This allowance estimate was prepared under ASC Topic 326, Subtopic 326-20 — Financial Instruments — Credit Losses (Measured at Amortized Cost) (ASU 2016-13). The standard requires estimates of lifetime expected credit losses based on historical experience, current conditions, and reasonable and supportable forecasts (ASC 326-20-30-7 through 30-9). The Loss Rate Method was applied, as contemplated by ASC 326-20-55-15 through 55-19.

The peer group was defined as FDIC-insured commercial banks with total assets between —M and —M (50–200% of the subject institution's assets), further filtered to institutions whose loan portfolio composition — measured as real estate, commercial & industrial, consumer, and agricultural loan concentrations as percentages of total loans — falls within ±— percentage points of the subject's own composition on each component. This filter ensures that peer loss experience reflects institutions with genuinely comparable credit risk profiles, not merely comparable scale. — institutions met both criteria. The subject institution's loan mix was: —.

Annual net charge-off ratios (NTLNLSR — net charge-offs as a percentage of average loans, from FFIEC Call Report Schedule RI-B, Part II) were collected for each peer over a —-year look-back period. Each peer bank's rate was averaged across available years before ranking. The applied loss rate at the —th percentile of this peer distribution (—%) represents management's assessment of the historical loss scenario corresponding to the selected loss environment. Source data: FDIC BankFind Suite (public domain).

ASC 326-20-30-8 requires the ACL estimate to reflect reasonable and supportable forecasts about the future. This estimate incorporates a forward-looking overlay computed from three independent signals, each sourced from publicly available Federal Reserve data (FRED) with no API key required. The overlay is additive to each segment's peer NCO base rate. Each signal is floored at zero — benign conditions do not reduce reserves below the peer-derived base.

• Signal 1 — FOMC Unemployment Projection (UNRATEMD): The FOMC Summary of Economic Projections (SEP) median unemployment forecast for the target year is compared to the current BLS unemployment rate (UNRATE). Each 1 percentage point of projected deterioration adds +12 basis points. Fires only when the FOMC projects rising unemployment.
• Signal 2 — Investment Grade Credit Spread (BAMLC0A0CM): The ICE BofA US Corporate Bond OAS (Option-Adjusted Spread) is compared to its full historical mean (computed from all available daily observations). Each 0.1% that the current spread exceeds the long-run mean adds +5 basis points. Spread widening leads bank NCO increases by approximately 1–3 quarters. Fires only when spreads are elevated above average.
• Signal 3 — Senior Loan Officer Survey C&I Tightening (DRTSCILM): The Federal Reserve's quarterly Senior Loan Officer Opinion Survey net percentage of banks reporting tighter commercial & industrial lending standards. Each 10% of net tightening adds +4 basis points. Tighter lending standards historically lead NCO increases by 2–4 quarters. Fires only when standards are tightening (positive net).

The forecast horizon is — quarters; for horizons of 6 or 8 quarters, FOMC projections for the following calendar year are used. The resulting R&S adjustment was —, producing segment annual loss rates of (peer NCO rate + —%). For periods beyond the supportable forecast horizon, the peer group's long-run historical distribution serves as the reversion basis, consistent with ASC 326-20-30-9.

Management must document the rationale for the selected forecast horizon at each measurement date and assess whether these three macro signals are the most relevant indicators for the institution's specific loan portfolio, markets, and borrower concentrations.

Loans were disaggregated into nine segments per FFIEC Call Report Schedule RC-C, Part I: Construction & Land Development (LNRECONS), CRE Nonfarm Nonresidential (LNRENRES), 1-4 Family Residential (LNRERES), Multifamily Residential (LNREMULT), Agricultural Real Estate / Farmland (LNREAG), Other Real Estate (LNREOTH), Commercial & Industrial (LNCI), Consumer (LNCON), and Agricultural Production (LNAG). Any balance not allocated to a named segment is reserved in an "Other Loans" category. Multipliers and WARM factors are disclosed in the segment detail table.

Each segment's annual loss rate equals the R&S-adjusted applied rate multiplied by an industry-average relative loss rate multiplier. Annual rate × WARM produces the lifetime loss rate per ASC 326-20-30-2. Reserve equals balance × lifetime rate.

