Social Security's special minimum benefit is declining in relative value, does not provide a full benefit equal to the poverty threshold, and reaches fewer beneficiaries each year. Members of Congress and other key policymakers have proposed several methods for revising the special minimum benefit, either as part of reforming Social Security more broadly or as stand-alone policy options. Most of the new options would index the benefit to wages, helping ensure its sustainability into the future. The options differ in how they define a “year of coverage,” how many years of coverage are required to be eligible for any benefit increase, and how much the full benefit increase should be. Those choices will determine who will receive the benefit increase and how adequate their benefit will be.
Glenn Springstead, Kevin Whitman, and Dave Shoffner are with the Office of Retirement Policy, Office of Retirement and Disability Policy, Social Security Administration. Questions about the analysis should be directed to the authors at (202) 358-6234, (202) 358-6317, and (202) 358-6210, respectively.
Acknowledgments: The authors thank Natalie Lu, Mark Sarney, Melissa Favreault, Kathleen Romig, Hilary Waldron, and Craig Feinstein for their helpful comments and suggestions.
The findings and conclusions presented in this brief are those of the authors and do not necessarily represent the views of the Social Security Administration.
|AWI||average wage index|
|CPI||consumer price index|
|HHS||Department of Health and Human Services|
|MINT||Modeling Income in the Near Term|
|NASI||National Academy of Social Insurance|
|OCACT||Office of the Chief Actuary|
|PIA||primary insurance amount|
|QC||quarter of coverage|
|WEP||Windfall Elimination Provision|
|YOC||year of coverage|
The special minimum benefit, an alternative primary insurance amount (PIA) intended to increase benefits for long-term low-wage earners, is projected to be functionally obsolete for retired workers beginning with those who will become eligible for benefits between 2017 and 2023 (Feinstein 2013). Researchers and policymakers have long expected that outcome because the special minimum benefit, which is price-indexed, has risen more slowly than the regular PIA, which is wage-indexed (Feinstein 2000).1 The special minimum benefit's declining relevance, coupled with proposals to bolster trust fund solvency that reduce benefits, has generated broad interest in new minimum-benefit policy options to improve low lifetime earners' retirement security.
To structure a minimum benefit, policymakers must decide on features such as the minimum benefit amount and whether a work history is required to qualify for it and, if so, what type of work history. For example, the minimum benefit amount can be set relative to the poverty level to ensure a desired level of income adequacy. In turn, the work history needed to receive that benefit amount can differ by duration (proposals often specify 30 work years), as well as by the yearly covered-earnings level needed to qualify as a “work year.” Options can also include prorating the minimum benefit amount for beneficiaries with shorter work histories. These factors can substantially influence the effects of different minimum benefit proposals on beneficiaries. This brief uses projections from Version 6 of the Modeling Income in the Near Term (MINT6) microsimulation model to show how various proposed minimum benefit policy levers could affect aged beneficiaries.
How the Current Special Minimum Benefit Works
Congress established the special minimum benefit in 1972 to target increased benefits to workers with low earnings over long careers. At that time, the program already had a “regular minimum benefit”—in place since Social Security's inception—but legislators were concerned that it provided a windfall to workers whose low lifetime earnings were due to sporadic work histories in covered employment, rather than to consistent low-wage work (Olsen and Hoffmeyer 2001/2002).2 The special minimum benefit was designed to address that concern.3
To be eligible for the special minimum benefit, a worker would need to accrue a certain level of lifetime Social Security–covered earnings as determined by a programmatic measure called the “year of coverage” (YOC), described later. The worker would need at least 11 YOCs to qualify, and the minimum benefit amount would increase with each YOC accrued from 12 to 30. The tabulation below presents illustrative special minimum benefit amounts for workers with 11, 20, and 30 YOCs as of December 2013. The values reflect the number of YOCs multiplied by $11.50, adjusted for post-1979 consumer price index (CPI) growth. The special minimum benefit is awarded to individuals only when it exceeds the regular PIA, which is the benefit an individual would receive if he or she started receiving retirement benefits at the full retirement age.
|YOCs||Monthly special minimum benefit as of December 2013 ($)|
|SOURCE: SSA (n.d. b).|
As of December 2012, about 64,000 beneficiaries in current-payment status received the special minimum benefit (SSA 2013, Table 5.A8). That figure represents a reduction of more than half since 2000, when there were more than 140,000 special minimum beneficiaries. Over the last decade, fewer than 1,000 beneficiaries each year, on average, have been newly awarded the special minimum benefit (Shelton 2011). All new special minimum beneficiaries since 1998 have qualified because they were subject to the Windfall Elimination Provision (WEP) (Feinstein 2013). The WEP reduces benefits for individuals receiving a pension based on employment not covered by Social Security.4 Those beneficiaries do not constitute the population policymakers originally targeted for the special minimum benefit.
