CIVB — Q2 2026 Financial Report Analysis

Civista Bancshares, the Sandusky, Ohio community bank (about $4.3 billion in assets, 44 locations across Ohio, southeastern Indiana and northern Kentucky), earned $14.3 million in Q2 2026, up 30.0% from $11.0 million a year earlier. Most of the gain came from cheaper funding: the bank used last summer's share sale and the deposits it gained in the November 2025 Farmers Savings Bank (FSB) acquisition to pay down expensive Federal Home Loan Bank (FHLB) borrowings. Diluted earnings per share still fell, to $0.69 from $0.71, because those two deals left about 34% more shares sharing the profit.

At a glance

• Net interest margin of 3.89%, up 0.25 points. Net interest margin is what a bank earns on its loans and securities minus what it pays for deposits and borrowings, as a share of its earning assets. Here it rose because funding costs fell (cost of funds down 0.37 points to 1.94%), not because loans paid more; loan yields actually slipped to 6.17% from 6.39%.
• Efficiency ratio of 58.2%, down from 64.5%. This is the share of revenue eaten by operating costs, so lower is better. Revenue grew 15.0% while non-interest expense grew only 4.3%.
• EPS down 2.8% despite profit up 30.0%. The weighted average share count went from 15.4 million to 20.7 million after the July 2025 offering (3.79 million shares at $21.25) and the 1.43 million shares issued for FSB.

Q2 2026 results

Source: Civista's Q2 2026 Form 10-Q and its July 23, 2026 earnings release. Revenue is our sum of the two reported income lines. The 10-Q gives the common equity tier 1 (CET1) ratio, the core capital measure regulators watch, for June 30, 2026 and December 31, 2025 only, not for June 2025.

Where the extra profit came from: the borrowing bill

The clearest single driver is in the average balance table. A year ago Civista was funding itself with an average $412.5 million of short-term FHLB advances costing 4.48%. In Q2 2026 that was down to $107.8 million at 3.12%, and the interest on those advances fell from $4.6 million to $0.8 million. That $3.8 million saving is about the same size as the entire $3.8 million rise in net interest income. Total interest expense fell $3.5 million, while total interest income rose only $0.3 million.

Two transactions paid for that. The July 2025 share offering raised about $75.7 million net, which the 10-Q says was "initially used to pay-down short-term FHLB advances". The FSB deal (closed November 6, 2025) brought $236.1 million of deposits and $106.2 million of loans, so it added more funding than lending. Period-end short-term FHLB advances fell from $433.5 million in June 2025 to $123.5 million, though they rose $23.5 million in Q2 2026 as deposits dipped.

Management credits the margin gain to "disciplined deposit pricing, a reduced reliance on higher-cost wholesale funding, and favorable repricing dynamics, partially offset by pressure from changes in asset mix." The asset side was the weak part. Loans yielded 6.17%, down from 6.39%, and the average yield on all earning assets fell to 5.67% from 5.84%.

