social Security Act of 1935 and have been affirmed in multiple amendments over the years to extend the program s reach and ensure its fiscal well-being. However, maintaining an acceptable balance between the two goals has always been a challenge. Social Security is given In 1939, the year the first much credit for a large Social Security benefit claim was filed, a startling reduction in older adult 78 percent of people ages poverty in the years 65 and over were considered since the first claim was poor (Smolensky, Danziger, filed in & Gottschalk, 1987). Social Security is given much credit for a large reduction in older adult poverty over subsequent decades. Nonetheless, an estimated 4.2 million older Americans were officially poor in 2015, or 8.8 percent of all persons ages 65 and older. Poverty rates are higher among older minority groups: 7.5 percent among Whites but 18.4 percent among both African Americans and American Indians/Alaska Natives, 17.5 percent among Hispanics, and 11.8 percent among Asians (US Census, 2016). Poverty varies significantly by gender as well as race and ethnicity, with women consistently disadvantaged relative to men. Current Population Survey data for 2015 (US Census, 2016) indicate that poverty is higher among older White females (8.9 percent) than White males (6.0 percent), for Black females (19.6 percent) compared with Black males (16.7 percent), for Native American females (24.3 percent) relative to Native American males (11.3 percent), for Hispanic females (20.1 percent) compared with Hispanic males (14.0 percent), and for Asian females (14.3 percent) compared with Asian males (8.7 percent). Additionally, the incidence of poverty rises with age, as older adults spend down their retirement resources and encounter higher costs of living, in part linked to greater utilization of medical services and long-term care (De Nardi, French, Jones, & McCauley, 2016; Fahle, McGarry, & Skinner, 2016). In 2014, 12.7 percent of persons ages 85 and older were poor compared with 8.8 percent of those ages (AARP, 2017). 2 A REVISED MINIMUM BENEFIT TO BETTER MEET THE ADEQUACY AND EQUITY STANDARDS IN SOCIAL SECURITY

7 Increased age also raises the incidence of widowhood, and single older adults are more likely to be poor than partners in couples 4.4 percent of married persons versus 12.1 percent of widow(er)s with higher rates for the divorced (14.7 percent) and never-married (21.4 percent) (US Census, 2016). Marriage for women tends to bring economic protections in retirement (e.g., related to the accumulation of greater wealth and shared living expenses); however, such advantages are not equally experienced across racial/ethnic groups, leaving women of color more vulnerable irrespective of marital status (Lin, Brown, & Hammersmith, 2017; Traub, Sullivan, Meschede, & Shapiro, 2017). Financial need persists in spite of Social Security. In 2015, 77.8 percent of persons ages 65 and older who were living alone and in poverty were recipients of OASI benefits, as were 61.6 percent of poor multiperson families in which the household head was age 65 or older (Social Security Administration, 2016). A REVISED MINIMUM BENEFIT TO BETTER MEET THE ADEQUACY AND EQUITY STANDARDS IN SOCIAL SECURITY 3

8 MULTIPLE MEASURES OF POVERTY The foregoing figures reference official poverty thresholds that are widely perceived to be an outdated measure of economic insecurity (e.g., Blank & Greenberg, 2008). Since 2010, the Census Bureau has also reported a Supplemental The United States has Poverty Measure designed one of the highest to be a more accurate gauge of economic well-being. rates of poverty among It incorporates the value persons ages 66 and of in-kind benefits (e.g., older living in upperincome countries: Supplemental Nutrition Assistance Program or SNAP benefits) as well 21.0 percent in as major household expenditures (e.g., for medical care). By this measure, older adult poverty is even higher: 13.7 percent in 2015 for all persons ages 65 and older as compared with 8.8 percent for the official poverty measure (Renwick & Fox, 2016). The Supplemental Poverty Measure is not used to set policy but does provide strong evidence of need as do other indices, such as the Elder Economic Security Standard Index (Mutchler, Li, & Xu, 2016), or Smeeding s (2016) proposed poverty threshold of half the median national household income, a measure widely used in international contexts. By this measure, the United States has one of the highest rates of poverty among persons ages 66 and older living in upper-income countries: 21.0 percent in That rate compares unfavorably with such countries as Canada (9.0 percent), France (3.6 percent), Germany (9.5 percent), the Netherlands (3.1 percent), and the United Kingdom (13.1 percent) (OECD, 2017). 4 A REVISED MINIMUM BENEFIT TO BETTER MEET THE ADEQUACY AND EQUITY STANDARDS IN SOCIAL SECURITY

