Early Retirement Safety Net Gap

Definition

The early retirement safety net gap is the structural disconnect between the Social Security early retirement program (Old-Age and Survivors Insurance [OASI] benefits available at age 62) and the formal disability programs (Disability Insurance [DI] and Supplemental Security Income [SSI]) that leaves a significant minority of health-impaired 62–64 year olds reliant on early retirement benefits as their only income floor. Two institutional barriers prevent this population from accessing disability programs: (1) DI requires a sufficient work history (insured status — 20 of the last 40 quarters), which many low-earning women do not have; (2) SSI's $2,000 asset means test excludes those with modest but nonzero savings. Early OASI benefits bridge this gap, functioning as an unofficial disability program for workers who are too impaired to work but structurally ineligible for formal disability programs.

Key Ideas

Scale: In the early 1990s, about 12% of 62–64 year old OASI beneficiaries simulated to meet the strict Social Security Administration (SSA) medical disability criteria (Leonesio and Wixon 2004). About as many 62–64 year olds classified as severely disabled received early OASI benefits as received DI or SSI.
The insured status barrier (DI): DI requires 20 quarters of covered earnings in the last 40 quarters (5 of the last 10 years). In 1990, only $\approx 50\%$ of women aged 62 had the quarters needed, vs. $\approx 80\%$ of men. Women who spent years out of the paid labor force — as caregivers, in part-time informal work, or in uncovered employment — fail this test despite having qualifying impairments.
The asset test barrier (SSI): SSI's $2,000 individual asset limit excludes people with modest savings who nevertheless have little income. Among the simulated SSA-disabled OASI subgroup, median financial assets were only $800 — many are close to the SSI threshold but not below it.
The gap population is highly vulnerable: Among OASI beneficiaries simulated to meet SSA disability criteria: poverty rate 25% (vs. 4% for healthy early retirees); median financial assets $800 (vs. $37,652); Social Security $\geq 50\%$ of family income for 58%; median lifetime indexed earnings $2,332/yr (vs. $9,318); 79% female; 53% widowed/divorced/separated; 53% with less than 12 years of schooling.

What Would Happen if Early Retirement Benefits Were Restricted

Leonesio and Wixon (2004) simulate the counterfactual in which OASI benefits are no longer available at age 62–64 for the affected population:

Outcome	Share of OASI beneficiaries	Number
No significant work limitation	75%	2,322k
Significant work limitation — eligible for SSI disability	5%	155k
Significant work limitation — eligible for DI	3%	92k
Significant work limitation — not eligible for either	17%	533k

The 17% — roughly half a million people — have qualifying health impairments but fall through both disability programs: they fail DI's insured-status test and SSI's asset test simultaneously. They would be left with no income support.

Why It Matters

For Early Retirement Age Policy

Proposals to raise the Social Security early eligibility age from 62 — motivated by longevity improvements and fiscal sustainability — implicitly assume that the affected population is healthy enough to continue working or can access alternative programs. Leonesio and Wixon's simulation directly falsifies this assumption for a large minority of early retirees. Without parallel reform of DI insured-status rules and the SSI asset test, raising the early retirement age shifts the cost of program failure onto the most economically precarious early retirees.

For DI Caseload Interpretation

Early retirement serves as a substitute for DI among 62–64 year olds with insufficient work histories, which means measured DI recipiency understates the true scale of health-related labor force exit at this age group. This is a mirror image of Autor and Duggan's (2003) finding that DI expansion lowers measured unemployment — here, OASI availability lowers measured DI receipt among near-elderly disabled workers who cannot access DI. Both programs function as partially substitutable labor force exit pathways. See DI Replacement Rate.

For Mortality Research

The severe health gradient within early retirees has direct implications for measured mortality. Early OASI beneficiaries who are severely disabled have much higher mortality than healthy early retirees, yet they appear in OASI mortality statistics — not DI mortality statistics — despite having health profiles similar to DI beneficiaries. This blurs comparisons between OASI and DI mortality and may lead to underestimation of health selection into early retirement. See Select and Ultimate Mortality Tables for the actuarial treatment of selection at age 62.

For the Gender Gap in DI Receipt

The DI insured-status barrier falls disproportionately on women. As female labor force participation has risen (by 2002, $\approx 64\%$ of women aged 62 had disability insured status vs. $\approx 50\%$ in 1990), this gap has narrowed but not closed. The rising female insured rate is itself one of the demographic forces Pattison and Waldron (2013) identify as a driver of DI enrollment growth — rising insured rates expand the DI-eligible pool. See DI Growth Decomposition.

The Post-Reform Stickiness Gap (Deshpande, Fadlon, and Gray 2021)

The 62–64 structural gap documented by Leonesio and Wixon (2004) is driven by institutional ineligibility: workers who are too impaired to work but cannot access DI (insufficient insured status) or SSI (assets above means test). Deshpande, Fadlon, and Gray (2021) identify a second, behavioral version of the gap that operates at the 65–66 boundary for post-reform workers (Full Retirement Age [FRA] = 66).

After the 1983 Amendments raised the FRA from 65 to 66, claiming behavior responded one-for-one — the claiming spike migrated from 65 to 66. But retirement behavior did not follow: the spike at 65 persisted unchanged across a decade of post-reform cohorts. The most supported mechanism is employer-level norms and incentive structures that anchor retirement at the old FRA. For post-reform workers who retire at 65 per employer conventions, the result is a gap year: no labor income (having left the labor force) and no Social Security income (the new FRA is 66). This income gap can last up to a year.

Workers with health impairments who face this gap year — who have already exited the labor force at 65 and have no near-term earnings capacity — are prospective DI applicants. The pathway from stickiness to DI application is compressed: they can no longer claim early retirement benefits without actuarial reduction at the pre-reform timing, they cannot claim Social Security at the customary age 65, and they cannot work due to impairment. The income gap makes DI application attractive (or necessary) for this subgroup.

This 65–66 behavioral gap is structurally distinct from the 62–64 structural gap:

62–64 gap: Driven by DI insured-status rules and SSI asset test; affects workers who have never had sufficient insured quarters (disproportionately women) or who have modest savings.
65–66 behavioral gap: Driven by employer retirement norms anchoring exit at the old FRA; affects post-reform workers regardless of insured status who retire at 65 but cannot claim SS for a year.

Both gaps produce a population relying on informal substitutes (savings, spousal income, DI) rather than the officially intended pathway. See Retirement Age Stickiness and Conditional DI Applicants.

Open Questions

How has the gap evolved post-Affordable Care Act (ACA)? The ACA's Medicaid expansion and marketplace subsidies changed the insurance dimension substantially for the 62–64 age group. The uninsured rates documented in Leonesio and Wixon (2004) — $\approx 25\%$ among the most health-impaired early retirees — may no longer hold.
What share of the gap population eventually applies for DI or SSI after early retirement benefit onset? If early retirement triggers a period of income stability that allows them to accumulate medical documentation, some may qualify on reapplication.
How does the gap interact with state variation in Medicaid and SSI supplementation? State-level generosity could substantially affect the effective safety net for this population.