Special minimum benefit receipt has declined because the benefit amount is indexed to price growth, while the regular PIA is indexed to wage growth. Average wages generally rise faster than prices because of productivity growth, and since the 1970s, that relationship has largely held true. The Social Security Administration's Office of the Chief Actuary (OCACT) found that the real-wage differential, which is the difference between price growth and wage growth, averaged 0.7 percentage points per year from 1970 through 2009 (Board of Trustees 2011).
The demise of the special minimum benefit is presaged by the trend in the number of YOCs effectively required to qualify. In 2000, OCACT estimated that workers gaining eligibility that year would need 27 YOCs to qualify for the special minimum benefit (Feinstein 2000). OCACT currently estimates that in the next few years, newly eligible beneficiaries will cross the threshold of needing more than 30 YOCs to qualify (Feinstein 2013). However, even that limited eligibility is theoretical at this point, as it has been more than 15 years since a new special minimum beneficiary was not subject to the WEP.
The special minimum benefit's structure and growing obsolescence reflect the complex relationship of multiple policy levers. The following section explores those levers in more detail, alongside projections of the shares of beneficiaries aged 62 or older in 2013 who exhibit certain characteristics that correspond with various eligibility thresholds.5 We focus on 2013 because it is the last year for which we have data for many of the variables we use in our analysis, such as the Census Bureau poverty threshold. However, readers should note that the 2013 beneficiary characteristics we present from MINT are projected.6
Minimum Benefit Policy Levers
The special minimum benefit incorporates the following core policy levers:
- a formula for determining the earnings level that constitutes a YOC,
- a YOC eligibility threshold required for any special benefit (subject to prorating),
- a YOC eligibility threshold required for the full special benefit,
- a full special benefit amount, and
- a method of indexing the benefit.
Table 1 summarizes the current-law values that correspond with each lever as of December 2013.
|YOC formula||15 percent of an annually adjusted earnings threshold a|
|YOCs needed for—|
|Any minimum benefit||11|
|Full minimum benefit||30|
|Full monthly benefit ($)||816|
|Method of indexing||Consumer price index (CPI)|
|SOURCE: SSA (n.d. a, n.d. b).|
|a. The threshold, called the “old-law contribution and benefit base,” accounts for changes in national average wages.|
Many proposals to revise or replace the special minimum benefit include the same policy levers while suggesting different values. For example, some proposals would retain a work requirement in order to target retired workers with low lifetime earnings, but the formulas they would use to establish a YOC vary. Other proposals would increase the full benefit amount to reduce poverty among older beneficiaries, as the current-law amount is lower than the poverty threshold.7 In addition, some proposals would add new policy levers to try to refine eligibility standards. Common examples include allowing years spent taking care of a young child to count toward the work requirement and downscaling the work requirement for persons who become disabled.
The following sections address each core lever and discuss policy alternatives proposed in five publicly available plans to revise the special minimum benefit. The plans we examine are from the following sources:
- The National Commission on Fiscal Responsibility and Reform (Fiscal Commission);
- The National Academy of Social Insurance (NASI);
- The Bipartisan Policy Center's Debt Reduction Task Force (Policy Center);
- U.S. Representative Paul Ryan (R-WI); and
- U.S. Representative Jason Chaffetz (R-UT).8
A worker earns a YOC for each year his or her covered earnings meet or exceed a threshold determined by a formula that is set by law. A key component of that formula is called the “old-law contribution and benefit base” (shortened to “old-law base,” or OLB) because it reflects the base that would be in place if the 1977 Amendments to the Social Security Act were not in effect (SSA n.d. a).9 For years worked before 1991, a worker needed earnings of at least 25 percent of the OLB to earn a YOC. In 1991, to expand eligibility, the threshold was lowered to 15 percent of the OLB (Feinstein 2000). In 2013, the OLB was $84,300; thus, 15 percent of that amount, or $12,645, was the earnings level required to qualify for a YOC in 2013.