9 NEW SOCIAL RISKS AND ECONOMIC INSECURITY IN OLD AGE Devised in the 1930s as the keystone in a social policy structure meant to protect workers and their families against financial risk, Social Security never covered all social groups adequately. Eighty years later it is less well equipped for the kinds of new social risks (Hacker, 2004) that have emerged in recent decades. An example is rising income inequality compounded by the decline of the classic lifelong career, replaced for many workers with shortened job tenures, contingent employment, and wage stagnation (Johnson, 2016). People with less income are more likely to become a caregiver for an aging family member, a situation disproportionally impacting women that often has negative effects on subsequent earnings (Lee, Tang, Kim, & Albert, 2015). Changing social and employment trends, coupled with the shortcomings of the privatesector retirement financing system, hinder the accumulation of retirement savings, particularly for families living in the lower half of the income distribution (Rhee & Boivie, 2015; Sullivan et al., 2015). Thus many workers are challenged to anticipate and meet their postretirement needs for social and financial support (Reinhard, Feinberg, Choula, & Houser, 2015). America s minority ethnic and racial populations, which have long faced systematic economic disadvantage resulting from discrimination, are disproportionately subject to these risks. Less housing equity, lower wages, and more limited access to pension plans than non-hispanic White adults translate to less wealth and retirement income security (Choi, Minority households Tang, & Copeland, 2017; Veghte, Schreur, & Waid, experienced widening 2016). Minority households inequality in income experienced widening and wealth in the years inequality in income and wealth in the years preceding preceding the Great the Great Recession of Recession of , , and data show the and data show the wealth gap continuing to widen in the postrecession wealth gap continuing period (Kochhar & to widen in the Fry, 2014; Vornovitsky, postrecession period. Gottschalk, & Smith, n.d.). As we know, about half of American workers do not participate in a pension plan at work, and those without coverage tend to be lower earners, part-time workers, and persons working for small employers (Rhee & Boivie, 2015; Zukin & Van Horn, 2015). A REVISED MINIMUM BENEFIT TO BETTER MEET THE ADEQUACY AND EQUITY STANDARDS IN SOCIAL SECURITY 5

10 Members of racial and ethnic minorities are more likely to work in employment sectors where retirement plans and other job benefits are less commonly offered, which contributes to slower wealth accumulation (Shapiro, Meschede, & Osoro, 2013) and greater dependency on Social Security for retirement income. Moreover, lower-wage work, which often comes with more periods of unemployment and greater physical demands, can place socially and economically disadvantaged workers at greater risk for disability, serious health problems, and death at earlier ages (Zajacova, Montez, & Herd, 2014), and these differences in healthy life expectancy have serious implications for the financial stability of these workers families and the Social Security system (Rockeymoore & Lui, 2011). One report by the General Accountability Office (GAO, 2016) observed that among males approaching retirement, those with lower incomes had between 3.7 and 12.7 fewer years of life expectancy than did higher-income males. The study estimated that lower-income males lost percent in the value of Social Security benefits relative to someone with average life expectancy, while higher-income males saw benefit increases of percent. A separate analysis (Committee, 2015) found a similar pattern in life expectancy for women, while Olshansky and colleagues (2012) found the effect to be exacerbated by race. Income-related disparities in life expectancy have the effect of eroding the progressive design of Social Security benefits and undermining the equity principle. What links these different forms of risk is the lag in social policies to lessen their impact on workers and their families (Bonoli, 2006). Given the lack of private-sector initiatives adequate to meet these challenges, retirees of the future are likely to be even more dependent on Social Security than are current cohorts (Munnell, Hou, Webb, & Li, 2016), making program adjustments to meet increased need imperative. 6 A REVISED MINIMUM BENEFIT TO BETTER MEET THE ADEQUACY AND EQUITY STANDARDS IN SOCIAL SECURITY

11 HOW DOES A MINIMUM BENEFIT WORK? There is currently a minimum OASI benefit, and there has been one since the earliest days of the program. Originally, it was intended to provide a threshold benefit for retired workers without reference to their years of employment, but by the 1970s policy makers had become concerned that most benefits were not reaching the intended population but instead were going to two groups seen as less deserving: (a) retirees with short records of covered employment but with full pensions from state and local jurisdictions not covered by Social Security (so-called double-dippers), and (b) homemakers supported by their spouses incomes. The minimum benefit was also criticized for giving the program a welfare aspect that was at odds with the program s mission of providing for retired workers (Staats, 1979, p. i). In 1977 Congress froze the value of the original minimum benefit and eliminated it 4 years later for new beneficiaries. It was supplanted by a new special minimum benefit intended for retirees with eleven or more years of qualified earnings whose Social Security benefit fell below the poverty level. These retirees were eligible for a benefit that was slightly higher than the regular benefit they would be entitled to based solely on their limited earnings histories, and benefit levels rose with each year of covered employment up to 30 years. This formula was intended to correct the perception that many, if not most, people receiving the [original] minimum benefit... have had little connection with employment covered under social security (US Senate, Committee on Finance, 1972, p. 154). Poor older adults with shorter work records could apply for assistance through By the 1970s policy the Supplemental Security makers had become Income program, enacted in concerned that most 1972 (Martin & Weaver, 2005). benefits were not The special minimum benefit reaching the intended was inflation indexed, and over the years its value population but instead declined relative to regular were going to two Social Security benefits, groups seen as less which are indexed to wages deserving. (and which have tended to increase more rapidly than prices), thus gradually reducing the pool of beneficiaries. Since 1998 no new beneficiary has been assigned the special minimum because that benefit would be lower than the regular benefit he or she would be entitled to receive (Social Security Administration, 2014). A REVISED MINIMUM BENEFIT TO BETTER MEET THE ADEQUACY AND EQUITY STANDARDS IN SOCIAL SECURITY 7

12 While these policy initiatives came up short, they suggest elements a successful initiative will require. Apart from financial need there must be a sufficient link to a worker s employment record and contributions to Social Security. However, the link should not be so stringent as to have little impact in reducing poverty among older persons (Herd, 2009). And in light of the 2016 Trustees report projecting wages to continue to rise over at least the next decade at a faster rate than prices (Trustees, 2016), benefits ought to be wage adjusted. 8 A REVISED MINIMUM BENEFIT TO BETTER MEET THE ADEQUACY AND EQUITY STANDARDS IN SOCIAL SECURITY

