Summary
A 48-page policy report from the independent Social Security Advisory Board (SSAB) examining the growing role of professional third-party representation at the initial disability claims stage — a level that decides ~70% of all claims finally. Drawing on Social Security Administration (SSA) administrative data (fiscal years [FY] 2007–2011), stakeholder interviews, and field visits, the report documents that attorney representation adds ~25 days to processing times while represented Supplemental Security Income (SSI) claims have allowance rates nearly double the unrepresented rate (60–64% vs. 32–34%). The main finding is that SSA data systems are too fragmented to permit rigorous analysis, and the Board's primary recommendation is that SSA build the infrastructure to answer basic questions about representation.
Key Claims
Scale of the Process
- In FY2011, SSA made 3.4 million initial disability medical decisions (up from 2.5M five years prior; average growth 6.3%/yr — well above population or morbidity growth rates).
- ~70% of all claims are finally decided at the initial level: 1.8M of 2.57M initial determinations in 2008 never reached a hearing, either because they were allowed or because denied claimants did not appeal.
- ~800,000 hearing decisions were made annually.
- Administrative spending on disability claims: $3.6 billion in FY2011 (up from $2.9B in FY2007); unit cost per initial claim ~$1,058.
- SSA employed ~82,000 staff (down from ~90,000 in 2010 due to hiring freeze); operational budget cut by $938M from FY2010–2012.
Representation Trends
- Representation at the initial level is growing: Social Security Disability Insurance (SSDI) represented claims rose from 9.8% (2007) → 14.3% (2011); SSI from 4.1% → 4.8%.
- Still, 86% of SSDI and 95% of SSI claimants are unrepresented at the initial level (vs. hearing level where representation is much more common).
- Among represented SSDI cases, attorneys account for 9 percentage points and non-attorneys 5 percentage points (of the 14% total). Among represented SSI cases, attorneys <1% and non-attorneys ~3%.
- Three types of professional representatives: attorneys, non-attorney for-profit representatives, and non-profit organizations (legal aid, social service agencies, hospital eligibility specialists).
Processing Times: Representation Slows Claims
- Unrepresented SSDI initial claims: ~100–109 days processing time (FY2009–2011).
- Attorney-represented SSDI: ~128–134 days (+25 days; ~23% longer).
- Non-attorney represented: ~117–125 days (intermediate).
- Same pattern holds for SSI.
- Allegation from agency employees: for-profit attorneys intentionally delay to maximize claimant back pay → larger 25%-of-backpay fee. Allegation unverifiable with available data.
Allowance Rates: Large Difference for SSI, Small for SSDI
- Overall SSDI initial allowance rate: 36–37%; represented SSDI: 37–41% — barely different.
- Overall SSI initial allowance rate: 32–34%; represented SSI: 60–64% — nearly double.
- Caveat: selection bias uncontrolled. Possible explanations: (1) representatives select stronger cases; (2) non-profit SSI specialists serve populations more likely to qualify; (3) more fully developed applications. No data to distinguish.
Representative Fee Payments
- SSA-administered fees to all representatives (all adjudicative levels): $1.17B (2007) → $1.74B (2010), a 48% increase.
- SSDI fee agreements: $1.39B in 2010 (+45% vs. 2007). SSI fee agreements: $280M in 2010 (+75% vs. 2007).
- Fee cap: lesser of 25% of past-due benefits or $6,000 (as of June 2009).
- SSA does not separately track fees by level of adjudication, making the initial-level share unquantifiable.
Data Infrastructure Failures
- SSA cannot track cases by type of representation — the most basic fact needed to assess the impact of representation.
- Data are spread across multiple incompatible databases built for production, not analysis.
- SSA could not provide: processing time broken down by represented vs. unrepresented at field office vs. Disability Determination Services (DDS) stages; allowance rates by impairment type and representation type; total fees at the initial level alone.
Key Recommendations
- Data collection: Build comprehensive tracking of representation type, claim type, allowance rates, and processing stages.
- Technology: Seamless online application for SSI (only SSDI could fully apply online as of 2012); allow representative software to interface with SSA systems (like IRS/TurboTax model).
- Policy: Standardize filing periods; require representatives to certify application completeness before field office processes it; review whether SSA should administer representative fees.
- Communications: Formal liaison between field offices and representative community; expand electronic folder access to initial claims level.
- Allegations investigation: Investigate whether deliberate delay, incomplete applications, and restricted agency-claimant contact are systemic.
Concepts Introduced or Extended
Entities Mentioned
Quotes
"Approximately 70 percent of all claims are decided at the initial level."
"Processing times are longest for represented claims — particularly when represented by attorneys. The difference in processing times between attorney representation and no third party representation is on average around 25 days (approximately a 23 percent difference)."
"Represented cases have higher allowance rates, significantly so for SSI, but barely so for SSDI cases."
My Take
The most useful data in this report for the Disability Insurance (DI) research literature are the allowance rate and processing time figures. The near-doubling of initial SSI allowance rates for represented claimants (60-64% vs. 32-34%) is striking enough to warrant serious investigation of selection mechanisms — it connects directly to Deshpande and Li (2019) on application costs and targeting, and to the broader literature on who gains access to disability benefits. The 25-day processing delay for attorney-represented claims is directly relevant to the "Does Delay Cause Decay" literature (Autor et al. 2015). The data limitations documented here also explain why many SSA administrative data papers in the wiki work around the absence of representation identifiers.