New minimum benefit proposals have offered a variety of YOC formulas. Some proposals would continue to use the OLB with a different percentage threshold. For example, the Policy Center proposed setting the threshold at 20 percent of the OLB.10 Alternatively, the Fiscal Commission, NASI, and Chaffetz proposals would adopt four “quarters of coverage” (QCs)— the earnings level used in determining eligibility in regular PIA calculations—as the YOC value.
The tabulation below compares the 2013 earnings levels that would have constituted a YOC under three alternative formulas. A minimum benefit proposal's YOC criteria can have substantial effects on eligibility. For example, with all else being equal, a minimum benefit provision that sets the YOC at four QCs will have wider reach than a provision requiring earnings of at least 20 percent of the OLB. The lower the earnings threshold used for the YOC measure, the higher the share of beneficiaries who will potentially be eligible for the minimum benefit.
|Formula||Earnings (2013 $)|
|15 percent of OLB (current law)||12,645|
|20 percent of OLB||16,860|
|SOURCE: Authors' calculations.|
Some plans would also expand the minimum benefit's reach by allowing years spent caring for a child to count as YOCs. For example, the Policy Center option would allow a maximum of eight childcare credits, earned in full-year increments, to count toward the YOC requirement. The Chaffetz plan would allow up to five childcare credits. Under both plans, both parents could earn a childcare credit if they had a child aged younger than 6 and did not have sufficient earnings to meet the plan's YOC requirement. Those policy choices would significantly enlarge the universe of potential minimum-benefit recipients. For example, among beneficiaries aged 62 or older in 2013, around half would have had at least 1 year that qualified for a childcare credit.
YOC Thresholds for Minimum Benefit Eligibility
Under current law, an individual needs 11 YOCs to qualify for any minimum benefit and 30 YOCs to receive the full minimum benefit.11 Many minimum-benefit proposals, including those of the Fiscal Commission and NASI, retain the current 11-YOC threshold for any minimum benefit and the 30-YOC threshold for the full minimum benefit. However, some plans have higher minimum YOCs to more narrowly target low earners with longer working careers. For example, the Policy Center option would require beneficiaries to have at least 20 YOCs to be eligible for a partial minimum benefit.12
Plans with broader YOC threshold ranges provide a greater number of possible minimum benefit levels. For example, a plan with an 11–30 YOC range introduces 20 possible minimum benefits, compared with 11 possible minimum benefit levels for plan using a 20–30 YOC range. Additionally, minimum benefit proposals with greater YOC ranges produce benefits that rise more gradually with each additional YOC, while remaining above the comparable benefit available under a plan with a narrower YOC range. Chart 1 provides an illustrative example using the current special minimum's full benefit amount and comparing the 11–30 and 20–30 YOC ranges.13
The percentage of beneficiaries with earnings records that meet or exceed the specific YOC threshold declines substantially as the threshold increases from 11 to 30 YOCs. Chart 2 illustrates that pattern, showing the projected share of the beneficiary population aged 62 or older in 2013 with lifetime earnings that would have met three different YOC thresholds (11, 20, and 30) under three alternative YOC formulas: four QCs, 15 percent of the OLB, and 20 percent of the OLB.
With a YOC defined as 20 percent of the OLB, 80 percent of the beneficiaries aged 62 or older in 2013 would have met the 11-YOC requirement, 66 percent would have met the 20-YOC requirement, and 49 percent would have met the 30-YOC requirement. The shares of that population that would meet the benefit-eligibility criteria are 11 to 14 percentage points higher in each YOC category when using the four-QC measure. However, even if four QCs—the lowest of the earnings levels in this analysis—were to constitute a YOC, more than one-third of the beneficiaries would still have been ineligible for a full special minimum benefit because they would have fewer than 30 YOCs.
Full Benefit Amount
The current special minimum benefit is equal to the number years of work (as measured by YOCs) multiplied by $11.50, with adjustments for post-1979 growth in the CPI. As of December 2013, the full monthly benefit was $816.