13 MINIMUM PENSION BENEFITS IN INTERNATIONAL CONTEXTS Most European Union member states have some form of basic or minimum pension for their retirement-age populations, and similar measures have been adopted in public pension systems in other parts of the world Australia, Canada, Chile, Korea, Mexico, New Zealand, Turkey, and others. These may or may not be linked to retiree work records, but they are distinguished from safety-net or last-resort benefits, which are provided when other measures are unavailable or insufficient to prevent severe economic need. Typically, the basic and minimum pensions provide higher benefits than do the safety-net measures. One estimate valued them collectively at about 29 percent of average national earnings in the Organisation for Economic Cooperation and Development (OECD) member countries (Pearson & Whitehouse, 2009). Most exist alongside second-tier occupational pensions and third-tier private savings. National minimum and basic pension schemes vary considerably in their design and application. For some, benefit eligibility is linked to length of residency in the providing country (e.g., Australia, Canada, Finland); in others it is based on workers years of tax contributions to the public system but not their level of earnings (e.g., Ireland, Luxembourg, Portugal). Typically, these are a flat-rate benefit (OECD, 2015). In some instances, a minimum benefit guarantees a retirement income at a certain level by supplementing other income sources if they do not reach a set threshold. There is considerable variability in these schemes involving eligibility, means testing, generosity and financing of benefits, interactions with In recent decades, second- and third-tier income, aging populations and role of the private sector, and other policy considerations. declining fertility have challenged retirement In recent decades, aging financing in many populations and declining fertility have challenged countries and prompted retirement financing in many widespread reforms countries and prompted to public pension widespread reforms to public pension systems, systems, including more including more privatization privatization of risk. of risk. In this context, olderage poverty has become an increasing concern, and creating or strengthening minimum pension guarantees is often proposed in response, including by A REVISED MINIMUM BENEFIT TO BETTER MEET THE ADEQUACY AND EQUITY STANDARDS IN SOCIAL SECURITY 9

14 such organizations as the World Bank and International Labour Organization (Orenstein, 2013). Many design choices are available to policy makers, and to date research is limited on what policy configurations can best meet complex policy goals. Robalino and Holzmann (2009), in a World Bank sponsored international survey of social pension systems, identify features that contribute to success in providing a minimum level of retirement income while attending to the values of adequacy and equity. The principles are familiar: encourage labor force participation and private retirement saving; discourage early retirement; do not create incentives through overly generous benefits for early retirement or inadequate saving. Many observers urge wage indexing of benefits rather than price indexing, as a way to preserve the value of benefits over time. In general, however, it appears that as long as a core commitment exists to alleviate poverty, specific policy details can be tailored to individual countries social and political cultures, economic capacities, and existing policy frameworks. 10 A REVISED MINIMUM BENEFIT TO BETTER MEET THE ADEQUACY AND EQUITY STANDARDS IN SOCIAL SECURITY

15 DESIGN FOR A NEW MINIMUM BENEFIT IN SOCIAL SECURITY The proposed policy links a new minimum benefit to a percentage of the federal poverty threshold and a retiring worker s years of covered labor force participation. Recognizing the limitations of the official poverty measure but also its common utilization in policy making, we propose cut points above the poverty threshold. Eligible retirees with 80 or more quarters (20 years) of covered employment and household income totaling less than 125 percent of the poverty threshold receive a supplement raising their income to that level. For workers with quarters (10 19 years) of covered employment, the qualifying household income level and supplement is set at 112 percent of the poverty threshold. At age 80, all low-income beneficiaries are moved to the 125 percent threshold in recognition of the financial stresses that accompany advanced age. These thresholds are proposed for simplicity, but (assuming no significant additional administrative burden) benefits could be stepped in such a way as to reward each added year of covered employment perhaps in increments of 1 2 percent per year, as other policy analysts have proposed (e.g., Favreault, 2009; Favreault, Sammartino, & Steuerle, 2002). The benefit level might start at 110 percent of poverty for someone with 10 years of covered employment and rise by 1.5 percent for each additional year worked, up to a 125 percent benefit for 20 or more years of work. For purposes of this analysis, we assume no changes to other elements of Social Security eligibility: no change in the early and full retirement ages (including no increase above age 67), no change in the spousal benefit, no change in years of covered employment required for benefit eligibility, and no change in the formula by which the regular OASI benefit is calculated. One unintended effect of the policy for low-income retirees could be reductions in eligibility for valuable social welfare benefits such as SNAP and Medicaid unless there were accompanying eligibility adjustments for those programs. Placing additional financial strain on poor retired workers from increases in housing or out-of-pocket health expenses would negate any poverty prevention from the revised minimum benefit. A REVISED MINIMUM BENEFIT TO BETTER MEET THE ADEQUACY AND EQUITY STANDARDS IN SOCIAL SECURITY 11

