Short answer: Among the 998 banks that reported an estimated uninsured deposits figure for the quarter ending June 30, 2026, the median was 32.08% of deposits, with the middle half between 23.76% and 42.10%. Read that with its coverage in mind. Those 998 banks are just under 24% of the 4,238 banks these benchmarks are computed over, and almost all of them are at or above $1 billion in assets, so this is a distribution for larger banks rather than for the industry. There is no share that is automatically too much. A bank at 45% funded by operating accounts is in a different position from a bank at 25% where a dozen depositors hold most of the balance.
What the line is, and what it is not
Estimated uninsured deposits is RCON5597 on Schedule RC-O of the call report. It is the bank's estimate of how much of its deposit base sits above the FDIC's $250,000 standard maximum deposit insurance amount, which applies per depositor, per insured bank, per ownership category. Since 2023 it has been a standing board question, and it is one of the few deposit figures an analyst or a large depositor can pull for free.
Be precise about what it does not carry. It is not a measure of deposits that will leave. Uninsured is a legal status, not a behavior, and the same status attaches to a $400,000 balance that clears payroll every other Friday and to one placed last quarter for twelve basis points. It says nothing about what the bank could raise if those balances moved. And it is not complete: most banks do not report it at all.
Only about a quarter of banks report it
This is the most important thing to know before quoting an industry figure for uninsured deposits. The line is required of banks at or above $1 billion in total assets, and smaller banks generally leave it blank. For the quarter ending June 30, 2026, 998 banks reported a usable figure out of the 4,238 banks in the benchmark population, just under 24%. The population that does report skews heavily large, so every quartile below describes the banks above the reporting threshold and not the industry.
The pattern by size makes the gap obvious. Coverage is essentially complete above $3 billion, where almost every bank in the band reports a figure. It is partial from $1 billion to $3 billion, where 566 banks report against a band holding more than 600. Below $1 billion it collapses: 11 reporting banks from $300 million to $1 billion, and fewer than eight in each band under $300 million, too few to publish a quartile.
| Bank size (total assets) | Banks reporting | Bottom quartile | Median | Top quartile |
|---|---|---|---|---|
| All banks reporting | 998 | 23.76% | 32.08% | 42.10% |
| Under $100M | Fewer than 8 | n/a | n/a | n/a |
| $100M - $300M | Fewer than 8 | n/a | n/a | n/a |
| $300M - $1B | 11 | 11.74% | 20.75% | 25.54% |
| $1B - $3B | 566 | 23.48% | 30.30% | 39.71% |
| $3B - $10B | 263 | 24.33% | 32.58% | 44.05% |
| $10B - $100B | 126 | 28.89% | 38.56% | 47.09% |
| Over $100B | 32 | 33.65% | 44.87% | 52.05% |
Source: BankingLens, computed from FFIEC call reports for the quarter ending June 30, 2026. Quartiles are calculated across the full population of filing banks in each size band, not sampled. Non-insured non-deposit trust companies are left out of every statistic: they take no deposits and make no loans, so a margin or a funding cost computed for them has no meaning.
Coverage note: estimated uninsured deposits (RCON5597) is reported by 998 of the 4,238 banks in the benchmark population for this quarter, almost all of them at or above $1 billion in total assets. These quartiles describe the banks that report the line. They are not an industry-wide distribution, and no row here should be read as a benchmark for a bank under $1 billion. Bands with fewer than eight reporting banks are not quoted.
Within the reporting population the uninsured share rises steadily with size, from a 30.30% median in the $1 billion to $3 billion band to 44.87% over $100 billion. The top quartile of the largest banks is 52.05%, so a quarter of banks over $100 billion have more than half their deposits above the limit. That is a structural consequence of banking large corporate and institutional customers, not a ranking of risk.
The $300 million to $1 billion row rests on 11 banks, enough for arithmetic and not enough for a benchmark. If your bank is under $1 billion there is no published peer quartile here, which does not stop the board from asking. Compute it from your own core system anyway. The first time you are asked should not be the week it matters.
It is an estimate, and the bank makes it
The word estimated is in the caption for a reason. Deposit insurance is determined per depositor, per ownership category, after aggregating every account that depositor holds at the bank within the same category. Core systems are organized around accounts and tax identification numbers, not around ownership categories, so the correct arithmetic is not what the system does by default.