New proposed minimum benefit amounts are frequently targeted to an explicit measure of income adequacy, most often a poverty threshold, and are more generous than the current-law special minimum. For example, proposals for a new full minimum benefit specify a percentage of the Census Bureau's poverty threshold for aged individuals (133 percent in the Policy Center plan and 120 percent in the Ryan plan) or of the Department of Health and Human Services (HHS) poverty guidelines (125 percent in the Fiscal Commission and NASI plans).14 In comparison, the current full special minimum benefit equals about 88 percent of the Census Bureau aged poverty threshold and about 85 percent of the HHS poverty guideline. Table 2 compares some of the full benefit targets in 2013 dollars.15 All proposed minimum benefit amounts exceed the current special minimum PIA level by around $300 to $400 per month.
|120 percent of Census Bureau aged poverty threshold||1,117||13,408|
|133 percent of Census Bureau aged poverty threshold||1,238||14,860|
|125 percent of HHS poverty guideline||1,197||14,363|
|SOURCE: Authors' calculations.|
As noted earlier, to qualify for the special minimum benefit, an individual's regular PIA must be lower than his or her corresponding minimum benefit amount. Thus, the share of beneficiaries aged 62 or older with potential eligibility for the special minimum benefit would be substantially higher if YOCs were based on four QCs than it would be under a measure such as 20 percent of the OLB. To illustrate, Table 3 shows, for beneficiaries aged 62 or older with 30 or more YOCs in 2013, the percentages with PIAs that fall below 125 percent of the HHS poverty guideline under two YOC formulas (20 percent of the OLB and four QCs). It also shows the current-law regular PIA values for those two YOC-formula groups at selected percentiles. Those values indicate the proportions in each group that might be eligible for a benefit increase under alternative full benefit amounts.
|Formula||Percentage of beneficiaries with PIAs lower than 125 percent of the HHS poverty guideline ($1,197)||Current-law PIA at selected percentiles ($)|
|20 percent of OLB||9.8||1,203||1,432||1,728||2,007||2,211|
|SOURCE: Authors' calculations using MINT6.|
Among beneficiaries with 30 YOCs at 20 percent of the OLB, 9.8 percent have regular PIAs below 125 percent of the monthly HHS poverty guideline ($1,197). The comparable figure is significantly higher when using the four-QC measure (24.3 percent). Similar figures appear for the other full minimum benefit targets discussed in this brief, which are all within $100 (per month) of 125 percent of the HHS poverty guideline.
We have analyzed the special minimum benefit policy levers as of 2013. However, the future growth rates for the full benefit amount and the YOC dollar thresholds are important to the long-term performance of any new minimum benefit policy.
The special minimum benefit amount currently increases with changes in the CPI. New minimum benefit options also generally include some method of indexing to increase the benefit amount over time. Indexing options include continuing to use the CPI, adopting the national average wage index (AWI), or employing another measure.
The indexing choice can have significant effects. For example, research has shown that indexing the full benefit to wage growth rather than price growth would contribute to reducing the poverty rate of Social Security beneficiaries (Favreault, Mermin, and Steuerle 2007). The indexing target can also affect the long-term viability of the minimum benefit. Assuming that average wage growth continues to outpace average price growth, indexing the benefit to prices may cause it to become obsolete over time, as is true of the current special minimum. Chart 3 projects minimum benefit values using the CPI and the AWI, assuming an income adequacy measure of 125 percent of the HHS poverty guideline for 2013–2050. The CPI-based special minimum benefit steadily lags behind the value of the AWI-based minimum benefit.
The projected obsolescence of Social Security's special minimum benefit, coupled with policy efforts to protect low lifetime earners while ensuring long-term system solvency, has spurred consideration of new minimum benefit options. Recent proposals have used a variety of policy levers to enhance retirement security for low lifetime earners. This policy brief describes many of those levers and their effects on aged beneficiaries.
Our projections suggest that even though few beneficiaries would receive a benefit increase under most current proposals, some specific policy levers could significantly influence eligibility. For instance, the percentage of beneficiaries potentially eligible for the minimum benefit will be higher if YOC thresholds equal four QCs rather than 20 percent of the OLB. Future analysis should continue to explore the distributional effects of different minimum benefit options. For example, we plan to release additional work using MINT data to focus on the effectiveness of various publicly proposed plans in reaching their ostensible target populations of low lifetime earners.
1 Hereafter we refer to the wage-indexed PIA as the “regular PIA,” in contrast to the price-indexed special minimum benefit. (The Social Security Administration formerly used a third PIA formula for workers who died or were first eligible for retirement benefits before 1979.)