16 HOW COULD A NEW MINIMUM BENEFIT BE FINANCED? We offer two measures to help finance the proposed change in benefits. One is to assess some employers an increased share of the Old-Age, Survivors, and Disability Insurance (OASDI) tax, raising their obligation from 6.2 percent to 8 percent of employee earnings. The employer would contribute the extra 1.8 percentage points for any employee for whom the employer pays the OASDI tax but does not contribute at least 3 percent of the employee s earnings to a qualified pension plan as defined in the Internal Revenue Code. An exception would be made for an employer whose 3 percent contribution would result in a pension contribution higher than the annual legal limit. Employers meeting their pension obligation would continue to contribute to OASDI at the 6.2 percent rate. Self-employed workers not participating in a qualified pension plan would also pay the extra amount. A second component of the proposal is to modify the calculation of the taxable earnings base set at $127,200 in Under current policy, the taxable base is adjusted each year proportionally to the change in average national earnings. For example, average earnings rose by 1.28 percent from 2012 to 2013; the so-called tax max rose proportionally, from $117,000 in 2014 to $118,500 in (The lag relates to the time required to gather and process the data used in the calculation.) In recent years, the taxable earnings base has been equal to approximately 2.5 times average annual earnings. We propose formally linking the amount of taxable earnings to a set multiple of average national earnings specifically, raising the annual taxable base to 3.0 times average national earnings. Under this suggestion, for example, the taxable earnings base in 2015 would have been $144,295 instead of $118,500. In addition, we propose that all earnings of 10 or more times the average national earnings be subject to the OASDI tax, in response to long-term disproportionate growth at the highest levels of the earnings distribution (Committee, 2015; Mishel & Kroeger, 2016). (Indeed, the rising share of earnings not subject to the OASDI tax has contributed to the long-term projected fiscal imbalance in the system [Diamond, 2005].) In 2015, that upper earnings threshold would have been $480,980. In other words, under this proposal earnings between 3 and 10 times the national average would not be subject to the tax. 12 A REVISED MINIMUM BENEFIT TO BETTER MEET THE ADEQUACY AND EQUITY STANDARDS IN SOCIAL SECURITY

17 THE DYNASIM ANALYSIS The Urban Institute s Dynamic Simulation of Income Model was used to simulate the implementation of the revised minimum benefit as part of the OASI program. The analysis assumed our basic, twopart design: a minimum benefit worth 125 percent of the poverty threshold for 20 or more years of covered employment and 112 percent of poverty for covered years. All simulations were performed by Karen Smith of the Urban Institute. In brief, DYNASIM uses characteristics of the US population from the Survey of Income and Program Participation to estimate annual changes in marriage, fertility, mortality, and divorce. Economic indicators reflect employment status, disability, retirement, hours worked, earnings, pension coverage and contributions. DYNASIM simulates the participation, benefits, and cost of the Social Security programs, and the model allows us to estimate the distributional effects of the revised minimum benefit on various socioeconomic groups through A detailed description of DYNASIM is provided elsewhere (see Favreault, Smith, & Johnson, 2015). Several assumptions are made in the simulation results discussed below. In order to offset the cost of the minimum benefit, we stipulated that employers not contributing at least 3 percent of employee earnings to a qualified pension account would be assessed a higher OASDI tax of 8 percent. In the simulation, some employers respond by increasing their defined-contribution pension payments while others pay the tax. In both cases, the models assume that employers reduce employee wages to keep total compensation unchanged. Because higher DYNASIM uses earners pay more in payroll characteristics of the taxes under our proposal, US population from the they also receive higher amounts in the average Survey of Income and indexed monthly earnings Program Participation to formula for calculating their estimate annual changes Social Security benefit. in marriage, fertility, Eligibility for the revised mortality, and divorce. minimum benefit is based on total income, and total income is calculated as the sum of a tax unit s earnings, Social Security benefits, defined-benefit pension income, taxable interest, dividends, rental income, and withdrawals from retirement accounts. The simulations preserve the 2015 ratio between the wage-indexed federal poverty A REVISED MINIMUM BENEFIT TO BETTER MEET THE ADEQUACY AND EQUITY STANDARDS IN SOCIAL SECURITY 13

18 threshold and the price-based annual wage index in order to maintain the value of the minimum benefit over time. Results of the simulation are reported in 2015 dollars for successive decades to The simulation assumes that promised OASI benefits can be paid in full through 2065 without any change to current eligibility criteria or benefit formulas, other than those we propose as part of the revised minimum benefit. It also assumes that the Social Security Trust Fund has sufficient assets to pay promised benefits throughout the simulation period. 14 A REVISED MINIMUM BENEFIT TO BETTER MEET THE ADEQUACY AND EQUITY STANDARDS IN SOCIAL SECURITY

19 KEY FINDINGS FROM THE DYNASIM SIMULATION In the following section, we briefly summarize selected key results of the DYNASIM projections, focusing on economic impacts for demographic groups identified by income, age, gender, marital status, and race/ethnicity. While not a comprehensive summary, these results suggest that the revised minimum benefit can help strengthen the Social Security program s commitment to adequacy. IMPACT ON POVERTY One result we can readily see is the impact of the initiative on the official poverty rate for the 62-and-older population, although our proposed benefit would affect only those of full retirement age (ages 66 or 67), providing a benefit higher than the official poverty threshold, an outcome that was not analyzed directly. As Figure 1 shows, the share of the population living in poverty falls from 9.1 percent in 2015 to 3.9 percent in This drop is somewhat larger than the projected decline assuming no change to OASI: from 9.1 percent in 2015 to 5.7 percent in FIGURE 1. Projected Official Poverty Rates among Individuals Ages 62 and Older under Current Law and the Revised Minimum Benefit Option, Source: Urban Institute DYNASIM3 A REVISED MINIMUM BENEFIT TO BETTER MEET THE ADEQUACY AND EQUITY STANDARDS IN SOCIAL SECURITY 15