The gap runs in both directions. A business with three operating accounts under one tax ID aggregates into a single $250,000 limit, so counting account by account overstates its coverage. A couple with individual, joint and revocable trust accounts can carry several separate limits at the same bank, and the same method understates coverage there. Fiduciary accounts with multiple beneficiaries have their own pass-through treatment. Some banks run the full aggregation; some approximate.
So the figure is comparable across banks in the rough sense, not to the basis point. Two banks with identical depositor bases can report a few points apart on method alone. It is also a quarter-end balance, and operating accounts swing with payroll and tax dates, so a one or two point move is more often timing, one large balance or a method change than a trend. Eight quarters of a bank's own number tells you more than one quarter against a peer group.
Why a high share is not automatically a problem
Take two banks that both report 40%. The first is a commercial bank whose uninsured balances are the operating accounts of local businesses: receipts, payroll funding, escrow, retainage. Those balances are above the limit because businesses run above the limit, not because the depositor was shopping. They are attached to ACH origination, remote deposit capture, a lockbox, and in many cases a loan. Moving them means changing where the payroll clears.
The second reports the same 40%, but the balances are a few dozen large accounts that arrived for rate with no operating tie. The two banks have the same number on Schedule RC-O and almost nothing else in common, and the call report does not separate them.
What it does show is the deposit mix, and the noninterest-bearing share is the closest public proxy for operating money. The median bank funds 20.98% of deposits with noninterest-bearing accounts, the middle half between 15.03% and 27.73%. A bank well above that range with a high uninsured share is more likely to be the first bank. A bank in the bottom quartile, under 15.03% noninterest-bearing, with 40% uninsured, is telling you what those balances are being paid for. It is a proxy, not proof, since noninterest-bearing balances can themselves be concentrated in a handful of names.
What matters alongside it
The uninsured share is one input. These are the others, and the first one is not on the call report at all.
- Concentration. The percentage says nothing about how many depositors it represents. A 35% uninsured share spread across 900 commercial relationships and the same 35% held by twelve are different exposures with identical disclosure. Only the bank can answer this, from its own deposit system.
- Liquidity, meaning what can be raised in a week. Loans to deposits is the crude version and still the useful one: the median bank is at 80.16%, the middle half between 65.73% and 90.94%, and the $3 billion to $10 billion band higher at an 88.31% median. A bank near the top of that range has less of its balance sheet in securities it can pledge or sell (loan-to-deposit benchmarks).
- What that liquidity is actually worth. Securities are the first line of defense, and their carrying value is not their pledge value. The median bank's AOCI is -4.95% of equity, with the bottom quartile at -13.05%. An underwater bond still pledges, but at market, and selling it realizes the loss. See how much bank capital is still underwater.
- Collateralized public funds. Public deposits above the limit count as uninsured on this line even though state law generally requires the bank to pledge collateral against them. That cuts both ways: those balances are not exposed the way the raw share implies, and the securities pledged against them cannot be pledged anywhere else. A bank with large public funds can show a high uninsured share and a smaller usable securities book than its balance sheet suggests.
- Brokered and network funding. The median bank reports no brokered deposits, but the top quartile is at 4.32% and the share climbs with size: a 1.22% median from $1 billion to $3 billion, 3.69% from $10 billion to $100 billion, 4.75% above $100 billion. A bank that has already replaced core deposits with brokered funding has spent part of its contingency capacity. Reciprocal networks cut the other way, lowering the reported uninsured figure by spreading balances across banks. If the number dropped sharply without the depositor base changing, that is usually where it went.
- Contingent funding that has been tested. Secured borrowing capacity is capacity only if the collateral is already at the Federal Home Loan Bank or the discount window and the mechanics have been exercised. An unused line and an untested line look the same on paper.
Read together, they give the shape of the funding. Here is where the median bank sits on each.
| Measure, all banks | Banks reporting | Bottom quartile | Median | Top quartile |
|---|---|---|---|---|
| Estimated uninsured share of deposits | 998 | 23.76% | 32.08% | 42.10% |
| Noninterest-bearing share of deposits | 4,238 | 15.03% | 20.98% | 27.73% |
| Brokered share of deposits | 4,238 | 0.00% | 0.00% | 4.32% |
| Loans to deposits | 4,238 | 65.73% | 80.16% | 90.94% |
| AOCI as a share of equity | 4,237 | -13.05% | -4.95% | -1.05% |
Source: BankingLens, computed from FFIEC call reports for the quarter ending June 30, 2026. Quartiles are calculated across the full population of filing banks in each size band, not sampled. Non-insured non-deposit trust companies are left out of every statistic: they take no deposits and make no loans, so a margin or a funding cost computed for them has no meaning.