2 The original “regular minimum benefit,” created by the Social Security Act of 1935, was set at $10 per month. Congress increased that amount with ad hoc changes throughout the program's early history (Schobel and McKay 1982). The regular minimum benefit provision was terminated for most workers who became eligible for retirement or disability benefits after December 1981, although for some workers serving as members of a religious order, the regular minimum benefit was not eliminated until January 1992. For further details, see section 716 of the Social Security Handbook (http://www.socialsecurity.gov/OP_Home/handbook/handbook.07/handbook-0716.html).
3 The terms “special minimum PIA” and “special minimum benefit” are interchangeable.
5 Note that for 2013, the maximum age for beneficiaries included in MINT data is 87. Despite that restriction, MINT captures most beneficiaries. For example, as of December 2012, almost 94 percent of retired-worker beneficiaries were aged 87 or younger (SSA 2013, Table 5.A1.1).
6 MINT6 includes data from the 2001 and 2004 Surveys of Income and Program Participation (SIPP) linked to Social Security administrative records for earnings and benefits through 2009 (Smith and others 2010).
7 Favreault, Mermin, and Steuerle (2006) note that the elderly poverty rate is lower than that for prime-age workers and children; yet for some aged subgroups, such as women and minorities, a larger minimum benefit could substantially reduce poverty.
8 For the full reports of the plans from the first three listed sources, respectively see Fiscal Commission (2010); Sullivan, Meschede, and Shapiro (2008); and Policy Center (2010). For OCACT memoranda analyzing the financial effects of the plans from the fourth and fifth listed sources, see SSA (2010 and 2011), respectively.
9 Because the OLB is substantially lower than the current-law base, its use provides a lower earnings threshold.
10 Although the Policy Center plan would have a higher YOC threshold than the current-law YOC threshold, its full benefit amount would also be higher.
11 Real wage growth has steadily increased the number of YOCs required to receive the special minimum benefit. Nevertheless, even in 1972, Congress estimated that a worker would need at least 23 YOCs to obtain a special minimum benefit higher than his or her regular PIA. Although such a high threshold may seem counterintuitive given the formulaic minimum of 11 YOCs, Congress specifically intended to target the benefit to workers with long histories of low-wage work (Olsen and Hoffmeyer 2001/2002), and most new minimum benefit proposals echo that emphasis.
12 Social Security's special minimum PIA eligibility requirements for disabled-worker benefits are the same as those for retired-worker benefits. However, given the truncated earnings histories of disabled beneficiaries, some minimum-benefit options adjust YOC requirements for disabled workers. Minimum-benefit options could (1) exclude disabled beneficiaries from eligibility altogether; (2) treat them the same as other beneficiaries, with the same YOC requirements; or (3) scale the YOC requirement to account only for years in which the disabled beneficiary was in the workforce.
13 The Ryan plan includes complex YOC, benefit level, and program interactions that do not align perfectly with the simplified lever presentation we use. Thus, we do not include the Ryan plan in the “YOC formula” and “YOC thresholds for minimum benefit eligibility” sections of this brief. SSA (2010) describes the Ryan plan's benefit enhancement for low earners as follows: “the PIA for a worker with 30 years of earnings at an average wage-indexed level equivalent to the full-time annual minimum wage for 2009 would receive an increase in the PIA sufficient to yield an adjusted PIA equal to 120 percent of the Federal poverty level for an aged individual… The percentage increase in PIA provided to a worker with fewer than 30 years of covered earnings would be reduced linearly, reaching no enhancement for the worker with 20 or fewer years of earnings. The year-of-work requirements above would be ‘scaled' to the length of the elapsed period from age 22 to benefit eligibility for workers who become disabled or die before reaching age 62. In addition, the percentage increase in PIA would be reduced proportionally for workers with higher [adjusted indexed monthly earnings (AIME)], reaching a zero increase for the worker with an AIME equal to twice the level of a 35-year steady full-time minimum wage earner.”
14 The federal government uses the Census Bureau thresholds for statistical purposes and the HHS guidelines for administrative purposes such as determining eligibility for certain programs. For descriptions of the thresholds and guidelines, see http://www.census.gov/hhes/www/poverty/data/threshld/index.html and http://aspe.hhs.gov/POVERTY/index.cfm, respectively.