20 INCOME CHANGE BY QUINTILE The impact of the revised minimum benefit is better observed by looking at the change in average net income. Table 1 reports the change in average net individual cash income by quintile for the years 2025 through Here the bottom quintile is the general focus and greatest beneficiary of our initiative. With the minimum benefit, this income group experiences an average income increase of 16.4 percent in 2025 that rises each decade to 33.9 percent by The second quintile also sees income increases, but of a lower magnitude, from 2.5 percent in 2025 to 11.3 percent by The remaining three quintiles also see small increases, with the smallest accruing to the top quintile, an increase of less than 1 percent across the 50-year span of the simulation. TABLE 1. Change in Mean Net per Capita Cash Income of Persons Ages 62 and over by Shared Income Quintiles, (2015 Dollars) Quintile Bottom 0 (0) 1,604 (16.4) 2,340 (23.2) 2,857 (27.8) 3,472 (30.8) 4,239 (33.9) 2nd 0 (0) 533 (2.5) 1,100 (5.2) 1,968 (9.1) 2,554 (11.1) 2,923 (11.3) 3rd 0 (0) 172 (0.6) 250 (0.8) 526 (1.5) 823 (2.3) 1,025 (2.6) 4th 0 (0) 27 (0.0) 145 (0.3) 307 (0.6) 506 (0.9) 691 (1.2) Top 0 (0) 93 (0.0) 93 (0.1) 110 (0.3) 149 (0.5) 657 (0.8) Source: Urban Institute DYNASIM3 INCOME CHANGE BY AGE We know that economic insecurity increases with advancing age. How do the oldest retirees fare under a revised minimum benefit? Table 2 shows mean per capita cash incomes rising for each age group from age 62 to ages 85 and over. Younger OASI beneficiaries see the lowest level of increase, which is not surprising because our benefit is not available until full retirement age. Per capita cash incomes rise by age group, so that across the 50-year span of the data, 62- to 69-yearolds see an income rise of 1.2 percent, 70- to 74-year-olds see an increase of 3.9 percent, 75- to 79-year-olds see a rise of 5.6 percent, 80- to 84-year-olds incomes increase by 7.7 percent, and people ages 85 and older see an increase of 10.0 percent. Incomes also rise with each decade across the 50-year time span for all but the youngest group. 16 A REVISED MINIMUM BENEFIT TO BETTER MEET THE ADEQUACY AND EQUITY STANDARDS IN SOCIAL SECURITY

21 TABLE 2. Change in Mean Net per Capita Cash Income of Persons Ages 62 and over by Age, (2015 Dollars) Age (0) 202 (0.6) 228 (0.6) 434 (1.1) 454 (1.0) 613 (1.2) (0) 568 (1.6) 986 (2.8) 1,355 (3.4) 1,660 (3.9) 1,836 (3.9) (0) 548 (1.6) 1,048 (3.3) 1,323 (3.7) 1,956 (5.2) 2,301 (5.6) (0) 694 (2.2) 972 (2.9) 1,632 (5.2) 2,271 (6.4) 2,856 (7.7) (0) 894 (3.3) 1,090 (3.5) 1,756 (5.7) 2,459 (7.6) 3,377 (10.0) Source: Urban Institute DYNASIM3 INCOME CHANGE BY SEX Table 3 shows the change in income for women and men by decade to Women show slightly higher percentage increases in income under the revised minimum benefit, although both genders evidence a steady rise over the 50-year period. The rise is smaller for males, from 1.1 percent in 2025 to 3.9 percent by 2065, whereas for females the increase begins in 2025 at 1.6 percent and rises to 4.9 percent by TABLE 3. Change in Mean per Capita Net Cash Income of Persons Ages 62 and over by Sex, (2015 Dollars) Gender Female 0 (0) 515 (1.6) 821 (2.4) 1,238 (3.5) 1,627 (4.2) 2,062 (4.9) Male 0 (0) 372 (1.1) 665 (1.9) 1,057 (2.8) 1,360 (3.3) 1,732 (3.9) Source: Urban Institute DYNASIM3 INCOME CHANGE BY MARITAL STATUS Marriage acts as a buffer against financial hardship, and in Table 4 we see nevermarried beneficiaries receiving the largest percentage increase in per capita net income of the four groups considered by marital status. With a 3.9 percent increase in 2025, they see a rise to 10.1 percent by The next-highest increase is experienced by divorced beneficiaries, whose incomes rise by an average of 3.1 percent by 2025 and continue rising to 6.9 percent in Widowed persons fare only slightly less well, with increases from 1.9 percent in 2025 to 6.3 percent by Married individuals see barely any increase: 0.4 percent in 2025 to 1.7 percent by A REVISED MINIMUM BENEFIT TO BETTER MEET THE ADEQUACY AND EQUITY STANDARDS IN SOCIAL SECURITY 17