Segment-specific peer NCO rates: Each segment's base rate is drawn from its own empirical peer distribution built from FDIC Call Report Schedule RI-B, Part II net charge-off fields (NTRECONS, NTRERES, NTREMULT, NTRENRES, NTREAG, NTCI, NTCON, NTAG). Segment NCO rate = NT_segment ÷ LN_segment (period-end balance). Only peer-years where the segment balance exceeds $1M and 0.5% of total loans are included. A minimum of 10 contributing peers is required; where this threshold is not met (displayed as ⚠ fallback), the portfolio rate × an industry-average multiplier is substituted and disclosed in the segment table.

Rate denominator note: NTLNLSR (total portfolio) and all segment NT fields are measured against average loans during the year. Reserves are applied to period-end balances (LNLSGR at December 31). For growing portfolios this produces a modestly conservative reserve. This is standard practice under the Loss Rate Method.

HELOC note: The 1-4 Family Residential segment (LNRERES) includes home equity lines of credit. LNREHEL is not separately available in the BankFind public dataset; HELOCs cannot be segmented. A WARM of 6.0 years is applied to the combined balance. Institutions with high HELOC concentrations should consider whether a shorter effective WARM is warranted and document that assessment.

An estimate of expected credit losses on unfunded loan commitments was prepared per ASC 326-20-55-56. The reserve is organized by commitment type (CRE construction, CRE nonfarm nonresidential, C&I revolving lines, agricultural operating lines, HELOC & consumer revolving, and other). For each category the expected funded amount equals the commitment balance multiplied by a credit conversion factor (CCF) representing the expected percentage to be drawn before or at default; that expected funded balance is then multiplied by the applicable segment annual loss rate from Section V (peer NCO rate plus R&S adjustment) using a one-year commitment WARM.

CCF defaults by type reflect interagency guidance and Basel III SA-CCR reference ranges: CRE construction 75%, CRE nonfarm nonresidential 50%, C&I revolving 40%, agricultural lines 55%, consumer revolving / HELOC 30%, other 50%. Management should substitute institution-specific historical utilization data when available.

Data source: RC-L commitment balances are retrieved from the FFIEC Central Data Repository (CDR) public SDF filing for the institution's most recent reporting period. CDR data covers RCONF164, RCONF165, RCON3814, RCON3815, RCONJ457, and RCONJ459 — the full set of RC-L unused commitment line items reported on the FFIEC 041/051 Call Report form. Where CDR data is unavailable, balances must be entered manually from the institution's filed RC-L schedule. The unfunded commitment reserve is recorded in Other Liabilities, not the ALLL, consistent with interagency guidance (SR 23-8 / CA 23-5).

Credits individually evaluated per ASC 326-20-35-1 are excluded from the collective assessment pools. Expected credit losses on individually evaluated credits are measured on a collateral-dependent basis (fair value of collateral less estimated selling costs), by discounting expected cash flows at the loan's original effective interest rate, or by reference to observable market prices, as appropriate. Total specific reserves entered: —. Supporting documentation for each specific reserve must be maintained contemporaneously in the credit file.

Qualitative factor adjustments were applied at the loan segment level per the five-factor scorecard completed by management. Each factor is scored 1–5 (1 = low risk, 3 = neutral, 5 = high risk). The average score across factors for each segment converts to a basis-point adjustment (−10 / −5 / 0 / +10 / +25 bps at scores 1–5) that is added to the segment's annual loss rate before WARM multiplication. The five factors assessed were:

• Credit quality trends — changes in past-due rates, nonaccrual loans, and classified/criticized volume relative to the prior period and peer institutions.
• Underwriting standards & policy — recent changes to LTV limits, debt service coverage requirements, guarantor requirements, or exception rates.
• Portfolio concentration & growth — concentration relative to capital, recent growth rates, and geographic or product-type concentrations.
• Collateral values & market conditions — trends in collateral valuations, local real estate market conditions, and liquidity of collateral in the relevant market.
• External economic & competitive environment — local employment, commodity prices (for agricultural segments), competitive pricing pressure, and other external factors not captured in the R&S macro overlay.

Total Q-factor reserve addition: —. Supporting documentation for each non-neutral score must be maintained contemporaneously and presented to auditors and examiners upon request.

• Historical peer rates are proxies for expected future losses; actual results may differ materially.
• Peer group is defined by asset size and loan mix; other risk differences may not be captured.
• WARM factors are industry averages; actual remaining maturities may differ.
• Segment loss multipliers are industry averages; institution-specific history may differ.
• Unfunded commitments are limited to LNCOMRE; other off-balance-sheet exposures require supplemental analysis.
• This analysis is not an audit, CPA engagement, or regulatory examination. Management retains full responsibility for the ACL.