Characteristics of Noninstitutionalized DI and SSI Program Participants, 2013 Update

Introduction

The Social Security Administration (SSA) produces several statistical publications based on the data used to administer the Disability Insurance (DI) and Supplemental Security Income (SSI) programs. Although these data are extensive, they do not capture many of the economic and demographic characteristics of program participants. To better understand those beneficiary populations, SSA matches information from its administrative records with data collected by the Census Bureau in the Survey of Income and Program Participation (SIPP) and presents the results in notes such as this. DeCesaro and Hemmeter (2008) contains tables describing the characteristics of SSI and DI participants based on 2002 data, Bailey and Hemmeter (2014) updates those tables with 2010 data, and this note updates the tables with 2013 data.

Data

The SIPP is a household survey of the noninstitutionalized resident population of the United States. The survey broadly measures the economic situations of households, paying particular attention to the role of government transfer and service programs. Within a given SIPP wave, each participating household is tracked for several years with interviews conducted at regular intervals. In addition to the core questions about income and program participation, each interview includes questions from a rotating module on a topic such as marital history or education.

The estimates in this note are based on samples of DI beneficiaries and SSI recipients from the public-use data file for the 2008 SIPP panel, the same file used in the 2014 note to produce the 2010 estimates. For ease of comparison, the chart and table numbers in this update correspond with those used in the previous edition.¹ However, the definitions of some of the characteristics differ from those used in the earlier tables. Those changes were necessary to avoid disclosing information about individual sample members.

SIPP data are matched to Social Security administrative records based on the respondent's validated Social Security number. For individuals with valid numbers, we use the administrative records to determine whether they received DI benefits or SSI payments and, if so, the amounts received. For sample members without available matching administrative records, we use self-reported values from the SIPP. We identify individuals as DI beneficiaries and SSI recipients if they received a payment in the final month of the 4-month SIPP reference period.² Consistent with the 2014 edition of this note, we define SSI receipt only in terms of federally administered payments; that is, we do not include SSI state supplements in our definition. In tabulations of SSI recipients' Medicaid coverage, we recode the SIPP Medicaid variable to reflect the automatic Medicaid coverage of SSI recipients in certain states under Section 1634 of the Social Security Act.

Our analysis is based on follow-up interviews of members of a sample initially chosen to be representative of the 2008 noninstitutionalized population. We use data collected from four rotation groups in wave 15 of the 2008 SIPP panel, covering January through July 2013. The first rotation group was interviewed in May and provided information for January through April. The second, third, and fourth groups were respectively interviewed in June, July, and August, and likewise provided information for the preceding 4 months.

We use the data for those 4-month reference periods to measure income, poverty, and SSI and DI benefit amounts. All other demographic characteristics are based on SIPP data for the fourth month of the reference period (that is, April, May, June, or July, depending on the rotation group).

The current sample includes 2,589 DI beneficiaries and 2,162 SSI recipients—a slight reduction from the sample size for our previous update (2,644 DI beneficiaries and 2,207 SSI recipients from waves 7 and 8 of the 2008 SIPP, covering September–December 2010). About three-quarters of our wave 15 sample members were included in the sample for that update. We did not fully assess the causes of this attrition, but some of it is presumably due to individuals aging out of DI eligibility and becoming retired-worker beneficiaries instead. Additionally, some individuals will have died over the course of the survey panel. Other individuals will have newly entered the study population after experiencing recent health or financial shocks. We cannot determine whether or to what extent these changes in our sample affect its representativeness of the full population of noninstitutionalized program participants.

We weight the SIPP results using Social Security administrative data on the total number of DI beneficiaries and SSI participants.³ Because SSA data include institutionalized beneficiaries and SIPP results include only the noninstitutionalized population, we adjust the SSA program totals by the estimated percentages of DI beneficiaries and SSI participants residing in institutions.⁴

Discussion of the Estimates

This section highlights selected estimates from the tables and charts and notes key changes from the 2010 estimates. Tables 1–7 present the estimates for DI beneficiaries; Tables 8–14 present the estimates for SSI recipients. Changes to the presentation since the last update, both affecting DI beneficiaries, consist of the addition of two personal-income categories in Table 1 and a new table showing a measure of income adequacy if disabled-worker benefits were to be reduced by 19 percent (Table 6B). A 19 percent reduction corresponds with the expected DI Trust Fund shortfall should no legislative change occur by late 2016.⁵

DI Beneficiaries

DI beneficiaries in this sample comprise disabled workers, disabled widow(er)s, and disabled adult children. Because 91 percent of the beneficiaries in this sample were disabled workers, the tables present information separately for that subgroup, although we do not discuss those results in the text below. For convenience and consistency with other publications, we refer to all these groups as DI beneficiaries regardless of whether their benefit is paid from the Disability Insurance or the Old-Age and Survivors Insurance Trust Fund, as long as their eligibility is based on disability.⁶