22 TABLE 4. Change in Mean per Capita Net Cash Income of Persons Age 62 and over by Marital Status, (2015 Dollars) Marital Status Married 0 (0) 127 (0.4) 228 (0.7) 399 (1.1) 501 (1.2) 754 (1.7) Widowed 0 (0) 659 (1.9) 1,036 (2.8) 1,586 (4.2) 2,205 (5.6) 2,736 (6.3) Divorced 0 (0) 997 (3.1) 1,532 (4.4) 2,034 (5.5) 2,476 (6.0) 2,979 (6.9) Never married 0 (0) 1,205 (3.9) 1,831 (5.5) 2,548 (7.7) 3,218 (8.8) 3,835 (10.1) Source: Urban Institute DYNASIM3 INCOME CHANGE BY RACE AND ETHNICITY Under our proposal, all four racial and ethnic groups analyzed are projected to experience steady net gains in mean per capita income. As Table 5 shows, White non-hispanics see the smallest gain, with a 3.4 percent increase over the 50-year time span of Black non-hispanic older adults see the largest increase, 8.3 percent, followed by Hispanics (5.6 percent) and the group other with 5.1 percent. TABLE 5. Change in Mean Net per Capita Income of Persons Ages 62 and over by Race and Ethnicity, (2015 Dollars) Race/Ethnicity White non-hispanic 0 (0) 310 (0.8) 557 (1.4) 932 (2.3) 1,261 (2.8) 1,644 (3.4) Black non-hispanic 0 (0) 888 (3.5) 1,388 (5.1) 1,955 (6.7) 2,544 (7.9) 2,917 (8.3) Hispanic 0 (0) 816 (4.1) 1,042 (4.5) 1,334 (5.2) 1,477 (4.9) 1,893 (5.6) Other 0 (0) 627 (2.3) 979 (3.1) 1,277 (3.8) 1,605 (4.4) 2,042 (5.1) Source: Urban Institute DYNASIM3 SOCIAL SECURITY BENEFITS IN PROPORTION TO OASI TAXES Besides the impact of a revised minimum benefit on various demographic groups, an additional factor to consider is the return in benefits to retirees relative to their contributions to Social Security in the form of payroll taxes. Table 6 predicts the net returns of the proposed minimum benefit for retirees of different birth cohorts ( ) and different lifetime earnings profiles. The table shows the bottom two earnings quintiles faring best, with benefit levels expected to exceed taxes paid into the Social Security system for every birth cohort. In all cohorts, however, retirees in the third, fourth, and 18 A REVISED MINIMUM BENEFIT TO BETTER MEET THE ADEQUACY AND EQUITY STANDARDS IN SOCIAL SECURITY

23 fifth earnings quintiles are estimated to see negative returns that is, to receive less in Social Security benefits than they paid in taxes into the system. The disparity is particularly noticeable for the top earnings quintile, which for each birth cohort is projected to receive in benefits only about half what it contributed. However, other projections, not shown here, indicate that across the 50 years of projected Social Security income, persons in the highest income quintile will receive annual benefits averaging in dollars about 125 percent more than those in the lowest quintile and about 30 percent more than those in the second quintile. TABLE 6. Median Ratio of per Capita Lifetime Social Security Benefits to per Capita Lifetime OASDI Payroll Tax at Age 65 by Birth Year and Shared Lifetime Earnings Quintile Birth Year All Own Lifetime Earnings Quintile Bottom 2nd 3rd 4th Top Source: Urban Institute DYNASIM3 OTHER RETIREMENT FINANCIAL ASSETS In addition to Social Security, retirees depend on personal resources for their income security. Table 7 indicates substantial differences by lifetime earnings quintile in estimated retirement assets that include individual retirement accounts, Keoghs, and employer defined-contribution plans. Here again, persons in the lowest quintile are at a substantial disadvantage: only percent will have any of these assets. Over the 50-year span of the simulation, that minority is projected to gain about $2,500 in asset value to $13,860. This compares with over 90 percent of individuals with retirement assets in the top earnings quintile, who gain just over $340,000, for a 2065 total of $545,493. Put another way, over the 50-year simulation, assets for savers in the top quintile are predicted to average 33 times those of the minority in the bottom quintile who possess any retirement assets. These figures again predict heavy dependency on income from Social Security for tomorrow s lower-income workers. A REVISED MINIMUM BENEFIT TO BETTER MEET THE ADEQUACY AND EQUITY STANDARDS IN SOCIAL SECURITY 19

24 TABLE 7. Mean per Capita Retirement Account Assets by Lifetime Shared Earnings Quintile, (2015 Dollars) Quintile Bottom 11,328 10,826 9,996 10,470 11,438 13,860 2nd 33,642 38,447 38,056 38,603 45,142 50,001 3rd 57,026 75,038 83,192 91, , ,657 4th 98, , , , , ,990 Top 205, , , , , ,493 Source: Urban Institute DYNASIM3 NET CASH INCOME Table 8 displays the differences in individual net cash income by shared income quintile across the 50 years of the simulation. In 2015, the top quintile has income that is 8.5 times that of the lowest quintile, an advantage it maintains throughout the period. This estimate also shows that while lowerincome groups are favored relative to the higher-income quintiles for indicators such as the ratio of return on OASI taxes paid or in income gains attributable to the revised minimum benefit, the higher-income quintiles continue to experience a strong advantage in overall economic well-being. TABLE 8. Average per Capita Net Cash Income among Persons Ages 62 and over by Shared Income Quintile, (2015 Dollars) Quintile Bottom 7,635 9,316 9,671 9,563 10,153 11,173 2nd 17,521 19,109 19,335 20,263 21,662 23,731 3rd 26,336 28,676 29,322 30,089 32,271 35,394 4th 38,249 41,648 43,717 45,453 49,116 53,607 Top 65,365 70,106 76,085 83,426 94, ,878 Source: Urban Institute DYNASIM3 20 A REVISED MINIMUM BENEFIT TO BETTER MEET THE ADEQUACY AND EQUITY STANDARDS IN SOCIAL SECURITY

