Deshpande, Gross, and Wang 2017 — The Effect of Government Cash Assistance on Household Financial Outcomes

Summary

Provides the first causal estimates of the effect of government cash assistance on household financial outcomes using the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) 1996 age-18 Supplemental Security Income (SSI) review discontinuity as an instrument, linked to near-census Public Access to Court Electronic Records (PACER) bankruptcy records. The central finding is counterintuitive: losing SSI benefits reduces household bankruptcy rates dramatically, concentrated entirely in the parents of removed youth. The proposed mechanism is a credit access channel — SSI's stable income stream makes recipient households creditworthy; losing SSI means losing credit access, which mechanically eliminates bankruptcy filings. The paper is explicitly labeled preliminary as of September 2017; credit bureau data to test the mechanism were pending.

Key Claims

• SSI removal reduces parent probability of ever filing for bankruptcy by 12 percentage points (pp) (instrumental-variable [IV] linear regression-discontinuity difference-in-differences [RD-DD]), an 80% decrease from the 14.6% control mean
• Parent number of bankruptcy filings falls by 0.26 (100% decrease from control mean of 0.22)
• No significant effect on SSI youth's own bankruptcy rates (3.0% control mean)
• Household (parent + child) ever-files: −9.2 pp (55% decline from 16.5% base)
• Effect is driven by Chapter 7 bankruptcies (low-income), not Chapter 13
• Results robust across polynomial orders (linear/quadratic/cubic/quartic) and to inclusion of covariates; RD-DD preferred over simple RD due to covariate imbalance under linear specification
• Proposed mechanism: SSI provides stable, predictable income → lenders extend credit to recipient households → losing SSI → losing credit access → mechanical collapse in bankruptcy rates (credit access hypothesis; untested as of 2017)
• Baseline sample: 81,800 SSI children with 18th birthday within ±37 weeks of August 22, 1996 cutoff; 63% male; 73% mental/intellectual diagnoses; avg parental earnings $9,600/year; 51% single-mother households; N=232,067 in RD-DD specification

Concepts Introduced or Extended

• SSI Children's Program
• Nonhealth Risk and DI Insurance Value

Entities Mentioned

• Manasi Deshpande
• Tal Gross
• Jialan Wang
• Social Security Administration

Quotes

"We find that household bankruptcy rates fall almost to zero after the household loses SSI. We hypothesize that households that lose cash assistance also lose their access to credit, and therefore bankruptcy rates fall mechanically."

"Thus, removing an 18-year-old from SSI virtually eliminates bankruptcy filings for the parents of that youth over the next 16 years."

My Take

The bankruptcy-collapse finding is striking and, if the credit access mechanism holds, important for understanding how SSI functions as de facto collateral for low-income households. However, as of September 2017 the paper is explicitly preliminary and incomplete — the credit bureau test of the mechanism was pending, and the paper cautions against citation. The quantitative estimates should be treated as provisional. A competing interpretation — that households simply stop incurring debt once income becomes uncertain — is not fully ruled out by the available data. The paper is most valuable as a complement to Deshpande, Gross, and Su (2021), which rigorously establishes the disability insurance (DI)–financial distress connection; this working paper extends those ideas to the SSI children population and the parental household unit.