What the headline numbers hide

• Last year's quarter contained a one-off that makes the comparison less clean. Converting the Civista Leasing and Finance division to a new core system in Q2 2025 added $1.6 million to that quarter's net interest income, cut $1.0 million from non-interest income and cut $0.3 million from expenses. Before tax, that left Q2 2025 about $0.9 million better off. Without the $1.6 million, Q2 2025 net interest income would have been about $33.2 million, so the underlying growth is closer to 16% than the reported 10.9%. In the other direction, the 36.7% jump in non-interest income is flattered: lease revenue rose $0.9 million year over year, and the company says it was "relatively unchanged" once the adjustment is removed. The real fee drivers were a $0.7 million rise in gains on loan sales and $0.3 million more in service charges.
• Growth came almost entirely from the acquisition. Loans rose $103.8 million year over year, and FSB brought $106.2 million at closing, so loans without FSB were roughly flat. Deposits rose $262.0 million against FSB's $236.1 million, about 0.8% of growth beyond the deal. Since December 31, 2025, loans are down $15.1 million (-0.5%). Q2 did add $25.2 million (+0.8%), mostly construction loans (+$11.2 million, seasonal) and home mortgages (+$14.5 million). Brokered deposits, a pricier source of money bought through intermediaries, fell $52.0 million in the half to $350.1 million.
• Per-share growth is negative because of dilution, not weak operations. Pre-tax income rose 33.2% to $17.2 million. A higher tax rate (16.7% versus 14.6%; the 10-Q says this happened because pre-tax income grew faster than tax-exempt items) cost about $0.4 million, or roughly $0.02 a share. Share count growth of about 34% did the rest. No shares were bought back in the first half. The same new equity is why return on equity fell to 10.23% from 11.02% and ROTCE fell to 13.72% from 16.48%. The bank earns more but has a much larger capital base.
• Reserves moved in opposite directions in Q1 and Q2, which explains the dip from last quarter. Net income was $0.7 million lower than Q1's $15.0 million. Q1 had a net provision benefit of about $0.6 million: first-half provision was $1.1 million and Q2's was $1.8 million. In Q2 the bank set aside $1.8 million against only $74 thousand of net charge-offs, so the loan reserve grew from $40.5 million to $41.7 million. Pre-provision net revenue, meaning earnings before credit costs and tax, actually rose to $18.9 million from $17.4 million in Q1 and $13.9 million a year earlier. Underlying earning power improved this quarter, not fell.
• Credit losses are tiny, but problem loans are higher than a year ago. Q2 net charge-offs of $74 thousand work out to about 0.01% of average loans annualized by our calculation (the company doesn't state a ratio), against about 0.13% a year earlier. Non-performing assets, loans that have stopped paying normally, were $30.5 million (0.71% of assets). That is down slightly from $31.3 million at year-end but up from $23.2 million in June 2025, with the jump coming in Q4 2025. Loans 90+ days past due rose from $4.1 million to $13.9 million in the half. The 10-Q attributes this mainly to one non-owner-occupied commercial real estate loan of about $7.9 million, which was already on non-accrual (interest no longer booked as income) at December 31. Its specific reserve was cut from $3.0 million to $2.0 million. The reserve covers 1.28% of loans, unchanged from both comparison dates.
• Earnings are backed by cash. Operating cash flow for the first half was $29.3 million, matching net income of $29.3 million.

Takeaway: Civista's 30% profit jump is a funding story, not a lending story. Equity raised in 2025 and FSB's deposits replaced 4.5% FHLB money, which lifted the margin to 3.89%. Loans without the acquisition were flat, and the new shares took all of the per-share benefit. Further EPS growth now depends on actually growing loans.

Capital and rate sensitivity

Capital is much stronger than a year ago. The Tier 1 leverage ratio (core capital against total assets) was 11.87% against 8.80% in June 2025, and tangible common equity was 10.23% of tangible assets against 6.70%. CET1 was 14.4%, up from 13.6% at year-end and far above the 4.5% minimum. Tangible book value per share rose 6.0% in six months to $20.43. The quarterly dividend is $0.18, a 26% payout ratio.

The bank's own rate model says it gains when rates rise and loses when they fall. A 100 basis point (1 percentage point) cut in rates would reduce projected net interest income by 3.09%, and a 100 bp rise would add 1.66%. With most of the expensive FHLB borrowing already repaid, there is less of that cost left to cut. Further margin gains would need deposit costs to keep falling, and those rose slightly in Q2 to 1.83% from 1.81% in Q1.

What to watch next

The filing gives no numeric guidance. Q3 2026 results are scheduled for October 22, 2026, the first quarter under new CEO Chuck Parcher, who took over from Dennis Shaffer on August 28. This is our first report on Civista. The questions for Q3:

1. Loan growth without the acquisition. Q2's +0.8% was helped by seasonal construction lending. Without real loan growth, cheaper funding is mostly a one-time gain.
2. Margin versus deposit costs. Cost of funds fell only 0.02 points from Q1, while deposit costs rose 0.02 points. A margin that stalls near 3.9% would be consistent with a funding tailwind that is mostly used up.
3. The $7.9 million commercial real estate loan and non-performing assets. Whether that loan is resolved or charged off, and whether non-performing assets stay near $30 million rather than climbing, will decide whether reserve building continues.

Our view: earning power per dollar of assets has clearly improved (return on assets 1.34% vs 1.06%). But with the extra shares now in the year-ago comparisons from Q3 onward, EPS growth from here has to come from the balance sheet growing, which the first half did not show.