25 CONCLUSION The DYNASIM data provide evidence that a revised minimum benefit in Social Security may substantially enhance the financial security of certain vulnerable groups of retirees. The data show a reduction in poverty among the older population, and lower-income groups see substantial extra income. As years go on, older retirees see higher benefits, women do slightly better than men, and the unmarried are helped more than marriage partners. Racial and ethnic minorities fare better than non-hispanic Whites. Lower-income groups also receive more value in Social Security benefits than they contribute in payroll taxes to the system. While vulnerable groups may be better off relative to their status at the 2015 baseline, the revised minimum benefit does little to alter the substantial absolute advantage in income and assets accruing to upper-income retirees. Indeed, DYNASIM assumes that disparities in assets and income now apparent among socioeconomic groups in the United States will widen over the coming decades. An analysis using simulation data has obvious limitations. While it is useful for predicting the impact of a single intervention in a complex, dynamic policy world that is temporarily held constant, we know that real-world events will look far different, especially over a 50-year time horizon. Other changes will need to be made to Social Security seeking to achieve other objectives, including the long-term solvency of the Trust Fund. Increases in the taxable earnings threshold that we draw on to help finance the minimum benefit may be needed for different priorities. The measure could also face difficult challenges The simulated data clearly politically, depending on show that a revised the overall climate in which minimum benefit could Social Security reforms are being considered. be an effective measure for alleviating economic Nevertheless, the simulated data clearly show that a insecurity among a wide revised minimum benefit range of population could be an effective groups. measure for alleviating economic insecurity among a wide range of population groups. For that reason it is well worth continuing to analyze possible designs and funding mechanisms and promoting awareness among policy makers of the minimum benefit s possibilities. A REVISED MINIMUM BENEFIT TO BETTER MEET THE ADEQUACY AND EQUITY STANDARDS IN SOCIAL SECURITY 21

26 REFERENCES AARP. (2017). Poverty, supplemental and official poverty measures by race/ethnicity, age and poverty measure (custom table). AARP Public Policy Institute DataExplorer. Accessed February 14, 2017, at policydata.aarp.org. Blank, R. M., & Greenberg, M. H. (2008). Improving the measurement of poverty. Policy Brief Washington, D.C.: Brookings Institution. Bonoli, G. (2006). New social risks and the politics of post-industrial social policies. In K. Armingeon & G. Bonoli (eds.), The politics of post-industrial welfare states: Adapting postwar social policies to new social risks, New York: Routledge. Choi, E., Tang, F., & Copeland, V. C. (2017). Racial/ethnic inequality among older workers: Focusing on whites, blacks, and Latinos within the cumulative advantage/ disadvantage framework. Journal of Social Service Research, 43(1), Committee on the Long-Run Macroeconomic Effects of the Aging U.S. Population Phase II; Committee on Population, Division of Behavioral and Social Sciences and Education; & Board on Mathematical Sciences and Their Applications, Division on Engineering and Physical Sciences (Committee). (2015). The growing gap in life expectancy by income: Implications for federal programs and policy responses. Washington, D.C.: National Academies Press. De Nardi, M., French, E., Jones, J. B., & McCauley, J. (2016). Medical spending of the U.S. elderly. Fiscal Studies, 37(3 4), Diamond, P. A. (2005). Saving Social Security: The Diamond-Orszag plan. The Economists Voice, 2(1), 1 8. Fahle, E., McGarry, K., & Skinner, J. (2016). Out-of-pocket medical expenditures in the United States: Evidence from the Health and Retirement Study. Fiscal Studies, 37(3 4), Favreault, M. M. (2009). A new minimum benefit for low lifetime earners. Washington, D.C.: Urban Institute. Favreault, M. M., Sammartino, F. J., & Steuerle, C. E. (2002). Social Security benefits for spouses and survivors: Options for change. In Favreault, Sammartino, & Steuerle (eds.), Social Security and the family: Addressing unmet needs in an underfunded system, Washington, D.C.: Urban Institute Press. Favreault, M. M., Smith, K. E., & Johnson, R. W. (2015). The dynamic simulation of income model (DYNASIM): An overview. Washington, D.C.: Urban Institute. Accessed April 16, 2017, at dynamic-simulation-income-model-dynasimoverview. 22 A REVISED MINIMUM BENEFIT TO BETTER MEET THE ADEQUACY AND EQUITY STANDARDS IN SOCIAL SECURITY