The bank counts differ because not every bank reports every line: estimated uninsured deposits is filed by 998 banks, almost all at or above $1 billion in assets, while the deposit mix and loan-to-deposit figures cover 4,238 banks and AOCI covers 4,237. For AOCI, negative figures are losses, so the bottom quartile is the deeper end. The uninsured row is not comparable to the rows beneath it as an industry figure.
What a banker should be able to answer about their own number
The percentage is not the deliverable. Being able to answer these without starting a project is.
- How did we compute it? A full ownership-category aggregation, or an account-level approximation, and has the method changed in the last eight quarters?
- How much of the uninsured balance is public funds, and what is pledged against them?
- How much is already insured through a reciprocal network, and what does that cost us?
- What do the largest 10 and largest 25 depositor relationships hold as a share of total deposits, and how much of that is uninsured?
- How much of the uninsured balance sits in accounts with an operating or service tie, and how much is rate-driven with no other relationship?
- If the largest 10 relationships left over 30 days, what do we sell or pledge, at what price, and what does that do to earnings and to AOCI?
- How much unused secured borrowing capacity exists today, tested, after the collateral already pledged to public funds and other obligations?
- Which way has the number moved over eight quarters, and did the depositor base move or did the method?
A bank that can answer those has a funding profile. A bank that can only produce the ratio has a number. The 2023 failures did not turn on the percentage anyone reported the quarter before; they turned on how fast concentrated balances could move and what the bank could raise against a securities book worth less than its carrying value. That is what the line stands in for, and it stands in poorly.
Frequently asked questions
How much uninsured deposits is too much for a bank?
There is no share that is automatically too much. Among the 998 banks that reported the line for the quarter ending June 30, 2026, the median was 32.08% of deposits and the middle half ran from 23.76% to 42.10%. What matters is what sits behind it: whether the balances are operating accounts tied to cash management or a small number of large rate-driven balances, how concentrated they are, and what the bank can raise in a week.
What is the estimated uninsured deposits line on a call report?
It is RCON5597 on Schedule RC-O, the bank's own estimate of the portion of its deposits that exceeds the FDIC's $250,000 standard maximum deposit insurance amount, which applies per depositor, per insured bank, per ownership category. The caption says estimated because the instructions ask for the bank's best estimate, not an audited figure, and it is a quarter-end balance rather than an average.
Do all banks report estimated uninsured deposits?
No. For the quarter ending June 30, 2026, 998 of the 4,238 banks in the benchmark population reported it, just under 24%. The line is required of banks at or above $1 billion in total assets, so coverage is essentially complete above $3 billion, partial in the $1 billion to $3 billion band, and thin below it, with fewer than eight reporting banks in each band under $300 million. Any quartile for this metric describes larger banks, not the industry.
Is a high uninsured deposit share dangerous by itself?
Not by itself. A commercial bank whose uninsured balances are payroll, receipts and escrow attached to cash management behaves very differently from a bank whose uninsured balances are a few dozen large accounts that came in for rate. The call report does not separate them, so the share has to be read with deposit mix, loans to deposits, securities values and depositor concentration.
How are uninsured deposits different from brokered deposits?
They measure different things. Uninsured is a legal status: the balance sits above the insurance limit, whoever placed it. Brokered describes how it was sourced. A deposit can be one, both or neither. The median bank carries no brokered funding, though the top quartile is at 4.32% of deposits and the median bank over $100 billion is at 4.75%.
Do collateralized public deposits count as uninsured?
Amounts above the insurance limit are uninsured for this line even when state law requires the bank to pledge collateral against them. That cuts two ways. Those balances are not exposed in the way the raw percentage implies, and the securities pledged against them cannot be pledged anywhere else, so a bank with large public funds can show a high uninsured share and a smaller usable securities book than it appears to have.
Where these numbers come from
Every figure on this page is computed from FFIEC call reports for the quarter ending June 30, 2026, across the 4,296 banks that filed for the quarter, less the 58 non-insured non-deposit trust companies that take no deposits and make no loans, which leaves the 4,238 banks these benchmarks are computed over. Nothing is modeled or sampled, and where a line is reported by only part of the industry we say so rather than extrapolating. You can see deposit mix, funding and securities figures for a real bank, ranked against its FFIEC peer group, on the sample scorecard, or open your own bank in the dashboard. Plans that cover your state or every bank start at $29 a month (pricing).