Call Reports: the public numbers behind a bank’s balance sheet

A familiar bank, an unfamiliar set of numbers

A hypothetical bank reports $60 million of net interest income in June, after $28 million in March. At first glance, business appears to have more than doubled. But the June figure covers six months: the second quarter contributed $32 million. That small distinction is central to understanding a Call Report. The same document mixes balances measured on one date with income accumulated over a period. [11]

Call Reports are the quarterly financial filings of U.S. banks and savings associations. Their formal name is the Consolidated Reports of Condition and Income. They give regulators and the public a common framework for understanding a bank’s assets, funding, earnings and risks, including institutions without publicly traded shares. The reporting scope and level of detail depend on the institution and its activities. [1]

The bank and its parent are different reporting subjects

A bank’s Call Report does not automatically describe everything owned by the bank’s holding company. The Federal Reserve’s FR Y-9C covers the consolidated holding-company organization, potentially including the parent, banks and nonbank businesses. It generally applies at $3 billion or more in consolidated assets, with exceptions. Parent debt or a sister business therefore cannot simply be treated as a liability or operation of the insured bank. [2]

For a publicly reporting company, the annual Form 10-K and quarterly Form 10-Q provide another view; the 10-K includes audited financial statements and a broader account of the business. These filings complement bank regulatory reports. A comparison only makes sense when the legal entity and consolidation boundary match. [13]

Why there are three main forms

The FFIEC 031 serves banks with foreign offices or certain related international operations. Despite its title, it also applies to domestic-only banks with at least $100 billion in consolidated assets and to advanced-approaches institutions under the capital rules. The bank’s own consolidated subsidiaries are within the report’s scope. Form number is therefore more than a domestic-versus-international label. [3]

The FFIEC 041 is the domestic-office form for banks outside the 031 requirements, unless they use the eligible small-bank alternative. The FFIEC 051 is that streamlined alternative for eligible domestic-only banks below $5 billion. Exceptions include specified capital and deposit-insurance-assessment categories, and a bank may elect or be required to use 041. A smaller form does not mean the bank has received a better supervisory assessment. [4][5]

The schedules connect the balance sheet to the business

Schedule RC is the balance sheet: assets on one side of the accounting equation, liabilities and equity on the other. Deposits are liabilities because the bank owes that money to customers. Schedule RI is the income statement, showing income and expenses. Supporting schedules explain what lies inside those totals. A profitable period and a strong end-of-period balance sheet describe related but different things. [6]

The recurring schedule letters form a compact map of the filing. [3]

Scroll horizontally to see all columns.

Loans and deposits reveal the business model

RC-C separates real-estate loans, commercial and industrial lending, consumer credit and other categories. Residential property, construction, commercial property, cards and auto lending have distinct lines. Its balances are before deducting the loan allowance; it includes held-for-sale loans but excludes trading assets. A category total is not necessarily identical to net loans on RC. [4]

RC-E separates transaction and nontransaction deposits and identifies components such as brokered deposits. RC-O contains an uninsured-deposit estimate for required filers. These are different classifications: a funding source, an account type and insurance coverage answer different questions. A large uninsured balance does not by itself predict how quickly customers will withdraw. [4]

Consider two imaginary banks with the same $3 billion loan total. One could hold mostly residential mortgages; the other could concentrate on construction and credit cards. Their identical headline size would conceal different borrowers and repayment patterns. Likewise, unchanged total deposits could mask a shift toward more expensive funding. The schedules expose the composition behind a headline without supplying a complete explanation of customer behavior.

Three clocks: quarter-end, year-to-date and annualized

Balance-sheet figures normally refer to the reporting date. Income and expense figures generally accumulate from January 1, even though reports arrive quarterly. June therefore usually means six months of income; September means nine. An isolated quarter comes from subtracting the previous quarter’s year-to-date amount within the same calendar year. December less September gives the fourth quarter. Subtracting last December from this March would mix two different annual reporting cycles. [11]

A rate needs a matching denominator. The UBPR methodology relates net interest income to average earning assets and net charge-offs to average loans. Quarterly averages commonly come from RC-K. An annualized quarter multiplies that quarter’s flow by four; a first-half year-to-date flow is multiplied by two. Annualization expresses a pace, not a forecast of the next twelve months. [7]

RC-K permits averages of daily balances or weekly Wednesday balances. Neither is necessarily the midpoint of the opening and closing balance. A late-quarter acquisition or deposit inflow makes that distinction particularly visible. [5]

Returning to the hypothetical bank, second-quarter net interest income is $60 million minus $28 million, or $32 million. With $4 billion of average interest-earning assets during that quarter, a simple annualized net interest margin is $32 million × 4 ÷ $4 billion = 3.20%. Dividing the six-month income by the quarter’s assets and multiplying by four would incorrectly produce 6.00%. These examples use four-times-quarterly annualization and no tax-equivalent adjustment. [7]

The allowance, provision and charge-off are different events

The allowance for credit losses is an accounting estimate of amounts not expected to be collected, rather than a separate pot of cash. Under CECL, estimates for covered loans incorporate historical experience, current conditions and reasonable, supportable forecasts over the remaining contractual term. The provision is the expense that changes the allowance. A charge-off removes an amount deemed uncollectible; recoveries offset earlier charge-offs. [8]