27 General Accountability Office (GAO). (2016, Apr. 4). Shorter life expectancy reduces projected lifetime benefits for lower earners. Report Washington, D.C.: Author. Accessed February 18, 2017, at Hacker, J. S. (2004). Privatizing risk without privatizing the welfare state: The hidden politics of social policy retrenchment in the United States. American Political Science Review, 98(2), Herd, P. (2009). The problem of poverty among older people in America: Options for reform. Journal of Poverty and Social Justice, 17(2), Johnson, R. W. (2016). Cumulative advantage and retirement security: What does the future hold? Public Policy and Aging Report, 26(2), Kochhar, R., & Fry, R. (2014, Dec. 12). Wealth inequality has widened along racial, ethnic lines since end of Great Recession. Washington, D.C.: Pew Research Center. Accessed February 17, 2017, at /12/12/racial-wealth-gaps-greatrecession/. Lee, Y., Tang, F., Kim, K. H., & Albert, S. M. (2015). The vicious cycle of parental caregiving and financial well-being: A longitudinal study of women. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 70(3), Lin, I. F., Brown, S. L., & Hammersmith, A. M. (2017). Marital biography, Social Security receipt, and poverty. Research on Aging, 39(1), Martin, P. P., & Weaver, D. A. (2005). Social Security: A program and policy history. Social Security Bulletin, 66(1), Mishel, L., & Kroeger, T. (2016, Oct. 27). Strong across-the-board wage growth in 2015 for both bottom 90 percent and top 1.0 percent. Working Economics Blog. Washington, D.C.: Economic Policy Institute. Accessed April 10, 2017, at strong-across-the-board-wage-growth-in for-both-bottom-90-percent-and-top-1-0-percent/. Munnell, A. H., Hou, W., Webb, A., & Li, Y. (2016). Pension participation, wealth, and income: Working Paper Chestnut Hill, Mass.: Boston College, Center for Retirement Research. Mutchler, J. E., Li, Y., & Xu, P. (2016). Living below the line: Economic insecurity and older Americans insecurity in the states Boston: Gerontology Institute, University of Massachusetts Boston. Olshansky, S. J., Antonucci, T., Berkman, L., Binstock, R. H., Boersch-Supan, A., Cacioppo, J., Carnes, B. A., et al. (2012). Differences in life expectancy due to race and educational differences are widening and many may not catch up. Health Affairs, 31(8), Orenstein, M. A. (2013). Pension privatization: Evolution of a paradigm. Governance, 26(2), Organisation for Economic Co-operation and Development (OECD). (2017). Poverty rate: Total, 0 17 year-olds, 66 year-olds or more, ratio, doi: /0fe1315d-en. Accessed on December 1, 2017, at data.oecd.org/inequality/poverty-rate.htm. A REVISED MINIMUM BENEFIT TO BETTER MEET THE ADEQUACY AND EQUITY STANDARDS IN SOCIAL SECURITY 23

28 Organisation for Economic Co-operation and Development (OECD). (2015). Pensions at a glance: OECD and G20 indicators. Paris: OECD Publishing. Pearson, M., & Whitehouse, E. (2009). Social pensions in high-income countries. In R. Holzmann, D. A. Robalino, & N. Takayama (eds.), Closing the coverage gap: The role of social pensions and other retirement income transfers, Washington, D.C.: International Bank for Reconstruction and Development, World Bank. Reinhard, S. C., Feinberg, L. F., Choula, R., & Houser, A. (2015, July). Valuing the invaluable: 2015 update. Washington, D.C.: AARP Public Policy Institute. Renwick, T., & Fox, L. (2016). The Supplemental Poverty Measure: Current Population Report P Washington, D.C.: Census Bureau. Rhee, N., & Boivie, I. (2015, March). The continuing retirement savings crisis. Washington, D.C.: National Institute on Retirement Security. Robalino, D. A., & Holzmann, R. (2009). Overview and preliminary policy guidance. In R. Holzmann, D. A. Robalino, & N. Takayama (eds.), Closing the coverage gap: The role of pensions and other retirement income transfers, Washington, D.C.: International Bank for Reconstruction and Development, World Bank. Rockeymoore, M. M., & Lui, M. (2011). Plan for a new future: The impact of Social Security reform on people of color. Washington, D.C.: Commission to Modernize Social Security. Shapiro, T., Meschede, T., & Osoro, S. (2013, Feb.). The roots of the widening racial wealth gap: Explaining the black-white economic divide. Research and Policy Brief. Waltham, Mass.: Institute on Assets and Social Policy, Brandeis University. Smeeding, T. M. (2016). Poverty measurement. In D. Brady & L. M. Burton (eds.), The Oxford handbook of the social science of poverty, New York: Oxford University Press. Smolensky, E., Danziger, S., & Gottschalk, O. (1987). The declining significance of age in the United States trends in the well-being of children and the elderly since Discussion Paper Madison, Wisc.: Institute for Research on Poverty, University of Wisconsin. Social Security Administration. (2016). Annual statistical supplement to the Social Security Bulletin, Report Washington, D.C.: Author. Social Security Administration. (2014). Special minimum benefit: Program explainer. Washington, D.C.: Office of Retirement Policy, Author. Accessed September 30, 2016, at program/special-minimum.html. Staats, E. B. (1979). Minimum Social Security benefit: A windfall that should be eliminated. Report to Congress by the Comptroller General. HRD Washington, D.C.: General Accounting Office. Sullivan, L., Meschede, T., Mann, A., Ramirez- Kuykendall, S., & Shapiro, T. (2015). Navigating an unclear path: Preparing for retirement in the 21st century. Leveraging Mobility Series. Waltham, Mass.: Institute on Assets and Social Policy, Brandeis University. 24 A REVISED MINIMUM BENEFIT TO BETTER MEET THE ADEQUACY AND EQUITY STANDARDS IN SOCIAL SECURITY