Although DI beneficiaries are eligible for Medicare, coverage does not begin until 24 months after the first month of benefit receipt. In 2013, 64 percent of DI beneficiaries reported Medicare coverage (alone or in combination with other coverage), 45 percent reported Medicaid coverage, 31 percent reported private health insurance, and 7 percent reported having no health insurance (Chart 1). About 27 percent of beneficiaries held joint Medicare and Medicaid coverage and 20 percent had only Medicare coverage (Table 1). Many of those who reported Medicaid coverage were likely to be among the 21 percent of disabled beneficiaries who also received SSI payments. Other forms of income included earnings (received by 11 percent of DI beneficiaries), property income from all assets (received by 26 percent), and other public assistance (received by 32 percent). The percentage of DI beneficiaries receiving property income and income from unspecified other sources declined from 2010 levels by 4 and 6 percentage points, respectively.

Chart 1.
DI beneficiaries, by type of health insurance coverage, 2013 (in percent)

SOURCE: Social Security administrative records matched to 2008 SIPP wave 15.

NOTES: Individuals may have more than one type of coverage.

Data are for the month preceding the SIPP interview date. Interviews took place May–August 2013.

About half of DI beneficiaries had 4-month personal income between $2,500 and $4,999, almost a quarter had income between $5,000 and $7,499, and nearly another quarter had income of $7,500 or more. DI benefits were an important income source, constituting at least 75 percent of personal income for 3 out of 5 DI beneficiaries (Table 2). A higher percentage of beneficiaries relied on DI benefits for 100 percent of their income in 2013 (36 percent) than did so in 2010 (31 percent).

Nearly half of DI beneficiaries had 4-month family income of less than $10,000 and more than half owned their own home (Table 3). About 34 percent of DI beneficiary households relied on food assistance through the Supplemental Nutrition Assistance Program (SNAP, formerly known as food stamps). Table 4 shows that the two largest sources of family income for DI beneficiaries were their Social Security benefits (58 percent) and earnings (24 percent).

Although 29 percent of DI beneficiaries had family income at or above 300 percent of the poverty threshold, nearly 20 percent were in poverty (Table 5). DI beneficiaries were more likely to have family income below the poverty threshold if they were women, younger than 40, black, divorced or separated, or never married. The proportions of DI beneficiaries with family income above 300 percent of the poverty threshold differed widely by educational attainment. Only 11 percent of those with fewer than 12 years of education had family income above 300 percent of the threshold, compared with 43 percent of those with 16 years or more of education (Chart 2).

Chart 2.
DI beneficiaries with family incomes at selected percentages of the poverty threshold, by years of education, 2013 (in percent)

Show as table

Table equivalent for Chart 2. DI beneficiaries with family incomes at selected percentages of the poverty threshold, by years of education, 2013 (in percent)

Years of education	Family income relative to poverty threshold
	Less than 100%	300% or more
0–11	31.6	10.9
12	21.8	24.5
13–15	15.9	36.1
16 or more	11.7	42.7

SOURCE: Social Security administrative records matched to 2008 SIPP wave 15.

NOTES: A “family” is two or more persons residing together who are related by birth, marriage, or adoption.

Data are for the 4-month period ending with the month preceding the SIPP interview date. Interviews took place May–August 2013.

As part of the social safety net, DI benefits help keep some individuals out of poverty. If DI benefits were removed from their income, half of disabled beneficiaries would have been in poverty in 2013 based on their family income level (Table 6A). With benefits included in family income, only 19 percent of disabled workers had income below the poverty threshold. If the DI Trust Fund were depleted and disabled-worker benefits were reduced by 19 percent, the poverty rate for disabled workers would rise from 19 percent to 26 percent (Table 6B).⁷ Receiving DI benefits reduced the aggregate poverty gap, or the aggregate difference between family income and the poverty threshold for those below the threshold, by 85 percent (Table 7). Beneficiaries' economic behavior would likely change if they did not receive disability benefits or received reduced benefits; thus, hypothetical estimates that simply exclude or reduce DI benefits do not fully capture the economic situation beneficiaries and their families would experience in the absence of the program or in the face of reduced benefits. Rather, these estimates represent how the program contributes to the reduction in poverty under existing patterns of behavior.

SSI Recipients

SSI recipients in this sample include aged, blind, and disabled individuals. The tables show results for three subgroups: children (younger than age 18), working-age adults (18–64), and the aged (65 or older).⁸ The maximum monthly SSI payment amount, called the federal benefit rate, can be reduced for individuals with countable earnings and unearned income. In 2013, the federal benefit rate was $710 for an individual and $1,066 for a couple in which both members were eligible.