In a simplified hypothetical quarter, a $36 million starting allowance plus a $6 million provision, less $4 million of charge-offs, plus $1.1 million of recoveries produces a $39.1 million ending allowance. This assumes no acquisition, accounting-transition or other adjustments. The provision is $6 million, net charge-offs are $2.9 million, and the allowance grew $3.1 million: three different answers to three different questions. Here, charge-offs use the existing allowance; subtracting both provision and charge-offs from income would count the loss twice. [8]

With $2.9 billion of average loans for that same hypothetical quarter, the annualized net charge-off rate is $2.9 million × 4 ÷ $2.9 billion = 0.40%. The denominator is average loans, not deposits, ending total assets or the allowance. [7]

Problem loans are a stock; charge-offs are a flow

RC-N separates loans that are 30–89 days past due and still accruing interest, loans at least 90 days past due and still accruing, and nonaccrual loans. Nonaccrual describes the accounting treatment of interest, not simply a late-payment bucket. The columns are distinct, so adding all past-due balances to nonaccrual without checking the definitions can produce a misleading measure. [3]

A common noncurrent-loan measure combines 90-days-or-more past-due loans still accruing with nonaccrual loans. In a hypothetical portfolio, $10 million in the first group and $45 million in the second gives $55 million, or 1.83% of $3 billion of quarter-end gross loans. Another $20 million that is 30–89 days late and still accruing is outside that definition. This is a point-in-time ratio and is not annualized. [9]

The label “nonperforming loans” warrants its own definition whenever a company or data provider uses it. Problem balances and realized losses can move differently: a loan may be charged off, sold, repaid or restored to performing status. A lower ending problem-loan balance is therefore not, by itself, evidence that borrowers became healthier. That is an interpretation requiring the movements behind the total. [9]

Capital ratios answer another set of questions

Regulatory capital ratios in RC-R should not be confused with an ordinary equity-to-assets percentage. Common equity tier 1, or CET1, compares a defined capital numerator with risk-weighted assets. The standard tier 1 leverage ratio instead uses tier 1 capital over average consolidated assets after specified deductions. Different numerators and denominators can make both percentages correct while giving different impressions of the same institution. [10]

For illustration, $400 million of CET1 against $3 billion of risk-weighted assets gives 13.33%. If tier 1 capital is also $400 million and the adjusted leverage denominator is $5 billion, the leverage ratio is 8.00%. Neither number alone establishes every applicable capital requirement or the bank’s overall condition. Banks electing the community bank leverage-ratio framework have different reporting requirements, so an absent CET1 field need not mean zero capital. [10][1]

A public record with revisions and blind spots

A blank field is not automatically zero: it may reflect reporting applicability or confidentiality. Banks amend filings, and UBPR calculations can also change. Comparing two downloads without their reporting dates, definitions and retrieval dates can mistake a revision for a new business development. [11]

The public Uniform Bank Performance Report, or UBPR, turns Call Report data into ratios and peer comparisons. It is ordinarily available shortly after the underlying filing, while peer figures arrive later. Its ongoing recalculation means the first available number is not necessarily the last. A bank’s public reporting history is a revisable dataset, not a permanently frozen scorecard. [12]

Reporting thresholds, mergers, loan sales, accounting changes and differences in business mix can interrupt comparisons. Standardized fields help, but the same ratio at a card specialist and a mortgage-focused bank does not necessarily describe the same economics. The form and instructions applicable to each reporting date define what was actually collected; the September 2026 forms underpin this guide’s schedule descriptions. [5]

Most importantly, public financial data are separate from confidential examination findings and supervisory ratings. A reader cannot reconstruct an official CAMELS rating from a handful of public ratios. Call Reports can reveal where earnings, funding or credit quality changed; they cannot substitute for every fact available to supervisors or tell the complete story of why management made its choices. [14]

Sources

1. Federal Reserve: Call Report coverage and reporting scopeOfficial sourceBack to text: ↑1↑2
2. Federal Reserve: FR Y-9C reporting scope, updated September 30, 2026Official sourceBack to text: ↑
3. FFIEC 031: September 2026 form, cover and Schedule RC-NOfficial source · PDFBack to text: ↑1↑2↑3
4. FFIEC 041: September 2026 form, schedules RC-C, RC-E and RC-OOfficial source · PDFBack to text: ↑1↑2↑3
5. FFIEC 051: September 2026 form, eligibility and reporting detailOfficial source · PDFBack to text: ↑1↑2↑3
6. FFIEC: current FFIEC 041 overview and instructionsOfficial sourceBack to text: ↑
7. FFIEC: UBPR Executive Summary methodology, March 3, 2025Official source · PDFBack to text: ↑1↑2↑3
8. Federal Reserve: interagency CECL questions and answersOfficial sourceBack to text: ↑1↑2
9. FFIEC: UBPR past-due and nonaccrual methodology, March 3, 2025Official source · PDFBack to text: ↑1↑2
10. Federal Reserve: Regulation Q, section 217.10, capital-ratio definitionsOfficial sourceBack to text: ↑1↑2
11. FFIEC: public-data and UBPR frequently asked questionsOfficial sourceBack to text: ↑1↑2↑3
12. FFIEC: UBPR availability and recalculationOfficial sourceBack to text: ↑
13. SEC Investor.gov: Form 10-K and related reportsFiling / reportBack to text: ↑
14. Federal Reserve and FDIC: confidential ratings and examination information, October 18, 2019Official releaseBack to text: ↑