Bipartisan Policy Center Debt Reduction Task Force. 2010. Restoring America's Future: Reviving the Economy, Cutting Spending and Debt, and Creating a Simple, Pro-Growth Tax System. Washington, DC: Bipartisan Policy Center. http://www.bipartisanpolicy.org/library/report/restoring-americas-future.
[Board of Trustees] Board of Trustees of the Federal Old-Age and Survivors Insurance and Federal Disability Insurance Trust Funds. 2011. The 2011 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Federal Disability Insurance Trust Funds. Washington, DC: Government Printing Office.
Favreault, Melissa M., Gordon B. T. Mermin, and C. Eugene Steuerle. 2006. “Minimum Benefits in Social Security.” Washington, DC: The Urban Institute. http://www.urban.org/publications/411406.html.
———. 2007. “Minimum Benefits in Social Security: Design Details Matter.” Older Americans Economic Security Brief No. 10. Washington, DC: The Urban Institute. http://www.urban.org/publications/311417.html.
Feinstein, Craig A. 2000. “Projected Demise of the Special Minimum PIA.” Actuarial Note No. 143. Social Security Administration, Office of the Chief Actuary. http://www.socialsecurity.gov/OACT/NOTES/note2000s/note143.html.
———. 2013. “Diminishing Effect of the Special Minimum PIA.” Actuarial Note No. 154. Social Security Administration, Office of the Chief Actuary. http://www.socialsecurity.gov/OACT/NOTES/pdf_notes/note154.pdf.
[Fiscal Commission] National Commission on Fiscal Responsibility and Reform. 2010. The Moment of Truth: Report of the National Commission on Fiscal Responsibility and Reform. http://www.fiscalcommission.gov/sites/fiscalcommission.gov/files/documents/TheMomentofTruth12_1_2010.pdf.
Olsen, Kelly A., and Don Hoffmeyer. 2001/2002. “Social Security's Special Minimum Benefit.” Social Security Bulletin 64(2): 1–15. http://www.socialsecurity.gov/policy/docs/ssb/v64n2/v64n2p1.pdf.
[Policy Center] See Bipartisan Policy Center Debt Reduction Task Force.
Schobel, Bruce D., and Steven F. McKay. 1982. “Characteristics of Newly Awarded Recipients of the Social Security Regular Minimum Benefit.” Social Security Bulletin 45(6): 11–19. http://www.socialsecurity.gov/policy/docs/ssb/v45n6/v45n6p11.pdf.
Shelton, Alison M. 2011. Social Security: The Minimum Benefit Provision. CRS Report No. R41518. Washington, DC: Congressional Research Service.
Smith, Karen E., Melissa M. Favreault, Barbara A. Butrica, and Philip Issa. 2010. Final Report: Modeling Income in the Near Term Version 6. Washington, DC: The Urban Institute. http://www.urban.org/UploadedPDF/412479-Modeling-Income-in-the-Near-Term-Version-6.pdf.
[SSA] Social Security Administration. 2010. “Estimated Financial Effects of Title IV of ‘The Roadmap for America's Future Act of 2010'—Legislation Introduced as Title IV of H.R. 4529 (111th Congress) on January 27, 2010, by Representative Paul Ryan.” http://www.socialsecurity.gov/oact//solvency/PRyan_20100427.pdf.
———. 2011. “Estimated Financial Effects of a Proposal to Restore Sustainable Solvency for the Social Security Program Requested by Representative Jason Chaffetz.” http://www.socialsecurity.gov/oact//solvency/JChaffetz_20111109.pdf.
———. 2013. Annual Statistical Supplement to the Social Security Bulletin, 2013. http://www.socialsecurity.gov/policy/docs/statcomps/supplement/2013/index.html.
———. No date a. “Old-Law Contribution and Benefit Base.” http://www.socialsecurity.gov/OACT/COLA/oldcbb.html.
———. No date b. “Social Security Benefit Tables: Special Minimum Benefit Tables.” http://www.socialsecurity.gov/oact/progdata/tableForm.html.
Sullivan, Laura, Tatjana Meschede, and Thomas M. Shapiro. 2008. “Enhancing Social Security for Low-Income Workers: Coordinating an Enhanced Minimum Benefit with Social Safety Net Provisions for Seniors.” Washington, DC: National Academy of Social Insurance. http://www.nasi.org/sites/default/files/research/Sullivan,_Meschede,_Shapiro_January_2009_Rocke.pdf.