The real brokered-deposit surge happened in 2023 — a year before the rule BPI is fighting existed

Direct answer

BPI has argued that the FDIC's since-withdrawn 2024 brokered-deposit reclassification proposal would have caused a ~110% increase in deposits reclassified as "brokered," based on a sample of its member banks. [C01] OptimaYield's own population-scale data shows real brokered-deposit reliance already rose far more than that at every bank-size cohort — 454% to 458% at the two mid-size cohorts — but that rise happened by Q3 2023, roughly a year before the proposal existed, and lines up with the March 2023 regional-bank funding-stress episode, not with anticipation of any specific rule. [C06, C07] This article cannot confirm or refute BPI's specific figure, which describes a hypothetical rule's modeled impact rather than a historical fact — but it can show what actually happened, and when.

Key findings

The real cycle, cohort by cohort

Every cohort's reliance on brokered deposits collapsed during the 2020-2022 period of extraordinarily cheap, abundant organic deposit inflow, then rose sharply as banks competed harder for stable funding once rates rose and the March 2023 bank-failure episode made funding stability a live supervisory concern. [C06] By the time the FDIC's brokered-deposit reclassification proposal existed in 2024, this rise had already happened and, at the $10-50B and $50-250B cohorts, had already begun to ease. [C07]

Every cohort's relative increase in brokered-deposit reliance since its 2021-2022 low exceeds BPI's own cited 110% figure — but the increase that actually happened predates the specific rule BPI's number is about by roughly a year.

Did the regulatory news cycle move the data at all?

No visible inflection appears at either the 2024 proposal's introduction or its March 2025 withdrawal. [C08] Reliance was declining through most of 2024 at the $10-50B and $50-250B cohorts — the opposite of what an anticipatory-reclassification response to a pending rule would look like — and continued on essentially the same gradual path through the March 2025 withdrawal with no discontinuity. Whatever is driving today's brokered-deposit levels, it is not visible evidence of banks reacting to this specific regulatory news cycle in either direction.

Named-institution trajectories: three different stories behind rising reliance

• Wells Fargo Bank — fell to a 0.49% trough in Q2 2022, then rose roughly twenty-fold to 9.75% by Q2 2026, just below its own 2016-2026 peak of 9.87% (Q1 2019). [C09] A large bank's reliance returning to, but not yet exceeding, a level it has visited before.
• Goldman Sachs Bank USA — fell steadily from 57.7% (Q1 2016) to a low of 17.5% (Q3 2023) as the bank built its Marcus retail-deposit franchise, then partially reversed to 21.9% by Q2 2026. [C10] A multi-year strategic shift toward retail funding, not yet fully sustained.
• Morgan Stanley Bank — rose from near zero (2016) to 33% (2020), a climb coincident with Morgan Stanley's 2020 acquisition of E*Trade and its brokerage sweep-deposit programs, then fell sharply to 5.8% (Q4 2021) — likely a sweep-deposit reallocation, not organic drift — before climbing again to 23.3% by Q2 2026. [C11] A named corporate event explains the discontinuity; the more recent climb is not yet explained by one.

The highest-ratio names are a business-model story, not a risk story

Reliance is far from universal: 24 of 158 institutions (15.2%) report zero brokered deposits, while 31 (19.6%) exceed 10% and 11 (7.0%) exceed 25%. [C03] BMW Bank of North America (84.8%), Toyota Financial Savings Bank (78.9%), and Sallie Mae Bank (40.6%) top the ranking by ratio. [C04] All three are captive or specialty finance banks — auto lending and student lending — without a retail branch network to gather deposits the way a community or regional bank does. High brokered-deposit reliance is structural to this business model, not a signal of funding stress, and should not be read the same way a sudden increase at a full-service retail bank would be.

Why this matters

This is a live regulatory and advocacy fight: BPI's own comment letters use the ~110% figure to argue the FDIC's now-withdrawn proposal would have imposed material reclassification costs, while the Road to Housing Act's §902 has separately expanded the reciprocal-deposit safe harbor Congress judged worth protecting (see PUBLICATION-TOPIC-BACKLOG.md). Whatever the merits of BPI's modeled counterfactual, the population-scale, already-realized picture is that brokered-deposit reliance rose sharply and for a different, dateable reason — the 2023 funding-stress episode — and any current policy debate about reclassification is playing out against a landscape that had already moved before the specific rule did.

Methodology

Reuses OptimaYield's existing, already-validated brokered_deposit_ratio metric contract (RC-E domestic RCON2365 ÷ RC RCON2200) via the shared deriveValidatedMetrics function, unmodified — called directly from the first version of this analysis, not added after a self-correction. Current-quarter population: every institution in the detailed ($10B+) Call Report layer with a derivable ratio for Q2 2026 (158 institutions). Cohort history reuses the already-computed percentile series in semantic.peer_statistics (42 quarters, 2016-03-31 through 2026-06-30). Named-institution trajectories query each bank's full available history directly. Live queries: scripts/analysis-brokered-deposit-population.ts and scripts/analysis-brokered-deposit-history.ts.

Limitations and counter-evidence

• This does not test BPI's specific claim. [C12] BPI's ~110% figure describes a modeled impact of a rule that never took effect, computed on BPI's own member sample — not a population-scale historical fact. The trough-to-peak increases found here describe something real that already happened, for a different, identifiable reason, not a confirmation or refutation of BPI's counterfactual.
• A high or rising ratio is not, by itself, a risk finding. The highest-ratio institutions are predominantly captive/specialty finance banks for which brokered funding is structural to the business model, not a distress signal.
• No cause is claimed for any bank's specific trajectory beyond the two named corporate events (Goldman's Marcus buildup, Morgan Stanley's E*Trade acquisition) that have independent, public confirmation; other movements are described, not explained.
• Not a frozen snapshot. Figures were pulled live from the production database rather than a versioned, hashed catalog artifact.

Glossary

• Schedule RC-E — the Call Report schedule covering deposit composition, including the brokered-deposits line.
• RCON2365 — the FFIEC field code for domestic brokered deposits.
• Brokered deposit — a deposit obtained through a third party (a deposit broker) rather than directly from the depositor, typically more rate-sensitive and less "sticky" than organic retail deposits.
• Reciprocal deposit — a specific category of brokered deposit exchanged between banks through a network so each depositor's funds stay within FDIC insurance limits at any one institution; subject to a separate statutory safe harbor.

Related on OptimaYield

• Q2 2026 uninsured-deposit concentration — this quarter's companion piece on deposit funding structure.
• Q2 2026 CRE concentration screen — this quarter's companion piece on asset-side concentration.
• The bank-profile pages for the named institutions above (not yet linked pending publication approval).

Official sources

• Bank Policy Institute, "The FDIC's Proposed Brokered Deposit Reclassification: An Empirical Evaluation."
• Troutman Pepper Locke, "Road to Housing Act: Brokered Deposit Reforms Offer Long-Awaited Clarity for Community Banks," July 2026.
• FFIEC, Consolidated Reports of Condition and Income instructions, Schedule RC-E.