Chart 3 summarizes educational attainment patterns. For both working-age and aged recipients, greater shares had at least 12 years of education in 2013 (72 percent and 60 percent, respectively) than did in 2010 (68 percent and 47 percent). Although education levels are lower for aged recipients than for working-age recipients, the gap has narrowed substantially, from 21 to 12 percentage points, since 2010. In 2013, about 29 percent of aged recipients had less than 9 years of education, compared with 10 percent of working-age recipients (Table 8). In 2010, the corresponding estimates were 37 and 12 percent.

Chart 3.
SSI recipients, by educational attainment and age group, 2013 (in percent)

SOURCE: Social Security administrative records matched to 2008 SIPP wave 15.

NOTE: Data are for the month preceding the SIPP interview date. Interviews took place May–August 2013.

Unlike Medicare coverage under DI, which begins only after 24 months in the program, Medicaid coverage begins immediately for SSI recipients in most states. In 2013, 96 percent of all SSI recipients reported Medicaid coverage and 39 percent reported Medicare coverage. The percentage of aged SSI recipients with Medicare coverage is undoubtedly high, although perhaps not quite the 99 percent reported in the SIPP.⁹ We note that a substantial proportion of aged SSI recipients reported Social Security benefits in the SIPP (76 percent); however, according to SSA records, only 56 percent of SSI recipients aged 65 or older had Social Security benefits (SSA 2015, Table 7.D2). Private health insurance covered about 9 percent of SSI recipients. Table 8 also shows that 50 percent of recipients reported receiving public assistance other than SSI (such as General Assistance) and 6 percent reported earnings. About 39 percent of SSI recipients reported receiving Social Security benefits and 6 percent reported other sources of income.

The majority of adult SSI recipients (52–53 percent, depending on the age group) had 4-month total personal income below $3,000 and most of those individuals had income between $2,500 and $2,999. We note that SSI payments for a recipient who qualified for the federal benefit rate of $710 in 2013 would amount to $2,840 over a 4-month period, which would fall within that range.

Majorities of SSI adult recipients were nonhomeowners and received SNAP food assistance (Table 9). Homeownership rates declined almost 20 percent from the 2010 levels for aged recipients (from 40 percent to 32 percent) and families with a child SSI recipient (from 30 percent to 25 percent). About three-quarters of working-age recipients resided in households where they were the sole SSI adult recipient. Similarly, for aged recipients, 67 percent resided in households where there was only one adult recipient.

Although family income (shown in Table 9) was generally higher than personal income (shown in Table 8) for SSI recipients, total 4-month family income was still less than $7,500 ($1,875 a month) for 57–60 percent of adult SSI recipients. Family income for families with a child SSI recipient was higher, with only about 30 percent falling below $7,500 across 4 months. Although the percentage of aged recipients relying on SSI for 100 percent of personal income (17 percent, Table 10) fell slightly from its 2010 level (19 percent), that share among working-age recipients rose by more than 9 percentage points, from nearly 34 percent in 2010 to 43 percent in 2013. Family income for SSI recipients primarily comprised SSI payments, Social Security benefits, and earnings (Table 11). SSI payments accounted for 43 percent of family income, followed by earnings, which accounted for 25 percent, and Social Security benefits, which accounted for another 23 percent. The share of family income contributed by earnings was higher for families with child SSI recipients than for working-age recipients (35 percent versus 21 percent). The majority of family income for SSI recipients aged 65 or older was provided by the combination of Social Security benefits (37 percent) and SSI payments (32 percent), although earnings also accounted for 26 percent.

A substantial minority (43–44 percent) of adult SSI recipients and 34 percent of families with a child SSI recipient had family income below the poverty level in 2013 (Table 12). Among those most likely to fall under the poverty threshold were recipients who were divorced or separated and recipients with 9–11 years of education. SSI payments helped some families move out of poverty, but most families remained within 150 percent of the poverty threshold. Without counting SSI payments in family income, the poverty rate would be 63 percent, compared with the actual rate (42 percent) when SSI payments are included (Table 13). The percentage of child SSI recipients with family income below the poverty line falls from 58 percent to 34 percent when SSI is counted. Thus, of children whose family income without SSI would be below the poverty line, about 41 percent are not in poverty when SSI payments are included in family income. The impact of SSI payments on the family income of working-age recipients is somewhat smaller, helping 36 percent move above the poverty threshold. SSI payments reduced the aggregate poverty gap by about 68 percent (Table 14).

In closing, we note that the Census Bureau substantially changed the SIPP beginning in 2014. SIPP interviews are now conducted yearly rather than every 4 months, and include somewhat different questions. The SIPP was redesigned to improve accuracy, reduce the burden to respondents and survey administration costs, and ensure that the data collected are relevant and timely (Census Bureau 2015). In 2014, SSA funded a supplemental interview fielded by the Census Bureau about 6 months after the first interview of the redesigned SIPP. Although SSA will explore the possibility of updating these tables using the redesigned SIPP, comparable estimates may not be forthcoming.